Comparing CLE-AdCSV applications using SA and TAC to determine the Fe binding characteristics of model ligands in seawater

. Competitive ligand exchange–adsorptive cathodic stripping voltammetry (CLE-AdCSV) is used to determine the conditional concentration ([L]) and the conditional binding strength (log 𝐾 (cid:3030)(cid:3042)(cid:3041)(cid:3031) ) of dissolved organic Fe-binding ligands, which 10 together influence the solubility of Fe in seawater. Electrochemical applications of Fe speciation measurements vary predominantly in the choice of the added competing ligand. Although different applications show the same trends, [L] and log 𝐾 (cid:3030)(cid:3042)(cid:3041)(cid:3031) differ between the applications. In this study, binding of two added ligands in three different common applications to three known types of natural binding ligands are compared. The applications are: 1) Salicylaldoxime (SA) at 25µM (SA25) and short waiting time, 2) SA at 5µM (SA5) and 3)2-(2-thiazolylazo)-ρ-cresol (TAC) at 10 µM, the latter two with overnight 15 equilibration. The three applications were calibrated under the same conditions, although having different pH values, resulting in the detection window centers (D) DTAC > DSA25 ≥ SA5 (as log D values with respect to Fe 3+ : 12.3>11.2≥ underestimated by TAC with a factor 0.5 to 1.05 for model A ligands and 0.52-0.7 for HS, model B ligands. The SA5 application both under- and over- estimated ligand concentrations (0.57-1.5 for model A ligands and 1.27-1.8 for model B ligands), but our kinetic studies suggest that true equilibrium may still be an issue for strong ligands. The SA25 application overestimated [L] by a factor 1.31-2.3 for model A ligands and 2.33-4.17 for model B ligands. Moreover, for all approaches, ligand specific interferences occurred, as for humic acids.


Introduction
The trace element Fe is an important micro-nutrient for phytoplankton (De Baar and La Roche, 2003;Achterberg et al., 2018;35 Lauderdale et al., 2020). Together with light it limits the growth of phytoplankton in 30 to 40% of the oceans (De Baar, 1990;Martin et al., 1990;Rijkenberg et al., 2018;Boyd et al., 2000;). One of the reasons for the limiting role of Fe is its low solubility in seawater, which can be enlarged at least tenfold by complexation with dissolved organic ligands (Liu and Millero, 2002).
The organic complexes of dissolved iron (DFe) in the oceans are important since these decrease the inorganic Fe concentration, and therefore reduce precipitation as Fe(oxy)hydroxides and adsorption onto particles (scavenging). Organic ligands can also 40 be oxidised under the influence of light and reduce Fe(III) into the labile but more bio-available Fe(II) via ligand-metal charge transfer reactions (Barbeau et al., 2001;Barbeau, 2006;. Furthermore, organic complexation of Fe can be expected to modify Fe bioavailability, although the relationship between DFe speciation and bioavailability appears to be complex ( Van den Berg 1995;Hutchins at al., 1999;Shaked et al., 2005;Salmon et al., 2006;Morrisey and Bowler, 2012;Gledhill and Buck, 2012;Shaked et al., 2020). The significance of Fe speciation to its biogeochemistry has led to incorporation 45 of chemical Fe speciation into global biogeochemical models, with varying levels of complexity Völker, 2011, 2015;Ye and Völker, 2017). Recent modelling work has also highlighted the potential importance of the physico-chemical environment on Fe speciation, in particular highlighting the role that pH plays in modifying Fe speciation (Ye et al., 2020).
The role of both pH and temperature is potentially of great significance considering climate change and ocean acidification.
Out of the natural ligand pool, the following Fe-binding organic ligands groups have been identified: 50 1. siderophores, being relatively strong Fe-binding ligands excreted by micro-organisms to bind Fe and make it bioavailable (Gledhill et al., 2004;Mawji et al., 2008Mawji et al., , 2011Boiteau and Repeta, 2015;Boiteau et al., 2018); 2. humic substances, a diverse group of large molecules, that include ligands with affinity for iron in a broad range possibly spanning from weak to as strong as some siderophores (Laglera et al., 2007, 2019, Su et al., 2018Slagter et al., 2019); 55 3. polysaccharides, a group of ligands binding Fe relatively weakly (Hassler et al, 2011(Hassler et al, , 2015, although stronger polysaccharides have also been reported (Norman et al. 2015).
Organic ligands increase the solubility and residence time of Fe. Although specific methods exist that focus on analyzing siderophores, humic materials and polysaccharides, the connection between the actual abundance of these groups and the overall Fe-binding capacity is often not well resolved. A few exceptions include Laglera et al. (2019) who determined 60 specifically humic Fe-binding ligands, Boiteau et al. (2018) who focused on DFe bound to siderophores and Bundy et al. (2014Bundy et al. ( , 2015 who combined methods to determine the abundance of specific groups. For approximately three decades, competitive ligand exchange-adsorptive cathodic stripping voltammetry (CLE-AdCSV) has been used to estimate the overall Fe-binding capacity of organic matter in seawater. The application of this method enlarged our knowledge on the marine chemistry of Fe and the results formed the base of the explanation why DFe depth profiles 65 deviated from those of other trace metals (Van den Berg, 1995;Rue Bruland, 1995;Hutchins et al., 1999;Croot et al., 2004;Laglera and van den Berg, 2009;Gledhill and Buck, 2012;Bundy et al., 2014;Buck et al., 2015;Hassler et al., 2019;Lauderdale et al., 2020). The technique estimates the conditional concentration of ligands in the sample ([ ]), and the conditional stability constant ( ) of their complexes without specifying the different contributions of specific ligands (Gledhill and van den Berg, 1994;Rue Bruland, 1995;Wu and Luther, 1995;Croot Johansson, 2000;Boye et al., 2001;Buck et al., 2007;Cabanes 70 et al., 2020;Ardiningsih et al., 2020). The term "conditional" is extremely important and means that the obtained results are specific to the composition of the sample matrix analysed (DFe, temperature, pH, ionic strength). Since [ ] also depends on the conditions like pH, salinity and dissolved organic matter, we will use the term conditional for both parameters. The results cannot therefore be considered as an absolute quantification of the properties of all the available Fe-binding sites and extrapolated to other conditions or matrices Gerringa, 2017, Town andvan Leeuwen, 2014). The technique uses 75 an added ligand (AL) with known concentration and conditional stability constant with Fe that form an electro active Fe-ligand complex that competes with the natural ligands present in a sample for Fe. The sample is equilibrated for a defined time period under controlled conditions of pH, light and temperature. The Fe bound to the artificial ligand is analyzed through its electroactive properties. By adding increasing amounts of Fe to subsamples, the competing natural organic ligands are titrated until the natural binding sites are no longer strong or abundant enough to compete successfully with the AL. The competition is 80 reflected by the increased proportion of Fe bound to the artificial ligand and from this the conditional concentration and binding strength can be calculated (Van den Berg 1982). Although the method does not provide information on the molecular composition of the binding sites, CLE-AdCSV does give information on the Fe binding capacity of seawater at the measurement pH, temperature and DFe concentration of the sample. Thus, an indication of the potential capacity for further Fe-binding in a particular sample can be assessed (van den Berg 1995; Tagliabue and Völker, 2011;Pham and Ito, 2019). 85 Application of CLE-AdCSV allowed in some samples the division of the overall ligand in two broad ligand groups as a function of their conditional stability constants indicated with 1, ( , ), for the relatively strong ligand group and with 2, ( , ), for the relatively weak ligand group (Rue and Bruland, 1995Bundy et al., 2014Bundy et al., , 2015. Four different ALs have been reported as forming effective electroactive complexes for the purposes of CLE-AdSCV: 1nitroso-2-napthol (NN) (Gledhill and van den Berg, 1994), salicylaldoxime (SA) (Rue and Bruland 1995), TAC 2-(2-90 thiazolylazo)-p-cresol (Croot and Johansson, 2000) and 2,3-dihydroxynaphthalen (DHN) ( Van den Berg, 2006). The two ALs, SA and TAC, are the usual selection in field studies Bruland, 1995, 1997;Croot and Johansson, 2000;Croot et al., 2004;Boye et al., 2005;Thuróczy et al., 2011aThuróczy et al., , 2011bThuróczy et al., , 2012Kondo et al., 2012;Bundy et al., 2014Bundy et al., , 2015Buck et al., 2015Buck et al., , 2018Gerringa et al., 2015Gerringa et al., , 2017Abualhaija et al., 2015;Klein et al., 2016;Slagter et al., 2017Slagter et al., , 2019 and basin scale data sets now exist for and [ ] obtained using these two ALs (Caprara et al., 2016;Cabanes et al. 2020;Schlitzer et al., 2018) 95 which provide an important resource for our understanding of iron biogeochemistry in the ocean (Boyd and Ellwood, 2010;Boyd and Tagliabue, 2015;Völker and Tagliabue, 2015;Tagliabue et al., 2016;Lauderdale et al., 2020). However, results of inter-comparisons of field data suggest that although trends may be similar for different ALs, the different methods may not be directly intercomparable as conditional [ ] differed significantly, with SA giving higher [ ] and often identifying more than one ligand group compared with TAC Buck et al., 2015). With SA, often two ligand groups can be 100 distinguished, while TAC distinguishes only one, complicating comparison of trends in . The obtained by TAC is in between the two values of the two groups obtained with SA. The question is therefore -what is the underlying cause of these differences? It was found urgent within the SCOR work group 139 to test or calibrate methods with model ligands.
Although there are studies that determined [ ] and for model ligands such as siderophores, with some success with respect to [ ] at least (Rue and Bruland, 1995;Buck et al. 2000;Witter e al. 2000;Croot and Johansson, 2000), a thorough 105 examination of multiple ligands and approaches that also sought to compare determined [ ] and with values calculated from thermodynamic constants has not been previously undertaken to our knowledge. In this work we chose to examine potential bias between ALs via a series of carefully controlled studies of selected Fe-binding ligands that are likely representative of those found in the marine environment. We chose to work with SA and TAC, and further compared two reported SA methods. Our three approaches comprised a) 10 µM TAC with overnight equilibration at pH = 8.05 (Croot and 110 Johansson, 2000), b) 25 µM SA (SA25) at pH = 8.2 with a short waiting time of 20 15 minutes for the competing reaction to occur Bruland 1995, Buck et al., 2007), and c) 5 µM SA (SA5) at pH = 8.2 with overnight equilibration as described by Abualhaija et al., (2015). Since all parameters derived in CLE-AdCSV are fundamentally dependent on the side reaction coefficient of the Fe-AL under the conditions of analysis, we calibrated each ligand in the same laboratory under comparable conditions for consistency and to avoid any issues of bias relating to the choice of calibrating ligand, the calculation methods 115 employed in the original papers and the choice of side reaction coefficient for Fe. We made some (arbitrary) choices on conditional binding constants between DTPA and Fe, however, we worked with one set of thermodynamic constants to make comparison between the methods consistent. We therefore press the point that the focus of the paper is on comparing the empirical outcome of the three applications and not on the accuracy of .
We used Diethylenetriaminepentaacetic acid (DTPA) as a model simple well-defined molecule, the naturally occurring phytic 120 acid, the hydroxamate siderophores desferrioxamine B, ferrioxamine E, ferrichrome, and the cathecholate siderophore virbriobactin, and fulvic and humic acids. Moreover, we carried out for the first time a specific study on the kinetics of complex formation and ligand exchange of Fe(SA)x complexes. All the experiments were performed in a single laboratory in order to minimize inter-laboratory variations in protocol, material and reagent variations. We begin with a short review of previous criticisms of the CLE-AdCSV approach, since this provides important context for our study. Our overall aim was to shed light 125 on the processes that lead to method discrepancies in the determination of natural iron ligand concentrations.

Potential origins of bias in the determination of binding parameters by CLE-AdCSV
CLE-AdCSV is based on many limitations and assumptions which have been discussed at some length in the literature (e.g. Apte et al., 1988;Turoczy and Sherwood, 1997;Town and Filella, 2000;Hudson et al., 2003;Croot and Heller, 2012;Laglera et al., 2013;Town and van Leeuwen, 2014;Gerringa et al., 2014;Laglera and Fillela, 2015;Pižeta et al., 2015;Turner et al., 130 2016;Gledhill and Gerringa, 2017). If the assumptions are sufficiently satisfied, the calculation of ligand complexation parameters like the conditional ligand concentration [ ]and the conditional stability constant log can be undertaken, usually using the Langmuir isotherm (e.g. Gerringa et al., 2014 and references herein).
Here we give a brief overview of the limitations and assumptions.
1. Thermodynamic equilibrium between Fe, the added ligand and natural ligands must be established. Failure to achieve 135 equilibrium can lead to incorrect estimates of [ ] and the conditional constants (Hudson, 1998;Gerringa et al., 2014;Town and van Leeuwen, 2014;Laglera and Filella, 2015). Non-equilibrium conditions arise if the electroactive complex is a reaction intermediate, if insufficient time is allowed for equilibration of the reactants, or if there is a fraction of DFe that is kinetically inert in the time scale of the equilibrium period (e.g. aged inorganic colloids or Fe(AL) complexes).
2. There must be a detectable level of competition between the added and natural ligands. The competitive interaction is 140 summarised by the side reaction coefficients for the natural and added ligands (αFeL, αFeAL respectively). The side reaction coefficient, which is often expressed as a logarithm, is defined as where [´] and [´] are the conditional concentration of the organic ligand not bound by Fe and the concentration of AL not bound by Fe respectively and ´ is the Fe concentration not bound to L. The side reaction coefficient of AL defines the center of the detection window or analytical window, which we defined here as D to prevent confusion between side reaction coefficients of added and natural ligands. The window is assumed to be approximately two to three orders of magnitude wide, 150 one to two orders of magnitude above and below D (Apte et al. 1988;Van den Berg and Donat, 1992;Milller andBruland, 1997, Laglera et al., 2013;Laglera and Fillela, 2015). In practice, the upper limit of D is defined by the analytical sensitivity of the AdCSV method, as it is bound by the limit of detection of FeAL. The lower limit of D is bound by ligands that are outcompeted by the AL within the range of Fe added during the titration. Since AdCSV is internally calibrated via standard additions, in practice the lower limit of the detection window is bound by the value of achieved when the analytical 155 response is deemed to be linear (Apte et al, 1988;Laglera and Fillela, 2015).
3. The concentration of the FeAL complex can be accurately determined at each titration point. AdCSV is internally calibrated via standard additions, D and values obtained for and [ ] are strongly influenced by our ability to accurately calculate the sensitivity (Turoczy andSherwood, 1997, Hudson et al., 2003;Pizeta et al., 2015).
4. Complexes of Fe with natural ligands cannot be electro-labile under the experimental conditions since this could result in 160 interferences with the actual detection of the Fe-AL complex (Yang and van den Berg, 2009;Laglera et al., 2011). 5. The equilibrated Fe-AL complex must be electro-active since it is the reaction on which the detection is based.
6. The AL should not react with the natural ligands altering or cancelling their binding ability.
In the last decade, many studies have questioned the compliance of the CLE-AdCSV methodology to these assumptions and 165 their influence on method discrepancies. Laglera et al. (2011) showed the inability of TAC to measure fulvic and humic acids as Fe-binding dissolved organic ligands, which might be due to either assumption 2 or 6. Humic substances are ubiquitous, they form large diverse molecules and are broadly recognized as Fe-binding ligands (Krachler et al, 2015;Su et al., 2018;Laglera et al 2019;Whitby et al., 2020, Yamashita et al., 2020. According to other work, TAC is able to detect at least part humics as Fe binding organic ligands (Batchelli et al., 2010;Slagter et al., 2017;Dulaquais et al., 2018). 170 The SA25 application has been criticized for not meeting assumption 1, SA25 has a waiting time of 15 to 20 minutes in contrast with the overnight equilibration used for the TAC and SA5 applications Van den Berg, 2014, Abualhaija et al., 2015;Slagter et al., 2019). Abualhaija and Van den Berg (2014) found that two Fe-SA complexes are formed, FeSA and Fe(SA)2, and only FeSA is electro-active. At higher [SA], the proportion of Fe(SA)2 increases and the analytical signal decreases resulting in a negative 175 relationship between sensitivity and competitive force. Finally, we would like to point out that the pH of the analysis may have a larger influence on the organic complexation of DFe than previously thought Avendaño et al., 2016;Ye et al., 2020). Andbut the same competition of OH ions in binding Fe, irrespective of the buffered pH values of the SA and TAC applications is sometimes used (8.2 and 8.05, respectively, which is a factor 1.4 different in terms of H + concentration).
This complicates a direct comparison of data even more. 180

Methods
The natural seawater used in the experiments consisted of mixed leftover filtered samples of the northern Western Atlantic GEOTRACES cruise GA02 (stored frozen at -20 •C) . A sample volume, assumed to be necessary for the following few days, was thawed, mixed, UV-irradiated to destroy the natural organic Fe-binding ligands, and stored in the 185 refrigerator. Consequently, one batch differs from others in with respect to the DFe content, and also potentially in other constituents, such as other trace metals. Since surface samples were not used we do not expect large differences in salinity, the average salinity was 35.09 ±0.61 (N=434), obtained as average of all samples >100 m depth taken for the ligand analysis in Gerringa et al. (2015). Samples for DFe analysis were taken from every batch. UV irradiated sea water was stored for 3 days at most. 190

Equipment and electrochemical parameters
We carefully followed procedures as described in the literature to ensure methodology was as close as possible to that originally described (Tables S1 and S2). Three different voltammetric setups were used (Table S1). A standard Metrohm set up was used for TAC. For use with SA, a separate Metrohm system was modified to allow for air purging whilst the mercury drop formation 195 was still executed under nitrogen pressure. Nitrogen did not leak into the headspace of the sample during the measurements in our Metrohm stand. However, when drops are formed pulses of nitrogen are released and end up in the headspace of the sample, purging with air would remove (at least part of) the nitrogen. To check a potential effect of this nitrogen a kinetic experiment with SA25 was executed. To five identical subsamples SA was added at the same time. These subsamples were each measured repetitively during one to several hours, one after the other. No effect was seen after sub sample replacements 200 ( Figure S1). We concluded that nitrogen from the stand did not influence the kinetic process, since the measured FeAL concentrations had a gradual change over time, independent of the subsample. For the other kinetic experiments with SA, BASi equipment was used (Table S1). The electrochemical settings used by Croot and Johansson (2000), Buck et al. (2007) and Abualhaija and van den Berg (2014) were used without alteration and are summarized in Table S2. Before use, all materials such as vials, bottles and cells, were conditioned overnight in UV-irradiated seawater with the prepared combinations of seawater and ligand. The conditioning procedure was performed at least three times for the analysis 215 with TAC and at least five times for analysis with SA. The cell with electrodes, stirrer and purge tube were kept overnight in low metal seawater. Before a titration started first two measurements were executed with seawater containing all chemicals but no Fe addition, these measurements served also as check for possible contamination of the cell. Hereafter, two zero additions were measured (see section 3.4), of which the second was used as start of the titration.
Before starting kinetic measurements, a 30 ml vial with the same content and treatment as the sample was used for three 220 analyses (thus three times 10 ml) in order to condition the cell wall, electrodes and stirrer.
The 200 ml bottles used for kinetic studies were conditioned with 6 nM Fe, in the absence of the added ligand. For tests with UV irradiated seawater without a model ligand, 200 ml bottles were conditioned by rinsing the bottle three times for two minutes with 30 ml of the test seawater. Since UV-irradiated seawater did not contain Fe-binding organic ligands, most of the added 6nM DFe would adsorb on the bottle walls or precipitate. 225 Samples were equilibrated according to the specific method descriptions, which was overnight equilibration for the TAC and 5 µM SA and 15 minutes for 25 µM SA Johansson, 2000, Buck et al., 2007;Abualhaija and Van den Berg, 2014).
The 15 minutes equilibrations were applied precisely using a stopwatch, whereas overnight equilibration resulted in a period of at least 14 hours.

UV irradiation
Samples, without any additions, were poured into 30 ml Nalgene FEP bottles and placed in in a custom-made UV box between 4 TUV 15W/G15 T8 fluorescent tubes (Phillips) for 4 h. These bottles were tested in the past for photo-oxidation of organics by Co analyses by ICPMS and Cu ligand titrations and proved to be UV permeable enough to destroy organic material (Rapp et al., 2017;Wuttig et al., 2019). Precipitates were not observed. After UV irradiation, samples were transferred into a trace-235 metal clean one-liter LDPE bottle and kept in the refrigerator.

Model Ligands
The following discrete synthetic ligands of known concentration (model A ligands) were used at a concentration of 2 nM, unless otherwise stated. No tests were undertaken to check the purity of the siderophores. The solutions were used within two 240 weeks after preparation, and kept in the refrigerator in the dark at 4•C, which should at least for DFOB be short enough to prevent degradation (Hayes et al., 1994). The aim of our research was to compare the three applications. Humics (model B ligands, 0.1 or 0.2 mg/liter) were added in a concentration to give an iron binding capacity of approximately 3 nM (Laglera and Van den Berg, 2009;Yang et al., 2017;Sukekava et al., 2018). The stoichiometry of the formed Fe-model ligand complexes differs for each model ligand. In order to simplify the comparison of binding strengths, stability constants are given 245 for a 1:1 stoichiometry: Model A ligands  Diethylenetriaminepentaacetic acid (DTPA C14H23N3O10, Sigma-Aldrich D6518-5G). DTPA was used to calibrate the added ligands via reverse titration according to methods described previously (Croot and Johansson, 2000). We calculated a conditional binding constant Log , of 19.01 using the ion pairing speciation 250 software visual MINTEQ (Gustafsson, 2012), disregarding the formation of FeOHDTPA. The Log , value was independent of the pH difference 8.05-8.2. This is 0.34 higher than the value (18.65) used by Croot and Johansson (2000) at I=0.7, pH =8.05 with the difference most likely arising as a result of the

Model B ligands
Humic substances are the heterogeneous mix of hydrophobic compounds originating from chemical and microbial transformation of living matter as it decays in the environment. 285  Fulvic Acids (FA) is the smaller and more soluble fraction of humic substances (Buffle, 1990). Therefore, this is not just the ligand but a series of compounds of which a fraction functions as iron binding ligands. (2009)  Standard II 2S101F). It must be noted that the batches are different between the above results and these Formatted: Font: 10 pt values can differ per produced batch, still we assumed 2.92 nM Equivalents (nM Eq) of ligand sites to be added with 0.2 mg SRFA per liter.

Laglera and Van den Berg
 Humic acids (HA) are the larger fraction of humic substances that precipitate at low pH (pH 2) (Buffle, 1990

AL Calibration
Seven or eight conditioned Teflon 30 ml vials were filled with 10 ml UV irradiated seawater spiked with buffer, 6 nM Fe 300 (Table S1) and increasing amounts of the calibrating ligand DTPA. For TAC the pH was 8.05 and for SA the pH was 8.2 according to the original method specifications (according to the NSB scale). The calibrations were repeated 4 times. The buffer used for all applications was ammonium borate (Abualhaija and van den Berg, 2014;Buck et al., 2007). Details can be found in the Supplementary Information. DTPA additions were 0, 10, 100, 200, 400, 1000, 2500 nM DTPA for TAC, 0, 1, 10, 40, 80, 100 and 200 nM DTPA for SA5 305 and 0, 1, 40, 100, 200, 400 and 1000 nM DTPA for SA25. Mixtures of UV seawater with buffer, DFe and DTPA were equilibrated for at least 8 h after which SA or TAC were added. New mixtures were equilibrated either overnight or for 15 min in the case of SA25, after which peak heights for FeAL were determined following the procedures described in Table S1. We calibrated SA25 using a short waiting time instead of 5 h of equilibration (Rue and Bruland, 1995;Buck et al., 2007) to ensure consistency with the approach applied to samples. For the calibration of TAC the normal Metrohm instrument was used, for 310 SA the BASi instrument (Tables S1, S2). The measurements were done in sequence of increasing DTPA concentrations, without rinsing cells in between.

Signal stability tests
As we found the CSV signal after SA addition lacked stability and decreased with time, we performed a series of experiments to find the cause. We tested the following aspects for both instruments: the influence of a purge step with air ( Figure S2), the 315 influence of the size of the mercury drop ( Figure S3, Table S3), the influence of the mercury puddle on the bottom of the cell, and the influence of the SA concentration. For the kinetic measurements (section 2.4) of the SA applications, both BASi and Metrohm stands were used. Details on procedures are given in the Supplementary Information file.

TAC 320
Fifteen vials were prepared with increasing Fe content, in a mixture of UV-irradiated seawater and model ligand (Table S1, Johannsson, 2000, Ardiningsih et al., 2020). Blanks were obtained by analysis in the absence of model ligands.

SA5
The application followed Abualhaija and van den Berg (2014), but used the above described BASi instrument. SA (added to a final concentration of 5 µM), buffer, Fe additions (Table S1) and samples were left to equilibrate overnight. 325

SA25
For SA25, 25 µM SAthe buffer and DFe were added one hour before analysis. SA was added to a final concentration of 25 µM separately to each vial, 15 minutes before the measurement.

Kinetic measurements 330
The samples contained either UV-irradiated seawater, or UV-irradiated seawater with a model ligand to which buffer and 6 nM Fe were added in a pre-conditioned bottle. If a model ligand was present, this was first allowed to equilibrate overnight with the buffer and 6 nM Fe. In samples with only UV-irradiated seawater, two approaches were followed: One in which Fe was added together with TAC or SA at t=0, and one in which Fe was equilibrated overnight prior to addition of TAC or SA.
In the latter case, there is the possibility that Fe-oxide precipitates were formed prior to the addition of TAC or SA and were 335 probably dissolved after the addition of TAC or SA. At t = 0, TAC or SA was added.
At t = 0, the first measurements were done as rapidly as possible until approximately t = 1 h, followed by subsequent measurements every 20 min, every 40 min and 1 h until either t = 4 h or t = 7 h. The number of analyses depended on the application and experiment duration (4 or 7 h) but contained a minimum of 14 duplicate measurements. In this way the FeAL(x) formation in time can be followed. However, the model ligand dissociation rates cannot be calculated from the rate of peak 340 increment because the addition of Fe (6 nM) was in excess of the model ligand concentration (2 nM Eq, if not indicated differently). Therefore, the excess Fe formed hydroxides and adsorbed on the cell and electrode surfaces. The increment of signal reflects the competition of TAC with all these iron species and not just with the model ligand.
Two protocols were followed ( Figure S4): 1. In-cell experiments: repeated scans of the same sample. In this way, contamination was prevented and more 345 measurements could be undertaken, especially at the start of the experiments. The total time of the experiments lasted 4 or 7 h. Samples of 30 ml were prepared of which the first 20 ml were used to condition the cell twice, after which, the last 10 ml was transferred to the cell and the experiment undertaken. The AL was added to the cell at t 2. Bottle experiments: scans were carried out on separate aliquots of one sample. In this experiment, a fresh aliquot of 10 ml was pipetted into the voltammetric cell for the determination of peak height at each time point. The total 355 sample volume was 200 ml and the AL was added at t = 0. In this experiment, accumulation of mercury at the bottom of the cell was limited. The experiment lasted for 4 or 7 h, consistent with the "in-cell" approach. The first 10 ml was transferred as quickly as possible into the preconditioned cell and the measurement started. In order to determine the amount of adsorbed Fe on the 250 ml bottle walls, the bottles were rinsed carefully with 5 ml of elution acid (1.5 M teflon distilled HNO3 that contained rhodium see below section 32.6 ICPMS analysis) and Fe 360 concentration in the acid rinse determined by ICPMS (section 3.7).

Calculations
The sensitivity, S, the ligand concentration [L] and conditional stability constant ( ) were calculated by direct non-linear fitting of the Langmuir isotherm  with inherent co-dependence of [L] and (Apte et al., 1988;365 Hudson et al., 2003;Gerringa et al., 2014) The inorganic side reactions of DFe with dissolved hydroxides, αFe, was calculated using the constants from Liu and Millero (2002) resulting in an inorganic alpha for Fe (αinorg) of logαinorg = 9.9 at pH = 8.05 and of logαinorg = 10.4 at pH = 8.2. These are slightly different from literature values of Croot and Johansson (2000) and Abualhaija and van den Berg (2014). The conditional binding strength of DTPA was obtained using Visual Minteq. We used an average seawater major ion composition, 370 and an average deep sea DFe concentration of 0.5 nM was chosen for these calculations. DTPA was added to the composition at the concentrations used, and the pH was fixed at values of 8.05 and 8.2. According to the VMinteq calculations, , = 27.3, the logarithm of side reaction coefficients for DTPA with major cations was 8.26, resulting in , = 19.01.

ICPMS analysis of dissolved Fe
Dissolved Fe was analysed with a Thermo Finnigan HR-ICPMS element 2 (for details see Middag et al., 2015;Gerringa et al., 2020). Briefly, seawater aliquots, with and without the addition of model ligands were concentrated using a seaFAST system after UV-destruction. The competition by DTPA causes a reduction in peak height compared to the situation without DTPA ( Figure S4S5). At equilibrium, dissolved Fe is distributed over the following species, For the application of TAC, the contribution of FeTAC is thought to be negligible with respect to the formation of Fe(TAC)2 (Croot and Johansson, 2000). For SA in the micromolar range, both FeSA2 and FeSA are formed, although only FeSA is the 405 electroactive species (Abualhaija and Van den Berg, 2014). Using (1) and (3)   The difference becomes larger when calculated with respect to inorganic Fe (Fe′) when using the pH adjusted values of logαinorg 420 = 9.9 for 8.05 and logαinorg = 10.4 for pH = 8.2 (Liu and Millero, 2002). For SA5 and SA25, the comparison between our data and literature values is thus offset with respect to Fe′, due to the application of logαinorg = 10.4. It is possible that the larger deviation in SA25 from previously reported values is partly due to the shorter waiting time used in our study,15 minutes instead of 5 hours (Rue and Bruland, 1995). However, the calibration should be executed according to the published protocol of the analyses. 425

Titrations
We present the concentrations of FeAL determined during the titrations of the selected ligands with the 3 different methods to allow direct comparison between the approaches (Figure 1, Tables 2  seawater can be calculated using the side reaction coefficient of 6.25 for DFO binding to Ca and Mg at pH 8.0 (Schijf and Burns, 2016;Wuttig et al., 2013) and the stability constant for FeDFO given by Hider and Kong (2010). For the complex 440 ligands, model B, like humic and fulvic acids, which contain multiple binding sites with a range of affinities, an average will be determined based on D of the method being applied (Tables 1 and 2). Both factors highlight important concepts that relate to the CLE-AdCSV approach in general, that need to be taken into consideration when interpreting derived from CLE-AdCSV titrations.  Ligand concentrations were highest with SA25 and lowest with TAC (Tables 2 and 3 Assuming the concentration of ligand sites per weight unit determined by Laglera and van den Berg (2009) and Sukekava et al. (2018) to be correct, the overestimation by SA25 was larger for model B ligands (Tables 2 and 3). The standard deviationdifference from the average value for duplicate measurements of [L] was 0.3 and 0.2 nM Eq of Fe for the TAC and SA5 application, respectively (excluding Ferrioxamine E because it was saturated with Fe, see below). The standard deviation with SA25 (N=5) was 0.3 nM Eq of Fe. In the following we will assume ±0.3 nM Eq of Fe as precision for [L]. The differences 465 between the applications are smaller when the αFeL values are compared (Table 3), which is understandable, since it is αFeL that is titrated and also because αFeL is calculated from the product of L not bound by Fe ([L′]) and and thus compensates for any codependence between and [L].
We 'back' calculated the titration curves using our present results, and [L], and presented this in log-log plots of [FeAL] versus total dissolved Fe together with the actual data points (Figure 2). For DTPA and desferrioxamine B we added the 470 theoretical titration curves that should be obtained given the calculated from the thermodynamic constants and the 2 nM of added model A ligands. We presented the back calculations in log-log plots in order to magnify the initial part of the titration (Figure 2).   (Apte et al., 1988), which is the case for the results from TAC and SA5 (by a factor of 0.56-0.87, Table 3). But the DTPA ligand concentration was overestimated by SA25 (by a factor of 1.31, Table 3, also see titration in Fig. 1). Indeed, in the log-log plots (Figure 2A) our results are well off from the theoretical titration line. Additionally, at the very low concentrations the data points of all applications deviate from the modeled curves.

Phytic acid 500
The differences between the applications are also not large for phytic acid, the two SA applications are even quite similar. The large difference for the -values of phytic acid estimated by TAC was not expected when comparing the two very similar titration curves. We found that small changes in determined FeTAC2 concentrations at low Fe additions could be responsible for this difference ( Figure 2B). is much lower than measured by Rue and Bruland (1995) ≥23, although they recovered 100% of the added 2.5 nM.
However, we note that they used another protocol and applied 4 minutes of nitrogen purging before every measurement which would have interfered with the signal stability according to Abualhaija  FeIII (Anderson, 1963). Colloid formation will interfere in several ways with the analysis by the loss of Fe, since formation of colloids results in a potentially inert fraction of Fe, the loss of phytic acid, and interference of the colloids on the mercury electrode surface. However, the formation of these colloids is dependent on the phytic acid concentration Purawatt et al. 2007) and at 2 nM phytic acid we do not expect colloids to be formed. The ratios of added [L] and obtained values by CLE-AdCSV by Witter et al. (2000) varied between 0.8 to 1.7, resembling our results, although the ratio 530 was 1 for both desferrioxamine B and ferrichrome. Thus, even for model ligands there is no consistence in the literature between ligands or methods suggesting problems in the standardization of the methodology. It is possible that the siderophores used are not of 100 % purity, which would result in a systematic underestimation of [L]. However, whilst it is interesting to note absolute values, our research focusses on differences between the three applications, which should not be impacted by any impurities. 535 The theoretical titration curve for desferrioxamine B has a relatively large offset at low concentrations compared to the modelled results ( Figure 2C). This indicates that the theoretical was not even approached by the three applications. That we obtained (and not for the first time) clearly curved titrations, where we should not within the applied D values, is hard to explain. One possible explanation could be a reaction taking place at the electrode surface in CSV promoting ligand exchange of Fe(III) siderophore complexes, which is producing a current. Another alternative explanation might be aluminium 540 competition (which is not accounted for by the thermodynamic constants) since Al complexes with siderophores are detected in MS analysis of samples (Gledhill et al., 2019). The Al content however, is unknown.
Ferrioxamine E was the only model ligand that was saturated with Fe prior to the start of the experiment. Moreover, none of the ALs should sequester Fe from ferrioxamine E, which is required in order to estimate because its is too high, Formatted: Font: Not Italic outside D (Apte et al., 1988;Hudson et al., 2003;Gerringa et al., 2014 Therefore, the titration of ferrioxamine E was more or less a standard addition and in theory [L] should be equivalent to the 555 determined DFe concentration. However, all three applications overestimated the ligand concentration, but the estimated [L] of ferrioxamine E compares very well for TAC and SA5, and even SA25. The relatively large difference in resulting between the two SA-applications for the hydroxamate siderophores was not expected since the D-values are close. The two most probable explanations are D and lack of equilibrium. Possibly the siderophores have αFeL at the borders of and >D and therefore a small decrease in D still had a consequence for the outcome of the calculations. The short waiting time may be the 560 other reason of the deviation of SA25, which we will discuss in the next section. The [L] of the catechol Vibriobactin was underestimated by SA5 and TAC and overestimated by SA25. We believe that this divergence was a combination of lack of equilibrium due to the high stability of Fe-vibriobactin complexes during the short equilibrium period of SA25 and consequent overestimation of [L], and possibly by the tendency of catecholates to oxidise or hydrolyse in water (Brickman and McIntosh, 1992), which could have resulted in partial loss of vibriobactin during overnight 565 equilibration.

Humic substances
Our titrations of FA and HA show remarkable differences between the applications, [L] by TAC was 13-25 % of [L] by SA25 (Table 3). TAC detected 55 -70% of FA and HA in contrast to an early report that TAC could not detect any portion of IHSS humic reference material (Laglera et al., 2011). Our FA and HA results are in line with the partial detection of HS by TAC 570 that was observed already by several field studies, where an increase in [L]TAC correlated with an increase in natural humics (Gerringa et al., 2017;Dulaquais et al., 2018;Slagter et al., 2017Slagter et al., ,2019Laglera et al., 2019). HS showed the largest deviations from the expected (literature) results of all tested ligands in [L]. Humic substances are ubiquitous in seawater (Laglera et al., 2009;Whitby et al., 2020;Yamashita et al., 2020) and potentially more representative of the dominant fraction of dissolved organic matter actually present in seawater than the model A ligands tested here, since although siderophores are detected in 575 seawater, they are typically only present at pM concentrations (Mawji et al., 2008(Mawji et al., , 2011Velasquez et al., 2016;Boiteau et al.,2018). The different results between the three applications could explain a major part in the offset between the TAC and SA methods in natural waters Slagter et al., 2019;Ardiningsih et al., 2021). Moreover, the deviation in obtained by TAC from the other two applications ( / up to 1.08 and / up to 1.1) is greater than for most model A ligands (Table 3). This is also likely to be linked to the heterogeneity of humic substances, which means 580 the detection window of each method will have a greater influence on the groups of binding sites titrated during the experiments. We cannot provide a definitive explanation for the [L] spread. TAC showed almost straight-line patterns for FA and HA ( Figure 1G, H), as in HS rich estuarine waters (Gerringa et al., 2007, Croot andJohansson 2000). This could be compatible with a fraction of HS being too strong and a fraction too weak to compete with TAC (both fractions would be at or beyond the upper and lower limits of D). There is an abundant presence of strong binding sites in HS that may not be 585 outcompeted by TAC, since desferrioxamine B could also not outcompete all HS binding sites in Arctic Ocean samples (Laglera et al., 2019). Another possible explanation to the similar recoveries for FA and HA, despite their reported different affinity for iron, is that TAC could form interactions with some of the binding groups of HS, cancelling their interaction with iron. In other words, the use of TAC would not obey Langmuir assumption 5. For the SA applications, [L] with SA25 seems to be substantially over the literature values in contrast to SA5. Titration data of HA with SA5, showed detectable levels of 590 Fe-AL at low total Fe concentrations, while for SA25 they could not be seen. Thus, the formation of the electro-active Fe-SA complex does not happen until after 6-7 nM Fe has been added (0.1 mg HA, or over 10 nM for 0.2 mg HA). This is most probably an effect of ongoing association and dissociation processes between Fe, SA and HA, i.e. a lack of equilibration.
Another explanation could be a substantial decrease of the sensitivity cause by adsorption of free and complexed humics onto the surface of the electrode, shielding the electrode form interaction with Fe(AL) complexes (Laglera et al., 2011(Laglera et al., , 2017. 595 Adsorption of humics at the mercury electrode has been extensively discussed by Buffle and co-authors (Buffle and Cominoli, 1981;Cominoli et al., 1980) and the drop of sensitivity for CLE-AdCSV was discussed in Laglera et al. (2011 and. The SA25 application does not obey Langmuir assumption 1. This might also explain the large data spread in figure 2H for SA25.

Overall
The log-log plots for DTPA and desferrioxamine B, between known and observed conditional stability constants show that the 600 data points obtained by TAC are closest to the theoretical curve of DTPA, and by SA5 are closest to desferrioxamine B (Figures   2A,C). However, the TAC application has the highest D and should therefore be better equipped to detect desferrioxamine B, and least equipped to detect DTPA. The different results between applications and, for DTPA and desferrioxamine B, between known and observed conditional stability constants are mostly due to data in the first curved part of the titration as shown in figure 1 and illustrated when compared with the theoretical titration curves (Figures 2A,C). At this part of the titration curve 605 peaks should in many cases be below the detection limit, thus the precision of these measurements is very low. The log log plots (Figures 2) emphasize the differences between expected and observed values. The observations seem to overestimate the FeAL at low metal additions. Possible reasons are: 1. electrochemical, for example a catalytic effect becoming more important at low concentrations and enhancing the signal or tiny peaks caused by impurities of the reagent or sitting in the methanol solvent as show in previous work with NN (Boye et 610 al., 2001), 2. concentrations are more likely to be overestimated near the detection limit, 3. desorption from conditioned cells and electrode surfaces is more significant at low concentrations.
Further analysis is required in order to resolve these possibilities and verify that the response of FeAL is linearly related to [Fe′] or [Fe 3+ ], even at very low Fe concentrations (< 0.5 -1 nM). Moreover, in order to obtain reliable estimates of [L] and 615 ) we suggest that samples should have [L′] greater than 2*DFe to ensure the titration starts at low enough [Fe′] (5 )and (6)).
It is also possible that reactions occurred during the cathodic scan, which could also explain the deviating results of [L] for Vibriobactin, which was underestimated by 52-62% using TAC and 60-70% using SA5. Free catechols can be electro-active (Fakhari et al., 2008) and even a small contribution to the CSV peak from the ligand side of the complex would lead to a 620 significant underestimation of the complexed fraction. A last explanation of underestimating ligand concentrations can be contamination after sampling for Fe determination by ICPMS took place. with our results suggest that cannot be estimated precisely. Here SA-applications result in worse estimates of due to a lack of data at FeL<L, and the possibility that the ligands are outside D, the detection window. As far as we know, four publications describing an intercomparison exist. Three of these compared SA25 and TAC Slagter et al., 2019), one compared SA5 and TAC (Ardiningsih et al., 2021). In all publications, [LSA] is larger than [LTAC] when the data 635 is fitted with a one ligand model). In Buck et al. (2016) it was concluded that, when using the same calculation method, comparison between the results of both applications seemed good with one exception that SA could measure a second ligand whereas TAC could not and therefore the total ligand concentration obtained with SA25 was always considerably larger (their figure 2E). This difference was attributed to an underestimation by TAC because TAC does not detect binding sites of humic substances and cannot discriminate a second ligand as well as SA25   No intercomparison between SA5 and SA25 has been undertaken since the SA5 application was published in 2014. The question remains as to why SA5 and SA25 in this work give different results. One explanation might be disequilibrium of the SA25 application. To further study the equilibration process, we executed some kinetic experiments. Metrohm stand used). For SA25, the peak decrease was high at t = 3 min (initial value) and drops sharply to the values close to the limit of detection within an hour in UV-irradiated seawater and UV+DTPA. For TAC we observed a slow increase of the Fe(TAC)2 concentration followed by an asymptotic change to a constant value, as expected for a product of a ligand exchange reaction tending towards equilibrium.
Kinetic experiments with the low concentrations of model ligands in a volume of 10 ml was not pursued with the SA 670 applications, only with TAC. In these experiments, equilibrium between TAC and model ligands was reached after approximately 6 h, as observed previously (Croot and Johansson, 2000), 4 h for most ligands and 6 h for desferrioxamine B ( Figure S76). Although the rate of Fe(TAC)2 formation changed with the type of model ligand, a steep increase where the relative weaker ligands were added, DTPA, FA and phytic acid, against a slow and steady increase where desferrioxamine B and vibriobactin were added, all model ligands followed the theoretical ligand exchange concentration evolution ( Figure S76). 675   increasing mercury at the cell bottom. On the contrary, the two SA-applications did show a decrease that was steeper and larger with increasing drop size. The decrease in the SA5 application was larger than in the SA25 application ( Figure   S3). A positive linear relationship (R 2 = 0.98) exists for SA25 between the decrease in peak height within 43 min and the volume of dispensed mercury at the bottom of the cell. However, no relation exists for SA5 although the reduction 715 is strongest in that application ( Figure S3). We tested whether SA was reversibly adsorbed to the mercury puddle at the bottom of the cell by transferring mercury accumulated under SA5 and SA25 protocols into a cell filled with 10 mL UV-irradiated seawater with 6nM added Fe. No SA was released from the mercury into the seawater as no peak could be detected when analysed with the normal AdCSV procedure. An explanation could be that only Fe(SA)2 adsorbs on the mercury causing a direct relationship with peak reduction and mercury volume at high SA. Since Fe(SA)2 might be 720 formed irreversibly, no release of SA into the solution that could lead to the formation of the electroactive FeSA species would be possible. It remains hard to explain the strong reduction of the peak height without a relationship with the mercury volume at SA5. 4. The lack of purging influences the conditions. The BASi electrode only allows either stirring or purging in an automatic measurement, and stirring is the normal practice. Purging with air should maintain a constant concentration of oxygen 725 and should increase the sensitivity and prevent decreasing peak heights with time (Abualhaija and Van den Berg, 2014).
We checked the effect of an air purge step on the SA measurements of both electrodes, BASi and Metrohm. The decrease in peak height with time was not influenced by an air purge step in both electrodes ( Figure S2).
We can conclude that the formation kinetics of Fe(TAC)2 using in-cell experiments with model ligands reached equilibrium within 8 hours. In-cell kinetic experiments with SA did not reach equilibrium and showed a continuous decay of the peak 730 height. This can be explained by a combination of processes like adsorption of Fe(SA)-complexes on dispensed mercury at the cell bottom, and formation of the irreversible (FeSA+SA forms FeSA2) and according to Abualhaija and Van den Berg (2014) to the non electroactive Fe(SA)2. The formation of irreversible species is not compatible with techniques such as CLE-AdCSV that require a dynamic equilibrium between competing ligands before analysis. 735

Bottle kinetics
The kinetic experiments were repeated extracting 10 mL aliquots from a 200 ml bottle. Experiments carried out were: UVirradiated seawater, UV-irradiated seawater with desferrioxamine B, UV-irradiated seawater with phytic acid and UVirradiated seawater with FA ( Figure 6). Some points have to be considered for interpretation. The conditioning procedure of the bottle in which the reaction takes place raises the question whether to condition with or without the AL. Addition of AL 740 should occur at t = 0, conditioning of the bottle with AL is thus not possible. Conditioning without AL with UV-irradiated seawater with 6 nM added DFe could cause Fe precipitation and/or adsorption on the bottle walls. DFe at the end of the experiment is probably higher than 6 nM due to Fe desorption from the bottle wall. The bottle kinetics were executed with UV-irradiated seawater, and UV-irradiated seawater with desferrioxamine B, phytic acid and FA as model ligands. Figure 6: Kinetic measurements for the formation of Fe-SA complexes, the change in peak height versus time. At t=0 the AL is added. In-cell means the whole experiment is done in the same 10 ml which was placed in the cell, dispensed mercury accumulates at the bottom of the cell. Bottle experiment means for every measurement a fresh 10 ml was taken from a large volume of sample.
Here the reaction takes place in the large volume, and AL was added at t=0. A and B: UV irradiated sea water (UV); C and D:

TAC
The results with TAC showed the same pattern as for the in-cell experiments an increase that levels off to equilibrium for UV 755 irradiated seawater, phytic acid and FA (Figure 6) characteristic of a ligand exchange reaction reaching a steady state. Higher equilibrium signals prove an extra Fe input from the bottle wall. The experiment with desferrioxamine B did not level off after 8 h and the slope was less steep. This concurs with our in-cell observation that desferrioxamine B dissociates more slowly.
That equilibrium is not reached even after 8 h explains, at least for this model ligand, the underestimation of [L] by TAC (Table   2; Croot and Johansson, 2000). It also indicates that with this protocol kinetics were not just dependent on ligand exchange 760 but also on Fe desorption and/or redissolution of precipitated Fe.

SA
The results of the bottle experiments for SA5 changed dramatically with respect to those of the in-cell protocol. They show that equilibrium appears to be achieved for phytic acid and fulvic acid, and the increase of signal over time in the first 1-2 h is 765 similar to that for TAC ( Figure 6). An equilibrium is reached after approximately 4 h with phytic acid and FA. More time, 8 h, is needed to reach an equilibrium with desferrioxamine B, in line with the slower formation of FeTAC2 in the presence of desferrioxamine B compared to the other model ligands (Figures 6, S65).
Equilibrium or a steady state is difficult to establish for SA25. A plateau is reached after approximately one hour. This seemingly depends on the added ligand, but after the plateau the signal decreases at a steady rate. We can conclude thus far 770 that the steep decrease shown in the BASi electrode in the in-cell kinetics does not happen with the bottle experiments. This result points to adsorption on the dispensed mercury puddle as the main cause of the disappearing signal. Equilibrium is not reached at short waiting times after addition of 25 µM SA. Therefore, the use of the Langmuir isotherm to calculate the ligand characteristics is not possible. We hypothesize that the distinction in more than one ligand group, which was often possible with this application could have been caused by the absence of equilibrium. 775

Conclusions
All applications have drawbacks, however the SA25 application clearly does not obey the main assumption of the Langmuir isotherm, no equilibrium is reached and therefore the results cannot be reliable. Next to this, the most upsetting conclusion is that the estimation of the conditional stability constant is a very rough estimate only and systematically biased by the We confirm the conclusion of Abualhaija and van den Berg (2014) that the SA concentration needs to be low ≤5µM to prevent formation of not electro-active Fe(SA)2. Our experiments suggest that the formation of Fe(SA)2 is irreversible, and thus does not obey the Langmuir equation.
Probably Possibly Fe(SA)2 also adsorbs on dispensed mercury on the bottom of the cell. When using a BASi electrode small mercury drops are to be preferred. 790 All methods have drawbacks, and the characterization of Fe-binding organic ligands at nM concentrations in seawater with a high ionic strength and at 40-80 µmol/kg DOC (Hansell et al. 2009) is and remains challenging. In case voltammetric methods are used, we advise the SA5-application, but should not be overinterpreted and issues relating to the technique fully acknowledge. In particular, it is hard to justify determination of equilibrium constants using an adsorption isotherm that assume equilibrium, if a signal is not stable and the experimental system is therefore not at equilibrium (for whatever reason), especially 795 bearing in mind the method specifically uses the term "equilibrium" in its title. Furthermore, it appears that apart from the constraint forced by D, these estimates have a much larger error than apparent from those obtained from the fits to the Langmuir isotherm, which may explain why direct links between speciation predicted from conditional parameters do not relate strongly to bioavailability (Shaked et al., 2020(Shaked et al., , 2021, although this could also be caused by the limited role assigned to iron reduction prior to bio-assimilation. Given that the research questions that are typically addressed when determining conditional 800 parameters from the Langmuir isotherm actually relate to how variations in the parameters impact on Fe speciation (e.g. abundance of Fe not bound to organic matter and thus assumed to be bioavailable), it would seem appropriate to broaden the methodology applied to these questions to other methods capable of e.g. estimating concentrations and distributions of the dominant groups of binding sites and/or the lability and solubility of Fe. For example, Whitby et al. (2020) and Laglera et al., (2019) have used alternative methods based on voltammetry to suggest that humic substances are more important than thought 805 before. Other recent studies have focused on the influence of the pH on metal organic complexation and vary the pH of titrations (Ye et al., 2020Zhu et al., 2021). Here we note that the acid-base chemistry of organic matter -which underpins metal binding -is severely understudied, but can nevertheless quantify the distribution and concentration of the total cation binding sites present, at least in the portion of dissolved organic matter that can be isolated from seawater (Lodeiro et al., 2020). With further knowledge of acid-base chemistry of dissolved organic matter and how it changes with inputs of fresh 810 material from biological activity, the total binding site concentration could be independently constrained and only binding affinities derived. Alternatively models can be applied that allow for the estimation of metal speciation for ambient conditions (e.g. Hiemstra and van Riemsdijk, 2006;Stockdale et al., 2015), with predictions of how binding to organic matter could influence e.g. iron solubility (Zhu et al., 2021). The wider application of methods employing cation chelating resins, such as diffusive thin film gradients (Zhang and Davison, 2015;Town et al., 2009;Bayens et al., 2018) might also offer alternative 815 insights into the lability of metals in seawater. Another way to apply voltammetric methods is to characterize metal-binding ligands by pseudovoltammetry (Luther et al, (2021), although unfortunately this method is not currently suitable for DFe.
Whilst continued application of the CLE-AdCSV approach will no doubt further develop our knowledge on how operationally defined ligand concentrations and stability constants vary in the ocean under a restrictive set of conditions apparently specific to one added ligand, new approaches to both the determination and the interpretation of metal binding to organic matter will 820 surely stimulate discussions in the field of the organic metal complexation and furthermore be likely lead to new insights.It is time to search for other methods to study the overall effect of organic ligands on iron speciation in marine waters. However, in case voltammetric methods are used, we advise the SA5-application but to be careful to draw conclusions from obtained -values, since apart of the constrainedt forced by D, these estimates have a much larger error than expected apparent from those obtained from the fits to the Langmuir isothermby fitting a Langmuir isotherm.It is time to search for other methods 825 to study the overall effect of organic ligands on iron speciation in marine waters.
It is clear that further work needs to be done to effectively contextualise the data base and robust quality control procedures are urgently required. We recommend that these procedures include the determination of a standard ligand with independently determined thermodynamic constants that is within the detection window of the applied methodology. Our work suggests that none of the ligands examined here are ideal for this, since they might have been outside the detection window, or do not have 830 available thermodynamic constants for comparison. Table captions   Table 1 Alpha values of the AL are the direct outcomes of the calibration exercises; therefore, these have a standard deviation added, which is the standard deviation around the mean of 4 calibrations. Since K and/or β are directly derived by dividing trough the AL concentration or squared concentration the standard deviations of    Table 3