Fast and precise method for HPLC–size exclusion chromatography with UV and TOC (NDIR) detection: Importance of multiple detectors to evaluate the characteristics of dissolved organic matter
Graphical abstract
Highlights
► A fast and precise HPLC–SEC system with both UVA and TOC detectors was developed. ► The system achieved shorter analytical time and lower sample volume. ► HMW-DOM with low UVA and LMW-DOM comprised of DOM from Lake Kasumigaura. ► LMW-DOM could be provided from sediment pore water.
Introduction
Dissolved organic matter (DOM) in natural aquatic systems is one of the largest active carbon reservoirs along with land plants and atmospheric carbon dioxide (Hedges, 1992). Increased attention to global carbon cycles in the last few decades has led to increased interest in DOM in natural aquatic systems because temperature increases and increases in the frequency of severe droughts are likely to increase DOM concentrations steadily in river waters (Worrall et al., 2004). DOM in natural aquatic systems is regarded as a source of organic pollution, as an energy source for the microbe-based aquatic food web, as a factor in the cycling of trace elements, and as an influence on the biological activity of phytoplankton and bacteria (Salonen et al., 1992). DOM affects the color of drinking water and leads to disinfection-byproduct formation (Amy et al., 1987) and membrane fouling during water treatment (Fonseca et al., 2007). Determining DOM sources and the factors regulating its production and degradation is a key to understanding both the carbon cycle and water quality in aquatic environments.
The chemical and physical properties of DOM, such as composition and molecular size, appear to affect its reactivity and degradability. For example, in phytoplankton, more than 80% of organic matter is composed of amino acids, sugars, and lipids, whereas these components make up <10% of deep ocean DOM (Benner, 2002). DOM freshly produced by phytoplankton is easily degraded, whereas DOM collected in the deep ocean is highly refractory (Barber, 1968, Hansell and Carlson, 1998). The molecular size of DOM is also an important determinant of DOM reactivity. In the last decade or so, our understanding of the size and reactivity of DOM has been completely revised (Amon and Benner, 1996). For example, small DOM used to be considered reactive, and large DOM to be refractory, but larger molecules are currently regarded as much more reactive than smaller molecules. In addition, owing to advancements in analytical techniques, we now know that aquatic humic substances, once thought to be over 100,000 Da, are actually smaller than 1000 Da (Chin et al., 1994).
The most widely used method for estimating the molecular size of DOM is size exclusion chromatography (SEC) with ultraviolet absorbance (UVA) detection. This method is relatively easy to determine the size of DOM, but since it only relies on UV absorbance, obtained results only show the qualitative properties of DOM because different chemical structures in organic carbon give different responses. Low-molecular-weight organic acids, which are recognized as a main cause of membrane fouling (Speth et al., 1998), may not be detected by this system. Therefore, quantitative analyses of DOC are important to thoroughly understand the molecular size distribution of DOM.
Huber and Frimmel, 1991, Huber and Frimmel, 1996 developed a system that combines high-performance size exclusion chromatography with UV absorbance (HPSEC–UVA) detection with a highly sensitive online organic carbon detector. Because this system allows detection of all organic carbon, the size of DOM can be estimated both qualitatively and quantitatively. Her et al., 2002, Her et al., 2003 further improved the HPSEC–UVA system with organic carbon detection. Now, the system has been developed with a combination of different carbon detectors. Besides the use of NDIR (Speth et al., 1998, Allpike et al., 2005) and conductivity (Allpike et al., 2007), Warton et al. (2008) designed to use a mass spectrometer for quantitative organic carbon detection, which would enable to measure carbon isotopes.
Chin et al. (1994) first examined the combination of columns and mobile phases. They concluded that a mobile phase with ionic strength equivalent to 0.1 M NaCl and a pH 6.8 and a silica column would give the most optimal condition that the apparent molecular size by PSS was the closest to that of humic substance. However, for HPLC–UVA with organic carbon detector systems, the use of NaCl would cause the production of chlorine gas, harmful product, during the oxidation processes. Chlorine gas also causes destruction of NDIR detectors. Therefore, the use of NaCl is not suitable for HPLC–UVA with organic carbon detector systems. The use of phosphate buffer appears to be the best choice for mobile phase.
The column used by Chin et al. (1994) was a silica-based column (Waters Protein-Pak 125). The silica-based columns are usually a very powerful SEC column to separate molecules. Therefore, they have been used for HPLC–SEC system with UVA detection for a long time. However, they may release organic matter from the column. The internal structure of the columns is packed with silica beads coated on organic silanol groups. These silanol groups are gradually peeled off from the silica beads, causing the contaminants for organic carbon detection. Allpike et al. (2007) reported that the carbon concentrations from the silica-based columns such as Tosoh TSKgel G3000SWXL were over the limit of their detector.
The recent HPLC–SEC systems with UVA and organic carbon detection comprise of a preparatory column packed with Toyo Pearl resin as well as phosphate buffer (Huber and Frimmel, 1991) and numerous publications. The Toyo Pearl resin is known as clean resin with almost no carbon released. However, each Toyo Pearl resin is relatively large (20–40 μm) compared to silica beads (∼5 μm) Therefore, the current system requires a relatively large sample volume (2 mL) and a long analytical time (>1 h). Our objectives in the current study were to develop a highly sensitive HPSEC system with both UV and non-dispersive infrared (NDIR) total organic carbon (TOC) detectors, a relatively low sample volume, and a fast analytical time and to determine possible source(s) of DOM at different sizes.
Section snippets
Sample collections and preparation
Water and sediment samples were collected from Lake Kasumigaura (Fig. 1), a shallow (maximum depth = 7 m) and highly eutrophic lake in Japan, during the monthly sampling carried out as part of the GEMS/Water Trend Monitoring Program of the National Institute for Environmental Studies (NIES) (GEMS/Water Website). Water samples were collected in a 1-L precombusted glass bottle (450 °C, 4 h) with a 2-m column sampler on July 17, 2007. The samples were immediately cooled in an ice cooler. Sediment cores
New SEC system
The purpose of establishing this SEC system was to provide the fast and precise analyses of both UVA and TOC detection. As pointed out by Allpike et al. (2007), silica-based columns including Tosoh TSKgel G3000SWXL release organic carbon into the mobile phase. The silica-based columns are packed with silica beads whose surface is coated by organic compounds such as silanol groups. These compounds interact with water, causing breakup from the silica beads and are released to the mobile phase.
Conclusion
A new type of size exclusion chromatography system with both UVA and total organic carbon (TOC) detectors has been developed to provide important information about molecular size distributions of dissolved organic matter qualitatively and quantitatively. Our newly developed provided reasonably precise analyses of size distribution of DOM with a relatively short time (∼35 min) and low sample volume (100 μL) as compared to other similar systems, which usually need a long analytical time (>90 min)
Acknowledgments
This work was supported by Environmental Technology Research Fund from Ministry of the Environment, Government of Japan (2006–2007) and by Grant-in-Aid for Scientific Research (No. 21241008) from the Japan Society for the Promotion of Science. Sampling was supported by the GEMS/Water Trend Monitoring Project at Lake Kasumigaura. We thank the members of the project for their cooperation.
References (31)
- et al.
A new organic carbon detector for size exclusion chromatography
Journal of Chromatography A
(2007) Chemical composition and reactivity
- et al.
Anaerobic carbon transformation: experimental studies with flow-through cells
Marine Chemistry
(2003) - et al.
A high-temperature catalytic-oxidation method for the determination of marine dissolved organic-carbon and its comparison with the UV photooxidation method
Marine Chemistry
(1993) - et al.
The abundance, distribution, and configuration of porewater organic colloids in recent sediments
Geochimica et Cosmochimica Acta
(1991) - et al.
High pressure size exclusion chromatography (HPSEC) of humic substances: molecular sizes, analytical parameters, and column performance
Chemosphere
(1999) Global biogeochemical cycles – progress and problems
Marine Chemistry
(1992)- et al.
Characterization of DOM as a function of MW by fluorescence EEM and HPLC–SEC using UVA, DOC, and fluorescence detection
Water Research
(2003) - et al.
Fractionation and characterization of dissolved organic matter in a shallow eutrophic lake, its inflowing rivers, and other organic matter sources
Water Research
(2001) - et al.
Size-exclusion chromatography with organic carbon detection using a mass spectrometer
Journal of Chromatography A
(2008)
Size exclusion chromatography to characterize DOC removal in drinking water treatment
Environmental Science and Technology
Bacterial utilization of different size classes of dissolved organic matter
Limnology and Oceanography
Comparing gel-permeation chromatography and ultrafiltration for the molecular-weight characterization of aquatic organic-matter
Journal of American Water Works Association
Dissolved organic carbon from deep waters resists microbial oxidation
Nature
Analysis and characterization of natural organic matters in freshwaters. I. Study of analytical techniques
Schweizerische Zeitschrift fur Hydrologie-Swiss Journal of Hydrology
Cited by (52)
Contribution of sedimentary organic matter to arsenic mobilization along a potential natural reactive barrier (NRB) near a river: The Meghna river, Bangladesh
2022, ChemosphereCitation Excerpt :The fluorescence spectroscopy data from these wavelengths was used to generate an excitation-emission matrix (EEM) for each sample. Size exclusion chromatography (SEC) using a high-performance liquid chromatography (HPLC) system (Agilent 160 Infinity system II, USA) with variable wavelength (VWD) at 254 nm was used to determine the molecular weight distribution of water-extractable SOM at Auburn University, following similar methods described elsewhere (Kawasaki et al., 2011). Details of these methods are described in the Supplementary Text.
Sorption and partitioning of perfluorooctanoic acid (PFOA) and perfluorooctane sulfonate (PFOS) onto sediments of Diep and Plankenburg river systems Western Cape, South Africa
2022, Environmental Technology and InnovationCharacterization of natural organic matter
2022, Natural Organic Matter in Water: Characterization, Treatment Methods, and Climate Change Impact, Second Edition