Elsevier

Water Research

Volume 134, 1 May 2018, Pages 301-310
Water Research

The impact of loading approach and biological activity on NOM removal by ion exchange resins

https://doi.org/10.1016/j.watres.2018.01.052Get rights and content

Highlights

  • Mass transfer conditions are critical when interpreting multiple loading tests.

  • Biological ion exchange (BIEX) achieved stable NOM removal for ∼1yr.

  • The long-term stability of BIEX is affected by the raw water characteristics.

  • If required, BIEX can be regenerated using standard regeneration approaches.

  • BIEX could replace BAC as the technology of choice for NOM removal.

Abstract

The present study investigated the impact of different loading approaches and microbial activity on the Natural Organic Matter (NOM) removal efficiency and capacity of ion exchange resins. Gaining further knowledge on the impact of loading approaches is of relevance because laboratory-scale multiple loading tests (MLTs) have been introduced as a simpler and faster alternative to column tests for predicting the performance of IEX, but only anecdotal evidence exists to support their ability to forecast contaminant removal and runtime until breakthrough of IEX systems. The overall trends observed for the removal and the time to breakthrough of organic material estimated using MLTs differed from those estimated using column tests. The results nonetheless suggest that MLTs could best be used as an effective tool to screen different ion exchange resins in terms of their ability to remove various contaminants of interest from different raw waters.

The microbial activity was also observed to impact the removal and time to breakthrough. In the absence of regeneration, a microbial community rapidly established itself in ion exchange columns and contributed to the removal of organic material. Biological ion exchange (BIEX) removed more organic material and enabled operation beyond the point when the resin capacity would have otherwise been exhausted using conventional (i.e. in the absence of a microbial community) ion exchange. Furthermore, significantly greater removal of organic matter could be achieved with BIEX than biological activated carbon (BAC) (i.e. 56 ± 7% vs. 15 ± 5%, respectively) when operated at similar loading rates. The results suggest that for some raw waters, BIEX could replace BAC as the technology of choice for the removal of organic material.

Introduction

Natural organic matter (NOM) is a complex mixture of various organic substances and is ubiquitous in surface waters (Leenheer and Croué, 2003, Matilainen et al., 2002, Pelekani et al., 1999; Singer, 1999). Its presence is of concern because it can impact the aesthetic quality of the water (e.g. colour, taste, odour), contribute to bacterial regrowth and biofilm formation in distribution systems, lead to the formation of disinfection by-products during treatment, and disrupt several water treatment processes (Leenheer and Croué, 2003, Kleiser and Frimmel, 2000, Singer, 1999, Van der Kooij, 1992).

Macroporous anionic ion exchange resins have received considerable attention as a relatively simple and cost-effective treatment process to remove NOM from surface waters (Bolto et al., 2002, Drikas et al., 2002, Fu and Symons, 1990, Hongve et al., 1999). Bench or pilot-scale column tests have historically been used to determine the removal efficiency of constituents of interest (e.g. NOM), the capacity (i.e. mass of NOM retained prior to breakthrough) and the operating conditions (i.e. hydraulic loading rate) required to meet a treatment objectives with ion exchange resins. Although simple in concept, column tests are relatively complex to set-up and are time-consuming, generally lasting weeks to months. Laboratory-scale Multiple Loading Tests (MLTs) have been introduced as a simpler and faster alternative to column tests (Boyer and Singer, 2006, Galjaard, 2010). In MLTs, new batches of raw water to be treated are repeatedly mixed (i.e. loaded) with a given amount of the ion exchange resin. The tests are simple to conduct and because of the high ratio of water to resin used, the tests can be completed in 2–3 days (Orica, 2007). The use of MLTs to assess the feasibility of ion exchange resins for NOM removal is particularly attractive for small and remote communities that do not have the technical and or financial resources to conduct column testing. Unfortunately, only anecdotal evidence exists to support the use of MLTs as an alternative to column tests to identify conditions required to meet treatment objectives using ion exchange resins. Also, some studies have suggested that removal of NOM during repeated loadings could change depending on the operating conditions (Drikas et al., 2011, Mergen et al., 2008, Walker and Boyer, 2011).

Mass transfer conditions could differ substantially in column tests (i.e. pack beds) and MLTs (i.e. fluidized beds) (Garic-Gruvolic et al., 2011). Differences in mass transfer could affect NOM removal kinetics and therefore the test outcomes. Times to breakthrough may also differ substantially between column tests and MLTs (i.e. weeks vs. days, respectively). In full-scale applications, resins are regenerated relatively frequently to ensure continuous and effective removal of the constituents of interest. However, during column tests, resins are not regenerated until breakthrough occurs. During this time, the resin could provide an environment suitable for microbial attachment and growth due to its macro-porous structure containing a high surface area (Flemming, 1987, Simpson, 2008). The presence of a microbial community could impact NOM removal and therefore the outcomes from column tests (Schulz et al., 2017).

The present study was designed to comprehensively address the current knowledge gap. The first objective was to determine if NOM removal and resin capacity obtained using column tests are similar to those obtained using MLTs. The second objective was to quantify the impact of microbial activity on NOM removal and resin capacity.

Section snippets

Ion exchange resin

Purolite®A860, a strongly basic macroporous anionic ion exchange resin (particle size: 0.3–1.2 mm), was used in all tests for which abiotic and biotic (i.e. with suppression of biological activity) ion exchange resins were considered. This resin has an exchange capacity of 0.8 mEg/mL and was selected because it was previously identified to be effective at removing organic material (i.e. Dissolved Organic Material – DOC) from surface waters (Bazri and Mohseni, 2014).

Biological activated carbon

Wood-base Picabiol® granular

Comparison of column tests and MLTs (abiotic conditions)

A side-by-side comparison of column tests and MLTs was performed to assess the removal efficiency and resin capacity with these test approaches. The tests were performed over a period equivalent to that corresponding to the treatment of approximately 3000 bed volumes, which corresponded to approximately 2 months of operation for the column tests, and 15 consecutive ‘loadings’ for the MLT. To consider the impact of raw water characteristics, test were conducted with JP and SR NOM water. Because

Conclusions

The overall trends for the removal and the time to breakthrough of organic material estimated using MLTs differed from those estimated using column tests. The differences were attributed to different loading approaches and microbial activity. Similar mass transfer conditions can likely be achieved in packed and fluidized beds; however, further work, beyond the scope of the present research, would be required. The results indicate that MLTs could be used as an effective tool to screen different

Acknowledgements

The authors would like to thank Fuhar Dixit from the University of British Columbia (Vancouver, BC) and Julie Philibert from Polytechnique Montreal for conducting the algal toxins analyses and the disinfection by-product formation potential tests, respectively. The authors also wish to thank Res’Eau WaterNet for providing funding for this project.

References (43)

Cited by (0)

View full text