Oxidation of Fe(II) in Natural Watersat High Nutrient ConcentrationsA R I D A N E G . G O N Z A L E Z ,J . M A G D A L E N A S A N T A N A - C A S I A N O , *N O R M A P E R E Z , A N DM E L C H O R G O N Z A L E Z - D A V I L A

Departamento de Quımica, Facultad de Ciencias del Mar,Universidad de Las Palmas de Gran Canaria, Campus deTafira, 35017 Las Palmas, Spain

Received March 26, 2010. Revised manuscript receivedJune 25, 2010. Accepted September 15, 2010.

The Fe(II) oxidation kinetic was studied in seawater enrichedwith nutrients as a function of pH (7.2-8.2), temperature (5-35°C), and salinity (10-36.72) and compared with the sameparameters in seawater media. The effect of nitrate (0-1.77× 10-3 M), phosphate (0-5.80 × 10-5 M) and silicate (0-2.84× 10-4 M) was studied at pH 8.0 and 25 °C. The experimentalresultsdemonstratedthatFe(II)oxidationwasfaster inhighnutrientconcentrations affecting the lifetime of Fe(II) in nutrient richwaters. Silicate displayed the most significant effects on the Fe(II)oxidation rate with values similar to those determined inseawater enriched with all the nutrients.A kinetic model was applied to the experimental results inorder to account for changes in the speciation and to computethe fractional contribution of each Fe(II) species to the totalrate constant as a function of pH. FeH3SiO4

+ played a key rolein the Fe(II) speciation, dominating the process at pH over8.4. At pH 8.0, FeH3SiO4

+ represented 18% of the total Fe(II)species. Model results show that when the concentration ofsilicate is 3 × 10-5 M as in high nutrient low chlorophyll areas,FeH3SiO4

+ contributed at pH 8.0 by 4% increasing the rate to11% at 1.4 × 10-4 M. The effect of nutrients, especially silicate,must be considered in any further study in seawater mediacultures and eutrophic oceanic areas.

Introduction

Iron is one of the most abundant elements on the Earth,although its concentration in seawater is very low (1). Ironspeciation is of ecological significance both in the openocean (2-5) and coastal waters (1, 6-11). Due to its key rolein photosynthesis and nitrogen assimilation, iron is espe-cially relevant for phytoplankton (12, 13).

Some work has been carried out to improve knowledgewith respect to the biogeochemical cycle of iron, by studyingthe Fe(II) oxidation process in seawater (14-21) at micro-molar and nanomolar concentrations. Iron can be presentin seawater as Fe(II) or Fe(III) and the concentrations oftotal dissolved Fe(II) and Fe(III) are influenced by the redoxcondition of the marine environments (22). There are differ-ent Fe(II) sources in oxygenic surface seawaters: wet deposi-tion (23-25), photochemical reduction in situ (26, 27), andbiological reduction (13, 28, 29). Once the Fe(II) is presentin seawater, the most widely accepted mechanism to describe

the oxidation of Fe(II) in natural waters is the Haber-Weissmechanism:

Fe(III) reduction with superoxide and the competition withthe copper present in natural waters should also be con-sidered (18, 21).

Natural waters can be ranked as a function of nutrientcontents (NO3

-, HPO42-, Si(OH)4) from oligotrophic (low

nutrient concentration; [NO3-] ) 0-6 µM, [HPO4

2-] ) 0-0.5µM, [Si(OH)4] ) 0-10 µM) to eutrophic (high nutrientconcentration; [NO3

-] ) 18-25 µM, [HPO42-] ) 0.8-2 µM,

[Si(OH)4] ) 20-60 µM) (30). When the nutrient concentra-tions increase excessively, eutrophication can take place ashas been described in coastal and estuarine areas (31) andlagoons (32). The enrichment of these nutrients may havesignificant effects on iron speciation, transformation andtransport, depending on local conditions (33). In culturemedia, as in f/2 medium, the concentration of nutrients canreach 8.82 × 10-4 M, 2.88 × 10-5 M, and 1.41 × 10-4 M forNO3

-, HPO42-, and SiO3

2-, respectively (34), and may affectthe Fe speciation and the redox chemistry.

This work studied the oxidation of Fe(II) using seawaterand seawater enriched with high nutrient concentration(nitrate, phosphate and silicate) (SEN) in air saturatedconditions under different pH, temperatures and salinitiesto improve our knowledge with respect to the iron chemistryin natural waters and cultures. Under such conditions, theapparent oxidation rate of Fe(II) can be defined by the eq 9(14)

In excess oxygen, the apparent rate constant can be expressedas a pseudofirst order rate constant

where k’ ) kapp[O2].A kinetic model (18) has been applied to the experimental

data in order to describe the role played by iron-nutrientspecies in the oxidation process of Fe(II).

Experimental SectionReagents. The stock solutions of nutrients were preparedusing sodium nitrate (Sigma) (882 mM), potassium hydrogenphosphate (Sigma) (28.8 mM) and sodium silicate (Sigma)(142 mM). Under the different physicochemical conditionsof this work, the concentration of each nutrient was kept at

* Corresponding author phone: +34-928-454-448; fax: +34-928-452-922; e-mail: jmsantana@dqui.ulpgc.es.

Fe(II) + O2 f Fe(III) + O2·- (1)

Fe(II) + O2·-98

2H+Fe(III) + H2O2 (2)

Fe(II) + H2O2 f Fe(III) + OH· + OH- (3)

Fe(II) + OH· f Fe(III) + OH- (4)

Fe(III) + O2·- f Fe(II) + O2 (5)

Fe(III) + 3OH- f Fe(OH)3(s) (6)

Cu(II) + O2·- f Cu(I) + O2 (7)

Cu(I) + O2·- + 2H+ f Cu(II) + H2O2 (8)

d[Fe/II)]/dt ) -kapp[Fe(II)][O2] (9)

d[Fe(II)]/dt ) -k′[Fe(II)] (10)

Environ. Sci. Technol. 2010, 44, 8095–8101

10.1021/es1009218 2010 American Chemical Society VOL. 44, NO. 21, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 9 8095

Published on Web 10/01/2010

8.82 × 10-4 M, 2.88 × 10-5 M, and 1.41 × 10-4 M for NO3-,

HPO42-, and SiO3

2-, respectively.The stock solutions of Fe(II) (4 × 10-4 M) were prepared

using ferrous ammonium sulfate hexahydrate (Sigma),acidified at a pH 2 with Suprapur HCl in NaCl 0.7 M. Theinitial concentration of Fe(II) was always kept at 25 nM.

Fe(II) oxidation was studied in seawater (SW) and sea-water enriched with nutrients (SEN). The seawater used inthis study was collected 200 m off-shore North-East of theisland of Gran Canaria Island at 8 m depth. Some studieswere carried out in artificial seawater (ASW) (35, 36). Thissolution was also enriched with high nutrient concentrations(ASEN) in order to study the differences with the sameparameters for seawater. All the chemicals used were traceanalysis grade.

Experimental Conditions. The Fe(II) oxidation wasstudied as a function of pH (7.2-8.2), temperature (5-35 °C)and salinity (10-36.72). The effect of individual nutrients(nitrate, phosphate, and silicate) was also studied. The rangeof nutrient concentrations varied from 0 to 1.77 × 10-3 M,5.80 × 10-5 M, and 2.84 × 10-4 M for NO3

-, HPO42-, and

SiO32-, respectively.

The pH measurements, in free hydrogen ion scale(pHF ) -log[H+]), were carried out potentiometrically withan Orion pH-meter using a combination electrode (RossCombination, glass body). The pH was adjusted automaticallyat the desired value using an automatic titrator system (Titrino719S, Methrom) with HCl 0.1 M. The effect of temperatureand salinity on the pK* of the Tris-buffers was considered inthis study (35). Tris-(hydroximethyl)aminomethane(tris)-artificial seawater buffers (0.005 mol ·kg-1 Tris and Tris-HClin artificial seawater) were used to calibrate the electrode.The temperature was controlled to (0.02 °C with a refriger-ated bath RB-5A in the range of 5-35 °C. The salinity of theseawater as determined using a salinometer (Portasal, 8410A)was 36.720. For the salinity studies, the dilutions were madewith Milli-Q ion exchange water (18 MQ). The experimentswere always carried out in saturated air conditions, bubblingpure air through the solution for one hour.

Oxidation Experiments. The oxidation experiments werecarried out in a 250 mL glass thermostatted reaction cell.After the solution was bubbled with pure air for one hour,the pH was adjusted at the desired value with an automatictitration system. The addition of Fe(II) to the samplecorresponded to the zero time of reaction. A blank withseawater and the nutrients and reagents for iron determi-nation was prepared in each case. Fe(II) concentrations weremeasured spectrophotometrically using a modified versionof the ferrozine method. The details of the method used havebeen given elsewhere (17, 37). In order to measure nanomolariron(II) concentrations, a 5 m long waveguide capillary flowcell (LWCFC) from World Precision Instruments was used.The Fe(II) rate constant kapp (kg mol-1min-1) was computedconsidering kapp ) k’/[O2 ] from eq 10 where the values of[O2] were determined from solubility equations (38). Thepseudofirst order behavior was observed for periods of overhalf-life time in all the studies considered with a R2 g 0.98.

Numerical Model. The Gepasi version 3.30 software wasused to simulate the chemical kinetics for all the reactants.The overall and individual rate constants ki were obtainedby adjusting the experimental Fe(II) concentrations/time pairof data to the model output as indicated elsewhere (17). Thekinetic model (18) was applied as a function of pH for bothseawater and seawater with high nutrient concentrations(SEN), considering all dissociation and equilibrium constantsfor the reaction of Fe(II) species with the different majorinorganic species in seawater. In accordance with theexperimental results obtained in the present article, the modelwas extended to include Fe(II)-silicate species.

Results and Discussion

The Fe(II) oxidation followed a pseudofirst order behaviorfor all the studies under the experimental conditions usedin the present work. The log kapp was always higher in thepresence of nutrients (Figure 1). The nutrient speciation wasstudied in the ocean (36), showing that silicates are usuallypresent as Si(OH)4 (96%) and Si(OH)3O- (4%) in aqueoussolutions. Nitrates are present as NO3

- and phosphates mostlyas HPO4

2- (79.2%) or PO43- (20.4%), the phosphates interact-

ing with Ca2+ and Mg2+. Phosphate and Mg2+ can be foundin the ocean as MgH2PO4

+ (7%), MgHPO4 (45.8%) and MgPO4-

(26.6%). In addition, phosphate and Ca2+ can be found asCaH2PO4

+ (0.7%), CaHPO4 (4.9%), and CaPO4- (73.2%) (36).

Effect of Nutrient Concentration. The effect of eachnutrient on the Fe(II) oxidation rate was determined inseawater enriched with different concentrations of NO3

-

(0-1.77 × 10-3 M), HPO42- (0-5.80 × 10-5 M), and Si(OH)4

(0-2.84 × 10-4 M). The selected range of concentrationscovered those found in poor nutrient surface ocean olig-otrophic waters, the HNLC waters (NO3

- ) 29 × 10-6 M,HPO4

2- ) 1.9 × 10-6 M and Si(OH)4 ) 30 × 10-6 M) (39, 40),the highly nutrient-rich waters from eutophication ([NO3

-]g 450 µM, [HPO4

2-] g 80 µM, [Si(OH)4] g 100 µM) (41) andwaters with f/2 medium prepared for cultures (SEN). Thenutrient concentrations added to natural seawater to preparethe SEN solution are those commonly used in f/2 mediawhen laboratory experiments with phytoplankton are con-sidered (34, 42, 43). Figure 1 showed the effect of eachindividual nutrient concentration on the Fe(II) oxidation rateconstant in seawater. The Fe(II) rate constant followed thesame behavior for the three nutrients used. The Fe(II)oxidation rate increased with the nutrient concentration toa maximum. The data have been fitted to an exponentialfunction reaching a maximum (eqs 11-13), where thestandard error of estimation was 0.013, 0.004, and 0.020 fornitrate, phosphate and silicate, respectively.

FIGURE 1. Effect of added individual nutrients (nitrate, phos-phate and silicate) on the apparent Fe(II) oxidation rate inseawater solution. The dots, squares and triangles are the ex-perimental data. Lines represent the polynomial fitting (eq11-13). The initial Fe(II) concentration was 25 nM, temperature,25 °C and pH fixed at 8.0.

log kapp,NIT ) 2.70 + 0.23(1 - e(-3391.28[NIT])) R2 ) 0.995(11)

8096 9 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 21, 2010

NIT, PHP and SIL are the total concentrations of nitrate,phosphate and silicate for each condition. The maximumvalues in log kapp were 2.93 ( 0.01, 2.95 ( 0.01, and 3.03 (0.01 (kg mol-1min-1) for NO3

-, HPO42-, and Si(OH)4, respec-

tively. The changes on the rate constant confirm the effectof nutrients on the Fe(II) oxidation.

The studies carried out for each nutrient (NO3- ) 8.82 ×

10-4 M, HPO42- ) 2.88 × 10-5 M and Si(OH)4 ) 1.41 × 10-4

M), were compared with studies carried out in SW and inSEN, at pH 8.0 and temperature 25 °C (Figure 2). The seawater,enriched with nitrate, phosphate or silicate, was labeled asSNIT, SPH, and SSIL, respectively. The Fe(II) rate constantincreased according to SW (log kapp ) 2.70 ( 0.01 kgmol-1min-1) <SNIT (log kapp ) 2.93 ( 0.02 kg mol-1min-1)< SPHP (log kapp ) 2.95 ( 0.05 kg mol-1min-1) < SSIL (log kapp

) 3.03 ( 0.03 kg mol-1min-1) < SEN (log kapp ) 3.04 ( 0.02kg mol-1min-1). SSIL and SEN showed similar values in logkapp (3.04 and 3.03 kg mol-1min-1) (Figure 3). Artificialseawater (ASW) was considered to be the reference value inFigure 3. Figure 3 depicts various Fe(II) oxidation scenariosas a function of the nutrient contents in environmentalwaters: oligotrophic waters (SW), theoretical high nutrientlow chlorophyll waters (HNLC), seawater enriched withnutrients (SEN) and artificial seawater enriched with nutrients(ASEN), both of them at the same concentration of the f/2media. Seawater with high concentrations of each individualnutrient was also considered. The slowest Fe(II) oxidationrate was found in SW and in HNLC seawater, with log kapp

) 2.70 kg mol-1min-1and 2.75 kg mol-1min-1, respectively.The fastest Fe(II) oxidation rate was determined in seawaterwith high nutrient concentrations, SEN (log kapp ) 3.04 kgmol-1min-1). The difference between SW and HNLC was 0.7min in half-life time for Fe(II), at pH 8.0 and T ) 25 °C. TheFe(II) oxidation rate constant for SSIL was similar to the valuefor ASEN (log kapp ) 3.03 kg mol-1min-1). This studyconfirmed that nutrients at high concentrations as occurs inthe culture media may play an important role on the Fe(II)rate constant, with silicate as the most active.

The oxidation of Fe(II) in SW was slightly slower than inASW (∆logkapp)-0.14). The difference observed is explained

by the effects of the presence of organic matter in the redoxchemistry of iron (17, 44). When the water is enriched withnutrients in both SEN and ASEN studies, the effects of thenutrients on the Fe(II) oxidation rate constant exceed thoseof the organic matter, making them negligible.

pH, Temperature, and Salinity Dependence. The Fe(II)oxidation rate as a function of pH was studied in SW and inSEN. ASW was also considered. The effect of pH was studiedin the range 7.2-8.2 (Figure 4). The rate constants in SENwere always higher than in SW. The pH dependence wasfitted to a second order polynomial equation for SW andSEN, respectively (eqs 14 and 15):

The standard error of estimation was 0.03 and 0.02 forSW and SEN.

The increase observed in the Fe(II) rate constant forseawater at high nutrient concentrations (SEN) represented12% at pH 8.0. As indicated above, the Fe(II) oxidation process

FIGURE 2. Fe(II) concentration in different media: seawater (SW),seawater enriched with high nutrient concentrations (SEN), sea-water enriched with nitrate (SNIT) (8.82 × 10-4 M), seawaterenriched with phosphate (SPHP) (2.88 × 10-5 M) and seawaterenriched with silicate (SSIL) (1.41 × 10-4 M). Initial Fe(II)concentration was 25 nM, temperature, 25 °C and pH fixed at 8.0.

log kapp,PHP ) 2.70 + 0.25(1 - e(-220810.50[PHP])) R2 ) 0.999(12)

log kapp,SIL ) 2.70 + 0.36(1 - e(-21188.17[SIL])) R2 ) 0.995(13)

FIGURE 3. Effect of different nutrient concentrations on theFe(II) oxidation rate constant at different solutions (∆log k )log ki - log kASW): seawater (SW), seawater enriched with highnutrient concentrations (SEN), seawater enriched with nitrate(SNIT), seawater enriched with phosphate (SPHP), seawaterenriched with silicate (SSIL), artificial seawater enriched withnutrients (ASEN), and HNLC simulated waters (HNLC). The rateconstant for artificial seawater, ASW, is considered as thereference value. Initial Fe(II) concentration was 25 nM, temper-ature, 25 °C and pH fixed at 8.0.

FIGURE 4. Effect of pH on the Fe(II) oxidation rate constant inSW, SEN, and ASW at 25 °C (initial Fe(II) concentration, 25 nM).

log kapp,SW ) 46.40 - 12.54pH + 0.88pH2R2 ) 0.998

(14)

log kapp,SEN ) 15.09 - 4.26pH + 0.34pH2R2 ) 0.994

(15)

VOL. 44, NO. 21, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 9 8097

increased in the presence of nutrients, indicating that theproducts resulting from the interaction among Fe2+ and NO3

-,HPO4

2-, and Si(OH)4 contribute efficiently to the Fe(II)oxidation process. The log kapp (2.70 ( 0.01 kg mol-1min-1)for SW at pH 8.0 and T ) 25 °C was comparable to the valuesin the natural Pacific Subarctic waters (log kapp ) 2.60 kgmol-1min-1 at pH 8.0 and 25 °C (45)), and with the GulfStream waters (log kapp ) 2.99 kg mol-1min-1, at pH 8.0 and25 °C (18)). The differences in the Fe(II) rate constant valuesare related to the original composition of the natural sea-water samples (14).

From pH 7.2 to pH 8.2, (Table 1) the t1/2 was reduced 1order of magnitude in both SW (from 37.1 to 3.3 min) andSEN (from 20 to 2.4 min). For SW and SEN media, t1/2

decreased by 17.2 min at pH 7.2 and 1.2 min at pH 8.2 (Table1). At the different pH conditions, the presence of thesenutrients decreased the lifetime of Fe(II) in the naturalenvironment. Figure 4 also depicts the pH effect on the Fe(II)oxidation rate in ASW. As was pointed out above, thedifferences are explained by the presence of organic matterin the SW media.

The study of the temperature dependences will enable usto know whether the oxidation process of Fe(II) in thepresence of high nutrient concentrations differs from thesame process in natural seawater. Temperature dependencewas studied from 5 to 35 °C, at 10 °C interval, in SW and SEN(Figure 5). The experimental data were fitted to SW and SENsolutions respectively (eq 16-17).

where T is temperature (K) and the standard error ofestimation was 0.04 and 0.02 for SW and SEN. The rateconstants were higher in SEN than in SW at each temperature.However, the plot of log kapp versus 1/T gave similar slopes,-4636 K-1 in SW and -4473 K-1 in the SEN solution. TheEnergy of Activation (Ea) was 88 ( 2 kJmol-1 and 86 ( 4kJmol-1 for SW and SEN, respectively. These values werecomparable to nanomolar Fe(II) oxidation experiments

measured in Gulf Stream seawater (18) and sub-Arctic Pacificwaters (45). These results indicated that the mechanismcontrolling the Fe(II) oxidation in seawater involved the samechemical process as in seawater with high nutrient contents,only accelerated by the presence of these nutrients, especiallyby the silicate. log kapp increased 0.05 units/degree in bothsolutions. The t1/2 values (Table 1) were 52.9 min at 5 °C and2.1 min at 35 °C for seawater. The t1/2 decreased from 27.6min at 5 °C to 1.4 min at 35 °C in the SEN sample. Thesedifferences represent a reduction of 46% in lifetime whenthe nutrients were present in solution, favoring the disap-pearance of Fe(II) from the natural waters and increasing itsreactivity.

The effect of salinity on the Fe(II) rate constant for SWand SEN was carried out by dilution with Milli-Q waters (18MΩ), keeping the nutrient concentrations constant in theSEN solutions. The range of salinity studied was 10-36.72(Figure 6). The effect of HCO3

- was corrected for bothsolutions (18) and fitted to 2 mM. The experimental resultsfor Figure 6 were fitted to SW and SEN solutions, respectively:

TABLE 1. Half-Life Time (t1/2) for Fe(II) Oxidation in SW and SEN (Nitrate (8.82 × 10-4M), Phosphate (2.88 × 10-5M) and Silicate(1.41 × 10-4M))

media pH temperature (°C) salinity log kapp (kgmol-1min-1) t1/2 (min-1)

seawater (SW)

7.2 25 36.7 1.96 37.17.5 25 36.7 2.07 28.67.8 25 36.7 2.39 14.08.0 25 36.7 2.70 6.78.2 25 36.7 3.02 3.38.0 5 36.7 1.64 52.98.0 15 36.7 2.21 17.48.0 35 36.7 3.28 2.18.0 25 10.0 3.03 2.78.0 25 16.0 2.92 3.58.0 25 20.0 2.88 4.08.0 25 30.0 2.77 5.5

seawater enriched with nutrients (SEN)

7.2 25 36.7 2.23 20.07.5 25 36.7 2.41 13.27.8 25 36.7 2.76 5.98.0 25 36.7 3.04 3.08.2 25 36.7 3.22 2.08.0 5 36.7 1.92 27.68.0 15 36.7 2.39 11.58.0 35 36.7 3.45 1.48.0 25 10.0 3.36 1.38.0 25 20.0 3.23 1.88.0 25 30.0 3.12 2.5

log kapp,SW ) 18.29 - 4636/T R2 ) 0.998 (16)

log kapp,SEN ) 17.97 - 4473/T R2 ) 0.992 (17)

FIGURE 5. Effect of temperature on the Fe(II) oxidation rateconstant in SW and SEN at pH 8.0 (initial Fe(II) concentration,25 nM).

8098 9 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 21, 2010

S corresponds to salinity with a standard error ofestimation for log kapp of 0.02 and 0.01, respectively. Theseequations included the effect of changes in the carbonateconcentrations and salinity. When the concentration ofbicarbonate was kept constant to 2 mM, log kapp increasedand the dependence was fitted to SW (eq 20) and SEN (eq21).

The standard error of estimation for both equations was0.01. The Fe(II) oxidation process was faster for SEN in allthe salinity range. The difference in log kapp for both mediastayed constant at 0.34 ( 0.02 kgmol-1min-1. The t1/2 (Table1) decreased from 6.7 to 2.7 min when the salinity changedfrom S ) 36.72 to S ) 10, in SW, whereas the t1/2 moved from3 to 1.3 min, in SEN.

The apparent rate constant (kg mol-1 min-1) can be fittedunder the experimental conditions of pH (free scale),temperature (Kelvin) and salinity, in SW (eq 22) and SENsamples (eq 23).

The standard error of estimation was 0.04 and 0.06. The R2

was 0.998 and 0.996, respectively. These equations can beused and applied in the environment, under the experimen-tal conditions used.

Kinetic Model. A kinetic model was applied to theexperimental data as a function of pH, in order to accountfor the speciation of Fe(II) and for the contribution of eachindividual species to the Fe(II) overall rate constant, inseawater at high nutrient concentration. According to theexperimental results, the Fe(II) oxidation rates in the SEN

solutions were controlled by the silicate effect. The differencebetween SW and SEN was constant in the pH-range studied(log kSEN - log kSW ) 0.34 ( 0.06 kg mol-1min-1), and thesilicate effect can be considered dominant over the entirepH range. Therefore, the base model, with the equilibriumand rate constants for all inorganic species involved in theprocess for seawater, from references (18) and (20) wereextended by including eqs 24-28 (46, 47).

The Fe(II) apparent rate constant is composed of severalindividual rate constants for Fe(II) species, which react withoxygen at different rates. The Fe(II) oxidation rate can bedetermined as a function of the weighted sum of the oxidationrates of the individual Fe(II) species:

FIGURE 7. Speciation of Fe(II) in seawater enriched with sili-cate (A) 1.41 × 10-4 M and (B) 3 × 10-5 M, at 25 °C and salin-ity 36.72.

FIGURE 6. Effect of salinity on the Fe(II) oxidation rate constantin SW and SEN at T ) 25 °C and pH 8.0 (initial Fe(II) concen-tration, 25 nM).

log kapp,SW ) 3.08 - 0.02S + 1.34 × 10-4S2R2 ) 0.993

(18)

log kapp,SEN ) 3.40 - 0.01S + 1.16 × 10-4S2R2 ) 0.999

(19)

log kapp,SW ) 3.21 - 0.02S + 1.61 × 10-4S2R2 ) 0.997

(20)

log kapp,SEN ) 3.51 - 0.02S + 1.16 × 10-4S2R2 ) 0.999

(21)

log kapp,SW ) 57.48 - 11.38pH + 0.81pH2 - 4532/T -0.02S + 2 × 10-4S2 (22)

logkapp,SW ) 28.67 - 3.72pH + 0.31pH2 - 4557/T -8.61 × 10-3S - 9.88 × 10-5S2 (23)

Si(OH)4 T Si(OH)3O- (24)

Mg2+ + Si(OH)3O- T MgH3SiO4+ (25)

Ca2+ + Si(OH)3O- T CaH3SiO4+ (26)

Fe3+ + Si(OH)3O- T FeH3SiO42+ (27)

Fe2+ + Si(OH)3O- T FeH3SiO4+ (this work) (28)

VOL. 44, NO. 21, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 9 8099

where Ri ) [FeXi]/[Fe(II)]T is the molar fraction of each Fe(II)species in the solution and also a function of the ionic media.k is the apparent overall rate constant (kg mol-1min-1) andki, the individual rate constants for the Fe(II) species.

Speciation of Fe(II) was included in the kinetic model.The Fe(II) speciation of seawater enriched with two differentsilicate concentrations are shown in Figure 7. The equilibriumconstant for eq 28 was estimated by the kinetic model as logK ) -4.40 and the rate constant for the species FeH3SiO4

+

fixed to log k ) 2.70 (kg mol-1 min-1). Considering a silicateconcentration of 1.41 × 10-4 M (Figure 7A) as a referencevalue in eutrophicated waters, the speciation of Fe(II) isdominated by Fe2+, Fe(CO3), FeH3SiO4

+, FeCl+, FeSO4,Fe(CO3)(OH)-. Fe2+ is the most important species in the Fe(II)speciation from pH 6 (58%) to 8.2 (26%). Fe(CO3) dominatesthe speciation at pH between 8.2 (26%) and 8.4 (30%). Therole of FeH3SiO4

+ increased from pH 6, which accounted for0.6% of the total Fe(II) speciation, converting it into thesecond most important species at pH 8.4 (30%), reaching pH8.5, 31%. FeCl+ and FeSO4 were also important species frompH 6 (25% and 16%, respectively) to 7.5 (17% and 12%,respectively). Fe(CO3)(OH)- began to be important to pHover 7.6, reaching 7% at pH 8.5.

At silicate concentrations of 3 × 10-5 M, typical for surfaceHNLC regions (Figure 7B), the FeH3SiO4

+ was still impor-

tant, rating 5% at pH 8.0 and 9% at pH 8.5.The other Fe(II)species followed a similar distribution. Therefore, FeH3SiO4

+

should be considered in the Fe(II) speciation in naturalwaters, especially in eutrophicated waters, where silicate canbe found at high concentrations.

The individual contribution to the Fe(II) overall kineticrate was computed from the results of the kinetic model andthe speciation of Fe(II) (eq 29). The results for each fractionalcontribution are shown as a function of pH in Figure 8. Thefractional contribution follows a similar distribution for bothhigh silicate concentrations (1.41 × 10-4 M) and HNLC typesilicate concentration (3 × 10-5 M) (Figure 8A and 8B,respectively). The contribution was controlled by Fe2+,Fe(OH)2, Fe(CO3)2

2-, Fe(OH)+, FeH3SiO4+, Fe(CO3)(OH)-, and

Fe(CO3). In SSIL media, the overall contribution was domi-nated by Fe2+ from pH 6 (97%) to 7.7 (22%). At pH over 7.7,the dominant species was Fe(OH)2, reaching 22% at pH 7.7and 58% at pH 8.5. The contribution of FeH3SiO4

+ wasimportant over the whole pH range. It was 2% at pH 6, 14%at pH 7.5 and 6% at pH 8.5. The contribution of FeH3SiO4

+

was the fourth most important at pH 8.0 (11% at SEN and4% at HNLC). These results confirm that nutrient concentra-tions, in particular silicate, play an important role in theoxidation of Fe(II) in seawater at high nutrient concentrations,making Fe(II) less available for biologically mediated pro-cesses in this aquatic media.

AcknowledgmentsThis study was supported by Project CTM2006-09857 ofMinisterio de Ciencia e Innovacion from Spain. A.G.G.participation was supported by the Grant BES-2007-15776of Ministerio de Ciencia e Innovacion.

Literature Cited(1) Johnson, K. S.; Chaves, F. P.; Friederich, G. E. Continental shelf

sediment as primary sources of iron for coastal phytoplankton.Nature. 1999, 398, 697–700.

(2) Martin, J. H.; Gordon, R. M.; Fitzwater, S. E. Iron in Antarcticwaters. Nature. 1990, 345, 156–158.

(3) Martin, J. H.; Coale, K. H.; Johnson, K. S.; Fitzwater, S. E.; Gordon,R. M.; Tanner, S. J.; Hunter, J. N.; Elrod, V. A.; Nowicki, J. L.;Coley, T. L.; Barber, R. T.; Lindley, S.; Watson, A. J.; Van Scoy,K.; Law, C. S.; Liddicoat, M. I.; Ling, R.; Stanton, T.; Stockel, J.;Collins, C.; Anderson, A.; Bidigare, R.; Ondrusek, M.; Latasa, M.;Millero, F. J.; Lee, K.; Yao, W.; Zhang, J. Z.; Friederich, G.;Sakamoto, C.; Chavez, F.; Buck, K.; Kolber, Z.; Greene, R.;Falkowski, P.; Chisholm, S. W.; Hoge, F.; Swift, R.; Yungel, J.;Turner, S.; Nightingale, P.; Hatton, A.; Liss, P.; Tindale, N. W.Testing the iron hypothesis in ecosystems of the equatorialPacific Ocean. Nature. 1994, 371, 123–129.

(4) Boyd, P. W.; Watson, A. J.; Law, C. S.; Abraham, E. R.; Trull, T.;Murdoch, R.; Bakker, D. C. E.; Bowie, A. R.; Buessler, K. O.;Chang, H.; Charette, M.; Croot, P.; Downing, K.; Frew, R.; Gall,M.; Hadfield, M.; Hall, J.; Harvey, M.; Jameson, G.; LaRoche, J.;Liddicoat, M.; Ling, R.; Maldonado, M. T.; McKay, R. M.; Nodder,S.; Pickmere, S.; Rintoul, S.; Safi, K.; Sutton, P.; Strzepek, R.;Tanneberger, K.; Turner, S.; Waite, A.; Zeldis, J. A mesoscalephytoplankton bloom in the polar Southern Ocean stimulatedby iron fertilization. Nature. 2000, 407, 695–702.

(5) Watson, A. J.; Bakker, D. C. E.; Ridgwell, A. J.; Boyd, P. W.; Law,C. S. Effect of iron supply on southern ocean CO2 uptake andimplications for glacial atmospheric CO2. Nature. 2000, 407,730–733.

(6) Hutchins, D. A.; DiTullio, G. R.; Zhang, Y.; Bruland, K. W. Aniron limitation mosaic in the California upwelling regime.Limnol. Oceanogr. 1998, 43, 1037–1054.

(7) Hutchins, D. A.; Franck, V. M.; Brzezinski, M. A.; Bruland, K. W.Inducing phytoplankton iron limitation in iron-replete coastalwaters with a strong chelating ligand. Limnol. Oceanogr. 1999,44, 1009–1018.

(8) Matsunaga, K.; Nishioka, J.; Kuma, K.; Toya, K.; Suziki, Y. Riverineinput of bioavailable iron supporting phytoplankton growth inKesennuma Bay (Japan). Water Res. 1998, 32, 3436–3442.

(9) Wells, M. L. Manipulation iron availability in nearshore waters.Limnol. Oceanogr. 1999, 44, 1002–1008.

FIGURE 8. Contribution of each individual specie on the totalFe(II) oxidation rate in seawater enriched with silicate (A) 1.41× 10-4 M and (B) 3 × 10-5 M, at 25 °C and salinity 36.72.

k ) kFe2+RFe2+ + kFeOH+RFeOH+ + kFe(OH)2RFe(OH)2

+

kFeHCO3+RFeHCO3

+ + kFe(CO3)RFe(CO3) +

kFe(CO3)22-RFe(CO3)2

2- + kFe(CO3)OH-RFe(CO3)OH- +

kFeCl+RFeCl+ + kFeSO4RFeSO4

+ kFeH3SiO4+RFeH3SiO4

+

(29)

8100 9 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 44, NO. 21, 2010

(10) Zhang, J. Evidence of trace metal limited photosynthesis ineutrophic estuarine and coastal waters. Limnol. Oceanogr. 2000,45, 1871–1878.

(11) Ozturk, M.; Steinnes, E.; Sakshaug, E. Iron speciation in theTrondheim Fjord from the perspective of iron limitation forphytoplankton. Estuarine, Coastal Shelf Sci. 2002, 55, 197–212.

(12) Sunda, W. G.; Huntsman, S. A. Interrelated influence of iron,light, and cell size on marine phytoplankton growth. Nature.1997, 390, 389–392.

(13) Maldonado, M. T.; Price, N. M. Reduction and transport oforganically bound iron by Thalassiosira oceanica (Bacillari-ophyceae). J. Phycol. 2001, 37, 298–309.

(14) Millero, F. J.; Sotolongo, S.; Izaguirre, M. The oxidation kineticsof Fe(II) in seawater. Geochim. Cosmochim. Acta 1987, 51, 793–801.

(15) Millero, F. J.; Izaguirre, M. Effect of ionic strength and ionicinteractions on the oxidation of Fe(II). J. Solut. Chem. 1989, 18,585–599.

(16) King, D. W.; Lounsbury, H. A.; Millero, F. J. Rates and mechanismof Fe(II) oxidation at nanomolar total iron concentration.Environ. Sci. Technol. 1995, 29, 818–824.

(17) Santana-Casiano, J. M.; Gonzalez-Davila, M.; Millero, F. J. Theoxidation of Fe(II) in NaCl-HCO3

- and seawater solutions inthe presence of phthalate and salicylate ions: a kinetic model.Mar. Chem. 2004, 85, 27–40.

(18) Santana-Casiano, J. M.; Gonzalez-Davila, M.; Millero, F. J.Oxidation of nanomolar levels of Fe(II) with oxygen in naturalwaters. Environ. Sci. Technol. 2005, 39, 2073–2079.

(19) Santana-Casiano, J. M.; Gonzalez-Davila, M.; Millero, F. J. Therole of Fe(II) species on the oxidation of Fe(II) in natural watersin the presence of O2 and H2O2. Mar. Chem. 2006, 99, 70–82.

(20) Gonzalez-Davila, M.; Santana-Casiano, J. M.; Millero, F. J.Competition between O2 and H2O2 in the Oxidation of Fe(II) inNatural Waters. J. Solution Chem. 2006, 35, 95–111.

(21) Rose, A. L.; Waite, T. D. Kinetic model for Fe(II) oxidation inseawater in the absence and presence of natural organic matter.Environ. Sci. Technol. 2002, 36, 433–444.

(22) Pankow, J. F. Kinetics for the Aquatic Environment. AquaticChemistry Concepts; Lewis Publishers: Chelsea, MI, 1991; pp1155-1164.

(23) Faust, B. C.; Zeep, R. G. Photochemistry of aqueous iron(III)-polycarboxylate complexes: Roles in the chemistry of atmo-spheric and surface waters. Environ. Sci. Technol. 1993, 27, 2517–2522.

(24) Millero, F. J.; Gonzalez-Davila, M.; Santana-Casiano, J. M.Reduction of Fe(III) with sulfite in natural waters. J. Geophys.Res. 1995, 100, 7235–7244.

(25) Kieber, R. J.; Williams, K.; Willey, J. D.; Skrabal, S.; Avery, G. B.Iron speciation in coastal rainwater: Concentration and deposi-tion to seawater. Mar. Chem. 2001, 73, 83–95.

(26) Wells, M. L.; Mayer, L. M.; Donard, O. F. X.; De Souza Sierra,M. M.; Ackelson, S. G. The photolysis of coloidal iron in theoceans. Nature. 1991, 353, 248–250.

(27) Johnson, K. S.; Coale, K. H.; Elrod, V. A.; Tindale, N. W. Ironphotochemistry in seawater from equatorial Pacific. Mar. Chem.1994, 46, 319–334.

(28) Kustka, A. B.; Shaked, Y.; Milligan, A. J.; King, D. W.; Morel,F. M. M. Extracellular production of superoxide by marinediatoms: Contrasting effects on iron redox chemistry andbioavailability. Limnol. Oceanogr. 2005, 50, 1172–1180.

(29) Rose, A. L.; Salmon, T. P.; Lukondeh, T.; Neilan, B. A.; Waite,T. D. Use of superoxide as an electron shuttle for iron acquisitionby the marine cyanobaterium Lyngbya majuscule. Environ. Sci.Technol. 2005, 39, 3708–3715.

(30) Levitus, S.; Conkright, M. E.; Reid, J. L.; Najjar, R. G.; Mantyla,A. Distribution of nitrate, phosphate and silicate in the worldoceans. Prog. Oceanogr. 1993, 31, 245–273.

(31) Rabalais, N. N.; Turner, R. E.; Dıaz, R. J.; Justic, D. Global changeand eutrophication of coastal waters. ICES J. Mar. Sci. 2009, 66,1528–1537.

(32) Windom, H. L.; Niencheski, L. F.; Smith, R. G. Biogeochemistryof nutrients and trace metals in the estuarine region of thePatos Lagoon (Brazil). Estuarine, Coastal Shelf Sci. 1999, 48,113–123.

(33) Ozturk, M.; Bizsel, N.; Steinnes, E. Iron speciation in eutrophicand oligotrophic Mediterranean coastal waters; Impact ofphytoplankton and protozoan blooms on iron distribution. Mar.Chem. 2003, 81, 19–36.

(34) Guillard, R. R. L. Culture of phytoplankton for feeding marineinvertebrates. In Culture of Marine Invertebrate Animals; Smith,W. L., Chanley, M. H., Eds.; Plenum Publishing Corp.: New York,1975; pp 29-60.

(35) Millero, F. J. The pH of estuarine waters. Limnol. Oceanogr.1986, 31, 839–847.

(36) Millero, F. J. Chemical Oceanography; Taylor and Francis Group:Boca Raton, FL, 2006; pp 297-324.

(37) Violler,E.; Inglett,P.W.;Hunter,K.;Roychuodhury,A.N.;Cappellen,P. The ferrozine method revisited: Fe(II)/Fe(III) determination innatural waters. Appl. Geochem. 2000, 15, 785–790.

(38) Benson, B.; Krause, D. The concentration and isotopic frac-tionation of oxygen dissolved in fresh water and seawater inequilibrium with the atmosphere. Limnol. Oceanogr. 1984, 29,620–632.

(39) Nolting, N. F.; Gerringa, L. J. A.; Swagerman, M. J. W.;Timmermans, K. R.; de Baar, H. J. W. Fe(III) speciation in thehigh nutrient, low chlorophyll Pacific region of the SouthernOcean. Mar. Chem. 1998, 62, 335–352.

(40) Boye, M.; van den Berg, C. M. G.; de Jong, J. T. M.; Leach, H.;Croot, P.; de Baar, H. J. W. Organic complexation of iron in thesouthern ocean. Deep-Sea Res., Part I. 2001, 48, 1477–1497.

(41) Maier, G.; Glegg, G. A.; Tappin, A. D.; Worsfold, P. J. The use ofmonitoring data for identifying factors influencing phytoplank-ton bloom dynamics in the eutrophic Taw Estuary, SW England.Mar. Pollut. Bull. 2009, 58, 1007–1015.

(42) Croot, P. L.; Moffet, J. W.; Brand, L. E. Production of extracellularCu complexing ligands by eucaryotic phytoplankton in responseto Cu stress. Limonol. Oceanogr. 2000, 45, 619–627.

(43) Shaked, Y.; Kustka, A. B.; Morel, F. M. M. A general kinetic modelfor iron acquisition by eukaryotic phytoplankton. Limnol.Oceanogr. 2005, 50, 872–882.

(44) Santana-Casiano, J. M.; Gonzalez-Davila, M.; Rodrıguez, M. J.;Millero, F. J. The effect of organic compounds in the oxidationkinetics of Fe(II). Mar. Chem. 2000, 70, 211–222.

(45) Roy, E. G.; Wells, M. L.; King, W. Persistence of iron(II) in surfacewaters of the western subarctic Pacific. Limnol. Oceanogr. 2008,53, 89–98.

(46) Smith, R. M.; Martell, A. E. Critical Stability Constants. InorganicComplexes; Plennum Press: New York, 1991; Vol. 4.

(47) Reardon, E. J. Complexing of silica by iron(III) in natural waters.Chem. Geol. 1979, 25, 339–345.

ES1009218

VOL. 44, NO. 21, 2010 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 9 8101