Urban wastewater treatment by Tetraselmis sp. …w3.ualg.pt/~jvarela/articles/pdfs/Schulzeetal2017a.pdfElemental analysis of N, H and C was assessed using an elemen-tal analyser (Vario

Bioresource Technology 223 (2017) 175–183

Contents lists available at ScienceDirect

Bioresource Technology

journal homepage: www.elsevier .com/locate /bior tech

Urban wastewater treatment by Tetraselmis sp. CTP4 (Chlorophyta)

http://dx.doi.org/10.1016/j.biortech.2016.10.0270960-8524/� 2016 Elsevier Ltd. All rights reserved.

⇑ Corresponding author.E-mail address: [email protected] (L. Barreira).

Peter S.C. Schulze a,b, Carolina F.M. Carvalho a, Hugo Pereira a, Katkam N. Gangadhar a,c, Lisa M. Schüler a,Tamára F. Santos a, João C.S. Varela a, Luísa Barreira a,⇑aCCMAR – Centre of Marine Sciences, University of Algarve, Campus de Gambelas, 8005-139 Faro, Portugalb FBA – Faculty of Biosciences and Aquaculture, Nord University, 8049 Bodø, Norwayc LEPABE – Laboratory of Engineering of Processes, Environment, Biotechnology and Energy, University of Porto, Rua Dr. Roberto Frias s/n, P-4200-465 Porto, Portugal

h i g h l i g h t s

� Tetraselmis sp. CTP4 is suitable forurban wastewater treatment.

� Wastewater N and P were 100%removed in batch and continuousconditions.

� Produced biomass displayed highprotein and carbohydrate contents.

� Aging cultures showed decreasedprotein and increased carbohydratecontents.

� Cells grown in low saline media hadlow PUFA but high EPA content.

g r a p h i c a l a b s t r a c t

a r t i c l e i n f o

Article history:Received 5 August 2016Received in revised form 6 October 2016Accepted 8 October 2016Available online 14 October 2016

Keywords:Wastewater treatmentNutrient removalTetraselmis sp.Fatty acidsBiochemical composition

a b s t r a c t

The ability of a recent isolate, Tetraselmis sp. CTP4, for nutrient removal from sewage effluents before andafter the nitrification process under batch and continuous cultivation was studied. Biomass productivitiesin both wastewaters were similar under continuous conditions (0.343 ± 0.053 g L�1 d�1) and nutrientuptake rates were maximal 31.4 ± 0.4 mg N L�1 d�1 and 6.66 ± 1.57 mg P-PO4

3� L�1 d�1 in WW beforenitrification when cultivated in batch. Among batch treatments, cellular protein, carbohydrate and lipidlevels shifted with aging cultures from 71.7 ± 6.3 to 29.2 ± 1.2%, 17.4 ± 7.2 to 57.2 ± 3.9% and 10.9 ± 1.7 to13.7 ± 4.7%, respectively. In contrast, CTP4 cultivated continuously in Algal medium (control) showedlower biomass productivities (0.282 g VSS L�1 d�1) although improved lipid content (up to 20% lipids)in batch cultivation. Overall, Tetraselmis sp. CTP4 is promising for WW treatment as a replacement ofthe costly nitrification process, fixating more nutrients and providing a protein and carbohydrate-richbiomass as by-product.

� 2016 Elsevier Ltd. All rights reserved.

1. Introduction

Combining wastewater (WW) treatment with microalgal pro-duction has already been researched since 1950s (Oswald et al.,1957) and has received increasing interest in science and industryover the last decades. Currently, microalgal basedWW treatment is

considered to be an economically and environmentally sustainableprocedure to remove dissolved nutrients from effluents and to pro-duce valuable biomolecules to offset water treatment costs. Partic-ularly, the usage of high rate algal ponds (HRAPs) fed bywastewater and/or CO2 exhausts is considered as the most promis-ing strategy to clean waste water streams and produce a feedstockfor microalgal based by-products such as biofuels (Craggs et al.,2014; Park et al., 2011). Although the main focus of using microal-gae has been the removal of nutrients (Christenson and Sims,

176 P.S.C. Schulze et al. / Bioresource Technology 223 (2017) 175–183

2011), the reintroduction of nutrients into the market as trans-formed bio-products would follow the circular economy principle,a requirement to sustain the present life standards in industrialnations (European commission 2015, IP/15/620). In Europe, onaverage, 0.51 kg P and 2.52 kg N per inhabitant and year are dis-charged in WW (EU-EEA, 2015). These nutrients are valuableand/or finite resources that can substitute expensive fertilizersfor production of crops and algae (Vaccari, 2009). Nitrogen andphosphorus are currently removed through treatments involvingbiological nitrification followed by denitrification or precipitation(US-EPA, 2013). The nitrification step receives ammonia-rich WWfrom a biological oxygen demand (BOD) removal step. Ammoniais transformed into nitrate by nitrifying bacteria under anoxygen-rich environment. Subsequently, the nitrate-rich effluententers the denitrification step where denitrifying bacteria trans-form nitrate into molecular nitrogen under anoxic conditions,which is stripped out as gas by gentle aeration. However, denitrify-ing bacteria require external carbon sources that are often of fossilorigin (e.g. methanol). WWs with high phosphorus concentrationsmust be treated through a phosphorus removal step before dis-charge into protected areas. Here, effluents coming from the deni-trification step are supplemented with flocculants (e.g. aluminiumsalts, lime stone) allowing the precipitation of phosphorus as insol-uble salts. However, these procedures present additional costs andcan cause deterioration of the biomass quality as a feedstock ofnutrients in a functional fertilizer. One reason for this is the con-tamination with toxic, metal-containing flocculants that remainbound to phosphorus (Christenson and Sims, 2011). Recent reportsshowed that different microalgal strains use these nutrients fromurban, industrial and aquaculture effluents to produce biomole-cules that can be further upgraded for the production of bioplas-tics, feed, or biodiesel, among others (Christenson and Sims,2011; Zeller et al., 2013). Moreover, contrary to standard biologicaltreatments, algae were found to further improve the final effluentquality through natural disinfection and incorporation of othercontaminants, such as heavy metals, pharmaceuticals and endo-crine disrupters (Correa-Reyes et al., 2007; Craggs et al., 2014;Devi et al., 2012). The separation of microalgae cells from the trea-ted water, however, remains a major bottleneck for the large-scaleimplementation of microalgal-based bioremediation facilities, ascurrent technologies (e.g. centrifugation and flocculation) for bio-mass recovery have high costs in terms of energy and/or chemicals(Christenson and Sims, 2011).

The present work aimed to treat urban sewage water (beforeand after nitrification processes) in batch and continuous cultiva-tion systems, using a novel, robust, euryhaline microalgal strain(Tetraselmis sp. CTP4) that naturally settles down to the bottomof the containing vessel, reducing significantly the dewateringcosts (Pereira et al., 2016). The biochemical composition (e.g. totallipids, proteins, fatty acid profile), best culturing strategies andapplications for the biomass to offset the costs of nutrient removalare also discussed.

2. Methods

2.1. Microalgal strain and growth conditions

Tetraselmis sp. CTP4 belongs to the T. striata/convolutae clade(Chlorophyta, Chlorodendrophyceae) and was previously isolatedfrom the Ria Formosa lagoon, located in the south of Portugal(Algarve) near a WW stream using a high throughput method com-bining a pre-enrichment step for robust microalgae, fluorescenceactivated cell sorting and direct plating onto 96-well plates asdescribed by Pereira et al. (2016).

Urban sewage effluent waters were collected at a WW treat-ment plant located in Quinta do Lago, Algarve, Portugal (WWTPQuinta do Lago) between May and July 2015. In this WWTP, urbanWW goes through primary/secondary screens and grit chambers; aprimary clarifier and aeration tank for removal of dissolved organiccarbon (DOC) and finally undergoes a nitrification/denitrificationstep prior to sterilization and discharge. For this study, WW sam-ples were collected before (Pre-N) and after (Post-N) the nitrifica-tion step, and CTP4 was cultivated in both effluents to allow adirect comparison. The effluents were not sterilized and werestored at 4 �C until use. As control, CTP4 was also grown in a pre-viously established standard medium composed of sterile seawaterfrom the Atlantic shoreline of Faro (Portugal, salinity of ca 35)enriched with Modified Algal Medium (MAM; 2 mM NaNO3,0.1 mM, KH2PO4, 20 lM EDTA-Na, 20 lM FeCl3*6 H2O, 2 lMMgSO4*7H2O, 1 lM ZnCl2, 1 lM ZnSO4, 1 lM MnCl2, 0.1 lM Na2-MoO4, 0.1 lM CoCl2, 0.1 lM CuSO4 (Pereira et al., 2016).

All cultures were initially grown in 5L reactors filled up to a vol-ume of 4L, until the stationary phase was reached. Experimentswere conducted in a climate chamber (Aralab Fitoclima s 600 PLclima plus 400), in triplicates, at 20 �C and exposed to a light inten-sity of 100 lmol photons s�1 m�2 using continuous illumination(Osram cool white 840). Cultures were aerated with 0.2 lm-filtered air with a flow rate of �1 L min�1. For continuous cultiva-tion conditions, cultures were grown in batch until the late expo-nential phase (biomass concentrations of �0.7 g L�1) with asubsequent shift to continuous growth. A hydraulic retention time(HRT) of 6.6 d�1 was set by controlling the influent flow rate. ThisHRT equalled 2 � l�1, where l is the average growth rate esti-mated during the batch growth among all treatments to obtainhighest biomass productivity (Ruiz et al., 2013a).

2.2. Analytical methods

At different time points during the batch and continuous cultur-ing period, aliquots of the cultures were harvested and centrifuged(5000g, t = 5 min) to separate algal biomass from the growth med-ium. The algal pellet was freeze dried and stored in a desiccatoruntil analysis, whereas the supernatant was filtered (pore£ = 1.2 lm) and stored at �20 �C for the analyses of the dissolvednutrients.

2.2.1. Monitoring of algal growthCell growth was monitored by cell counting (Neubauer cham-

ber), optical density (OD750) and volatile suspended solids concen-tration (VSS), which is equivalent to ash free dry weight (AFDW).Algal suspension was filtered (pore £ = 0.47 lm) and VSS wasdetermined using 0.5 M sodium bicarbonate as washing reagentand processed according to standard methods. Briefly, 10–30 mgbiomass was dried at 105 �C for 24 h to obtain total suspendedsolids (TSS). Subsequently, the moisture free biomass was inciner-ated at 560 �C for 8 h and the ash content was determined. The VSSof the samples was calculated by subtracting the moisture weightand the TSS from the initial biomass weight. Upon plotting VSS andcell concentration (cells mL�1) against the OD750 of different exper-iments and time points (n = 93), significant linear correlations werefound and used to estimate VSS (rP 0.97, p < 0.01) and cellularconcentrations (rP 0.90, p < 0.01) on a daily basis (Eqs.(1) and(2)):

VSS ðg L�1Þ ¼ 1:991 � OD750 � 0:075 ð1Þ

Cell concentration ðCells mL�1Þ ¼ 3:57 � 106 � OD750 þ 1:67 � 105

ð2Þ

Table 1Nutrients concentration (mg L�1) in Algal medium (MAM) and in the effluent samplescollected before (Pre-N) and after (Post-N) the nitrification process.

Nutrients MAMa Pre-Nb Post-Nb

TN 28.3 27.6 ± 2.2 33.4 ± 6.7N-NH4

+ N.A. 26.9 ± 1.9 4.3 ± 2.8N-NO3

� 28.3 N.A. 22.2 ± 2.2N-NO2

� N.A 0.7 ± 0.3 6.9 ± 1.7P-PO4

3� 3.1 6.7 ± 1.2 6.4 ± 0.9

a Calculated from ALGAL formula; see Section 2.1.b Average from 3 batches of WW received during June and July; TN: total

nitrogen.

P.S.C. Schulze et al. / Bioresource Technology 223 (2017) 175–183 177

Elemental analysis of N, H and C was assessed using an elemen-tal analyser (Vario EL iii, Elementar Analysensysteme GmbH, Ger-many). Protein determination was carried out according to amodified Lowry et al. (1951) method as detailed in Pomory(2008). Upon plotting protein levels obtained by the Lowry methodagainst the N content in Algal biomass (%N), a significant linearrelationship was obtained (r = 0.97, p < 0.01) and protein contentin VSS could be calculated according Eq. (3):

%Protein ¼ 14:49þ 6:66 �%N ð3ÞTotal lipids were determined according to a modified Bligh and

Dyer (1959) method as previously described in Pereira et al.(2012). Briefly, lipids from dried biomass were homogenized usingan Ultra Turrax (IKA) and extracted with a mixture of chloroformand methanol (2:1). Phase separation was achieved by centrifuga-tion, and the chloroform phase was then removed, evaporated, andthe dried residue weighed. Carbohydrates were determined bysubtraction.

Culture dominance was assessed by flow cytometry in a BectonDickinson FACS Aria II (BD Biosciences, Erembodegem, Belgium)using FACSDiva software (version 6.1.3). The chlorophyll fluores-cence signal was recorded in the FL3 channel (695/40) after excita-tion with the blue (488 nm) laser.

2.2.2. FAME profileFatty acids (FA) were determined according to a modified

Lepage and Roy (1984) method previously described in Pereiraet al. (2012). FA were extracted and converted to the correspondingfatty acid methyl ester (FAME) by a direct transesterification reac-tion using acetyl chloride and methanol (1:20 v/v) followed byhexane extraction. The extracted FAME were analysed on a BRU-KER GC–MS (SCION 456 GC, TQ Mass Selective Detector) equippedwith a WCOT fused silica column (30 m � 0.25 mm internal diam-eter, 0.25 lm film thickness, Rxi�-5Sil MS,Cat. 13623 Restek) usinghelium as carrier gas (flow rate 1 mL min�1). The temperature pro-gram was 60 �C (1 min), 30 �C min�1 to 120 �C, 4 �C min�1 to250 �C, and 20 �C min�1 to 300 �C (4 min). Injection temperaturewas 300 �C using splitless mode. For identification and quantifica-tion of FAME the total ion mode was used. Temperatures of thetransfer line and ion source were 250 �C and 220 �C, respectively.

2.2.3. Nutrients analysisWater samples were defrosted and nutrients determined by

means of spectrophotometric methods described in APHA (1999)using a microplate reader (Biotek Sinergy 4). Ammonium-basedN (N-NH4

+) was determined by the blue-indophenol method(4500-NH3 F), nitrate-based N (N-NO3

�) by the UV method (4500-NO3

� B) and nitrite-based N (N-NO2�) according to the sufanilamide

method (4500-NO2� B). Dissolved phosphate based P (P-PO4

3�) con-centration was determined by the ascorbic acid method (4500-P E).Because of the high concentrations of ions and organic moleculesin the water samples, which have been reported to interfere withthe used UV N-NO3

� detection method (e.g. NO2�, Cl�, organic mat-

ter), a validation of the obtained data was performed via acadmium-based nitrate detection method (4500-NO3

� E).

2.3. Data analysis

Growth kinetics and nutrient consumption of cultures wereassessed using the models proposed by Ruiz et al. (2013a,b). Statis-tical analyses were performed using GraphPad Prism 6 softwareand Addinsoft XLSTAT statistics software (release 2014.5.03). Ananalysis of variance (one-way ANOVA) was performed to detecteffects of treatments or time on the response variable (e.g. totallipid or protein levels in biomass or nutrient concentrations). Lin-ear correlations were assessed via a two-tailored Pearson’s test (r).

3. Results and discussion

3.1. Medium composition

In order to investigate the removal capacities of N in theforms of ammonium (NH4

+), ammonia (NH3) and nitrate (NO3�),

CTP4 was grown in non-sterilized WW before (Pre-N) and after(Post-N) the nitrification process. Considering the nutrient loads(NH4

+, NO3� and PO4

3�) obtained throughout the experimental per-iod, the WW samples were classified as weak to medium pol-luted effluents (Table 1; Christenson and Sims, 2011). Total N(TN) and P (TP) concentrations in both WW were on averagedouble of that in MAM. However, the nutrient concentrationsvaried among the three batches of WW samples obtained fromthe WWTP Quinta do Largo and during the experimental period(Table 1). In Pre-N WW, on average, almost �97% of total N wasin the form of total ammonium, measured as N-NH4

+ and only�3% was N-NO2

�. In Post-N WW most of the TN was in the formof N-NO3

� (�67%); however, considerable amounts of N-NO2�

(�12% of TN) and NH4+ (�21% of TN; Table 1) were detected.

Such high levels of NH4+ can collapse the bacterial consortia pre-

sent in the subsequent denitrification step in WWTPs (Zhouet al., 2011), an event that would cause the release of such N-species into the environment, which are toxic to many aquaticanimals (Soucek and Dickinson, 2012).

Generally, urban WW and MAM are considered to have a non-limiting pool of all micronutrients necessary for plant and algalgrowth (Christenson and Sims, 2011; Fabregas et al., 1984;Jönsson et al., 2004), thus allowing non-nutrient limited growthof CTP4. Despite a possible pH rise due to photosynthetically CO2

uptake causing a shift of NH4+ to N-NH3 in Pre-N or high loads of

N-NO2� in Post-N media, CTP4 was successfully cultivated for

15 days under batch conditions and continuously for 21 days inAlgal medium, Pre-N and Post-N, respectively.

3.2. Growth

CTP4 was successfully cultivated under batch and continuousculturing conditions in both WW samples collected before andafter the nitrification process (Fig. 1). In the culture grown inPre-N WW a short lag phase (�3 days) was observed (Fig. 1A)and VSS contents of the culture were always lower than in the cul-ture grown in Post-N WW.

Pre-N WW was characterized by a rather elevated concentra-tion of ammonia. In the obtained water samples, total ammoniacould be considered as the sum of in-equilibrium levels of toxicammonia (NH3) and ammonium (NH4

+) (Peccia et al., 2013). Duringthe experiment, the pH of the growth media rose from 7.1 to 8.6during the active growth phases, which might have caused a disso-ciation of �10% total ammonium into NH3, leading to concentra-tions of up to 3 mg N-NH3 L�1. This concentration may have

Fig. 1. Growth curves and VSS productivities of Tetraselmis CTP4 cultivated in synthetic algal medium (black cycles, left bar), and in wastewater collected before nitrification(Pre-N; red triangle, middle bar) and after nitrification (Post-N; green square, right green bar) under batch (A) and continuous (B) cultivation conditions. Different lettersabove bars indicate statistical differences among the groups. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of thisarticle.)

Fig. 2. Flow cytometry dot plots of inner cell complexity (side scatter; SSC-A) versusautofluorescence of chlorophyll (FL3) of Tetraselmis CTP4 cultivated in synthetic algalmedium (A), and in wastewater collected before nitrification (B) and after nitrifica-tion (C) under continuous cultivation conditions. (For interpretation of the referencesto color in this figure legend, the reader is referred to the web version of this article.)

178 P.S.C. Schulze et al. / Bioresource Technology 223 (2017) 175–183

limited algal growth as observed in the early growth stage in Pre-Ntreatment (Fig. 1; Borowitzka and Moheimani, 2013). In a real sit-uation, for instance if HRAPs are used for WW treatment, the pHcould be lowered through CO2 supplementation, which would leadalso to better productivities (Faria et al., 2012). Nevertheless, thepH was never higher than the maximum permitted by law (9.0in discharged waters; DL 246/98, Ministério do Ambiente,Portugal, 1998).

Under batch conditions (Fig. 1A), cultures grown in WW col-lected before and after the nitrification process (0.099 and0.110 g VSS L�1 d�1, respectively) displayed significantly higherproductivities compared to cells grown in MAM (0.086 g VSS L�1 -d�1). The maximum specific growth rate (l) was similar in bothWW treatments (average: 0.27 ± 0.02 d�1) but statistically highercompared to that of MAM (0.23 ± 0.01 d�1). The same trend wasfound for maximum VSS: similar among Pre-N and Post-N treat-ments (1.89 ± 0.07 g and 1.88 ± 0.07 g VSS L�1, respectively), butlower in cells cultivated in MAM (1.60 ± 0.14 g VSS L�1).

Similarly to what was observed in the batch cultivation, undercontinuous conditions, CTP4 showed highest VSS production ratesin the Post-N treatment (0.351 ± 0.057 g VSS L�1 d�1), followed byPre-N (0.335 ± 0.060 g VSS L�1 d�1) and MAM(0.282 ± 0.066 g VSS L�1 d�1). Generally, similar productivitieswere reported for microalgae growing in weak to medium pollutedurbanWW (Christenson and Sims, 2011). In the continuous growthexperiment, the growth rate was highest in the Post-N treatment(0.49 ± 0.02 d�1) and MAM (0.46 ± 0.01 d�1) and lowest in Pre-N(0.41 ± 0.01 d�1, p < 0.01). However, VSS productivities were aboutthree times higher compared to CTP4 grown in batch (Fig. 1B).

Despite the natural occurrence of bacteria and fungi in the non-sterile wastewater samples used, microscopical observationsshowed that CTP4 remained as the dominant species throughoutthe experiments in all media used. In order to confirm the resultsobtained by microscopy, samples from the last day of the continu-ous experiment were analyzed by flow cytometry (Fig. 2). The pre-sented dot plots relate the relative inner cell complexity (sidescatter; SSC-A) with the autofluorescence of chlorophyll (FL3) ofcells. The control (MAM) sample revealed a common pattern forunialgal cultures, with a clear distinction between the bacterial(blue) and Tetraselmis sp. CTP4 (red) populations (Fig. 2A).Although both samples from the pre-N (Fig. 2B) and post-N(Fig. 2C) WWs displayed a higher number of events between bothpopulations (green), Tetraselmis sp. CTP4 was clearly the dominantpopulation in all treatments.

P.S.C. Schulze et al. / Bioresource Technology 223 (2017) 175–183 179

3.3. Nutrient removal

Nutrient recovery from wastewater using microalgae dependsnot only on biomass productivity but also, and most importantly,on the intracellular storage capacity of nutrients by the alga. Espe-cially the capacity of the algae to take up nutrients in the dark(Ruiz et al., 2013b). In the present study, CTP4 recovered26.6 ± 0.2, 23.3 ± 2.4 and 31.4 ± 0.4 mg TN L�1 and 2.67 ± 0.11,7.13 ± 0.93 and 4.65 ± 0.03 mg P-PO4

3� L�1 from MAM, Pre-N andPost-N treatments under batch conditions (Fig. 3). Total N removalrate was significantly different among all treatments. The highestremoval rates were found in Pre-N (3.94 ± 0.52 mg N L�1 d�1) fol-lowed by MAM (3.46 ± 0.3 mg N L�1 d�1) and Post-N(2.79 ± 0.36 mg N L�1 d�1) treatments. Interestingly, N uptake rates(Pre-N: 31.4 ± 0.4, SAM: 26.1 ± 0.2 and Post-N:23.3 ± 2.4 mg N L�1 d�1) correlate significantly (r = 0.867, p < 0.05)with the initial total N (TN) concentration in the growth media(see Table 1). Nitrite levels in MAM increased from0.44 ± 0.18 mg N-NO2

� to 3.55 ± 0.25 mg N-NO2� within 18 h, and

gradually decreased towards zero afterwards. Such increase of N-NO2

� was probably caused by enhanced nitrate reductase activityinduced by high light intensity, as may occur at early growth stagesdue to low cell densities (Albert et al., 2013). However, throughoutculture growth, NO2

� and NO3� levels were completely depleted

from the media upon day 8.In the batch experiment, P removal rates by CTP4 were highest

in Pre-N followed by Post-N and MAM (6.66 ± 1.57, 1.92 ± 0.15 and0.51 ± 0.02 mg P-PO4

3� L�1 d�1, respectively; Fig. 3D). Monitoring ofpH revealed increases in all cultures, from the initial �7.1 to �8.6at day 3, which are probably linked to CO2 uptake from the med-

Fig. 3. Uptake of (A) nitrate (N-NO3�), (B) ammonium (N-NH4

+) and (C) nitrite (N-NO2�) an

Algal medium (MAM; black cycles) and before nitrification (Pre-N; red triangles) andreferences to color in this figure legend, the reader is referred to the web version of thi

ium due to microalgal photosynthetic activity. An additionalexperiment showed that, a pH of 8.0 caused salt precipitation inMAM, probably of Ca2+ and Mg2+ phosphates (Santos et al.,2012), whereas a further pH rise resulted in more precipitation.Thus, P might have not been fully available for uptake causing slowuptake rate by CTP4 cultivated in MAMs and became a limitingnutrient. As mentioned above, CO2 supplementation may lead toa decrease in pH level and thus preventing the precipitation ofphosphate salts. In addition, the high salinities occurring in MAMmight have contributed to an inhibition of P uptake by CTP4 dueto high ionic strength of Na+ cations that can inhibit alkaline phos-phatase activity as reported for Gracilaria tenuistipitata (Lee et al.,1999).

In the continuous experiment, concentrations of NO3�, NO2

�, NH4+

and PO43� were measured in the overflowWW and remained below

the detection limit the whole time of the experiment. Calculationsfrom daily supplemented medium (0.6 L d�1; HRT 6.6 d) and theaveraged initial nutrients (N, P) in media (Table 1) revealed thatCTP4 removed 1.86 ± 0.01, 4.04 ± 0.74 and 3.84 ± 0.55 mg P-PO4

3� L�1 d�1, as well as 17.0 ± 0.1, 16.6 ± 1.3 and 20.0 ± 4.0 mgTN L�1 d�1 from MAM, Pre-N and Post-N media, respectively.

High NO2� levels are typically caused by environmental condi-

tions favouring the growth of ammonium oxidising bacteria(AOB) rather than nitrite oxidising bacteria (NOB). Such conditionsinclude temperatures higher than 25 �C as well as low dissolvedoxygen, pH and C:N ratios (Zhou et al., 2011). Interestingly, algaetake up available dissolved NO2

� and promote unfavourable condi-tions for AOB through their photosynthetic activity, which result inhigh pH, low phosphate levels and high dissolved oxygen concen-trations (Zhou et al., 2011). Furthermore, during the nitrification

d (D) phosphorus (P-PO43�) by Tetraselmis CTP4 cultivated under batch conditions in

after nitrification (Post-N; green squares) wastewater. (For interpretation of thes article.)

180 P.S.C. Schulze et al. / Bioresource Technology 223 (2017) 175–183

and denitrification processes, N2O is released into the environ-ment, which is a greenhouse gas 298 times more powerful thanCO2 (Albert et al., 2013; Alcántara et al., 2015; Zhou et al., 2011).Hence, microalgal-based WW treatments should be placed beforethe nitrification process in the treatment of WW, because N-NH4

+

is the most abundant N source at this step. This recommendationhas the additional advantage of eliminating the oxidation fromNH4

+ to NO3� (nitrification) and the subsequent reduction of NO3

�

to NH4+ upon the photoautotrophic assimilation of nitrate

(Alcántara et al., 2015). The present study showed that CTP4 caneffectively uptake NH4

+ as N source. This observation thus indicatesthat CTP4 can be a promising biological agent to replace the com-mon secondary treatment, fixing nutrients from effluents in har-vested biomass and minimizing releases of NO2

�, N2O and NH4+

(Alcántara et al., 2015) into the environment. Considering thatthe nitrification and denitrification steps employed by the WWTPof Quinta do Lago represent 40% of the overall treatment costs,the implementation of a CTP4-based WW treatment would be verybeneficial for this WWTP.

During the experimental period, from May to August 2015, theeffluent of this WWTP displayed COD concentration of45.1 ± 9.3 mg O2 L�1, 12.2 ± 4.5 mg TN L�1 and 5.1 ± 1.2 mg TP L�1.For this WWTP, TN and COD values were within the legislation(max. 15 mg TN L�1; 125 mg O2 L�1 COD), whereas TP values wereabove (2 mg TP L�1). Certainly, many WWTPs have problems dur-ing hot summer or cold winter months to maintain viability ofnitrifying and denitrifying bacteria, resulting in high TN concentra-tions. Interestingly, the replacement of the traditional denitrifica-tion and nitrification by a microalgal-based treatment will helpto increase dissolved O2 concentration in the effluent, decreasingthe chemical and biological oxygen demand (COD, BOD) and thusreducing the toxicity of the effluent (Valderrama et al., 2002).Improvements of the efficiency of WW treatment system basedon CTP4 may include CO2 injection, which helps to increase nutri-ent removal but also biomass productivity (Faria et al., 2012) or theusage of blue light in the wavelengths�390–450 nm as reported toincrease N and P uptake in many species (Schulze et al., 2014,2016). In addition, the identification of the optimal micronutrientrequirements of CTP4 to balance metabolic intracellular pathwaysfor optimum nutrient removal and biomass production is ongoingand will be published elsewhere.

3.4. Biochemical composition of biomass

In the batch experiment, protein contents (Fig. 4A) in harvestedbiomass significantly decreased 3-fold from day 3 to day 15 underbatch conditions. Conversely, carbohydrate contents (Fig. 4C)increased roughly 3-fold over time. The initial protein, carbohy-drate and total lipid levels at day 0 were obtained from the CTP4grown in MAM used as inoculum. The sharp shifts of protein, lipidand carbohydrate contents between day 0 and day 3 demonstratethe biochemical response of CTP4 from nutrient-depleted toreplete conditions. Kim et al. (2016) observed the same trend,decreasing protein and increasing carbohydrate levels in Tetrasel-mis sp. exposed to either N replete or depleted conditions. Suchaccumulation of proteins under batch conditions together withthe sharp decrease of carbohydrates in between may be relatedto a fast uptake of N and its incorporation into amino acids suchas glutamine (Lourenço et al., 2004). A significant increase of totallipids (Fig. 4B) from about 10.2 ± 2.1% on day 3 up to 19.8 ± 1.7% onday 15 could be observed for MAM, a trend that was not found inWW, whose lipid contents remained at 9.8 ± 2.0% of VSS through-out the experimental period.

Unlike cultures under batch conditions, the biochemical compo-sition of CTP4 grown inWW under continuous conditions (Fig. 4D–F) did not vary significantly. However, significant differences were

found among cultures in MAM and in WW. Protein levels werehighest in MAM (42.2 ± 5.2%) compared to WW treatments (aver-age: 35.4 ± 4.0%), whereas carbohydrates where highest in WW(average: 57.2 ± 4.3%) compared to MAM (46.9 ± 3.6%). As in thebatch experiment, total lipid content was highest in CTP4 grownin MAM (11.0 ± 2.5%) when compared to both WW (average:7.5 ± 1.8%).

Taking the daily production rate of tested biochemical compo-nent into account, no statistical differences among treatments con-cerning protein and lipid productivity were found, averaging at121 ± 23 and 27.6 ± 8.3 mg L�1 d�1, respectively. However, therewere differences in carbohydrate productivity among the differenttreatments: 132 ± 27, 186 ± 32 and 205 ± 30 mg L�1 d�1 for MAM,Pre-N and Post-N treatments, respectively.

The significant increase of total lipids and decrease of carbohy-drates in MAM-grown CTP4 under batch conditions from day 3 to15 might correspond to nutrient limitations induced by high salin-ity, as already discussed above. Indeed, salt was reported toincrease lipid contents in Tetraselmis spp. and in other algae(Fon-Sing and Borowitzka, 2015; Takagi and Yoshida, 2006); how-ever, the regulatory mechanism behind it remains to be found. Pro-duction of sugars and lipids are competing processes within themicroalgal metabolism. The accumulation of lipids can be attainedat the expense of decreasing carbohydrate levels and low biomassproduction, as previously reported for Tetraselmis sp. (Kim et al.,2016). This development can be favoured by P or N limitation(Mutlu et al., 2013), high salinities (Asulabh et al., 2012; Panchaet al., 2015; Renaud and Parry, 1994) or other environmental fac-tors that inhibit cell growth (Park et al., 2011; Sharma et al.,2012). This would also explain the low lipid levels in CTP4 culti-vated in WW both under batch and continuous conditions. P limi-tation due to precipitation may have also contributed to lipidaccumulation (Ruangsomboon et al., 2013) or reduced growthwhen salinities and/or precipitating ions levels such as Mg+, Ca+,CO3

2� were high in the medium (e.g. MAM), as that can cause stressin algae (Guihéneuf and Stengel, 2013; Mohan and Devi, 2014).

3.5. FAME

The FAME profile of CTP4 cultures grown under different batchconditions was significantly affected by time and type of culturemedia (Table 2). The full data set is provided in supplemental data(Table S1). The major saturated fatty acid (SFA), monounsaturatedfatty acid (MUFA) and polyunsaturated fatty acid (PUFA) in CTP4were palmitic (C16:0), oleic (C18:1) and linoleic (C18:2; LA) acids,respectively. Palmitic acid increased from 29.35 ± 0.49% on day 3towards 39.16 ± 0.73% of total FA (TFA) on day 15 in MAM, whichwas on average 10% lower compared to both WW treatments. LAdecreased over time in MAM from 23.43 ± 0.43 to 8.30 ± 0.43% ofTFA and the abundances detected were always higher in MAMcompared to both WW treatments (e.g. 2-fold increase at day15). CTP4 grown in MAM displayed approximately the double ofhexadecatrienoic (C16:3) and hexadecadienoic (C16:2) acids, butonly half of the amount of eicosapentaenoic acid (EPA, C20:5),when compared to both WW among all time points. However,throughout all growth stages, the PUFA/SFA ratios in control cul-tures were on average 1.5 times higher compared to those of bothWW treatments (see also Table S1). In all treatments, the degree ofsaturation increased significantly with the cultivation time, a trendsimilar to Tetraselmis sp. exposed to N-depleted conditions (Kimet al., 2016).

No significant shifts of the FAME profile were observed duringthe experimental period in CTP4 grown in WW continuously(Table 2). However, similarly to the batch experiment, palmiticacid was lower in CTP4 grown in MAM compared to both WW(31.3 ± 2.6% and 39.8 ± 2.3% of TFA, respectively), but was probably

Fig. 4. Contents of proteins, total lipids and carbohydrates (CBH) in VSS biomass obtained from Tetraselmis CTP4 cultivated under batch (A–C) and continuous (D–F)conditions in Algal medium (MAM; black cycles) and before nitrification (Pre-N; red triangles) and after nitrification (Post-N; green squares) wastewater. (For interpretationof the references to color in this figure legend, the reader is referred to the web version of this article.)

Table 2Fatty acid profile of Tetraselmis CTP4 grown under batch and continuous conditions in Algal medium (MAM) and before (Pre-N) and after (Post-N) wastewater. Fatty acid profilesare shown only for the most discriminative time points, whereas the time averaged (day ø) profile is given when no statistical differences among time points were observed.Different letters within each cultivation method indicate significant differences, and standard deviations are provided in parenthesis. Detailed fatty acid profiles are provided inTable S1 in supplemental data.

Batch Continuous

Treatment(Salinity) Timepoint

MAM (35 ‰)day 3

Pre-N (5 ‰)day 3

Post-N (5 ‰)day 3

MAM (35 ‰)day 15

Pre-N (5 ‰)day 15

Post-N (5‰)day 15

Pre-N (5–1 ‰)day ø

Post-N (5–1 ‰)day ø

MAM (35 ‰)day ø

C16:0 (%) 29.35d (0.49) 35.36c (1.31) 35.14c (1.64) 39.16b (0.73) 44.78a (1.16) 44.72a (1.12) 39.87a,b (2.56) 39.79b (2.12) 33.80b (5.28)C16:1 (%) 7.57c (0.39) 7.10c (0.40) 11.06a,b

(0.60)10.54b (0.02) 12.88a (1.91) 12.68a (0.47) 9.59b,c (1.20) 10.14b (1.31) 11.16a,b

(1.88)C16:2 (%) 3.93a (0.13) 1.57b (0.14) 1.04b (0.08) 4.06a (0.19) 1.88b (0.16) 1.80b (0.03) 2.31a(0.29) 2.20a (0.18) 3.72a (1.30)C16:3 (%) 7.09a (0.07) 2.31b (0.13) 3.55b (0.12) 3.62b (0.10) 1.69b (0.09) 1.73b (0.13) 3.24b (0.72) 3.53b (0.68) 5.28a,b (1.30)C18:0 (%) 1.31a (0.49) 1.77a (0.21) 1.78a (0.09) 1.87a (0.25) 1.55a (0.13) 1.46a (0.13) 1.78a (0.26) 1.55a (0.21) 1.45a (0.36)C18:1 (%) 19.26c (0.63) 23.47b (0.48) 22.15b (0.34) 24.76a (0.42) 25.64a (1.54) 25.63a (1.64) 24.18a,b (1.15) 25.36a (0.76) 23.03b (1.57)C18:2(%) 24.43a (0.43) 13.93b (1.25) 14.07b (0.87) 8.30c (0.43) 4.16d (0.98) 4.25d (0.28) 11.20b (1.83) 9.64b,c (1.53) 14.07b (4.03)C20:4 (%) 2.01a (0.05) 2.82a (0.12) 2.42a (0.29) 1.51a (0.10) 1.59a (0.04) 1.16a (0.07) 1.87a (0.45) 1.72a (0.14) 1.99a (0.45)C20:5 (%) 3.79c (0.31) 7.86a (0.63) 5.87b (0.84) 3.30c (0.31) 5.24b (0.28) 4.75b (0.33) 3.89a (0.52) 3.49a (0.41) 3.19a (0.84)P

SFA (%) 30.66d (0.34) 37.13c (1.52) 36.92c (1.61) 41.03b (0.97) 46.32a (1.29) 46.18a (1.25) 41.65a (2.50) 41.34a (2.21) 35.24b (5.60)P

MUFA (%) 26.83e (0.25) 30.58d (0.73) 33.21c (0.37) 35.30b (0.44) 38.52a (2.74) 38.30a (1.22) 33.77b (2.07) 35.50a (1.24) 34.19a,b

(1.79)P

PUFA (%) 40.24a (0.76) 28.49b (2.23) 26.95b (1.93) 20.79c (0.55) 14.56d (0.78) 13.68d (0.47) 21.88c (2.73) 20.91c (2.13) 28.18b (6.67)PUFA/SFA (�) 1.31a (0.03) 0.77b (0.09) 0.73b (0.08) 0.51c (0.03) 0.31d (0.02) 0.30d (0.02) 0.53a (0.09) 0.51a (0.08) 0.84b (0.30)

P.S.C. Schulze et al. / Bioresource Technology 223 (2017) 175–183 181

not significantly affected by the salinity in the control treatment(Table S1).

3.6. Valorisation of produced CTP4 biomass

The biomass produced can be easily recovered from the treatedwastewater using common settlers. Pereira et al. (2016) pointedout that 80% of CTP4 biomass could be recovered within 6 h in20% of the initial water volume, which then can be dewateredmoreefficiently compared to other non-autonomously settling algae.Such rapid settling is clearly an advantage using a monocultureof CTP4 as a mix of several algae often displays higher settlingtimes (Mennaa et al., 2015). In addition, some authors reported

higher nutrient removal rates and more valuable biomass withconstant quality for single species treatment compared to algalassemblages (Mennaa et al., 2015; Tang et al., 1997). The harvestedCTP4 biomass from the continuous experiment consists mostly ofproteins and carbohydrates and low lipid levels. Proteins frommicroalgae are promising sources for bioplastics (Zeller et al.,2013), whereas carbohydrates can be upgraded to bioethanol as avaluable biofuel or biogas (Harun et al., 2010). Nonetheless, thefatty acid profile of CTP4 grown under MAM is highly suitable forbiodiesel production due to its high contents of C16 and C18 FAMEas well as the low PUFA/SFA ratio (Pereira et al., 2016; Zhou et al.,2012). However, biomass obtained from WW or low saline growthmedia showed elevated arachidonic acid (C20:4n-6) and C20:5

182 P.S.C. Schulze et al. / Bioresource Technology 223 (2017) 175–183

PUFAs, higher than recommended for biodiesel (Zhou et al., 2012).In addition, the low total lipid contents in microalgal biomassobtained fromWW treatments may not allow a sustainable biodieselextraction. Pereira et al. (2016) proposed a two-stage system withoptimum growth conditions in the first phase and N-starved condi-tions in a second phase, which caused lipid accumulation of up to30% in CTP4 dry weight with a low unsaturation profile. Hence,the combination of wastewater treatment, as a supply of nutrientrich culture media, with such two-stage cultivation method andbiorefinery scheme could be used to offset wastewater treatmentcosts. Other uses such as feed or food were not considered due tothe possible contamination of the biomass with toxins (e.g. pharma-ceuticals and endocrine disrupting compounds) and metals, whichare contaminants that commonly occur in wastewater. On the con-trary, the use of the produced biomass for biofuels will involve thedestruction of such contaminants by combustion and energyproduction.

4. Conclusions

The present study has shown that Tetraselmis sp. CTP4 is apromising microalga to treat wastewater before and after the nitri-fication process. Nutrients in effluents were completely removedand the produced biomass can be used as source of carbohydratesand proteins, which can be used for bioethanol or bioplastics pro-duction, respectively. Further studies should include the wastewa-ter treatment of CTP4 in scaled open pond systems and thedefinition of optimum hydraulic retention times. In addition, thepotential of CTP4 to decrease chemical and biological oxygendemands, heavy metal loads or pharmaceutical contaminants in awater body remains to be investigated.

Acknowledgements

We would like to thank Águas do Algarve for providing theurban wastewater samples used in the experiments and for theanalysis of nitrate levels in distinct water samples. H.P. (SFRH/BD/105541/2014) is a PhD student funded by the Portuguese Foun-dation for Science and Technology (FCT). K.N.G. (SFRH/BPD/81882/2011) and L.C. (IF/00049/2012) were supported byFCT as a post-doctoral research fellow and as FCT InvestigatorProgramme recipient, respectively.

Appendix A. Supplementary data

Supplementary data associated with this article can be found, inthe online version, at http://dx.doi.org/10.1016/j.biortech.2016.10.027.

References

Albert, K.R., Bruhn, A., Ambus, P., 2013. Nitrous oxide emission from Ulva lactucaincubated in batch cultures is stimulated by nitrite, nitrate and light. J. Exp. Mar.Biol. Ecol. 448, 37–45.

Alcántara, C., Muñoz, R., Norvill, Z., Plouviez, M., Guieysse, B., 2015. Nitrous oxideemissions from high rate algal ponds treating domestic wastewater. Bioresour.Technol. 177, 110–117.

APHA., 1999. Standard methods for the examination of water and wastewaterAmerican Public Health Association (APHA), American Water WorksAssociation (AWWA) and Water Environment Federation (WEF).

Asulabh, K., Supriya, G., Ramachandra, T., 2012. Effect of salinity concentrations ongrowth rate and lipid concentration in Microcystis sp., Chlorococcum sp. andChaetoceros sp. National Conference on Conservation and Management ofWetland Ecosystems. School of Environmental Sciences, Mahatma GandhiUniversity, Kottayam, Kerala.

Bligh, E.G., Dyer, W.J., 1959. A rapid method of total lipid extraction andpurification. Canadian J. Biochem. Physiol. 37, 911–917.

Borowitzka, M.A., Moheimani, N.R., 2013. Sustainable biofuels from algae. Mitig.Adapt. Strategies Glob. Change 18, 13–25.

Christenson, L., Sims, R., 2011. Production and harvesting of microalgae forwastewater treatment, biofuels, and bioproducts. Biotechnol. Adv. 29, 686–702.

Correa-Reyes, G., Viana, M.T., Marquez-Rocha, F.J., Licea, A.F., Ponce, E., Vazquez-Duhalt, R., 2007. Nonylphenol algal bioaccumulation and its effect through thetrophic chain. Chemosphere 68, 662–670.

Craggs, R., Park, J., Heubeck, S., Sutherland, D., 2014. High rate algal pond systemsfor low-energy wastewater treatment, nutrient recovery and energyproduction. New Zeal. J. Bot. 52, 60–73.

Devi, M.P., Subhash, G.V., Mohan, S.V., 2012. Heterotrophic cultivation of mixedmicroalgae for lipid accumulation and wastewater treatment during sequentialgrowth and starvation phases: effect of nutrient supplementation. Renew.Energ. 43, 276–283.

DL n� 236/98, 1-8-1998, Ministério do Ambiente. Diário da República: I Série A, No176 (1998). Portugal.

EU-EEA., 2015. Emission Intensity of the Domestic Sector in Europe. EuropeanEnvironment Agency.

Fabregas, J., Abalde, J., Herrero, C., Cabezas, B., Veiga, M., 1984. Growth of the marinemicroalga Tetraselmis suecica in batch cultures with different salinities andnutrient concentrations. Aquaculture 42, 207–215.

Faria, G.R., Paes, C., Castro, D., Tinoco, N.A., Barbarino, E., Lourenço, S.O., 2012.Effects of the availability of CO2 on growth, nutrient uptake, and chemicalcomposition of the marine microalgae Chlorella sp. and Nannochloropsis oculata,two potentially useful strains for biofuel production. Int. Res. J. Biotechnol. 3,65–75.

Fon-Sing, S., Borowitzka, M., 2015. Isolation and screening of euryhaline Tetraselmisspp. suitable for large-scale outdoor culture in hypersaline media for biofuels. J.Appl. Phycol., 1–14

Guihéneuf, F., Stengel, D.B., 2013. LC-PUFA-enriched oil production by microalgae:accumulation of lipid and triacylglycerols containing n-3 LC-PUFA is triggeredby nitrogen limitation and inorganic carbon availability in the marinehaptophyte Pavlova lutheri. Mar. Drugs 11, 4246–4266.

Harun, R., Danquah, M.K., Forde, G.M., 2010. Microalgal biomass as a fermentationfeedstock for bioethanol production. J. Chem. Technol. Biot. 85, 199–203.

Jönsson, H.S.A., Vinnerås, B, Salomon, E., 2004. Guidelines on the use of urine andfaeces in crop production. EcoSanRes Publication Series Report 2004–2, ISBN:91 88714 94 2.

Kim, G., Bae, J., Lee, K., 2016. Nitrate repletion strategy for enhancing lipidproduction from marine microalga Tetraselmis sp. Bioresour. Technol.

Lee, T.-M., Tsai, C.-C., Shih, M.-C., 1999. Induction of phosphorus deficiency andphosphatase activity by salinity (NaCl) stress in Gracilaria tenuistipitata(Gigartinales, Rhodophyta). Phycologia 38, 428–433.

Lepage, G., Roy, C.C., 1984. Improved recovery of fatty acid through directtransesterification without prior extraction or purification. J. Lipid Res. 25,1391–1396.

Lourenço, S.O., Barbarino, E., Lavín, P.L., Lanfer Marquez, U.M., Aidar, E., 2004.Distribution of intracellular nitrogen in marine microalgae: calculation of newnitrogen-to-protein conversion factors. Eur. J. Phycol. 39, 17–32.

Lowry, O.H., Rosebrough, N.J., Farr, A.L., Randall, R.J., 1951. Protein measurementwith the Folin phenol reagent. J. Biol. Chem. 193, 265–275.

Mennaa, F.Z., Arbib, Z., Perales, J.A., 2015. Urban wastewater treatment by sevenspecies of microalgae and an algal bloom: biomass production, N and P removalkinetics and harvestability. Water Res. 83, 42–51.

Mohan, S.V., Devi, M.P., 2014. Salinity stress induced lipid synthesis to harnessbiodiesel during dual mode cultivation of mixotrophic microalgae. Bioresour.Technol. 165, 288–294.

Mutlu, Y.B., Isçk, O., Uslu, L., Koç, K., Durmaz, Y., 2013. The effects of nitrogen andphosphorus deficiencies and nitrite addition on the lipid content of Chlorellavulgaris (Chlorophyceae). Afr. J. Biotechnol. 10, 453–456.

Oswald, W., Gotaas, H., Golueke, C., Kellen, W., Gloyna, E., Hermann, E., 1957. Algaein waste treatment [with discussion]. Sewage Ind. Wastes 29, 437–457.

Pancha, I., Chokshi, K., Maurya, R., Trivedi, K., Patidar, S.K., Ghosh, A., Mishra, S.,2015. Salinity induced oxidative stress enhanced biofuel production potential ofmicroalgae Scenedesmus sp. CCNM 1077. Bioresource Technol. 189, 341–348.

Park, J., Craggs, R., Shilton, A., 2011. Wastewater treatment high rate algal ponds forbiofuel production. Bioresour. Technol. 102, 35–42.

Peccia, J., Haznedaroglu, B., Gutierrez, J., Zimmerman, J.B., 2013. Nitrogen supply isan important driver of sustainable microalgae biofuel production. TrendsBiotechnol. 31, 134–138.

Pereira, H., Barreira, L., Figueiredo, F., Custódio, L., Vizetto-Duarte, C., Polo, C., Rešek,E., Engelen, A., Varela, J., 2012. Polyunsaturated fatty acids of marinemacroalgae: potential for nutritional and pharmaceutical applications. Mar.Drugs 10, 1920–1935.

Pereira, H., Gangadhar, K.N., Schulze, P.S.C., Santos, T., Bruno de Sousa, C., Schueler,L.M., Custódio, L., Malcata, F.X., Gouveia, L., Varela, J.C.S., Barreira, L., 2016.Isolation of a euryhaline microalgal strain, Tetraselmis sp. CTP4, as a robustfeedstock for biodiesel production. Sci. Rep. 6, 35663. http://dx.doi.org/10.1038/srep35663.

Pomory, C.M., 2008. Color development time of the Lowry protein assay. Anal.Biochem. 378, 216–217.

Renaud, S., Parry, D., 1994. Microalgae for use in tropical aquaculture II: effect ofsalinity on growth, gross chemical composition and fatty acid composition ofthree species of marine microalgae. J. Appl. Phycol. 6, 347–356.

Ruangsomboon, S., Ganmanee, M., Choochote, S., 2013. Effects of different nitrogen,phosphorus, and iron concentrations and salinity on lipid production in newlyisolated strain of the tropical green microalga, Scenedesmus dimorphus KMITL. J.Appl. Phycol. 25, 867–874.

P.S.C. Schulze et al. / Bioresource Technology 223 (2017) 175–183 183

Ruiz, J., Álvarez-Díaz, P., Arbib, Z., Garrido-Pérez, C., Barragán, J., Perales, J., 2013a.Performance of a flat panel reactor in the continuous culture of microalgae inurban wastewater: prediction from a batch experiment. Bioresour. Technol.127, 456–463.

Ruiz, J., Arbib, Z., Álvarez-Díaz, P., Garrido-Pérez, C., Barragán, J., Perales, J., 2013b.Photobiotreatment model (PhBT): a kinetic model for microalgae biomassgrowth and nutrient removal in wastewater. Environ. Technol. 34, 979–991.

Santos, A., Janssen, M., Lamers, P., Evers, W., Wijffels, R., 2012. Growth of oilaccumulating microalga Neochloris oleoabundans under alkaline–salineconditions. Bioresour. Technol. 104, 593–599.

Schulze, P.S.C., Barreira, L.A., Pereira, H.G.C., Perales, J.A., Varela, J.C.S., 2014. Lightemitting diodes (LEDs) applied to microalgal production. Trends Biotechnol. 32,422–430.

Schulze, P.S.C., Pereira, H.G.C., Santos, T.F.C., Schueler, L., Guerra, R., Barreira, L.A.,Perales, J.A., Varela, J.C.S., 2016. Effect of light quality supplied by light emittingdiodes (LEDs) on growth and biochemical profiles of Nannochloropsis oculataand Tetraselmis chuii. Algal Res. 16, 387–398.

Sharma, K.K., Schuhmann, H., Schenk, P.M., 2012. High lipid induction in microalgaefor biodiesel production. Energies 5, 1532–1553.

Soucek, D., Dickinson, A., 2012. Acute toxicity of nitrate and nitrite to sensitivefreshwater insects, mollusks, and a crustacean. Arch. Environ. Con. Tox. 62,233–242.

Takagi, M., Yoshida, T., 2006. Effect of salt concentration on intracellularaccumulation of lipids and triacylglyceride in marine microalgae Dunaliellacells. J. Biosci. Bioeng. 101, 223–226.

Tang, E.P., Vincent, W.F., Proulx, D., Lessard, P., de la Noüe, J., 1997. Polarcyanobacteria versus green algae for tertiary waste-water treatment in coolclimates. J. Appl. Phycol. 9, 371–381.

US-EPA. 2013. Wastewater Treatment Fact Sheet: External Carbon Sources forNitrogen Removal. United States Environmental Protection Agency.

Vaccari, D.A., 2009. Phosphorus: a looming crisis. Sci. Am. 300, 54–59.Valderrama, L.T., Del Campo, C.M., Rodriguez, C.M., de-Bashan, L.E., Bashan, Y., 2002.

Treatment of recalcitrant wastewater from ethanol and citric acid productionusing the microalga Chlorella vulgaris and the macrophyte Lemna minuscula.Water Res. 36, 4185–4192.

Zeller, M.A., Hunt, R., Jones, A., Sharma, S., 2013. Bioplastics and their thermoplasticblends from Spirulina and Chlorella microalgae. J. Appl. Polym. Sci. 130, 3263–3275.

Zhou, Y., Oehmen, A., Lim, M., Vadivelu, V., Ng, W.J., 2011. The role of nitrite and freenitrous acid (FNA) in wastewater treatment plants. Water Res. 45, 4672–4682.

Zhou, W., Li, Y., Min, M., Hu, B., Zhang, H., Ma, X., Li, L., Cheng, Y., Chen, P., Ruan, R.,2012. Growing wastewater-born microalga Auxenochlorella protothecoidesUMN280 on concentrated municipal wastewater for simultaneous nutrientremoval and energy feedstock production. Appl. Energ. 98, 433–440.