comparative assessment of leca and spartina maritima to ... ?· to remove emerging organic...

Download Comparative assessment of LECA and Spartina maritima to ... ?· to remove emerging organic contaminants…

Post on 19-Jul-2018




0 download

Embed Size (px)



    Comparative assessment of LECA and Spartina maritimato remove emerging organic contaminants from wastewater

    Ana Rita Ferreira1 & Paula Guedes1 & Eduardo P. Mateus1 & Alexandra B. Ribeiro1 &Nazar Couto1

    Received: 8 June 2016 /Accepted: 12 January 2017 /Published online: 18 January 2017# Springer-Verlag Berlin Heidelberg 2017

    Abstract The present work aimed to evaluate the capacity ofconstructed wetlands (CWs) to remove three emerging organ-ic contaminants with different physicochemical properties:caffeine (CAF), oxybenzone (MBPh), and triclosan (TCS).The simulated CWs were set up with a matrix of light expand-ed clay aggregates (LECA) and planted with Spartinamaritima, a salt marsh plant. Controlled experiments werecarried out in microcosms using deionized water and waste-water collected at a wastewater treatment plant (WWTP), withdifferent contaminant mass ranges, for 3, 7, and 14 days. Theeffects of variables were tested isolatedly and together (LECAand/or S. maritima). The presence of LECA and/or S.maritima has shown higher removal (around 6197%) of li-pophilic compounds (MBPh and TCS) than the hydrophiliccompound (CAF; around 1985%). This was attributed to thefact that hydrophilic compounds are dissolved in the watercolumn, whereas the lipophilic ones suffer sorption processespromoting their removal by plant roots and/or LECA. In thecontrol (only wastewater), a decrease in the three contaminantlevels was observed. Adsorption and bio/rhizoremediation arethe strongest hypothesis to explain the decrease in contami-nants in the tested conditions.

    Keywords Constructed wetlands . Spartinamaritima .

    LECA . Emerging organic contaminants .Wastewater


    Pharmaceuticals and personal care products (PPCPs) are dailydischarged into the aquatic systems due to the general ineffi-ciency of the conventional processes to remove emerging or-ganic contaminants in wastewater treatment plants (WWTP).As a consequence, some PPCPs have been detected in aquaticenvironment at concentrations of about microgram per liter(Heberer 2002; Fent et al. 2006). Even in vestigial concentra-tions, these contaminants are of concern due to their potentialecological and environmental impacts (Pal et al. 2010).

    One important aspect to decrease the mass of organic con-taminants discharged into surface waters is the optimization ofWWTP processes. Constructed wetlands (CWs), based onnatural wetland systems, are an attractive technology for theremoval of PPCPs from wastewaters due to their easy opera-tion and maintenance, low-energy requirements, high rates ofwater recycling, and potential for providing significant wild-life habitat (Matamoros and Bayona 2006; Vymazal 2010;Zhang et al. 2014).

    CWs are constituted by support matrix and plant species(Hijosa-Valsero et al. 2010). The substrates of CWs can benatural (e.g., gravel and sand) or an artificially modified nat-ural material, like light expanded clay aggregates (LECA).LECA is considered a low-cost sorbent and exhibits a highsorption capacity for pharmaceutical compounds (removals of4897%) (Dordio et al. 2009, 2010) and herbicide (re-moval of 5697%) (Dordio and Carvalho 2013). Despitethe potential of LECA to remove these lipophilic compounds(Log Kow between 1.37 and 3.97), its capacity is unknown tosimultaneously remove a mixture with lipophilic and

    Responsible editor: Philippe Garrigues

    * Ana Rita

    * Alexandra B.

    1 CENSE, Departamento de Cincias e Engenharia do Ambiente,Faculdade de Cincias e Tecnologia, Universidade Nova de Lisboa,Campus de Caparica, 2829-516 Caparica, Portugal

    Environ Sci Pollut Res (2017) 24:72087215DOI 10.1007/s11356-017-8452-4

  • hydrophilic compounds. Also, there is limited knowledgeabout LECA efficiency in removing a successive arrival ofPPCPs to CWs.

    Different salt marsh plant species, such as Phragmitesaustralis and Typha latifolia, have been used for pharmaceu-tical removal in CWs (Dordio et al. 2010; Carvalho et al.2012). Other plant species may be used for this purpose beinga valuable option to enlarge the range of applications in CWs.Spartina maritima has high tolerance to different salinitylevels (e.g., 0, 171, and 510 mM NaCl) and capacity to reactto climate changes (Adams and Bate 1995; Mateos-Naranjoet al. 2010) being an option to be used inWWTPs. S.maritimacontributes to heavy metal removal (Reboreda and Caador2007; Duarte et al. 2009), but as far as we could ascertain,there is limited information about its potential to removePPCPs.

    The aim of the present work was to evaluate the capacity ofCWs having LECA as a support medium and planted with S.maritima to remove from wastewater a mixture of lipophilicand hydrophilic compounds. The target compounds were caf-feine (CAF), oxybenzone (MBPh), and triclosan (TCS), cho-sen based on their worldwide consumption, physicochemicalproperties, and chemical classes (Table 1). A fed-batch modewas simulated with pulses of contaminants additionthroughout the experiment (controlled experimentalmicrocosm).

    Materials and methods

    Reagents and material

    CAF (90%), MBPh (90%), TCS (Irgasan, 97%) were pur-chased from Sigma-Aldrich (Steinheim, Germany). Methanol(MeOH), acetonitrile (ACN), and acetone (ACE) were HPLCgrade from Riedel-de Han (Germany). All aqueous solutionswere prepared withMilli-Q water. The standard solutions withthe three compounds CAF, MBPh, and TCS were prepared inmethanol (MeOH).

    Light expanded clay aggregates (LECA), which were usedfor the support matrix of the CWs, had a size range between 3and 8 mm and were supplied by ARGEX (Portugal). LECAwas washed several times with Milli-Q water and the liquidphase of the assay (deionized water or wastewater), to reducethe amount of fine materials and embed in the matrix.

    Wastewater collection

    The wastewater samples were collected after the secondarysettling tank in a WWTP from guas de Lisboa e Vale doTejo Group located in Quinta do Conde, Sesimbra, Portugal(38 34 13 N, 9 2 7 W). The WWTP has the capacity totreat in the project horizon 19,300 m3 day1 of urban waste-water, corresponding to about 94,000 equivalent inhabitants.

    Table 1 The physicochemical properties of the studied PPCPs

    Compound CAF MBPh TCS

    Function Central nervous system stimulant

    Sunscreen agent Antiseptic


    Molecular weight(g mol1)

    194.19 228.24 289.55

    Solubility in water(mg L1 at 25 C)

    pKa (24 C)a





    10 (at 20 C)


    Log Kowb

    0.07 3.79 4.76

    References:; www.SigmaAldrich.coma Logarithm of acid dissociation constantb Logarithm of the octanol-water partition coefficient

    Environ Sci Pollut Res (2017) 24:72087215 7209


  • The samples presented a neutral pH (7.01), and the parametersused to characterize wastewater quality like organic matterand suspended solids were evaluated at the WWTP laborato-ries at the time of collection. Parameters were within thePortuguese legal limits for discharge into water bodies (DL236/98). An initial screening of the studied contaminants wascarried out in the wastewater used in the assays, and all werebelow the analytical detection limits (0.55 g L1 of CAF,0.80 g L1of MBPh, and 3.3 g L1 of TCS) (Guedes2015b).

    Plant collection

    The exemplars of S. maritima were collected in Tagus riverestuary (38 36 59.39 N; 9 02 33.41 W). This species isabundant in the intertidal and has a wide geographic distribu-tion in temperate zones. This species belongs to a lower zoneof a salt marsh having characteristics to a partial or total sub-mergence, high soil salinity, and soil anoxia (Mateos-Naranjoet al. 2010).

    S.maritimawas collected in May, June, and October 2014.After collection, plants were immediately transported to thelaboratory in controlled conditions (refrigerated and avoidinglight exposure). The roots were thoroughly washed to removeany sediment particles attached to their surface, submersed(approx. 1 min) in a solution with sodium hypochlorite(0.5%) to stop microbial action, and rinsed with deionizedwater (approx. 30 s). Experiments were then started.

    Experimental design

    To simulate CWs, two matrices were used in the experimentalassays:

    i. Water (W)aiming to evaluate the removal efficiency in asimpler aquatic medium (deionized water) minimizingmatrix interferences

    ii. Wastewater (WW)simulating more realistic conditions

    and three main variables were tested for both matrices, Wand WW:

    (a) LECA (unplanted LECA, L)(b) S. maritima (only plant, P)(c) S. maritima and LECA (planted LECA, PL)

    The experimental design of the work is shown in Table 2.W were operated in a batch mode, i.e., only one PPCP

    spiking dose at time 0. WW were carried out in fed-batchmodes, i.e., PPCPs injected in pulses at specific times. Bothsystems were carried out under laboratory conditions at roomtemperature (22 2 C).

    W assays were divided into the following:

    W1: Three days to assess L removal efficiency in short pe-riods of time

    W2: Seven days to assess L, P, and PL removal efficiency inthe hydraulic retention time used for both mesocosmstudies and full-scale CWs (Weber and Legge 2011)

    Both assays (W1 andW2) were spiked with 0.2 mg of eachcontaminant and carried out under dark conditions.

    WW assays were divided into the following:

    W1: Seven days to assess L, P, and PL removal efficiencysimulating successive arriving of contaminants (i.e.,controlled experimentalmicrocosm spiked every 2 days(0, 2,