Elsevier

Marine Chemistry

Volume 217, 20 November 2019, 103719
Marine Chemistry

Electrodialysis as a sample processing tool for bulk organic matter and target pollutant analysis of seawater

https://doi.org/10.1016/j.marchem.2019.103719Get rights and content

Highlights

  • Natural organic matter and target pollutants are retained during electrodialysis.

  • Molecule polarity, size, charge and wall affinity influence the ED recovery.

  • A refined supplied current and an elevated final salinity improve ED recoveries.

  • During subsequent mass spectrometry, matrix-dependent ion suppression was reduced.

Abstract

Electrodialysis (ED) is an advancing seawater sample processing tool that enables the separation of analytes from the often interfering salt matrix. In this study, we present the evaluation of a laboratory scale ED system for both dissolved organic matter (DOM) and target pollutant analysis of seawater.

The developed sample processing protocol yields reproducible data and was found to be robust towards moderate changes in sample composition. At the final salinity of 0.1, the average recovery of DOM in the form of dissolved organic carbon, nitrogen and phosphorus (DOC, DON and DOP) was 44, 53 and 89%, respectively. DOM loss occurred mainly in the late stage of the ED process.

When investigating specific ED processing parameters, it was discovered that the initial sample salinity does not influence DOM recovery. The final salinity, by contrast, is a dominant influence factor on DOM recovery. Furthermore, DOC and DOP recoveries could be improved by 8% by refining the electrical current in the ED cell. Surprisingly, adjustments of the sample pH did not lead to any improvements in DOM recovery.

The experiments with target analytes showed that the recovery of individual molecules is determined by their n-octanol water partition coefficients logKow. High recoveries > 80% were achieved for compounds with medium logKow of −1 to 3. Hydrophobic compounds with logKow > 3 were lost through surface adsorption to the system walls and tubing. Small, polar and charged compounds with logKow < −1 are prone to loss via ED membrane passage, which occurred predominantly in the late stage of the ED process. Consequently, sample processing with ED was deemed beneficial for the LC-MS or GC–MS analysis of polar target compounds, because they are often difficult to enrich from seawater. Furthermore, during LC-MS or GC–MS analyses, matrix-dependent ion suppression was reduced in ED isolates, giving rise to increased signal responses of 25 to 620%, which resulted in improved instrumental sensitivity.

Introduction

Dissolved organic compounds in the marine environment constitute a complex pool of matter that is continuously influenced and transformed through terrestrial and atmospheric interactions as well as biogeochemical cycling (Moran et al., 2016; Ridgwell and Arndt, 2015). In the past decades, there have been numerous versatile efforts to characterize this dissolved organic matter (DOM) pool (Minor et al., 2014; Mopper et al., 2007; Nebbioso and Piccolo, 2013; Repeta, 2015; Seidel et al., 2017), since it is vital for our understanding of dynamics and drivers of biogeochemical processes and their interrelationships on different spatial and temporal scales. Moreover, in the era of rapidly increasing world population and industrialization, research on the anthropogenic influence on marine DOM brought on by climate change and pollution becomes increasingly important (Porcal et al., 2009; Remucal, 2014; Wenk et al., 2011; Zhuang and Yang, 2018).

DOM can be characterized trough a multitude of different approaches that range from comprehensive to selective techniques. An overview of the DOM characteristics can be gained by analyzing the bulk elemental composition in the form of dissolved organic carbon (DOC), nitrogen (DON) and phosphorus (DOP) (Hansell et al., 2009; Letscher et al., 2013; McGill, 1964) or carbon and nitrogen isotope signatures (Sanderman et al., 2009). The molecular characterization of DOM can be carried out by investigating functional group abundances, e.g. with nuclear magnetic resonance (NMR) or Fourier transform infrared- (FTIR) spectroscopy (Abdulla et al., 2010; Helms et al., 2015), compound class distributions, e.g. with high resolution mass spectrometry (HRMS) (Chen et al., 2014), or the concentrations of single target compounds (amino acids, sugars, lipids etc.) (Kaiser and Benner, 2012; Meyers-Schulte and Hedges, 1986; Panagiotopoulos et al., 2013; Wakeham et al., 2003; Yamashita and Tanoue, 2003). Traditionally, compounds of anthropogenic origin, e.g. pollutants, are not considered during DOM analysis; however, they have become part of the natural DOM pool and are usually analyzed with the same target methodological approaches as natural compounds.

To date, the bulk elemental composition of DOM (DOC, DON and DOP) has been subject to several studies carried out across the globe (Hansell et al., 2009; Hansell et al., 2012; Karl and Björkman, 2015; Letscher et al., 2013; McGill, 1964; Sipler and Bronk, 2015). Molecular characterization of DOM has been less comprehensive, which is mainly owed to its complexity and the large number of individual compounds within the DOM pool (Hertkorn et al., 2008). Nonetheless, as a result of promising analytical advances, for example in the fields of HRMS (Helms et al., 2015; Seidel et al., 2017), NMR- (Mao et al., 2012; Young and Ingall, 2010), FTIR (Abdulla et al., 2010; Sleighter et al., 2008) or UV/VIS (Helms et al., 2008; Macdonald and Minor, 2013) spectroscopy, it was recently estimated that 60–70% of DOM has undergone some form of structural characterization (Repeta, 2015).

When aiming for a molecular characterization of DOM, researchers are faced with the task of isolating and concentrating DOM, to make it accessible to the respective analysis techniques. The separation from the salt matrix present in seawater usually poses the major challenge, since salt concentrations outweigh DOM concentrations by several orders of magnitude (Minor et al., 2014; Repeta, 2015). For spectroscopic or spectrometric analysis, salt removal is necessary, because relatively pure DOM isolates are required for a successful analysis. Several different techniques for DOM isolation have been introduced; however, they usually induce an isolation bias (Repeta, 2015). Solid phase extraction (SPE) recovers all components of DOM that can be adsorbed to the employed sorbent. This usually is the hydrophobic fraction of DOM (Dittmar et al., 2008; Li et al., 2017; Wünsch et al., 2018). Common sorbents are based on octadecyl-bonded silica (C-18), cross-linked polystyrene (XAD) or styrene-divinylbenzene (PPL) (Mopper et al., 2007). Ultrafiltration recovers all DOM components larger than the pore size of the utilized membrane. Therefore, DOM isolates obtained via ultrafiltration usually contain the high molecular weight fraction of DOM (Panagiotopoulos et al., 2013; Walker et al., 2011). However, some low molecular weight compounds can also be recovered through adsorption to the membrane, which is commonly made of cellulose or polysulone (Repeta, 2015).

A relatively novel approach for the isolation of DOM is the use of electrodialysis (ED) (Gurtler et al., 2008; Sillanpää and Shestakova, 2017; Vetter et al., 2007). An ED cell consists of alternatingly positioned anion- and cation exchange membranes between a set of electrodes. The sample (diluate) and a receiving solution (concentrate) are circulated through the alternating interspaces between the membranes. An electric field is applied perpendicular to the membrane surfaces, which causes salt ions in the diluate to be transported through the membranes and into the concentrate. A third solution (electrode rinse) is circulated along the electrodes to carry off oxygen and hydrogen gas formed due to water splitting at the electrodes.

ED is commonly used in drinking water production and industrial processes (Sillanpää and Shestakova, 2017), but it has only found limited application in environmental sciences, so far. ED was first used for the isolation of DOM coupled to a reverse osmosis (RO) unit (Bell et al., 2017; Gurtler et al., 2008; Vetter et al., 2007). With this technique, DOM can be simultaneously isolated and concentrated with recoveries ranging from 40 up to 95% (Chambers et al., 2016). ED/RO isolates of DOM were shown to usually be representative of the source material and highly suitable for analysis with NMR or HRMS techniques (Bell et al., 2017; Chen et al., 2014; Koprivnjak et al., 2009; Young and Ingall, 2010). This highlights that the isolation bias of ED/RO is greatly reduced in comparison to SPE or ultrafiltration techniques. Nonetheless, both the loss of hydrophobic DOM compounds due adsorption to membranes or system walls and the loss of hydrophilic compounds due to membrane passage at lowered salt concentrations were shown to occur during ED/RO (Bell et al., 2017; Chambers et al., 2016; Koprivnjak et al., 2009; Vetter et al., 2007). Chambers et al. (2016) showed that ED can also be used on a smaller scale without an RO unit. They concentrated DOM from ED-processed samples with lyophilization.

In addition to its benefits in bulk analysis, we hypothesized that ED could also be useful for target analysis of compounds in the marine environment. Such compounds are often analyzed with liquid or gas chromatography coupled to tandem mass spectrometry (LC-MS/MS or GC–MS/MS) (Bjorklund et al., 2016; Habedank et al., 2017; Koponen et al., 2015; Skeff et al., 2015). However, since the concentration of many target analytes in seawater lies well below instrumental limits of detection (LOD), analyte enrichment is required. For polar compounds, SPE is commonly used in this step (Fisch et al., 2017; Hanke et al., 2008; Svahn and Björklund, 2019). However, the salt matrix in seawater can also significantly hinder the SPE enrichment of polar target analytes due to competition of salt and analytes for binding sites on the solid phase (Corbera et al., 2005; Lohrer et al., 2019).

The aim of the current study was to evaluate a newly set up laboratory scale electrodialysis system as a pre-processing tool for bulk and target analysis of seawater. The removal of the salt matrix in the form of many major seawater ions was examined. Moreover, the recovery of DOM in the form of DOC, DON and DOP was investigated. The influence of initial salinity, supplied current and sample pH on recoveries was tested. Finally, the retention of many different target compounds, mainly pesticides, covering a broad range of logKow values was investigated in order to assess the suitability of ED for coupling to target analyses with LC-MS/MS and GC–MS/MS.

Section snippets

Chemicals

Purified water (18.2 MΩ) was obtained through a MilliQ IQ 7000 system from Merck Millipore (Schwalbach, Germany). Artificial sea salt (Tropic Marin Sea Salt Classic) was purchased from Tropic Marin AG (Hünenberg, Switzerland). Sodium sulphate and sodium hydroxide were obtained from Merck (Darmstadt, Germany). 5 M NaOH was prepared with MilliQ water. Hydrochloric acid (Normapur, 32% and Titrinorm, 5 M) was purchased from VWR (Darmstadt, Germany). The 32% HCl was used to prepare pH 2 HCl with

Removal of the inorganic seawater matrix

For a comprehensive understanding of the utilized electrodialysis system, the removal of the inorganic seawater matrix was analyzed. For this, ED was conducted and subsamples were analyzed for a selection of inorganic seawater ions (n = 4, Fig. 1, Tables S1, S6).

The initial concentrations of the analyzed matrix components showed variation below 20%, except for phosphate, whose concentration in surface water varied between 0.03 and 0.43 μmol/L depending on the season (Table S2).

At the final

Analyte recovery during ED – loss mechanisms and influence factors

In general, different mechanisms through which the loss of organic and inorganic compounds from a sample in an electrodialysis system occurs can be described. On the one hand, compounds can transfer through the ion exchange membranes and into the concentrate solution; on the other hand, compounds can adsorb to membrane surfaces and system walls. In the first case, the respective molecules/ions (1) have to be transported into the boundary layer at the membrane surface and (2) need to be able to

Conclusion

Within this work, the drivers of compound loss during the electrodialysis of seawater were studied through a combination of bulk and target analysis of ED isolates. In addition, the desalination process itself was investigated through the analysis of matrix components. It was found that small, charged compounds and highly hydrophobic compounds are both prone to loss from the samples via membrane passage and adsorption to system walls, respectively. Recoveries of bulk and target compounds can be

Acknowledgements

This work was supported by the funding line strategic networks of the Leibniz Association within the scope of the Leibniz ScienceCampus Phosphorus Research Rostock. The authors further appreciate funding for instrumentation from the German Federal Ministry of Education and Research through the project PROSO (MARE: N 0307779A).

The authors would like to thank the members of the IOW marine chemistry section, namely Jenny Jeschek, Birgit Sadkowiak, Christoph Kamper, Lisa Rönspieß and Malte

References (62)

  • J.F. Koprivnjak et al.

    Chemical and spectroscopic characterization of marine dissolved organic matter isolated using coupled reverse osmosis–electrodialysis

    Geochim. Cosmochim. Acta

    (2009)
  • H.-J. Lee et al.

    Determination of the limiting current density in electrodialysis desalination as an empirical function of linear velocity

    Desalination

    (2006)
  • Y. Li et al.

    How representative are dissolved organic matter (DOM) extracts? A comprehensive study of sorbent selectivity for DOM isolation

    Water Res.

    (2017)
  • D.A. McGill

    The distribution of phosphorus and oxygen in the Atlantic Ocean as observed during the I.G.Y., 1957–1958

    Prog. Oceanogr.

    (1964)
  • C. Panagiotopoulos et al.

    Molecular level characterization of methyl sugars in marine high molecular weight dissolved organic matter

    Mar. Chem.

    (2013)
  • D.J. Repeta

    Chemical characterization and cycling of dissolved organic matter

  • A. Ridgwell et al.

    Why dissolved organics matter

  • R.E. Sipler et al.

    Dynamics of dissolved organic nitrogen

  • W. Skeff et al.

    Glyphosate and AMPA in the estuaries of the Baltic Sea method optimization and field study

    Mar. Pollut. Bull.

    (2015)
  • T. Vetter et al.

    Combining reverse osmosis and electrodialysis for more complete recovery of dissolved organic matter from seawater

    Separ. Purif. Technol.

    (2007)
  • S.G. Wakeham et al.

    Hydroxy fatty acids in marine dissolved organic matter as indicators of bacterial membrane material

    Org. Geochem.

    (2003)
  • B.D. Walker et al.

    Large-volume ultrafiltration for the study of radiocarbon signatures and size vs. age relationships in marine dissolved organic matter

    Geochim. Cosmochim. Acta

    (2011)
  • U.J. Wünsch et al.

    Quantifying the impact of solid-phase extraction on chromophoric dissolved organic matter composition

    Mar. Chem.

    (2018)
  • Y. Yamashita et al.

    Distribution and alteration of amino acids in bulk DOM along a transect from bay to oceanic waters

    Mar. Chem.

    (2003)
  • Y. Zhang et al.

    Separation of nutrient ions and organic compounds from salts in RO concentrates by standard and monovalent selective ion-exchange membranes used in electrodialysis

    J. Membrane Sci.

    (2009)
  • H.A.N. Abdulla et al.

    Using two-dimensional correlations of 13C NMR and FTIR to investigate changes in the chemical composition of dissolved organic matter along an estuarine transect

    Environ. Sci. Technol.

    (2010)
  • D.W. Bell et al.

    Isolation and molecular characterization of dissolved organic phosphorus using electrodialysis-reverse osmosis and solution 31P-NMR

    Limnol. Oceanogr. Meth.

    (2017)
  • E. Bjorklund et al.

    Pharmaceutical residues affecting the UNESCO biosphere reserve Kristianstads Vattenrike wetlands: sources and sinks

    Arch. Environ. Contam. Toxicol.

    (2016)
  • L.R. Chambers et al.

    Enhanced dissolved organic matter recovery from saltwater samples with electrodialysis

    Aquat. Geochem.

    (2016)
  • D.A. Cowan et al.

    Effect of turbulece on limiting current in electrodialysis cells

    Ind. Eng. Chem.

    (1959)
  • T. Dittmar et al.

    A simple and efficient method for the solid-phase extraction of dissolved organic matter (SPE-DOM) from seawater

    Limnol. Oceanogr. Meth.

    (2008)
  • Cited by (8)

    • The challenge of detecting the herbicide glyphosate and its metabolite AMPA in seawater – Method development and application in the Baltic Sea

      2021, Chemosphere
      Citation Excerpt :

      For this, a Baltic Sea water sample was collected at the Heiligendamm pier in Mecklenburg-Western Pomerania, Germany (54°08′46.7″ N, 11°50′36.1″ E) in June 2019. The sample was filtered (0.7 μm) and processed with electrodialysis, as described in Wirth et al. (2019). At specific residual salinities, 1 L subsamples were taken from the system.

    • Methodological aspects of methylphosphonic acid analysis: Determination in river and coastal water samples

      2020, Talanta
      Citation Excerpt :

      The water sample collected at station Heiligendamm (HD) was subjected to electrodialysis (ED), in order to reduce their salt content and possibly improve the recovery of the subsequent SPE step. The procedure was conducted with an ED system from Deukum GmbH (Frickenhausen, Germany) as described by Wirth et al. [26]. In an electrodialysis cell, anion- and cation exchange membranes are alternatingly positioned between a set of electrodes.

    View all citing articles on Scopus
    View full text