15 Judd Et Al, Wat Res (Algae)

7/26/2019 15 Judd Et Al, Wat Res (Algae)

1/12

See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/281291731

Algal remediation of CO2 and nutrientdischarges: A review

ARTICLE in WATER RESEARCH AUGUST 2015

Impact Factor: 5.53 DOI: 10.1016/j.watres.2015.08.021

CITATION

1

READS

90

5 AUTHORS, INCLUDING:

Simon jon Judd

Cranfield University; Qatar University

172PUBLICATIONS 5,227CITATIONS

SEE PROFILE

Hussein Znad

Curtin University

40PUBLICATIONS 587CITATIONS

SEE PROFILE

Available from: Simon jon Judd

Retrieved on: 08 February 2016
https://www.researchgate.net/profile/Simon_Judd2?enrichId=rgreq-fcdc016c-932c-490a-80ae-05327a151375&enrichSource=Y292ZXJQYWdlOzI4MTI5MTczMTtBUzoyODM2ODIzMzkzNDQzODlAMTQ0NDY0NjU0ODk3Ng%3D%3D&el=1_x_4https://www.researchgate.net/profile/Simon_Judd2?enrichId=rgreq-fcdc016c-932c-490a-80ae-05327a151375&enrichSource=Y292ZXJQYWdlOzI4MTI5MTczMTtBUzoyODM2ODIzMzkzNDQzODlAMTQ0NDY0NjU0ODk3Ng%3D%3D&el=1_x_5https://www.researchgate.net/profile/Simon_Judd2?enrichId=rgreq-fcdc016c-932c-490a-80ae-05327a151375&enrichSource=Y292ZXJQYWdlOzI4MTI5MTczMTtBUzoyODM2ODIzMzkzNDQzODlAMTQ0NDY0NjU0ODk3Ng%3D%3D&el=1_x_5https://www.researchgate.net/?enrichId=rgreq-fcdc016c-932c-490a-80ae-05327a151375&enrichSource=Y292ZXJQYWdlOzI4MTI5MTczMTtBUzoyODM2ODIzMzkzNDQzODlAMTQ0NDY0NjU0ODk3Ng%3D%3D&el=1_x_1https://www.researchgate.net/profile/Hussein_Znad?enrichId=rgreq-fcdc016c-932c-490a-80ae-05327a151375&enrichSource=Y292ZXJQYWdlOzI4MTI5MTczMTtBUzoyODM2ODIzMzkzNDQzODlAMTQ0NDY0NjU0ODk3Ng%3D%3D&el=1_x_7https://www.researchgate.net/institution/Curtin_University2?enrichId=rgreq-fcdc016c-932c-490a-80ae-05327a151375&enrichSource=Y292ZXJQYWdlOzI4MTI5MTczMTtBUzoyODM2ODIzMzkzNDQzODlAMTQ0NDY0NjU0ODk3Ng%3D%3D&el=1_x_6https://www.researchgate.net/profile/Hussein_Znad?enrichId=rgreq-fcdc016c-932c-490a-80ae-05327a151375&enrichSource=Y292ZXJQYWdlOzI4MTI5MTczMTtBUzoyODM2ODIzMzkzNDQzODlAMTQ0NDY0NjU0ODk3Ng%3D%3D&el=1_x_5https://www.researchgate.net/profile/Hussein_Znad?enrichId=rgreq-fcdc016c-932c-490a-80ae-05327a151375&enrichSource=Y292ZXJQYWdlOzI4MTI5MTczMTtBUzoyODM2ODIzMzkzNDQzODlAMTQ0NDY0NjU0ODk3Ng%3D%3D&el=1_x_4https://www.researchgate.net/profile/Simon_Judd2?enrichId=rgreq-fcdc016c-932c-490a-80ae-05327a151375&enrichSource=Y292ZXJQYWdlOzI4MTI5MTczMTtBUzoyODM2ODIzMzkzNDQzODlAMTQ0NDY0NjU0ODk3Ng%3D%3D&el=1_x_7https://www.researchgate.net/profile/Simon_Judd2?enrichId=rgreq-fcdc016c-932c-490a-80ae-05327a151375&enrichSource=Y292ZXJQYWdlOzI4MTI5MTczMTtBUzoyODM2ODIzMzkzNDQzODlAMTQ0NDY0NjU0ODk3Ng%3D%3D&el=1_x_5https://www.researchgate.net/profile/Simon_Judd2?enrichId=rgreq-fcdc016c-932c-490a-80ae-05327a151375&enrichSource=Y292ZXJQYWdlOzI4MTI5MTczMTtBUzoyODM2ODIzMzkzNDQzODlAMTQ0NDY0NjU0ODk3Ng%3D%3D&el=1_x_4https://www.researchgate.net/?enrichId=rgreq-fcdc016c-932c-490a-80ae-05327a151375&enrichSource=Y292ZXJQYWdlOzI4MTI5MTczMTtBUzoyODM2ODIzMzkzNDQzODlAMTQ0NDY0NjU0ODk3Ng%3D%3D&el=1_x_1https://www.researchgate.net/publication/281291731_Algal_remediation_of_CO2_and_nutrient_discharges_A_review?enrichId=rgreq-fcdc016c-932c-490a-80ae-05327a151375&enrichSource=Y292ZXJQYWdlOzI4MTI5MTczMTtBUzoyODM2ODIzMzkzNDQzODlAMTQ0NDY0NjU0ODk3Ng%3D%3D&el=1_x_3https://www.researchgate.net/publication/281291731_Algal_remediation_of_CO2_and_nutrient_discharges_A_review?enrichId=rgreq-fcdc016c-932c-490a-80ae-05327a151375&enrichSource=Y292ZXJQYWdlOzI4MTI5MTczMTtBUzoyODM2ODIzMzkzNDQzODlAMTQ0NDY0NjU0ODk3Ng%3D%3D&el=1_x_2


2/12

Review

Algal remediation of CO2 and nutrient discharges: A review

Simon Judd a,c, *, Leo J.P. van den Broeke a, Mohamed Shurair a , Yussuf Kuti a,Hussein Znad b

a Department of Chemical Engineering, Qatar University, Qatarb Department of Chemical Engineering, Curtin University, Australiac Craneld Water Science Institute, Craneld University, UK

a r t i c l e i n f o

Article history:

Received 25 April 2015

Received in revised form

4 July 2015

Accepted 10 August 2015

Available online 28 August 2015

Keywords:

Algae

Photobioreactor

CO2Nutrients

Wastewaters

a b s t r a c t

The recent literature pertaining to the application of algal photobioreactors (PBRs) to both carbon dioxidemitigation and nutrient abatement is reviewed and the reported data analysed. The review appraises the

inuence of key system parameters on performance with reference to (a) the absorption and biologicalxation of CO2 from gaseous efuent streams, and (b) the removal of nutrients from wastewaters. Keyparameters appraised individually with reference to CO2 removal comprise algal speciation, light in-tensity, mass transfer, gas and hydraulic residence time, pollutant (CO2 and nutrient) loading,

biochemical and chemical stoichiometry (including pH), and temperature. Nutrient removal has beenassessed with reference to hydraulic residence time and reactor conguration, along with C:nutrient

ratios and other factors affecting carbon xation, and outcomes compared with those reported forclassical biological nutrient removal (BNR).

Outcomes of the review indicate there has been a disproportionate increase in algal PBR researchoutputs over the past 5e8 years, with a signicant number of studies based on small, bench-scale sys-

tems. The quantitative impacts of light intensity and loading on CO 2uptake are highly dependent on thealgal species, and also affected by solution chemical conditions such as temperature and pH. Calculationsbased on available data for biomass growth rates indicate that a reactor CO 2residence time of around 4 h

is required for signicant CO2removal. Nutrient removal data indicate residence times of 2e5 days arerequired for signicant nutrient removal, compared with


3/12

2.4. Biochemical and chemical stoichiometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 361

3. Nutrient abatement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 362

4. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 364

Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 364

References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 364

1. Introduction

1.1. Algae for carbon dioxide mitigation

Mitigation of carbon dioxide through its capture, utilisation and

storage has undergone rapid development over the past 20 years,with research and development originally precipitated by therealisation of the impact of CO2as the single largest contributor toglobal warming (Hoyt, 1979). Various measures exist for CO2 miti-

gation generally and utilisation specically (Fig. 1), includingenhanced oil and gas recovery (EOR and EGR, the latter includingcoal-bed methane e ECBM), CO2conversion to chemical feedstock

and fuels, biological conversion (photosynthesis), and CO2 miner-alisation for the production of materials (Laumb et al., 2013; Hasanet al., 2014). These methods are primarily focussed on CO2 uti-lisation following capture; only biological conversion is capable ofdirect CO2 mitigation. Utilisation of CO2 as a feedstock for other

production processes, however, offers opportunities to offset partof the signicant capital investment associated with capturing theCO2.

The use of algae for CO2capture and utilisation offers a number

of benets over alternative methods for CO2mitigation. Firstly, the

method is inherently efcient and sustainable, analogous to con-

ventional biological wastewater treatment, since the biologicalprocess requires only the food source (the carbon) and ambienttemperatures and daylight to be sustained. The main product, thealgal biomass, has a market valueand canalso be reused for biofuel,

including biodiesel, biomethane and biohydrogen (Brennan andOwende, 2010; Scott et al., 2010; Cho et al., 2011 ), animal feed(Chauton et al., 2015) or other high-value products (Borowitzka,2013; Lopes da Silva et al., 2014). The latter include proteins and

various types of pigment (chlorophyll, carotenoids), and productsof a signicant global market size such as fatty acids. Indeed, it hasbeen noted (Lundquist et al., 2010; Batten et al., 2013) that the

economic case for PBR technology relies largely on the cost bene

toffered by the generation of these high-value products: PBRsappear to be uneconomical, even under the most favoured condi-tions, solely for pollutant removal from aqueous and gaseous wastestreams (Acien Fernandez et al., 2012a,b).

PBR technology has other attractive features. The reactor isrelatively uncomplicated e at its most basic level simply a pondsystem e is robust to changes in CO2 load and is fully scale-able. Theprocess technology design is exible, can use almost any source of

CO2and can be integrated and/or combined with other processes e

Fig. 1. CO2 utilisation, adapted fromLaumb et al. (2013).

S. Judd et al. / Water Research 87 (2015) 356e366 357
http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-


4/12

including wastewater treatment for organic carbon and nutrient

removal. Against this, the relatively slow rate of CO2 assimilation(compared with conventional biological treatment processes for

organic carbon and nutrient removal from municipal wastewaters)means that comparatively large land areas are required.

1.2. Algae for nutrient removal

The growing of algae from a municipal wastewater feed fortreatment purposes was investigated as early as the 1950s(Oswald et al., 1953), with the concept of using wastewater as a

medium for algae-based biofuel production reported in theseminal close-out report for the Aquatic Species Program (ASP)conducted from 1978 to 1996 (Sheehan et al., 1998). The use ofalgae for mitigation of the nutrients phosphate, nitrate and

ammonia in wastewater treatment has been also the subject ofstudy since around the mid-1970's (Bosch et al., 1974; Yun et al.,1977) as a means of combatting eutrophication (Gavrilescu andChisti, 2005; Liang et al., 2013). Whilst there are establishedbiological methods for nitrogen and phosphorus (N and P)

removal from wastewater, so-called biological nutrient removal(BNR), this classically demands supplementary sludge transferbetween aerobic, anoxic and anaerobic regions (Van Loosdrechtet al., 1997; Mulkerrins et al., 2004). More recently, however,

BNR has been demonstrated in a single process step through thedevelopment of very specic biochemical conditions through

extensive acclimation and rigorous process control (Daigger and

Littlejohn, 2014). There is nonetheless often an additionalrequirement for chemical dosing with iron or aluminium-based

coagulants to obtain the required P removal (De Gregorio et al.,2010; Li and Brett, 2012). The use of PBRs for the duty ofnutrient removal provides an economical and environmentally

sustainable alternative, combined as it is with bioenergy and bio-products production and CO2 mitigation (Sheehan et al., 1998;Clarens et al., 2010; Zhou et al., 2011, 2012a).

1.3. Research trends in algae

An indication of the relative scientic importance of the twodifferent aspects of algal PBRs can be surmised through the use ofsearch engines for examining scientic publications databases,

such as SCOPUS and Web of Knowledge. Searches of keywordsappearing in such databases for search terms based on algae(including micro-algae), water (including wastewater) and carbondioxide (including CO2 and ue gas) can be used to identify the

number of relevant papers.A consideration of all research papers dating back to the mid-

1960s reveals research articles encompassing water and algae(water algae) to be about twice as numerous and those based onwater carbon dioxide and ten times more in number than algae carbon dioxide (Fig. 2a). The application of algae bioreactors forcarbon capture appears to be a relatively recent area of study, with

Fig. 2. Summary of research articles in key subject areas: (a) total, and (b) by year, according to SCOPUS, based on the search terms ofalga(includingmicro-alga and microalga),

water(includingwastewaterand efuent) and carbon dioxide(including CO2 and ue gas).

Fig. 3. Research topics associated with algae

carbon dioxide

water articles (from SCOPUS).

S. Judd et al. / Water Research 87 (2015) 356e366358
http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-


5/12

a concerted research effort only evident from 1990 onwards(Fig. 2b). Whilst the number of research articles whose keywordsencompass all three topics are small compared to those for theindividual sets, the publication rate appears to be rapidly

increasing. The number of research articles relating to algae, waterand CO2as discrete topics have all increased at a compound annualgrowth rate of 6.4e6.7% per year on average since the mid-1960s,with those based on water unsurprisingly far outnumbering those

focused on the other two subject areas (Fig. 2a). Those encom-passing either all three topics (algae carbon dioxide water) or

just algae carbon dioxide, whilst much smaller in number haveboth increased at a growth rate of 25e30% per year since around

2007 (Fig. 2b) e a growth rate four times higher than that of theindividual sets.

An indication of the primary topics of interest within the 321-fold papers based on all three topics can be provided by a Wordlediagram (www.wordle.com) constructed from the keywords of thearticles (Fig. 3). The key search terms (Fig. 2) were excluded fromthe Wordle analysis, and the words manually normalised ( Santos

et al., 2011) as follows:

removal of all upper-case letters; conversion of plurals to singular;

aggregation of all types of PBR conguration into a single term

photobioreactor;

aggregation of all types of strains of an algal species, e.g.

Chlorella vulgaris or Chlorella sp., into a single term, e.g.

chlorella;

delineation of the terms bioenergy, biogas, biodieseland

biofuel.

According toFig. 3published work has been based primarily onthe Chlorella genusand onPBRs,as observedin recent reviewsof algalbiomass production and CO2 xation (Ho et al., 2011; Zhao and Su,

2014; Zeng et al., 2015). There has also been a preponderance ofbiofuel-related papers in this area(Brennan and Owende, 2010; Scottet al., 2010; Lopes da Silva et al., 2014), underlining one of the keyattractions of algal-based mitigation technologies. Chlorella is fav-

oured due to its high growth rate (or productivity) e up to 1.2 g/(L.d)under optimum conditions (Cheng et al., 2006; Chiu et al., 2008) eandability to assimilate CO2 at relatively high concentrations e upto100%, according to Concaset al., 2012. However, biologicalxation of

CO2 is highlydependenton operatingconditions such as CO2 loading,pH, temperature, light intensity, and medium composition,with rsttwo of these being inter-related (Section2.4).

Various algal reactor technologies have been investigated, with

designs based on either open or closed systems (Table 1) and with

various congurations of the latter (Fig. 4). The general trend is forincreasing intensivity with increasing complexity of design and/or

ow channels. Designs have included revolving systems (Gross and

Wen, 2014), air-lift (Cattaneo et al., 2003; Pirouzi et al., 2014) andmembrane-sparged systems (Fan et al., 2008), all ostensiblydesigned to improve CO2 mass transfer (Section 2.3) and thusproductivity.

The PBR process may then be operated in either batch orcontinuous mode, with the algal biomass being recovered as auseful product. As with conventional sewage treatment, biomass-water separation then takes place either by simple sedimenta-

tion, the predominantly preferred method (Milledge and Heaven,2013; Sirin et al., 2013), or occasionally by membrane separation(Gao et al., 2015; Drexler and Yeh, 2014; Marbelia et al., 2014). For

steady-state systems the hydraulic and solids (or biomass) reten-tion times, and thus the algal biomass concentration in the reactor,can be controlled prior to harvesting of the algae, which accountsfor 20e30% of the total costs (Barros et al., 2015). However, the

research has predominantly been based on batch systems.It is of interestto assessthe stateof the art of algal PBRtechnology

as it relates to both carbon capture and water treatment, and specif-ically nutrient removal. Thus far the precise facets required of the

technology for accomplishing these two key aims have not beensummarised in a single review, and yet evidence suggests (Fig. 2b)that these two applications have been of increasing signicance inrecent years, with interest possibly originally precipitated by land-

mark international legislation, such as the 1992 United NationsFramework Convention on Climate Change (UNFCCC) and the sub-

sequent 1997 Kyoto Protocol. Publications of interest can be roughlydividedintothosebased solely on carbonand thoseaiming topurify a

wastewater stream. Although there are additionally a signicantnumber of papers focused on the production of biofuel and otherhigh-value products, particularly since the turn of the decade(Brennan and Owende, 2010; Scott et al., 2010; Borowitzka, 2013;

Markou and Nerantzis, 2013; Lopes da Silva et al., 2014; Chautonet al., 2015), this areawas considered outside thescopeof this review.

2. Carbon capture

The retention of carbon dioxide in a reactor is dependent on (a)mass transfer of the CO2 from the gas to liquid phase and (b)assimilation of theCO2 by the algae, with eitherone or both of these

parameters being a function of the light intensity, CO2 loading,

Fig. 4. Algal PBR system congurations.

Table 1

PBR system facets, adapted fromSudhakar et al., 2011andBermudez et al., 2014.

Parameter Open Closed

Design complexity Lower Higher

Control Poor Good

Cost Lower Higher

Water losses High Low

Typical biomass concentration Low, 0.1e0.2 g/l High: 2e8 g/l

Temperature control Dif cult Easily controlled

Species control Dif cult Simple

Contamination High risk Low risk

Light utilisation Poor Very high

CO2losses to atmosphere High (up to 38%a) Almost none

Typical growth rate (g/m2/day ) Lo w: 10e25 Variable:1e500Area requirement Large Smaller

Depth/diameter of water 0.3 m 0.1 m

Surface: volume ratio ~6 60e400

Cleaning None Required

Bimass quality Variable Reproducible

Harvesting efciency Low High

Harvesting cost Higher Lower

Mo st co stly operating functio n Mixing Oxygen and

temperature control

Hydrodynamic str ess on algae Very low Low- moderateGas transfer control Low High

a

Douchal et al., 2005.

http://-/?-http://-/?-http://-/?-http://www.wordle.com/http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://www.wordle.com/http://-/?-http://-/?-http://-/?-


6/12

biomass concentration and volume, biomass retention time, algal

species, and solution chemistry (and specically the pH and tem-perature). The system therefore has numerous variables, and in-dividual experimental studies have not always provided all the

relevant system parameter values.Studies where a mass balance has been conducted (e.g.Chiang

et al., 2011) indicate that most of the CO2 uptake is assimilated asalgal cells rather than unbound organics or extracellular polymeric

substances. As a result, CO2 uptake is generally determined solelyfrom biomass generation through biochemical stoichiometry(Section2.3), rather than through determination of CO2mass owacross the system. CO2 uptake as a proportion of the supplied CO2 is

largely dependent on algal growth rate, such that in practice thereis requirement for sufcient reactor capacity (in terms of the overallCO2 retention time) for CO2 assimilation. Many studies are based on

single, bench-scale reactors of a few 100 mL volume (e.g. Tang et al.,2011), such that only a small percentage of the CO 2 is removed.However, studies conducted on larger-scale batch systems wherethe volume provided is sufcient for more signicant capture,either for large sealed systems (Li et al., 2013) multi-stage reactors

(Lam and Lee, 2013; Cheng et al., 2013) and/or reactors with recycle

ows (Lam and Lee, 2013), suggest that high removals are attain-able (Table 2). There is nonetheless currently a general paucity ofpilot and demonstration scale programmes exploring the most

encouraging of the bench-scale ndings regarding the mostpromising of the algal strains identied at bench scale with mini-mal overall % CO2 removal.

Table 2 indicates that biomass production is generally in the

range 0.26e0 . 7 g L 1 d1 formoderateto high (2e20%) feed gas CO2concentrations. Applying a mass ratio of 1.9:1 CO

2:biomass carbon

(Section 2.4) implies that the percentage carbon dioxide xed in the

biomass for operation at room temperature is given by:

%F100% 110PV=CQ 100% 110Pt=C (1)

wherePis the biomass productivity in g L1 d1,Vis the operatingreactor volume,Cis the %CO2in the feed gas stream,Qthe gas owrate in L d1 and t the gas residence time in days. Thus, for 100%removal, a moderate-to-high productivity of 0.5 g L1 d1 (Table 3)

and a feed CO2concentration of 10%, t 10/(110 0.5)0.18 d, or4.4 h. Productivity is also a function of feed gas CO2concentration,though trends do not appear to be consistent across all studies(Table 2).

Options for enhancing process intensivity include increasinglight intensity, enhancing mass transfer and adjusting the chemicalconditions. These are each considered in turn below.

2.1. Light intensity

Data for CO2 xation associated with specic light intensities, asprovided by a number of authors for a range of algal species, issomewhat varied (Fig. 5). The data reveals no overall pattern be-tween CO2 xation rates and light intensity either across different

algal species or across different studies for the same algal species(e.g.Chlorella vulgarisor Anabaenasp.). On the other hand, withinindividual studies under the same controlled conditions (Table 3) itis evident that there is the expected increase in xation and

biomass productivity with light intensity and/or exposure, untilreaching a maximum associated with light saturation (Chiang et al.,2011; Sanchez-Fernandez et al., 2012; Ho et al., 2012; Gonalves

et al., 2014). Batch tests conducted on four different algal species(C. vulgaris, P. subcapitata, Scenedesmus. salina, and M. aeruginosa)

Table 2

Calculated and reported CO2 xation data, 2010 onwards.

Algal species CCO2% Qg, mL/min P, g/L/d V, L % CO2xed Note Ref

Scenedesmus obliquus 10 200 0.29 0.2 0.002 Tang et al., 2011

Chlorella pyrenoidosa 10 200 0.26 0.2 0.002 Tang et al., 2011

Chlorella vulgaris 5 0.16e0.8 25 2ee12 1 Lam and Lee, 2013

Chlorella PY-ZU1 15 30 0.95 0.3e4.2 2e86 3 Cheng et al., 2013

Chlorella vulgaris 15 25e50 0.24e0.35 8 36e56 2 Li et al., 2013

Scenedesmussp. 10.6 100,000 0.43 20,000 66 4 De Godos et al., 2014Botryococcus braunii 5 n.a. 0.50 8 88 Sydney et al., 2010

Chlorella vulgaris 5 n.a. 0.25 8 87 Sydney et al., 2010

Dunaliella tertiolecta 10 n.a. 0.27 8 80 Sydney et al., 2010

Anabaenasp. 5e15 0.04* 0.65e0.8 5 90 Chiang et al., 2011

Spirulina platensis 2.5 200 0.99 1 n.a 5 Chen et al., 2013

CCO2%CO2in feed gas;PBiomass production rate;QgFeed gas ow rate;VReactor volume;% CO2xed100(1-CCO2out/CCO2in). *vvm e volume of gasow per minute per volume of

liquid.

Notes.1. multi-stage w. recycle.

2. closed raceway pond.

3. multi-stage.4. raceway pond, 100 m2, 0.4 m deep.

5. at-type photobioreactor.

Table 3

Reported CO2 xation rates, Anabaena sp.

Lightintensity

mmol m2 s1

CO2 xn.Rate g L1 d1

HRT, d Max. biomassconcn, g L1

Inlet CO2%v/v Flow ratevvm

g CO2gbiomass1 d1

Refs

900 1.45 2e3 3 0.03* 0.2 0.48 Sanchez-Fernandez, 2009

0e460 0.43 3.3 0.76 10.6 ~3 104 ~1 De Godos et al., 2014

250 0.65e0.8 5 0.58e1.2 5e15, 10 0.04 0.67e1.12 Chiang et al., 2011

650 0.16e0.58 0.7e6 0.35e0.95 0.03* 0.13e0.75 0.17e1.7 Sanchez-Fernandez et al., 2012

975 0.25e0.65 0.7e6 0.45e1.35 0.03* 0.13e0.75 0.18e1.44

1625 0.36e1 0.7e6 0.5e2 0.03* 0.13e0.75 0.18e2

*Atmospheric level; HRT hydraulic residence time; vvm volume gas per volume liquid per minute.

Most common data underscored.

http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-


7/12

suggest that an approximate trebling of light intensity (from36 mmol m2 s1) provides a 70e90% increase in growth rate and a35e45% increase in biomass productivity and CO2 uptake(Gonalves et al., 2014). However, further increases in light in-

tensity may then inhibit and diminish the CO2 xation rate andbiomass productivity (Ho et al., 2012).

2.2. Hydraulic residence time (HRT) and loading

Quantitative trends in CO2 uptake and associated productivityfor continuous reactors are highly dependent on both the HRT andCO2loading. Decreasing the HRT, whilst detrimental to the biomassconcentration, has nonetheless been shown to producea maximum

in CO2 xation rate and biomass productivity (Sanchez-Fernandezet al., 2012) whilst generally being detrimental to removal ofliquid-based contaminants such as nutrients (Section 3). Similarly,the percentage CO2 xation appears to reach a maximum with CO2specic loading (i.e. the mass ow rate of CO2) when loading ischanged either by increasing the ow rate (Kargupta et al., 2015) orfeed concentration (Chiang et al., 2011), according to batch reactor

studies. Since this trend has been reported for three differentspecies (Chlorella pyrenoidosa and Scenedesmus abundans by Kar-gupta et al., and Anabaena sp. by Chiang et al.), it is apparentlyindependent of microbiology and instead must presumably relate

to some physicochemical facet of the system.However, it is rare that all process parameter values inuencing

CO2 capture have been reported. Many studies are based on batchsystems, such that the impacts of the harvesting of the algaeproduct and/or the recovery of the water are not evident, yet it hasbeen demonstrated that the HRT impacts on CO2uptake (Sanchez-

Fernandez et al., 2012). Similarly, batch operation implies thatsteady-state conditions are not always reached, which has possibleimplications regarding the overall biochemical and chemical stoi-chiometry and specically the pH-dependent carbonate equilibria

(Section2.4).

2.3. Mass transfer

Reported mass transfer coefcient (kLa) values for recent studies

(Table 4) vary signicantly, as expected, according to the systemhydrodynamics. Whilst CO2 uptake has been reported to increasewith increased mass transfer coefcient (Fan et al., 2008), it isunclear as to whether a full-scale process is likely to be mass

transfer limited, given the length of the total CO2eliquid contacttime required for biomass growth. Mass transfer is more critical inconventional aerobic treatment, of industrial wastewaters inparticular, due to the higher carbon loads and the lower oxygen

solubility compared with CO2.

2.4. Biochemical and chemical stoichiometry

The general chemical formula of biomass, COmHnNoPp, takes

values of 0.242e0.485, 1.65e2.11, and 0.110e0.159 for m, n and orespectively, with p being around 0.1 for algal biomass ( Tsygankovet al., 2002; Cheng et al., 2006; Ho et al., 2011; Chiang et al., 2011;

Concas et al., 2012; Zhao and Su, 2014). There is then a 1:1 stoi-chiometric ratio of CO2 carbon to algal carbon. Values of m-p appearto be comparable with those for biomass associated municipalwastewater treatment (Fig. 6). However, with wastewater treat-

ment a complete carbon mass balance can be conducted bycomparing the decrease in organic carbon substrate expressed asbiochemical or chemical oxygen demand (BOD and COD) across thesystem with the increase in biomass generated. For algal bio-

reactors a mass balance can only be achieved by monitoring the

Fig. 5. Reported CO2 xationrates for variousalgal species (A.N. Aphanothece microscopica Nageli, A.s.Anabaena sp., C.v. Chlorella vulgaris, Cm.s. Chlorococcum sp.,D.s. Dunaliella salina,

S.s. Synechocystis sp., C.s. Chlorella sp. (Bhola et al., 2011; Chai et al., 2012; Douskova et al., 2009; Jacob-Lopes et al., 2009; Kim et al., 2012; Martinez et al., 2012; Ryu et al., 2009).

Table 4

Reported mass transfer values for PBRs.

Reactor conguration kLa, h1 Reference

External loop airlift 17e24 Pirouzi et al., 2014Membrane-sparged tubular reactor 250e430 Fan et al., 2008

Coarse bubble sparged reactor 20e65 Fan et al., 2008

Membrane cont actor react or 2.5e30 Fan et al., 2008Tube 18 Fernandez et al., 2012

Column up to 23 Cervantes et al., 2013

Raceway pond up to 9.6 Li et al., 2013

http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-


8/12

inlet and outlet CO2 concentrations. Whilst this has been conducted

by some authors (Jacob-Lopes et al., 2010; Chiang et al., 2011;Kargupta et al., 2015) it is more usual for CO2 xation to be infer-red from biomass production alone on the basis of, according todata in Fig. 6, the carbon contributing ~50% of the biomass by

weight. From the 1:1 stoichiometry, this infers a weight ratio of~2:1 CO2:biomass, with a value of 1.88 often chosen (Chisti, 2007).Against this, a review of reported algal carbon content byVan DenHende et al. (2012)conducted across a number of different genera

revealed this parameter to vary between 36% for Dunaliella tertio-

lecta (Sydney et al., 2010) to 65% (Chae et al., 2006) for Euglenagracilis.

Determining CO2 uptake from biochemical stoichiometry isacceptable provided that the assumed biostoichiometryapplies and

that the net loss is entirely by assimilation. CO2 dissolution ordesorption on the other hand is evidenced by a change in pH, asimplied by carbonate equilibria (Brezonik and Arnold, 2011)

wherein the pH changes with the carbon dioxide to bicarbonate(HCO3

e) ratio in accordance with the equation:

pHlog

hHCO3

i

CO2 6:38 (2)

This then demands that pH is monitored to allow the distinctionbetween CO2 uptake by assimilation and by dissolution, particularlyin batch systems for which it has been demonstrated (Lam and Lee,

2013; Kargupta et al., 2015) that the solution carbonate concen-tration impacts signicantly on CO2uptake.

3. Nutrient abatement

Algal-based PBRs offer a direct alternative to classical BNR.

Given that both are biological processes, the key contributing fac-tors with reference to their respective efcacies are (a) % nutrientremoval, (b) retention time, (c) specic energy demand, (d) wastegeneration, and (e) the requirement for ancillary operations or

consumables. BNR is integrated with an aerobic process whichprovides organic carbon removal but demands signicant energyfor process aeration of the aerobic tank. The PBR process, on theother hand, provides carbon dioxide sequestration and also added

value through the end algal products, but COD removal may vary

from >90% (Zhou et al., 2012a) to almost none (Arbib et al., 2013)

depending largely on the food:microorganism (F:M) ratio. A fullcost benet analysis is therefore challenging and very sensitive toassumptions made concerning the algal biomass processing andend product value. On the other hand, comparisons can be madebased solely on the wastewater treatment technology for contin-

uous systems (including semi-continuous technologies such as thesequencing batch reactor, SBR).

There has been signicant interest in the application of PBRs tonutrient abatement in recent years (Table 5) encompassing a vari-

ety of municipal wastewaters of various strengths, from secondaryefuent (Gao et al., 2015; Arbib et al., 2013) to primary clarier

efuent (Sutherland et al., 2014a,b,c,d), and anaerobic digester su-pernatant (Lee et al., 2015; Zhou et al., 2012a,b). According to this

data, the mean hydraulic retention time (HRT), when reported,generally ranges between 2 and 5 days largely irrespective oftechnology conguration. This compares to total HRTs in the region

of 7e15 h (Table 6) for the BNR process, of which 30e70% is asso-ciated with the anaerobic (An) and anoxic (Ax) zones required forphosphorus (P) removal and denitrication (or nitrate removal)respectively. Nutrient removal is dependent on a number of pa-

rameters, including the nutrient balancing (the P:N:C ratio), thedissolved oxygen (DO) concentration in the different zones, the pH,and the temperature. Whilst these parameters have been widelyexplored for the BNR process, the key parameter of nutrient

balancing appears to have largely overlooked in PBRs but has been

shown to signi

cantly increase N removal (Michels et al., 2014).Algal PBR performance in terms of N and P removal appearscomparable with that of the classical BNR process for a fully opti-

mised process. However, ranges reported are much more scatteredfor the PBR process, with N and P removals as low as 47% and 12% Nand P removal respectively reported (Table 5) compared with cor-responding values of 73% and 67% for the BNR process (Table 6). For

the most germane direct comparison between the BNR(Vaiopoulou and Aivasidis, 2008; Puig et al., 2008; Liu et al., 2008)and the HRAP (Sutherland et al., 2014a,b,c,d) suspended growthprocesses challenged with municipal wastewater, the respective

removal ranges are 73e87% N/67e98% P for the BNR vs. 59e79% N/12e79% P forthe HRAP. The corresponding HRT values are 6.6e15 hvs. 2e9 days. The algal process is therefore up to 15 times slowerand is less robust in removing nutrient than the classical BNR one.

Fig. 6. Stoichiometric ratios of elements in biomass, averaged values from published data ( Rittmann and McCarthy, 2001; Tsygankov et al., 2002; Cheng et al., 2006; Ho et al., 2011;

Chiang et al., 2011; Concas et al., 2012; Zhao and Su, 2014 ).

http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-


9/12

Moreover, a classical BNR plant employs a tank depth of ~5 m,

compared to

3 g/L, and between 8 and 15 g/L for a membranebioreactor, Judd, 2014) combined with the concomitant slowerbiokinetics of the algal system. Against this, very high removals

Table 5

PBR nutrient removal papers from 2012 onwards, predominantly continuous systems.

Species Wastewater(municipal)

Technology SS, g/L Product-ivity

g/m3/d

TPinmgL1 TNinmgL

1 TPoutmgL1 TNout

mgL1%N rem %P rem HRT, d Reference NB

S.c. Secondary Tub A-L 0.6e0.8 18e21 1.6e2.3 24e29 0.1e0.5 0.2e4 86e95 69e94 5 Arbib et al.,

2013

S.c. Secondary HRAP 0.2e0.3 5e8 1.6e2.3 24e29 0.5e1.1 5e12 62e77 51e63 10 Arbib et al.,

2013C.v. Secondary, anal. BMPBR 1.37 72 0.8 15 0.35 0.02 2.6 0.6 83 4 86 2 2 Gao et al.,

2015

C.v. Secondary, anal. MPBR 0.95 50 0.8 15 0.34 0.01 5.3 1.0 64 6 85 3 2 Gao et al.,

2015

- Primary, settled HRAP 0.10e0.26a 15e48 0.9e3.6 20e31 0.7e2.1 4e14 47 e79 20e49 5.5e9 Sutherland

et al., 2014a

a

P.b.,D.o. Primary, settled HRAP 0.27 nr 4 0.2 35 5 2.9e3.4 12e15 56e67 15e28 2 Sutherland

et al., 2014b

b

M.p. Primary, settled HRAP 0.18e0. 23 14e17 3.2e6.3 20e40 0.7e2.7 5e12 74e75 58e79 4 Sutherland

et al., 2014c

M.p. Primary, settled HRAP 0.095e0.19 nr 4.6e7.2 35e54 e e 59e79 12e34 4e9 Sutherland

et al., 2014d- Primary, settled Biolm ~0.3 2 1b 7 3 66 16 1.1e1.7 13e25 70 8 85 9 3.1e5.2 Posadas

et al., 2013

c

C.p. Primary, settled P. plate 0.35e0.8 42e60 8 56 e - 72 92 0.64 Ramos-Tercero

et al., 2014

C.v. Anal. Col MPBR 0.2

e

0.75 60 max 1.7e

2.2 7.5e22 e

- >95

max >

95 max 2e

5 Marbeliaet al., 2014 d

C.v. Anal. Col 0.2 max 33 max 1.7e2.2 7.5e22 e - ~85max >95 max 5 Marbelia

et al., 2014

e

T.s. Fish farm eff. Tube, batch 0.5 350 5 41 e - 49 99 e Michels

et al., 2014

O.m.,S.c.,C.v. Tertiary Batch 0.29e0.31 e e - e - >99 max >99 max >4d Ji et al., 2013

A.p. AD supernatant Batch 2.5 e e - e - 69e74 25e75 e Zhou et al.,

2012a

f

KEY.

Species Feedwater

A.p. Auxenochlorella protothecoides AD anaerobic digester

C.a. Coelastrum Anal. analogue

C.k. Chlorella kessleri Reactor conguration

C.p. Chlorella protothecoides BMPBR biolm membrane bioreactor

C.v. Chlorella vulgaris Col column

D.o. Desmodesmus opoliensis HRAP high-rate algal ponds

M.p. Mucidosphaerium pulchellum MPBR membrane photobioreactorO.m. O. multisporus P. plate parallel plate

P.b. Pediastrum boryanum Tub A-L tubular air-lift

S.c. Scenedesmus obliquus

T.s. Tetraselmis suecica

Italicised data refer to ammonia-N rather than TN.a Seasonally dependent pond organic matter: highest concentration in summer, lowest in winter: 7 e18 C temperature range. % removal decreases with increasing load.b pH dependency of P removal.c Units of g m2 d1, i.e. with reference to biolm area rather than reactor volume; 0.4e1.7 and 0.1e0.34 g m2 d1 of N and P loading respectively.d MPBR: N removal decreases from >95% to ~30% (P rem from ~95 to 50) as HRT decreases from 5 to 2 days; max productivity of 60 at 2 days HRT.e PBR: N removal decreases from ~85% to ~75% (P rem from 95% to 35%) as HRT decreases from 5 to 2 days; max productivity of 33 at 5 days HRT.f Removal loading-dependent.

Table 6

BNR papers, municipal wastewater.

Feed Hydraulic residence time, h % Nutrient removal Ref

An Ax Ae Total N P Notes

Real 5.4 1.35 2.25 9 73 67 Mean removals Vaiopoulou and Aivasidis, 2008

Analogue 2 4 8 14 83e88 68e80 Combined optimum Brown et al., 2011

Analogue 2 2 4 8 90 99 Optimum Liu et al., 2013Real e 3.1e4.6 6e9 9.3e15 75 98 Optimum Zeng et al., 2011

Real 1.8 2.6 2.2 6.6 87 94 Optimum Puig et al., 2008

Analogue 3 3 3 9 73e77 87e95 pH dependent Liu et al., 2008

http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-http://-/?-


10/12

(>99%) and/or reduced HRTs (16 h) have been reported for the

advancedPBR process congurations of column (Marbelia et al.,

2014) and parallel plate biolm (Ramos-Tercero et al., 2014) re-actors respectively, with further HRT reductions (


11/12

Chai, X., Zhao, X., Baoying, W., 2012. Bioxation of carbon dioxide by Chlorococcumsp. in a photobioreactor with polytetrauoroethene membrane sparger. Afr. J.Biotechnol. 11 (29), 7445e7450.

Chauton, M.S., Reitan, K.I., Norsker, N.H., Tveteras, R., Kleivdal, H.T., 2015. A techno-economic analysis of industrial production of marine microalgae as a source ofEPA and DHA-rich raw material for aquafeed: research challenges and possi-bilities. Aquaculture 436, 95e103.

Chen, C.-Y., Kaob, P.-C., Tsai, C.-J., Lee, D.-J., Chang, J.-S., 2013. Engineering strategiesfor simultaneous enhancement of C-phycocyanin production and CO2 xationwith Spirulina platensis. Bioresour. Technol. 145, 307e312.

Cheng, L., Zhang, L., Chen, H., Gao, C., 2006. Carbon dioxide removal from air bymicroalgae cultured in a membrane-photobioreactor. Sep. Purif. Technol. 50,324e329.

Cheng, J., Huang, Y., Feng, J., Sun, J., Zhou, J., Cen, K., 2013. Improving CO 2 xationefciency by optimizing ChlorellaPY-ZU1 culture conditions in sequential bio-reactors. Bioresour. Technol. 144, 321e327.

Chiang, C.-L., Lee, C.-M., Chen, P.-C., 2011. Utilization of the cyanobacteriaAnabaenasp. CH1 in biological carbon dioxide mitigation processes. Bioresour. Technol.102, 5400e5405.

Chisti, Y., 2007. Biodiesel from microalgae. Biotechnol. Adv. 25, 294e306.Chiu, S.-Y., Kao, C.-Y., Chen, C.-H., Kuan, T.-C., Ong, S.-C., Lin, C.-S., 2008. Lipid

accumulation and CO2utilization ofNannochloropsis oculatain response to CO2aeration. Bioresour. Technol. 99 (9), 3389e3396.

Cho, S., Luong, T.T., Lee, D., Oh, U.-K., Lee, T., 2011. Reuse of efuent water from amunicipal wastewater treatment plant in microalgae cultivation for biofuelproduction. Bioresour. Technol. 102, 8639e8645.

Clarens, A.F., Resurreccion, E.P., White, M.A., Colosi, L.M., 2010. Environmental lifecycle comparison of algae to other bioenergy feedstocks. Environ. Sci. Technol44 (5), 1813e1819.

Concas, A., Lutzu, G.A., Pisu, M., Cao, G., 2012. Experimental analysis and novelmodeling of semi-batch photobioreactors operated with Chlorella vulgaris andfed with 100% (v/v) CO2. Chem. Eng. J. 213, 203e213.

Daigger, G.T., Littlejohn, H.X., 2014. Simultaneous biological nutrient removal, astate-of-the-art review. Water Env. Res. 86 (3), 245e257.

De Godos, I., Mendoza, J.L., Acien, F.G., Molina, E., Banks, C.J., Heaven, S., Rogalla, F.,2014. Evaluation of carbon dioxide mass transfer in raceway reactors formicroalgae culture using ue gases. Bioresour. Technol. 153, 307e314.

De Gregorio, C., Caravelli, A.H., Zaritzky, N.E., 2010. Performance and biologicalindicators of a laboratory-scale activated sludge reactor with phosphatesimultaneous precipitation as affected by ferric chloride addition. Chem. Eng. J.165 (2), 607e616.

Douchal, J., Straka, F., Lvansky, K., 2005. Utilization of ue gas for cultivation ofmicroalgae (Chlorella sp.) in an outdoor open thin-layer photobioreactor. J. Appl.Phycol. 17, 403e412.

Douskova, I., Doucha, J., Livansky, K., Machat, J., Novak, P., Umsova, D., Zachleder, V.,Vitova, M., 2009. Simultaneous ue gas bioremediation and reduction ofmicroalgal biomass production costs. Appl. Microb. Biotechnol. 82 (1), 179e185.

Drexler, I.L.C., Yeh, D.H., 2014. Membrane applications for microalgae cultivationand harvesting, a review. Rev. Environ. Sci. Biotechnol. 13 (4), 487e504.Fan, L., Zhang, Y., Chen, H., 2008. Evaluation of a membrane-sparged helical tubular

photobioreactor for carbon dioxide bioxation by Chlorella vulgaris. J. Membr.Sci. 325, 336e345.

Fernandez, I., Acien, F.G., Fernandez, J.M., Guzman, J.L., Magan, J.J., Berenguel, M.,2012. Dynamic model of microalgal production in tubular photobioreactors.Bioresour. Technol. 126, 172e181.

Filippino, K.C., Mulhollanda, M.R., Bott, C.B., 2015. Phycoremediation strategies forrapid tertiary nutrient removal in a waste stream. Algal Res. 11, 125 e133.

Gao, F., Yang, Z.H., Li, C., Zeng, G.M., Ma, D.H., Zhou, L., 2015. A novel algal biolmmembrane photobioreactor for attached microalgae growth and nutrientsremoval from secondary efuent. Bioresour. Technol. 179, 8e12.

Gavrilescu, M., Chisti, Y., 2005. Biotechnology e a sustainable alternative forchemical industry. Biotechnol. Adv. 23, 471e499.

Gonalves, A.L., Sim~oes, M., Pires, J.C.M., 2014. The effect of light supply onmicroalgal growth, CO2uptake and nutrient removal from wastewater. EnergyConvers. Manag. 85, 530e536.

Gross, M., Wen, Z., 2014. Yearlong evaluation of performance and durability of a

pilot-scale revolving algal biolm (RAB) cultivation system. Bioresour. Technol.171, 50e58.

Hasan, M.M.F., Boukouvala, F., First, E.L., Floudas, C.A., 2014. National wide, regionaland statewide CO2capture, utilization and sequestration supply chain networkoptimization. Ind. Eng. Chem. Res. 53, 7489e7506.

Ho, S.-H., Chen, C.-Y., Lee, D.-J., Chang, J.-S., 2011. Perspectives on microalgal CO 2-emission mitigation systemse a review. Biotechnol. Adv. 29, 189e198.

Ho, S.-H., Chen, C.-Y., Chang, J.-S., 2012. Effect of light intensity and nitrogen star-vation on CO2 xation and lipid/carbohydrate production of an indigenousmicroalga Scenedesmus obliquus CNW-N. Bioresour. Technol. 113, 244e252.

Hoyt, D.V., 1979. An empirical determination of the heating of the Earth by thecarbon dioxide greenhouse effect. Nature 282 (5737), 388e390.

Jacob-Lopes, E., Revah, S., Hernandez, S., Shirai, K., Franco, T.T., 2009. Developmentof operational strategies to remove carbon dioxide in photobioreactors. Chem.Eng. J. 153, 120e126.

Jacob-Lopes, E., Scoparo, C.H.G., Queiroz, M.I., Franco, T.T., 2010. Biotransformationsof carbon dioxide in photobioreactors. Energy Convers. Manag. 51, 894e900.

Ji, M.K., Abou-Shanab, R.A.I., Kim, S.H., Salama, E.S., Lee, S.H., Kabra, A.N., Lee, Y.S.,Hong, S., Jeon, B.-H., 2013. Cultivation of microalgae species in tertiary

municipal wastewater supplemented with CO2 for nutrient removal andbiomass production. Ecol. Eng. 58, 142e148.

Judd, S., 2014. Industrial MBRs. Judd and Judd Ltd., Craneld, UK.Kargupta, W., Ganesh, A., Mukherji, S., 2015. Estimation of carbon dioxide seques-

tration potential of microalgae grown in a batch photobioreactor. Bioresour.Technol. 180, 370e375.

Keeley, J., Jarvis, P., Judd, S.J., 2012. The cost of coagulant recovery using membraneprocesses. Desalination 287, 132e137.

Kim, W., Park, J.M., Gim, G.H., Jeong, S.H., Kang, C., Kim, D.J., Kim, S.W., 2012.Optimization of culture conditions and comparison of biomass productivity of

three green algae. Bioprocess Biosys. Eng. 35 (1e

2), 19e

27.Lam, M., Lee, K., 2013. Effect of carbon source towards the growth of Chlorella

vulgaris for CO2 bio-mitigation and biodiesel production. Int. J. Greenh. GasControl 14, 169e176.

Laumb, J.D., Kay, J.P., Holmes, M.J., Cowan, R.M., Azenkeng, A., Heebink, L.V.,Hanson, S.K., Jensen, M.D., Letvin, P.A., Raymond, L.J., 2013. Economic andmarket analysis of CO2 utilization technologies e focus on CO2 derived fromNorth Dakota lignite. Energy Procedia 37, 6987e6998.

Lee, C.S., Lee, S.-A., Ko, S.-R., Oh, H.-M., Ahn, C.-Y., 2015. Effects of photoperiod onnutrient removal, biomass production, and algal-bacterial population dynamicsin lab-scale photobioreactors treating municipal wastewater. Water Res. 68,680e691.

Li, B., Brett, M.T., 2012. The impact of alum based advanced nutrient removal pro-cesses on phosphorus bioavailability. Water Res. 46 (3), 837e844.

Li, S., Luo, S., Guo, R., 2013. Efciency of CO2 xation by microalgae in a closedraceway pond. Bioresour. Technol. 136, 267e272.

Liang, Z., Liu, Y., Ge, F., Xu, Y., Tao, N., Peng, F., Wong, M., 2013. Efciency assessmentand pH effect in removing nitrogen and phosphorus by algae-bacteria com-bined system of Chlorella vulgaris and Bacillus licheniformis. Chemosphere 92,1383

e1389.

Liu, H., Sun, L., Xia, S., 2008. An efcient DPB utilization process: the modied A2Nprocess. Biochem. Eng. J. 38, 158e163.

Liu, G., Xu, X., Zhu, L., Xing, S., Chen, J., 2013. Biological nutrient removal in acontinuous anaerobiceaerobiceanoxic process treating synthetic domesticwastewater. Chem. Eng. J. 225, 223e229.

Loosdrecht, V., Hooijmans, M.C.M., Brdjanovic, D., Heijnen, J.J., 1997. Biologicalphosphate removal processes. Appl. Microbiol. Biotech. 48 (3), 289 e296.

Lopes da Silva, T., Gouveia, L., Reis, A., 2014. Integrated microbial processes forbiofuels and high value-added products: the way to improve the cost effec-tiveness of biofuel production. Appl. Microbiol. Biotechnol. 98, 1043e1053.

Lundquist, T.J., Woertz, I.C., Quinn, N.W.T., Benemann, J.R., 2010. A Realistic Tech-nology and Engineering Assessment of Algal Biofuel Production. Energy Bio-sciences Institute, University of California, Berkeley, California. Oct 2010.

Marbelia, L., Bilad, M.R., Passaris, I., Discart, V., Vandamme, D., Beuckels, A.,Muylaert, K., Vankelecom, I.F.J., 2014. Membrane photobioreactors for inte-grated microalgae cultivation and nutrient remediation of membrane bio-reactors efuent. Bioresour. Technol. 163, 228e235.

Markou, G., Nerantzis, E., 2013. Microalgae for high-value compounds and biofuelsproduction: a review with focus on cultivation under stress conditions. Bio-technol. Adv. 31, 1532e1542.

Martinez, L., Moran, A., Garcia, A.I., 2012. Effect of light on Synechocystis sp. andmodelling of its growth rate as a response to average irradiance. J. Appl. Phycol.24 (1), 125e130.

Michels, H.A.M., Vaskoska, M., Vermue, M.H., Wijffels, R.H., 2014. Growth ofTetra-selmis suecica in a tubular photobioreactor on wastewater from a sh farm.Water Res. 65, 290e296.

Milledge, J.J., Heaven, S., 2013. A review of the harvesting of micro-algae for biofuelproduction. Rev. Environ. Sci. Biotechnol. 12, 165e178.

Mulkerrins, D., Dobson, A.D.W., Colleran, E., 2004. Parameters affecting biologicalphosphate removal from wastewaters. Environ. Intern. 30 (2), 249e259.

Oswald, W.J., Gotaas, H.B., Ludwig, H.F., Lynch, V.,1953. Algae symbiosis in oxidationponds: photosynthetic oxygenation. Sew. Ind. Waste 25 (6), 692e705.

Pirouzi, A., Nosrati, M., Shojaosadati, S.A., Shakhesi, S., 2014. Improvement of mixingtime, mass transfer, and power consumption in an external loop airlift photo-bioreactor for microalgae cultures. Biochem. Eng. J. 87, 25 e32.

Posadas, E., Garca-Encina, P.A., Soltau, A., Domnguez, A., Daz, I., Mu ~noz, R., 2013.

Carbon and nutrient removal from centrates and domestic wastewater usingalgalebacterial biolm bioreactors. Bioresour. Technol. 139, 50e58.

Puig, S., van Loosdrecht, M.C.M., Colprim, J., Meijer, S.C.F., 2008. Data evaluation offull-scale wastewater treatment plants by mass balance. Water Res. 42,4645e4655.

Ramos-Tercero, E.A., Sforza, E., Morandini, M., Bertucco, A., 2014. Cultivation ofchlorella protothecoides with urban wastewater in continuous photobioreactor:biomass productivity and nutrient removal. Appl. Biochem. Biotechnol. 172,1470e1485.

Rittmann, B.L., McCarthy, F., 2001. Environmental Biotechnology: Principles andApplications. McGraw Hill, NY.

Ryu, H.J., Oh, K.K., Kim, Y.S., 2009. Optimization of the inuential factors for theimprovement of CO2 utilization efciency and CO2 mass transfer rate. J. Ind.Eng. Chem. 15, 471e475.

Sanchez-Fernandez, J.F., 2009. Utilization of the cyanobacteria Anabaena sp. ATCC33047 in CO2removal processes. Bioresour. Technol. 100 (23), 5904e5910.

Sanchez-Fernandez, J.F., Gonzalez-Lopez, C.V., Acien, F.G., Fernandez-Sevilla, J.M.,Molina, E., 2012. Utilization of Anabaena sp. in CO2 removal processes:modelling of biomass, exopolysaccharides productivities and CO

2xation rate.

http://refhub.elsevier.com/S0043-1354(15)30172-X/sref17http://refhub.elsevier.com/S0043-1354(15)30172-X/sref17http://refhub.elsevier.com/S0043-1354(15)30172-X/sref17http://refhub.elsevier.com/S0043-1354(15)30172-X/sref17http://refhub.elsevier.com/S0043-1354(15)30172-X/sref17http://refhub.elsevier.com/S0043-1354(15)30172-X/sref17http://refhub.elsevier.com/S0043-1354(15)30172-X/sref17http://refhub.elsevier.com/S0043-1354(15)30172-X/sref17http://refhub.elsevier.com/S0043-1354(15)30172-X/sref17http://refhub.elsevier.com/S0043-1354(15)30172-X/sref17http://refhub.elsevier.com/S0043-1354(15)30172-X/sref18http://refhub.elsevier.com/S0043-1354(15)30172-X/sref18http://refhub.elsevier.com/S0043-1354(15)30172-X/sref18http://refhub.elsevier.com/S0043-1354(15)30172-X/sref18http://refhub.elsevier.com/S0043-1354(15)30172-X/sref18http://refhub.elsevier.com/S0043-1354(15)30172-X/sref19http://refhub.elsevier.com/S0043-1354(15)30172-X/sref19http://refhub.elsevier.com/S0043-1354(15)30172-X/sref19http://refhub.elsevier.com/S0043-1354(15)30172-X/sref19http://refhub.elsevier.com/S0043-1354(15)30172-X/sref19http://refhub.elsevier.com/S0043-1354(15)30172-X/sref19http://refhub.elsevier.com/S0043-1354(15)30172-X/sref19http://refhub.elsevier.com/S0043-1354(15)30172-X/sref19http://refhub.elsevier.com/S0043-1354(15)30172-X/sref19http://refhub.elsevier.com/S0043-1354(15)30172-X/sref19http://refhub.elsevier.com/S0043-1354(15)30172-X/sref21http://refhub.elsevier.com/S0043-1354(15)30172-X/sref21http://refhub.elsevier.com/S0043-1354(15)30172-X/sref21http://refhub.elsevier.com/S0043-1354(15)30172-X/sref21http://refhub.elsevier.com/S0043-1354(15)30172-X/sref21http://refhub.elsevier.com/S0043-1354(15)30172-X/sref22http://refhub.elsevier.com/S0043-1354(15)30172-X/sref22http://refhub.elsevier.com/S0043-1354(15)30172-X/sref22http://refhub.elsevier.com/S0043-1354(15)30172-X/sref22http://refhub.elsevier.com/S0043-1354(15)30172-X/sref22http://refhub.elsevier.com/S0043-1354(15)30172-X/sref22http://refhub.elsevier.com/S0043-1354(15)30172-X/sref22http://refhub.elsevier.com/S0043-1354(15)30172-X/sref22http://refhub.elsevier.com/S0043-1354(15)30172-X/sref22http://refhub.elsevier.com/S0043-1354(15)30172-X/sref22http://refhub.elsevier.com/S0043-1354(15)30172-X/sref22http://refhub.elsevier.com/S0043-1354(15)30172-X/sref22http://refhub.elsevier.com/S0043-1354(15)30172-X/sref23http://refhub.elsevier.com/S0043-1354(15)30172-X/sref23http://refhub.elsevier.com/S0043-1354(15)30172-X/sref23http://refhub.elsevier.com/S0043-1354(15)30172-X/sref23http://refhub.elsevier.com/S0043-1354(15)30172-X/sref23http://refhub.elsevier.com/S0043-1354(15)30172-X/sref23http://refhub.elsevier.com/S0043-1354(15)30172-X/sref24http://refhub.elsevier.com/S0043-1354(15)30172-X/sref24http://refhub.elsevier.com/S0043-1354(15)30172-X/sref25http://refhub.elsevier.com/S0043-1354(15)30172-X/sref25http://refhub.elsevier.com/S0043-1354(15)30172-X/sref25http://refhub.elsevier.com/S0043-1354(15)30172-X/sref25http://refhub.elsevier.com/S0043-1354(15)30172-X/sref25http://refhub.elsevier.com/S0043-1354(15)30172-X/sref25http://refhub.elsevier.com/S0043-1354(15)30172-X/sref25http://refhub.elsevier.com/S0043-1354(15)30172-X/sref25http://refhub.elsevier.com/S0043-1354(15)30172-X/sref25http://refhub.elsevier.com/S0043-1354(15)30172-X/sref25http://refhub.elsevier.com/S0043-1354(15)30172-X/sref26http://refhub.elsevier.com/S0043-1354(15)30172-X/sref26http://refhub.elsevier.com/S0043-1354(15)30172-X/sref26http://refhub.elsevier.com/S0043-1354(15)30172-X/sref26http://refhub.elsevier.com/S0043-1354(15)30172-X/sref26http://refhub.elsevier.com/S0043-1354(15)30172-X/sref26http://refhub.elsevier.com/S0043-1354(15)30172-X/sref27http://refhub.elsevier.com/S0043-1354(15)30172-X/sref27http://refhub.elsevier.com/S0043-1354(15)30172-X/sref27http://refhub.elsevier.com/S0043-1354(15)30172-X/sref27http://refhub.elsevier.com/S0043-1354(15)30172-X/sref28http://refhub.elsevier.com/S0043-1354(15)30172-X/sref28http://refhub.elsevier.com/S0043-1354(15)30172-X/sref28http://refhub.elsevier.com/S0043-1354(15)30172-X/sref28http://refhub.elsevier.com/S0043-1354(15)30172-X/sref28http://refhub.elsevier.com/S0043-1354(15)30172-X/sref28http://refhub.elsevier.com/S0043-1354(15)30172-X/sref29http://refhub.elsevier.com/S0043-1354(15)30172-X/sref29http://refhub.elsevier.com/S0043-1354(15)30172-X/sref29http://refhub.elsevier.com/S0043-1354(15)30172-X/sref29http://refhub.elsevier.com/S0043-1354(15)30172-X/sref30http://refhub.elsevier.com/S0043-1354(15)30172-X/sref30http://refhub.elsevier.com/S0043-1354(15)30172-X/sref30http://refhub.elsevier.com/S0043-1354(15)30172-X/sref30http://refhub.elsevier.com/S0043-1354(15)30172-X/sref30http://refhub.elsevier.com/S0043-1354(15)30172-X/sref30http://refhub.elsevier.com/S0043-1354(15)30172-X/sref30http://refhub.elsevier.com/S0043-1354(15)30172-X/sref31http://refhub.elsevier.com/S0043-1354(15)30172-X/sref31http://refhub.elsevier.com/S0043-1354(15)30172-X/sref31http://refhub.elsevier.com/S0043-1354(15)30172-X/sref31http://refhub.elsevier.com/S0043-1354(15)30172-X/sref31http://refhub.elsevier.com/S0043-1354(15)30172-X/sref32http://refhub.elsevier.com/S0043-1354(15)30172-X/sref32http://refhub.elsevier.com/S0043-1354(15)30172-X/sref32http://refhub.elsevier.com/S0043-1354(15)30172-X/sref32http://refhub.elsevier.com/S0043-1354(15)30172-X/sref32http://refhub.elsevier.com/S0043-1354(15)30172-X/sref32http://refhub.elsevier.com/S0043-1354(15)30172-X/sref32http://refhub.elsevier.com/S0043-1354(15)30172-X/sref32http://refhub.elsevier.com/S0043-1354(15)30172-X/sref32http://refhub.elsevier.com/S0043-1354(15)30172-X/sref33http://refhub.elsevier.com/S0043-1354(15)30172-X/sref33http://refhub.elsevier.com/S0043-1354(15)30172-X/sref33http://refhub.elsevier.com/S0043-1354(15)30172-X/sref33http://refhub.elsevier.com/S0043-1354(15)30172-X/sref33http://refhub.elsevier.com/S0043-1354(15)30172-X/sref33http://refhub.elsevier.com/S0043-1354(15)30172-X/sref33http://refhub.elsevier.com/S0043-1354(15)30172-X/sref34http://refhub.elsevier.com/S0043-1354(15)30172-X/sref34http://refhub.elsevier.com/S0043-1354(15)30172-X/sref34http://refhub.elsevier.com/S0043-1354(15)30172-X/sref34http://refhub.elsevier.com/S0043-1354(15)30172-X/sref35http://refhub.elsevier.com/S0043-1354(15)30172-X/sref35http://refhub.elsevier.com/S0043-1354(15)30172-X/sref35http://refhub.elsevier.com/S0043-1354(15)30172-X/sref35http://refhub.elsevier.com/S0043-1354(15)30172-X/sref35http://refhub.elsevier.com/S0043-1354(15)30172-X/sref35http://refhub.elsevier.com/S0043-1354(15)30172-X/sref36http://refhub.elsevier.com/S0043-1354(15)30172-X/sref36http://refhub.elsevier.com/S0043-1354(15)30172-X/sref36http://refhub.elsevier.com/S0043-1354(15)30172-X/sref36http://refhub.elsevier.com/S0043-1354(15)30172-X/sref36http://refhub.elsevier.com/S0043-1354(15)30172-X/sref36http://refhub.elsevier.com/S0043-1354(15)30172-X/sref36http://refhub.elsevier.com/S0043-1354(15)30172-X/sref36http://refhub.elsevier.com/S0043-1354(15)30172-X/sref36http://refhub.elsevier.com/S0043-1354(15)30172-X/sref36http://refhub.elsevier.com/S0043-1354(15)30172-X/sref37http://refhub.elsevier.com/S0043-1354(15)30172-X/sref37http://refhub.elsevier.com/S0043-1354(15)30172-X/sref37http://refhub.elsevier.com/S0043-1354(15)30172-X/sref37http://refhub.elsevier.com/S0043-1354(15)30172-X/sref38http://refhub.elsevier.com/S0043-1354(15)30172-X/sref38http://refhub.elsevier.com/S0043-1354(15)30172-X/sref38http://refhub.elsevier.com/S0043-1354(15)30172-X/sref38http://refhub.elsevier.com/S0043-1354(15)30172-X/sref38http://refhub.elsevier.com/S0043-1354(15)30172-X/sref38http://refhub.elsevier.com/S0043-1354(15)30172-X/sref38http://refhub.elsevier.com/S0043-1354(15)30172-X/sref38http://refhub.elsevier.com/S0043-1354(15)30172-X/sref39http://refhub.elsevier.com/S0043-1354(15)30172-X/sref39http://refhub.elsevier.com/S0043-1354(15)30172-X/sref39http://refhub.elsevier.com/S0043-1354(15)30172-X/sref39http://refhub.elsevier.com/S0043-1354(15)30172-X/sref40http://refhub.elsevier.com/S0043-1354(15)30172-X/sref40http://refhub.elsevier.com/S0043-1354(15)30172-X/sref40http://refhub.elsevier.com/S0043-1354(15)30172-X/sref40http://refhub.elsevier.com/S0043-1354(15)30172-X/sref40http://refhub.elsevier.com/S0043-1354(15)30172-X/sref40http://refhub.elsevier.com/S0043-1354(15)30172-X/sref40http://refhub.elsevier.com/S0043-1354(15)30172-X/sref41http://refhub.elsevier.com/S0043-1354(15)30172-X/sref41http://refhub.elsevier.com/S0043-1354(15)30172-X/sref41http://refhub.elsevier.com/S0043-1354(15)30172-X/sref41http://refhub.elsevier.com/S0043-1354(15)30172-X/sref41http://refhub.elsevier.com/S0043-1354(15)30172-X/sref41http://refhub.elsevier.com/S0043-1354(15)30172-X/sref42http://refhub.elsevier.com/S0043-1354(15)30172-X/sref42http://refhub.elsevier.com/S0043-1354(15)30172-X/sref42http://refhub.elsevier.com/S0043-1354(15)30172-X/sref42http://refhub.elsevier.com/S0043-1354(15)30172-X/sref42http://refhub.elsevier.com/S0043-1354(15)30172-X/sref42http://refhub.elsevier.com/S0043-1354(15)30172-X/sref43http://refhub.elsevier.com/S0043-1354(15)30172-X/sref43http://refhub.elsevier.com/S0043-1354(15)30172-X/sref43http://refhub.elsevier.com/S0043-1354(15)30172-X/sref43http://refhub.elsevier.com/S0043-1354(15)30172-X/sref43http://refhub.elsevier.com/S0043-1354(15)30172-X/sref43http://refhub.elsevier.com/S0043-1354(15)30172-X/sref44http://refhub.elsevier.com/S0043-1354(15)30172-X/sref44http://refhub.elsevier.com/S0043-1354(15)30172-X/sref44http://refhub.elsevier.com/S0043-1354(15)30172-X/sref44http://refhub.elsevier.com/S0043-1354(15)30172-X/sref44http://refhub.elsevier.com/S0043-1354(15)30172-X/sref44http://refhub.elsevier.com/S0043-1354(15)30172-X/sref44http://refhub.elsevier.com/S0043-1354(15)30172-X/sref44http://refhub.elsevier.com/S0043-1354(15)30172-X/sref44http://refhub.elsevier.com/S0043-1354(15)30172-X/sref44http://refhub.elsevier.com/S0043-1354(15)30172-X/sref45http://refhub.elsevier.com/S0043-1354(15)30172-X/sref45http://refhub.elsevier.com/S0043-1354(15)30172-X/sref45http://refhub.elsevier.com/S0043-1354(15)30172-X/sref45http://refhub.elsevier.com/S0043-1354(15)30172-X/sref46http://refhub.elsevier.com/S0043-1354(15)30172-X/sref46http://refhub.elsevier.com/S0043-1354(15)30172-X/sref46http://refhub.elsevier.com/S0043-1354(15)30172-X/sref46http://refhub.elsevier.com/S0043-1354(15)30172-X/sref47http://refhub.elsevier.com/S0043-1354(15)30172-X/sref47http://refhub.elsevier.com/S0043-1354(15)30172-X/sref47http://refhub.elsevier.com/S0043-1354(15)30172-X/sref48http://refhub.elsevier.com/S0043-1354(15)30172-X/sref48http://refhub.elsevier.com/S0043-1354(15)30172-X/sref48http://refhub.elsevier.com/S0043-1354(15)30172-X/sref48http://refhub.elsevier.com/S0043-1354(15)30172-X/sref48http://refhub.elsevier.com/S0043-1354(15)30172-X/sref48http://refhub.elsevier.com/S0043-1354(15)30172-X/sref48http://refhub.elsevier.com/S0043-1354(15)30172-X/sref49http://refhub.elsevier.com/S0043-1354(15)30172-X/sref49http://refhub.elsevier.com/S0043-1354(15)30172-X/sref49http://refhub.elsevier.com/S0043-1354(15)30172-X/sref50http://refhub.elsevier.com/S0043-1354(15)30172-X/sref50http://refhub.elsevier.com/S0043-1354(15)30172-X/sref50http://refhub.elsevier.com/S0043-1354(15)30172-X/sref50http://refhub.elsevier.com/S0043-1354(15)30172-X/sref50http://refhub.elsevier.com/S0043-1354(15)30172-X/sref51http://refhub.elsevier.com/S0043-1354(15)30172-X/sref51http://refhub.elsevier.com/S0043-1354(15)30172-X/sref51http://refhub.elsevier.com/S0043-1354(15)30172-X/sref51http://refhub.elsevier.com/S0043-1354(15)30172-X/sref52http://refhub.elsevier.com/S0043-1354(15)30172-X/sref52http://refhub.elsevier.com/S0043-1354(15)30172-X/sref52http://refhub.elsevier.com/S0043-1354(15)30172-X/sref52http://refhub.elsevier.com/S0043-1354(15)30172-X/sref52http://refhub.elsevier.com/S0043-1354(15)30172-X/sref52http://refhub.elsevier.com/S0043-1354(15)30172-X/sref53http://refhub.elsevier.com/S0043-1354(15)30172-X/sref53http://refhub.elsevier.com/S0043-1354(15)30172-X/sref53http://refhub.elsevier.com/S0043-1354(15)30172-X/sref53http://refhub.elsevier.com/S0043-1354(15)30172-X/sref53http://refhub.elsevier.com/S0043-1354(15)30172-X/sref53http://refhub.elsevier.com/S0043-1354(15)30172-X/sref54http://refhub.elsevier.com/S0043-1354(15)30172-X/sref54http://refhub.elsevier.com/S0043-1354(15)30172-X/sref54http://refhub.elsevier.com/S0043-1354(15)30172-X/sref54http://refhub.elsevier.com/S0043-1354(15)30172-X/sref54http://refhub.elsevier.com/S0043-1354(15)30172-X/sref54http://refhub.elsevier.com/S0043-1354(15)30172-X/sref54http://refhub.elsevier.com/S0043-1354(15)30172-X/sref54http://refhub.elsevier.com/S0043-1354(15)30172-X/sref54http://refhub.elsevier.com/S0043-1354(15)30172-X/sref54http://refhub.elsevier.com/S0043-1354(15)30172-X/sref55http://refhub.elsevier.com/S0043-1354(15)30172-X/sref55http://refhub.elsevier.com/S0043-1354(15)30172-X/sref55http://refhub.elsevier.com/S0043-1354(15)30172-X/sref55http://refhub.elsevier.com/S0043-1354(15)30172-X/sref55http://refhub.elsevier.com/S0043-1354(15)30172-X/sref55http://refhub.elsevier.com/S0043-1354(15)30172-X/sref56http://refhub.elsevier.com/S0043-1354(15)30172-X/sref56http://refhub.elsevier.com/S0043-1354(15)30172-X/sref56http://refhub.elsevier.com/S0043-1354(15)30172-X/sref57http://refhub.elsevier.com/S0043-1354(15)30172-X/sref57http://refhub.elsevier.com/S0043-1354(15)30172-X/sref57http://refhub.elsevier.com/S0043-1354(15)30172-X/sref57http://refhub.elsevier.com/S0043-1354(15)30172-X/sref57http://refhub.elsevier.com/S0043-1354(15)30172-X/sref57http://refhub.elsevier.com/S0043-1354(15)30172-X/sref57http://refhub.elsevier.com/S0043-1354(15)30172-X/sref57http://refhub.elsevier.com/S0043-1354(15)30172-X/sref57http://refhub.elsevier.com/S0043-1354(15)30172-X/sref58http://refhub.elsevier.com/S0043-1354(15)30172-X/sref58http://refhub.elsevier.com/S0043-1354(15)30172-X/sref58http://refhub.elsevier.com/S0043-1354(15)30172-X/sref58http://refhub.elsevier.com/S0043-1354(15)30172-X/sref58http://refhub.elsevier.com/S0043-1354(15)30172-X/sref58http://refhub.elsevier.com/S0043-1354(15)30172-X/sref58http://refhub.elsevier.com/S0043-1354(15)30172-X/sref59http://refhub.elsevier.com/S0043-1354(15)30172-X/sref59http://refhub.elsevier.com/S0043-1354(15)30172-X/sref59http://refhub.elsevier.com/S0043-1354(15)30172-X/sref59http://refhub.elsevier.com/S0043-1354(15)30172-X/sref59http://refhub.elsevier.com/S0043-1354(15)30172-X/sref59http://refhub.elsevier.com/S0043-1354(15)30172-X/sref59http://refhub.elsevier.com/S0043-1354(15)30172-X/sref60http://refhub.elsevier.com/S0043-1354(15)30172-X/sref60http://refhub.elsevier.com/S0043-1354(15)30172-X/sref60http://refhub.elsevier.com/S0043-1354(15)30172-X/sref60http://refhub.elsevier.com/S0043-1354(15)30172-X/sref60http://refhub.elsevier.com/S0043-1354(15)30172-X/sref60http://refhub.elsevier.com/S0043-1354(15)30172-X/sref61http://refhub.elsevier.com/S0043-1354(15)30172-X/sref61http://refhub.elsevier.com/S0043-1354(15)30172-X/sref61http://refhub.elsevier.com/S0043-1354(15)30172-X/sref62http://refhub.elsevier.com/S0043-1354(15)30172-X/sref62http://refhub.elsevier.com/S0043-1354(15)30172-X/sref62http://refhub.elsevier.com/S0043-1354(15)30172-X/sref62http://refhub.elsevier.com/S0043-1354(15)30172-X/sref63http://refhub.elsevier.com/S0043-1354(15)30172-X/sref63http://refhub.elsevier.com/S0043-1354(15)30172-X/sref63http://refhub.elsevier.com/S0043-1354(15)30172-X/sref64http://refhub.elsevier.com/S0043-1354(15)30172-X/sref64http://refhub.elsevier.com/S0043-1354(15)30172-X/sref64http://refhub.elsevier.com/S0043-1354(15)30172-X/sref64http://refhub.elsevier.com/S0043-1354(15)30172-X/sref64http://refhub.elsevier.com/S0043-1354(15)30172-X/sref64http://refhub.elsevier.com/S0043-1354(15)30172-X/sref64http://refhub.elsevier.com/S0043-1354(15)30172-X/sref65http://refhub.elsevier.com/S0043-1354(15)30172-X/sref65http://refhub.elsevier.com/S0043-1354(15)30172-X/sref65http://refhub.elsevier.com/S0043-1354(15)30172-X/sref65http://refhub.elsevier.com/S0043-1354(15)30172-X/sref65http://refhub.elsevier.com/S0043-1354(15)30172-X/sref66http://refhub.elsevier.com/S0043-1354(15)30172-X/sref66http://refhub.elsevier.com/S0043-1354(15)30172-X/sref66http://refhub.elsevier.com/S0043-1354(15)30172-X/sref66http://refhub.elsevier.com/S0043-1354(15)30172-X/sref66http://refhub.elsevier.com/S0043-1354(15)30172-X/sref66http://refhub.elsevier.com/S0043-1354(15)30172-X/sref67http://refhub.elsevier.com/S0043-1354(15)30172-X/sref67http://refhub.elsevier.com/S0043-1354(15)30172-X/sref67http://refhub.elsevier.com/S0043-1354(15)30172-X/sref67http://refhub.elsevier.com/S0043-1354(15)30172-X/sref67http://refhub.elsevier.com/S0043-1354(15)30172-X/sref67http://refhub.elsevier.com/S0043-1354(15)30172-X/sref67http://refhub.elsevier.com/S0043-1354(15)30172-X/sref67http://refhub.elsevier.com/S0043-1354(15)30172-X/sref67http://refhub.elsevier.com/S0043-1354(15)30172-X/sref68http://refhub.elsevier.com/S0043-1354(15)30172-X/sref68http://refhub.elsevier.com/S0043-1354(15)30172-X/sref68http://refhub.elsevier.com/S0043-1354(15)30172-X/sref69http://refhub.elsevier.com/S0043-1354(15)30172-X/sref69http://refhub.elsevier.com/S0043-1354(15)30172-X/sref69http://refhub.elsevier.com/S0043-1354(15)30172-X/sref70http://refhub.elsevier.com/S0043-1354(15)30172-X/sref70http://refhub.elsevier.com/S0043-1354(15)30172-X/sref70http://refhub.elsevier.com/S0043-1354(15)30172-X/sref70http://refhub.elsevier.com/S0043-1354(15)30172-X/sref71http://refhub.elsevier.com/S0043-1354(15)30172-X/sref71http://refhub.elsevier.com/S0043-1354(15)30172-X/sref71http://refhub.elsevier.com/S0043-1354(15)30172-X/sref71http://refhub.elsevier.com/S0043-1354(15)30172-X/sref72http://refhub.elsevier.com/S0043-1354(15)30172-X/sref72http://refhub.elsevier.com/S0043-1354(15)30172-X/sref72http://refhub.elsevier.com/S0043-1354(15)30172-X/sref72http://refhub.elsevier.com/S0043-1354(15)30172-X/sref72http://refhub.elsevier.com/S0043-1354(15)30172-X/sref72http://refhub.elsevier.com/S0043-1354(15)30172-X/sref72http://refhub.elsevier.com/S0043-1354(15)30172-X/sref72http://refhub.elsevier.com/S0043-1354(15)30172-X/sref73http://refhub.elsevier.com/S0043-1354(15)30172-X/sref73http://refhub.elsevier.com/S0043-1354(15)30172-X/sref73http://refhub.elsevier.com/S0043-1354(15)30172-X/sref73http://refhub.elsevier.com/S0043-1354(15)30172-X/sref74http://refhub.elsevier.com/S0043-1354(15)30172-X/sref74http://refhub.elsevier.com/S0043-1354(15)30172-X/sref74http://refhub.elsevier.com/S0043-1354(15)30172-X/sref74http://refhub.elsevier.com/S0043-1354(15)30172-X/sref74http://refhub.elsevier.com/S0043-1354(15)30172-X/sref75http://refhub.elsevier.com/S0043-1354(15)30172-X/sref75http://refhub.elsevier.com/S0043-1354(15)30172-X/sref75http://refhub.elsevier.com/S0043-1354(15)30172-X/sref76http://refhub.elsevier.com/S0043-1354(15)30172-X/sref76http://refhub.elsevier.com/S0043-1354(15)30172-X/sref76http://refhub.elsevier.com/S0043-1354(15)30172-X/sref76http://refhub.elsevier.com/S0043-1354(15)30172-X/sref76http://refhub.elsevier.com/S0043-1354(15)30172-X/sref76http://refhub.elsevier.com/S0043-1354(15)30172-X/sref76http://refhub.elsevier.com/S0043-1354(15)30172-X/sref76http://refhub.elsevier.com/S0043-1354(15)30172-X/sref76http://refhub.elsevier.com/S0043-1354(15)30172-X/sref76http://refhub.elsevier.com/S0043-1354(15)30172-X/sref76http://refhub.elsevier.com/S0043-1354(15)30172-X/sref76http://refhub.elsevier.com/S0043-1354(15)30172-X/sref77http://refhub.elsevier.com/S0043-1354(15)30172-X/sref77http://refhub.elsevier.com/S0043-1354(15)30172-X/sref77http://refhub.elsevier.com/S0043-1354(15)30172-X/sref77http://refhub.elsevier.com/S0043-1354(15)30172-X/sref77http://refhub.elsevier.com/S0043-1354(15)30172-X/sref77http://refhub.elsevier.com/S0043-1354(15)30172-X/sref77http://refhub.elsevier.com/S0043-1354(15)30172-X/sref77http://refhub.elsevier.com/S0043-1354(15)30172-X/sref77http://refhub.elsevier.com/S0043-1354(15)30172-X/sref77http://refhub.elsevier.com/S0043-1354(15)30172-X/sref78http://refhub.elsevier.com/S0043-1354(15)30172-X/sref78http://refhub.elsevier.com/S0043-1354(15)30172-X/sref78http://refhub.elsevier.com/S0043-1354(15)30172-X/sref78http://refhub.elsevier.com/S0043-1354(15)30172-X/sref78http://refhub.elsevier.com/S0043-1354(15)30172-X/sref78http://refhub.elsevier.com/S0043-1354(15)30172-X/sref78http://refhub.elsevier.com/S0043-1354(15)30172-X/sref78http://refhub.elsevier.com/S0043-1354(15)30172-X/sref78http://refhub.elsevier.com/S0043-1354(15)30172-X/sref78http://refhub.elsevier.com/S0043-1354(15)30172-X/sref78http://refhub.elsevier.com/S0043-1354(15)30172-X/sref78http://refhub.elsevier.com/S0043-1354(15)30172-X/sref78http://refhub.elsevier.com/S0043-1354(15)30172-X/sref78http://refhub.elsevier.com/S0043-1354(15)30172-X/sref78http://refhub.elsevier.com/S0043-1354(15)30172-X/sref78http://refhub.elsevier.com/S0043-1354(15)30172-X/sref78http://refhub.elsevier.com/S0043-1354(15)30172-X/sref78http://refhub.elsevier.com/S0043-1354(15)30172-X/sref78http://refhub.elsevier.com/S0043-1354(15)30172-X/sref78http://refhub.elsevier.com/S0043-1354(15)30172-X/sref78http://refhub.elsevier.com/S0043-1354(15)30172-X/sref78http://refhub.elsevier.com/S0043-1354(15)30172-X/sref77http://refhub.elsevier.com/S0043-1354(15)30172-X/sref77http://refhub.elsevier.com/S0043-1354(15)30172-X/sref77http://refhub.elsevier.com/S0043-1354(15)30172-X/sref77http://refhub.elsevier.com/S0043-1354(15)30172-X/sref77http://refhub.elsevier.com/S0043-1354(15)30172-X/sref77http://refhub.elsevier.com/S0043-1354(15)30172-X/sref76http://refhub.elsevier.com/S0043-1354(15)30172-X/sref76http://refhub.elsevier.com/S0043-1354(15)30172-X/sref76http://refhub.elsevier.com/S0043-1354(15)30172-X/sref76http://refhub.elsevier.com/S0043-1354(15)30172-X/sref76http://refhub.elsevier.com/S0043-1354(15)30172-X/sref76http://refhub.elsevier.com/S0043-1354(15)30172-X/sref75http://refhub.elsevier.com/S0043-1354(15)30172-X/sref75http://refhub.elsevier.com/S0043-1354(15)30172-X/sref74http://refhub.elsevier.com/S0043-1354(15)30172-X/sref74http://refhub.elsevier.com/S0043-1354(15)30172-X/sref74http://refhub.elsevier.com/S0043-1354(15)30172-X/sref74http://refhub.elsevier.com/S0043-1354(15)30172-X/sref74http://refhub.elsevier.com/S0043-1354(15)30172-X/sref73http://refhub.elsevier.com/S0043-1354(15)30172-X/sref73http://refhub.elsevier.com/S0043-1354(15)30172-X/sref73http://refhub.elsevier.com/S0043-1354(15)30172-X/sref73http://refhub.elsevier.com/S0043-1354(15)30172-X/sref72http://refhub.elsevier.com/S0043-1354(15)30172-X/sref72http://refhub.elsevier.com/S0043-1354(15)30172-X/sref72http://refhub.elsevier.com/S0043-1354(15)30172-X/sref72http://refhub.elsevier.com/S0043-1354(15)30172-X/sref72http://refhub.elsevier.com/S0043-1354(15)30172-X/sref72http://refhub.elsevier.com/S0043-1354(15)30172-X/sref71http://refhub.elsevier.com/S0043-1354(15)30172-X/sref71http://refhub.elsevier.com/S0043-1354(15)30172-X/sref71http://refhub.elsevier.com/S0043-1354(15)30172-X/sref71http://refhub.elsevier.com/S0043-1354(15)30172-X/sref70http://refhub.elsevier.com/S0043-1354(15)30172-X/sref70http://refhub.elsevier.com/S0043-1354(15)30172-X/sref70http://refhub.elsevier.com/S0043-1354(15)30172-X/sref69http://refhub.elsevier.com/S0043-1354(15)30172-X/sref69http://refhub.elsevier.com/S0043-1354(15)30172-X/sref69http://refhub.elsevier.com/S0043-1354(15)30172-X/sref68http://refhub.elsevier.com/S0043-1354(15)30172-X/sref68http://refhub.elsevier.com/S0043-1354(15)30172-X/sref68http://refhub.elsevier.com/S0043-1354(15)30172-X/sref67http://refhub.elsevier.com/S0043-1354(15)30172-X/sref67http://refhub.elsevier.com/S0043-1354(15)30172-X/sref67http://refhub.elsevier.com/S0043-1354(15)30172-X/sref67http://refhub.elsevier.com/S0043-1354(15)30172-X/sref67http://refhub.elsevier.com/S0043-1354(15)30172-X/sref66http://refhub.elsevier.com/S0043-1354(15)30172-X/sref66http://refhub.elsevier.com/S0043-1354(15)30172-X/sref66http://refhub.elsevier.com/S0043-1354(15)30172-X/sref66http://refhub.elsevier.com/S0043-1354(15)30172-X/sref65http://refhub.elsevier.com/S0043-1354(15)30172-X/sref65http://refhub.elsevier.com/S0043-1354(15)30172-X/sref65http://refhub.elsevier.com/S0043-1354(15)30172-X/sref65http://refhub.elsevier.com/S0043-1354(15)30172-X/sref64http://refhub.elsevier.com/S0043-1354(15)30172-X/sref64http://refhub.elsevier.com/S0043-1354(15)30172-X/sref64http://refhub.elsevier.com/S0043-1354(15)30172-X/sref64http://refhub.elsevier.com/S0043-1354(15)30172-X/sref64http://refhub.elsevier.com/S0043-1354(15)30172-X/sref63http://refhub.elsevier.com/S0043-1354(15)30172-X/sref63http://refhub.elsevier.com/S0043-1354(15)30172-X/sref63http://refhub.elsevier.com/S0043-1354(15)30172-X/sref62http://refhub.elsevier.com/S0043-1354(15)30172-X/sref62http://refhub.elsevier.com/S0043-1354(15)30172-X/sref62http://refhub.elsevier.com/S0043-1354(15)30172-X/sref62http://refhub.elsevier.com/S0043-1354(15)30172-X/sref61http://refhub.elsevier.com/S0043-1354(15)30172-X/sref61http://refhub.elsevier.com/S0043-1354(15)30172-X/sref61http://refhub.elsevier.com/S0043-1354(15)30172-X/sref60http://refhub.elsevier.com/S0043-1354(15)30172-X/sref60http://refhub.elsevier.com/S0043-1354(15)30172-X/sref60http://refhub.elsevier.com/S0043-1354(15)30172-X/sref60http://refhub.elsevier.com/S0043-1354(15)30172-X/sref60http://refhub.elsevier.com/S0043-1354(15)30172-X/sref60http://refhub.elsevier.com/S0043-1354(15)30172-X/sref59http://refhub.elsevier.com/S0043-1354(15)30172-X/sref59http://refhub.elsevier.com/S0043-1354(15)30172-X/sref59http://refhub.elsevier.com/S0043-1354(15)30172-X/sref58http://refhub.elsevier.com/S0043-1354(15)30172-X/sref58http://refhub.elsevier.com/S0043-1354(15)30172-X/sref58http://refhub.elsevier.com/S0043-1354(15)30172-X/sref58http://refhub.elsevier.com/S0043-1354(15)30172-X/sref58http://refhub.elsevier.com/S0043-1354(15)30172-X/sref57http://refhub.elsevier.com/S0043-1354(15)30172-X/sref57http://refhub.elsevier.com/S0043-1354(15)30172-X/sref57http://refhub.elsevier.com/S0043-1354(15)30172-X/sref57http://refhub.elsevier.com/S0043-1354(15)30172-X/sref56http://refhub.elsevier.com/S0043-1354(15)30172-X/sref56http://refhub.elsevier.com/S0043-1354(15)30172-X/sref56http://refhub.elsevier.com/S0043-1354(15)30172-X/sref55http://refhub.elsevier.com/S0043-1354(15)30172-X/sref55http://refhub.elsevier.com/S0043-1354(15)30172-X/sref55http://refhub.elsevier.com/S0043-1354(15)30172-X/sref55http://refhub.elsevier.com/S0043-1354(15)30172-X/sref55http://refhub.elsevier.com/S0043-1354(15)30172-X/sref54http://refhub.elsevier.com/S0043-1354(15)30172-X/sref54http://refhub.elsevier.com/S0043-1354(15)30172-X/sref54http://refhub.elsevier.com/S0043-1354(15)30172-X/sref54http://refhub.elsevier.com/S0043-1354(15)30172-X/sref54http://refhub.elsevier.com/S0043-1354(15)30172-X/sref54http://refhub.elsevier.com/S0043-1354(15)30172-X/sref54http://refhub.elsevier.com/S0043-1354(15)30172-X/sref54http://refhub.elsevier.com/S0043-1354(15)30172-X/sref53http://refhub.elsevier.com/S0043-1354(15)30172-X/sref53http://refhub.elsevier.com/S0043-1354(15)30172-X/sref53http://refhub.elsevier.com/S0043-1354(15)30172-X/sref53http://refhub.elsevier.com/S0043-1354(15)30172-X/sref53http://refhub.elsevier.com/S0043-1354(15)30172-X/sref52http://refhub.elsevier.com/S0043-1354(15)30172-X/sref52http://refhub.elsevier.com/S0043-1354(15)30172-X/sref52http://refhub.elsevier.com/S0043-1354(15)30172-X/sref52http://refhub.elsevier.com/S0043-1354(15)30172-X/sref52http://refhub.elsevier.com/S0043-1354(15)30172-X/sref51http://refhub.elsevier.com/S0043-1354(15)30172-X/sref51http://refhub.elsevier.com/S0043-1354(15)30172-X/sref51http://refhub.elsevier.com/S0043-1354(15)30172-X/sref50http://refhub.elsevier.com/S0043-1354(15)30172-X/sref50http://refhub.elsevier.com/S0043-1354(15)30172-X/sref50http://refhub.elsevier.com/S0043-1354(15)30172-X/sref50http://refhub.elsevier.com/S0043-1354(15)30172-X/sref49http://refhub.elsevier.com/S0043-1354(15)30172-X/sref48http://refhub.elsevier.com/S0043-1354(15)30172-X/sref48http://refhub.elsevier.com/S0043-1354(15)30172-X/sref48http://refhub.elsevier.com/S0043-1354(15)30172-X/sref48http://refhub.elsevier.com/S0043-1354(15)30172-X/sref48http://refhub.elsevier.com/S0043-1354(15)30172-X/sref48http://refhub.elsevier.com/S0043-1354(15)30172-X/sref47http://refhub.elsevier.com/S0043-1354(15)30172-X/sref47http://refhub.elsevier.com/S0043-1354(15)30172-X/sref47http://refhub.elsevier.com/S0043-1354(15)30172-X/sref46http://refhub.elsevier.com/S0043-1354(15)30172-X/sref46http://refhub.elsevier.com/S0043-1354(15)30172-X/sref46http://refhub.elsevier.com/S0043-1354(15)30172-X/sref46http://refhub.elsevier.com/S0043-1354(15)30172-X/sref45http://refhub.elsevier.com/S0043-1354(15)30172-X/sref45http://refhub.elsevier.com/S0043-1354(15)30172-X/sref45http://refhub.elsevier.com/S0043-1354(15)30172-X/sref44http://refhub.elsevier.com/S0043-1354(15)30172-X/sref44http://refhub.elsevier.com/S0043-1354(15)30172-X/sref44http://refhub.elsevier.com/S0043-1354(15)30172-X/sref44http://refhub.elsevier.com/S0043-1354(15)30172-X/sref44http://refhub.elsevier.com/S0043-1354(15)30172-X/sref43http://refhub.elsevier.com/S0043-1354(15)30172-X/sref43http://refhub.elsevier.com/S0043-1354(15)30172-X/sref43http://refhub.elsevier.com/S0043-1354(15)30172-X/sref43http://refhub.elsevier.com/S0043-1354(15)30172-X/sref43http://refhub.elsevier.com/S0043-1354(15)30172-X/sref42http://refhub.elsevier.com/S0043-1354(15)30172-X/sref42http://refhub.elsevier.com/S0043-1354(15)30172-X/sref42http://refhub.elsevier.com/S0043-1354(15)30172-X/sref42http://refhub.elsevier.com/S0043-1354(15)30172-X/sref42http://refhub.elsevier.com/S0043-1354(15)30172-X/sref41http://refhub.elsevier.com/S0043-1354(15)30172-X/sref41http://refhub.elsevier.com/S0043-1354(15)30172-X/sref41http://refhub.elsevier.com/S0043-1354(15)30172-X/sref41http://refhub.elsevier.com/S0043-1354(15)30172-X/sref40http://refhub.elsevier.com/S0043-1354(15)30172-X/sref40http://refhub.elsevier.com/S0043-1354(15)30172-X/sref40http://refhub.elsevier.com/S0043-1354(15)30172-X/sref40http://refhub.elsevier.com/S0043-1354(15)30172-X/sref40http://refhub.elsevier.com/S0043-1354(15)30172-X/sref40http://refhub.elsevier.com/S0043-1354(15)30172-X/sref39http://refhub.elsevier.com/S0043-1354(15)30172-X/sref39http://refhub.elsevier.com/S0043-1354(15)30172-X/sref39http://refhub.elsevier.com/S0043-1354(15)30172-X/sref39http://refhub.elsevier.com/S0043-1354(15)30172-X/sref38http://refhub.elsevier.com/S0043-1354(15)30172-X/sref38http://refhub.elsevier.com/S0043-1354(15)30172-X/sref38http://refhub.elsevier.com/S0043-1354(15)30172-X/sref38http://refhub.elsevier.com/S0043-1354(15)30172-X/sref37http://refhub.elsevier.com/S0043-1354(15)30172-X/sref37http://refhub.elsevier.com/S0043-1354(15)30172-X/sref37http://refhub.elsevier.com/S0043-1354(15)30172-X/sref36http://refhub.elsevier.com/S0043-1354(15)30172-X/sref36http://refhub.elsevier.com/S0043-1354(15)30172-X/sref36http://refhub.elsevier.com/S0043-1354(15)30172-X/sref36http://refhub.elsevier.com/S0043-1354(15)30172-X/sref36http://refhub.elsevier.com/S0043-1354(15)30172-X/sref36http://refhub.elsevier.com/S0043-1354(15)30172-X/sref36http://refhub.elsevier.com/S0043-1354(15)30172-X/sref36http://refhub.elsevier.com/S0043-1354(15)30172-X/sref36http://refhub.elsevier.com/S0043-1354(15)30172-X/sref35http://refhub.elsevier.com/S0043-1354(15)30172-X/sref35http://refhub.elsevier.com/S0043-1354(15)30172-X/sref35http://refhub.elsevier.com/S0043-1354(15)30172-X/sref35http://refhub.elsevier.com/S0043-1354(15)30172-X/sref34http://refhub.elsevier.com/S0043-1354(15)30172-X/sref34http://refhub.elsevier.com/S0043-1354(15)30172-X/sref34http://refhub.elsevier.com/S0043-1354(15)30172-X/sref33http://refhub.elsevier.com/S0043-1354(15)30172-X/sref33http://refhub.elsevier.com/S0043-1354(15)30172-X/sref33http://refhub.elsevier.com/S0043-1354(15)30172-X/sref33http://refhub.elsevier.com/S0043-1354(15)30172-X/sref32http://refhub.elsevier.com/S0043-1354(15)30172-X/sref32http://refhub.elsevier.com/S0043-1354(15)30172-X/sref32http://refhub.elsevier.com/S0043-1354(15)30172-X/sref32http://refhub.elsevier.com/S0043-1354(15)30172-X/sref32http://refhub.elsevier.com/S0043-1354(15)30172-X/sref31http://refhub.elsevier.com/S0043-1354(15)30172-X/sref31http://refhub.elsevier.com/S0043-1354(15)30172-X/sref31http://refhub.elsevier.com/S0043-1354(15)30172-X/sref31http://refhub.elsevier.com/S0043-1354(15)30172-X/sref31http://refhub.elsevier.com/S0043-1354(15)30172-X/sref30http://refhub.elsevier.com/S0043-1354(1

Documents

15 Judd Et Al, Wat Res (Algae)