Elsevier

Marine Chemistry

Volume 220, 20 March 2020, 103754
Marine Chemistry

Assessment of the stability, sorption, and exchangeability of marine dissolved and colloidal metals

https://doi.org/10.1016/j.marchem.2020.103754Get rights and content

Highlights

  • This study tests the effect of storage conditions on Fe, Cu, Ni, Zn, Cd, Pb, Mn, Co.

  • Adsorption of trace metals to bottle walls is most significant for Fe, Co, and Zn.

  • Desorption of trace metals from bottle walls is incomplete for Fe and Co over 20 weeks.

  • Size partitioning of Fe and Co can be preserved under frozen conditions.

  • Metal colloids will quickly re-aggregate after removal through ultrafiltration.

Abstract

The size partitioning of dissolved trace metals is an important factor for determining reactivity and bioavailability of metals in marine environments. This, alongside the advent of more routine shipboard ultrafiltration procedures, has led to increased attention in determining the colloidal phase of metals such as Fe in seawater. While clean and efficient filtration, prompt acidification, and proper storage have long been tenets of trace metal biogeochemistry, few studies aim to quantify the kinetics of colloidal exchange and metal adsorption to bottle walls during storage and acidification. This study evaluates the effect of storage conditions on colloidal size partitioning, the kinetics of colloid exchange over time, and the timescale of bottle wall adsorption and desorption for dissolved Fe, Cu, Ni, Zn, Cd, Pb, Mn and Co. We report that preservation of dissolved size partitioning is possible only for Fe and only under frozen conditions. All metals except Mn and Cd show regeneration of the colloidal phase following its removal in as short as 14 h, validating the importance of prompt ultrafiltration. Adsorption of metals to bottle walls is a well-known sampling artifact often cited for Fe and assumed to be potentially significant for other metals as well. However, only Fe and Co showed significant proclivity to adsorption onto low density polyethylene bottle walls, sorbing a maximum of 91 and 72% over 40 months, respectively. After 20 weeks of acidification neither Fe nor Co desorbed to their original concentrations, leading to an acidified storage recommendation of 30 weeks prior to analyses following storage of unacidified samples for long periods of time. This study provides empirical recommendations for colloidal and dissolved trace metal methodology while also paving the way for much-needed future methods testing.

Introduction

Colloids represent a dynamic class of compounds that exert control on the fate (transport, reactivity, and bioavailability) of trace metals in seawater. A colloid is distinguished from a truly dissolved species (here called “soluble” species) based on its size, where the transition from soluble to colloidal compounds theoretically occurs when the internal characteristics of a compound become significantly differentiated from the solution such that an interface is established (Wells, 2002). This surficial interface is critical for the adsorption of trace metals, and given the larger relative surface area of colloids compared to their particulate analogs, colloids can serve as an important adsorptive sink for metals from the dissolved phase. Moreover, given the physical inclination of colloids to aggregate (Honeyman and Santschi, 1988), adsorption followed by colloidal aggregation can serve as an important output vector of metals from the ocean when they ultimately sink as particles to the sediments. Thus, colloids play an important role in marine biogeochemistry as an intermediary in the continuum of size fractions in seawater and, as aggregators, as a part of the “scavenging” removal of elements from the ocean.

Additionally, the reactivity and bioavailability of trace metals in the ocean is closely tied to their physicochemical speciation. For iron (Fe), which is the best studied of the micronutrient metals, species in the smallest soluble size fraction (typically <3 nm) have been established as the most bioavailable to marine phytoplankton, although larger Fe colloids (3–200 nm) are also thought to be bioavailable, depending on their chemical composition (Chen and Wang, 2001; Chen et al., 2003; Hassler et al., 2011). The colloidal distributions of other metals have been investigated in some earlier studies but remain poorly constrained in open ocean regions, with significant variability dependent on ultrafiltration method used and region studied (Buesseler et al., 1996; Wen et al., 1999; Doucet et al., 2007). However, it remains clear that the size and chemical composition of colloids play important roles in the bioavailability and reactivity of dissolved trace metals in seawater.

Over the last few decades, several isolation methods have been applied to the sampling of marine colloids: solid-phase extraction (Louchouarn et al., 2000), flow field-flow fractionation (FFFF) (Stolpe et al., 2010; Baalousha et al., 2011), chromatography (Minor et al., 2002), gel filtration and stirred-cell ultrafiltration (Guo and Santschi, 2007), as well as more recent methods such as Vivaspin centrifuge ultrafilters (Schlosser et al., 2013). However, ultrafiltration methods such as Anopore filtration (0.02 μm pore size cutoff) and the widely used cross flow filtration (1–1000 kDa) are now among the more common methods for metal colloid studies as they allow large quantities of water to be filtered at once over a large range of sizes (Buesseler et al., 1996; Guo and Santschi, 2007; Fitzsimmons and Boyle, 2014b). These have led to a multitude of global ocean investigations of the size partitioning of dissolved Fe (synthesized in (Fitzsimmons and Boyle, 2014a; von der Heyden and Roychoudhury, 2015)), which have shown that the size partitioning varies spatially and is critically dependent on which pore size cutoff is used, since the size distribution of Fe colloids is itself dynamic (Stolpe et al., 2010; Baalousha et al., 2011). Unfortunately, few studies have compared colloidal metal concentrations across a range of ultrafiltration pore sizes in seawater at the same sites (Larsson et al., 2002; Ingri et al., 2004; Fitzsimmons and Boyle, 2014a), likely because of the time consuming nature of ultrafiltration itself. This would help reveal the true size distribution of marine colloidal metals.

As is true for any operationally-defined method, it is important to rigorously calibrate the method and test for potential artifacts so that different users can expect to acquire the same results. For colloid ultrafiltration methods, methodological artifacts can arise after initial particulate removal via bulk filtration (>0.2 μm pore size), which is required since particles can clog ultrafilters and/or serve as sorptive surfaces, both before and during the slow ultrafiltration process. Some of these artifacts were reviewed during an intercomparison exercise early in the development of colloidal methods (Buesseler et al., 1996) as well as during later studies (Chen et al., 2004; Schlosser et al., 2013; Fitzsimmons and Boyle, 2014b). For metals, the most significant concerns addressed were contamination and recovery differences across ultrafilters (Reitmeyer et al., 1996). While a subset of these ultrafilters have been vetted and are in regular use today (Larsson et al., 2002; Fitzsimmons and Boyle, 2014a; Fitzsimmons et al., 2015b), potential artifacts still exist and warrant further constraint, including the underexplored kinetics of colloidal exchange with other solution components and/or with bottle walls, which are especially critical when it can take tens of minutes to many hours to ultrafilter a single seawater sample.

Previous studies have demonstrated that sorption and desorption from colloidal complexes is highly dependent on pH, ionic strength of the solution, temperature, and particularly the composition of particles, leading to exchange on the timescale of hours to days (Scheinost et al., 2001; Roberts et al., 2003). Marine colloids are thought to be primarily organic in nature, comprising a combination of humic-type substances and products of biological activity such as exopolymeric substances (Gaffney et al., 1996; Ron and Rosenberg, 2001) and, more generically, dissolved organic matter (DOM) (Santschi, 2018). Dissolved organic carbon (DOC) in particular is largely colloidal in size and, as such, is subject to rapid turnover, forming an important part of the global carbon cycle (Baskaran et al., 1992; Benner et al., 1992; Santschi et al., 1995; Guo and Santschi, 1997; Opsahl and Benner, 1997; Wen et al., 1997). If metals are bound to or otherwise incorporated within these organic colloids, they would be more vulnerable to exchange over short timescales.

Sorption of trace metals to sub-sampling bottle walls during storage also represents a potential artifact affecting measurements in seawater. The loss of various trace metals to bottle walls during storage is well documented (Massee et al., 1981), references therein), leading to the standardization of storage procedures that typically involve acidification of samples prior to analysis (Pellenbarg and Church, 1978; Bruland and Rue, 2001; Lohan et al., 2005). Furthermore, the bottle material and physicochemical speciation of each metal are known to affect the overall bottle wall sorption loss (Starik et al., 1963; James and Healy, 1972a; Pellenbarg and Church, 1978; Fischer et al., 2007). While previous studies have elucidated losses of dissolved, soluble, and colloidal Fe to bottle wall sorption (Schlosser et al., 2011; Fitzsimmons and Boyle, 2012), no recent rigorous kinetic testing has been performed to examine adsorptive loss of other trace metals to bottle walls. Moreover, trace metals such as manganese (Mn), cobalt (Co), nickel (Ni), copper (Cu), zinc (Zn), and cadmium (Cd) each have different inorganic speciation in seawater compared with Fe (Byrne, 2002), as well as varying degrees of complexation to ligands (Bruland et al., 2013). Thus, we might expect that adsorption and subsequent desorption of these metals would be different from Fe and dependent on their speciation at seawater pH, leading to important considerations of the timing of sample acidification and analysis, and for colloids, the slow ultrafiltration process. Naturally, these concerns raise questions of whether seawater samples can be preserved for ultrafiltration back in the lab, when more time and supplies would allow for better streamlining of the ultrafiltration process.

Here, we report the results of four experiments designed to answer questions related to the kinetics of colloidal metal exchange and adsorption to bottle walls, as well as the efficacy of preserving the colloidal size partitioning of metal samples: 1) a “Colloid Preservation Experiment” to assess the ability to preserve multi-metal marine colloidal size partitioning under room temperature and frozen conditions over several weeks, 2) a “Colloid Exchangeability Experiment” to assess the timescale of colloid re-formation rates from the soluble phase at room temperature over several days, 3) a “Bottle Adsorption Kinetics Experiment” to assess the timescale of adsorption onto low density polyethylene (LDPE) bottle walls using two different bottle volumes, and 4) a “Bottle Desorption Kinetics Experiment” to assess the timescale of desorption from the same LDPE bottle walls. We analyzed these experimental treatments for their Fe concentrations but also, for the first time, extended our measurements to the size partitioning of Mn, Co, Ni, Cu, Zn, Cd, and lead (Pb). The goal of these experiments was to provide recommendations regarding sample storage prior to acidification, constraints on how long samples can be stored before ultrafiltration, and the influence of the ultrafiltration processing time itself, in order to avoid artifacts during future field studies.

Section snippets

Sample collection

Water used for these experiments was collected on three research cruises: 1) in September 2015 during the U.S. Arctic GEOTRACES GN01 cruise (Station 52: 77.50°N, 148.01°W, 100 m depth) aboard the USCGC Healy for the Bottle Adsorption/Desorption Kinetics Experiments, 2) in January 2016 in the Southern Ocean during the Palmer Long-Term Ecosystem Research (Pal-LTER) cruise aboard the ARSV Laurence McKinley Gould (Station 600.040: -64.93°N, 64.40°W, 1320 m depth for the Colloid Preservation

Colloid preservation experiment: Can soluble/colloidal size partitioning be preserved over time by storing seawater frozen or at room temperature?

The goal of this experiment was to assess whether the natural size partitioning of marine metals into soluble and colloidal fractions could be preserved prior to ultrafiltration over a timescale of weeks under room temperature or frozen conditions. This would allow seawater to be collected in the field, stored using an ideal preservation method, and then ultrafiltered back in the laboratory when more time allows for optimal ultrafiltration conditions. This experiment was completed twice: first

Discussion and implications

Based on the experimental results described above, we first make several recommendations for the collection and handling of dissolved and particularly colloidal trace metals samples in seawater. Then, we discuss some implications for the physicochemical speciation of dissolved metals in the ocean.

We recommend that:

  • 1.

    Seawater should only be preserved for Fe size partitioning, not the size partitioning of any of the other metals, and only if stored seawater samples are preserved frozen at −20 °C

Conclusions

The accurate determination of trace metal concentrations and physicochemical speciation in seawater has progressed over the last few decades as clean sampling and storage procedures have been increasingly optimized. However, there are still challenges and unanswered questions regarding filtration and sample processing that warrant further examination, particularly in the context of multi-elemental analyses. For instance, while ultrafiltration to determine the colloidal concentrations of trace

Acknowledgements

We would like to thank the Captain and crew of the USCGC Healy, the ARSV Gould, and the R/V Trident; Dave Kadko, Greg Cutter, Hugh Ducklow, and Robert Sherrell for cruise leadership and sampling opportunities; Gabi Weiss, Simone Moos, Amber Annett, Janelle Steffen, and Nate Lanning for sample collection at sea and processing ashore; and Luz Romero for assistance with ICP-MS analyses. This work was supported by NSF OCE 1434493 and 1713677 to JNF, NSF OCE 1355833 to WML (Healy), a Texas A&M T3

References (82)

  • A.C. Fischer et al.

    On the relevance of iron adsorption to container materials in small-volume experiments on iron marine chemistry: 55Fe-aided assessment of capacity, affinity and kinetics

    Mar. Chem.

    (2007)
  • J.N. Fitzsimmons et al.

    Both soluble and colloidal iron phases control dissolved iron variability in the tropical North Atlantic Ocean

    Geochim. Cosmochim. Acta

    (2014)
  • J.N. Fitzsimmons et al.

    Daily to decadal variability of size-fractionated iron and iron-binding ligands at the Hawaii Ocean Time-series Station ALOHA

    Geochim. Cosmochim. Acta

    (2015)
  • J.N. Fitzsimmons et al.

    Partitioning of dissolved iron and iron isotopes into soluble and colloidal phases along the GA03 GEOTRACES North Atlantic transect

    Deep Sea Res. Part II Topical Stud. Oceanogr.

    (2015)
  • L.J.A. Gerringa et al.

    Fe-binding dissolved organic ligands near the Kerguelen archipelago in the Southern Ocean (Indian sector)

    Deep-Sea Res. II Top. Stud. Oceanogr.

    (2008)
  • B.P. von der Heyden et al.

    A review of colloidal iron partitioning and distribution in the open ocean

    Mar. Chem.

    (2015)
  • J. Ingri et al.

    Size distribution of colloidal trace metals and organic carbon during a coastal bloom in the Baltic Sea

    Mar. Chem.

    (2004)
  • R.O. James et al.

    Adsorption of hydrolyzable metal ions at the oxide—water interface. I. co(II) adsorption on SiO2 and TiO2 as model systems

    J. Colloid Interface Sci.

    (1972)
  • R.O. James et al.

    Adsorption of hydrolyzable metal ions at the oxide—water interface. III. A thermodynamic model of adsorption

    J. Colloid Interface Sci.

    (1972)
  • M.E. Lagerström et al.

    Automated on-line flow-injection ICP-MS determination of trace metals (Mn, Fe, Co, Ni, Cu and Zn) in open ocean seawater: application to the GEOTRACES program

    Mar. Chem.

    (2013)
  • M.C. Lohan et al.

    Determination of iron and copper in seawater at pH 1.7 with a new commercially available chelating resin, NTA Superflow

    Anal. Chim. Acta

    (2005)
  • R. Massee et al.

    Losses of silver, arsenic, cadmium, selenium and zinc traces from distilled water and artificial sea-water by sorption on various container surfaces

    Anal. Chim. Acta

    (1981)
  • E. Minor et al.

    Estuarine/marine UDOM as characterized by size-exclusion chromatography and organic mass spectrometry

    Mar. Chem.

    (2002)
  • J.W. Moffett et al.

    Cu complexation by organic ligands in the sub-arctic NW Pacific and Bering Sea

    Deep-Sea Res. I Oceanogr. Res. Pap.

    (2007)
  • V.E. Oldham et al.

    Soluble Mn(III)–L complexes are abundant in oxygenated waters and stabilized by humic ligands

    Geochim. Cosmochim. Acta

    (2017)
  • R.E. Pellenbarg et al.

    Storage and processing of estuarine water samples for trace metal analysis by atomic absorption spectrometry

    Anal. Chim. Acta

    (1978)
  • R.T. Powell et al.

    Colloidal trace metals, organic carbon and nitrogen in a southeastern US estuary

    Mar. Chem.

    (1996)
  • R. Reitmeyer et al.

    Colloidal aluminum and iron in seawater: an intercomparison between various cross-flow ultrafiltration systems

    Mar. Chem.

    (1996)
  • D.R. Roberts et al.

    Kinetics and mechanisms of Zn complexation on metal oxides using EXAFS spectroscopy

    J. Colloid Interface Sci.

    (2003)
  • H.B. Ross

    The importance of reducing sample contamination in routine monitoring of trace metals in atmospheric precipitation

    Atmos. Environ.

    (1986)
  • E.L. Rue et al.

    Complexation of iron (III) by natural organic ligands in the central North Pacific as determined by a new competitive ligand equilibration/adsorptive cathodic stripping voltammetric method

    Mar. Chem.

    (1995)
  • P.H. Santschi

    Marine colloids, agents of the self-cleansing capacity of aquatic systems: historical perspective and new discoveries

    Mar. Chem.

    (2018)
  • P.H. Santschi et al.

    Isotopic evidence for the contemporary origin of high-molecular weight organic matter in oceanic environments

    Geochim. Cosmochim. Acta

    (1995)
  • C. Schlosser et al.

    Effects of iron surface adsorption and sample handling on iron solubility measurements

    Mar. Chem.

    (2011)
  • B. Stolpe et al.

    Size and composition of colloidal organic matter and trace elements in the Mississippi River, Pearl River and the northern Gulf of Mexico, as characterized by flow field-flow fractionation

    Mar. Chem.

    (2010)
  • C.-E. Thuróczy et al.

    Key role of organic complexation of iron in sustaining phytoplankton blooms in the Pine Island and Amundsen polynyas (Southern Ocean)

    Deep-Sea Res. II Top. Stud. Oceanogr.

    (2012)
  • C. Van den Berg et al.

    Determination of interactions of nickel with dissolved organic material in seawater using cathodic stripping voltammetry

    Sci. Total Environ.

    (1987)
  • L.-S. Wen et al.

    An ultraclean cross-flow ultrafiltration technique for the study of trace metal phase speciation in seawater

    Mar. Chem.

    (1996)
  • L.-S. Wen et al.

    Interactions between radioactively labeled colloids and natural particles: evidence for colloidal pumping

    Geochim. Cosmochim. Acta

    (1997)
  • L.-S. Wen et al.

    Estuarine trace metal distributions in Galveston Bay: importance of colloidal forms in the speciation of the dissolved phase

    Mar. Chem.

    (1999)
  • L.-S. Wen et al.

    The role of organic carbon, iron, and aluminium oxyhydroxides as trace metal carriers: comparison between the Trinity River and the Trinity River estuary (Galveston Bay, Texas)

    Mar. Chem.

    (2008)
  • Cited by (27)

    • A simulation from offsite disturbance experiments on the metal resuspension process in the seafloor of the Western Pacific

      2023, Chemosphere
      Citation Excerpt :

      Based on the clay XRD results derived herein, the significant differences in the compositions of illite and kaolinite at the two stations may have different impacts on the behaviors of metals, such as Pb, Zn, and Cd, after the disturbance. It is commonly believed that colloids play an important role in marine biogeochemistry as intermediaries in the continuum of size fractions in seawater and as aggregators that partly “scavenge” the removal of elements from the ocean (Jensen et al., 2020). The concentration of large colloids with 0.4 ∼ 1-μm particle sizes in oceans was found to be 106∼107/mL, while the concentration ratios of smaller colloids (<0.2 μm) were higher than 109/mL (Wells and Goldberg, 1991, 1992).

    • Fine particles and pyrogenic carbon fractions regulate PAH partitioning and burial in a eutrophic shallow lake

      2022, Environmental Pollution
      Citation Excerpt :

      The dominance of the dissolved phase here was consistent with the results of most previous studies (Ya et al., 2017a, 2017b). The dissolved phase defined here also contained ultrafine particles or colloids <0.7 μm in diameter with a strong capacity to adsorb organic contaminants (Jensen et al., 2020). The significantly higher proportion of particles results in a higher concentration of PAHs (ng L−1) at a particle size of 2.7–10 μm than others.

    View all citing articles on Scopus
    View full text