Bioresource Technology 101 (2010) 8649–8657

Contents lists available at ScienceDirect

Bioresource Technology

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

Ammonia removal from anaerobic digestion effluent of livestock waste usinggreen alga Scenedesmus sp.

Jongmin Park a, Hai-Feng Jin a, Byung-Ran Lim b, Ki-Young Park c, Kisay Lee a,*

a Dept. of Environmental Engineering and Biotechnology, Myongji University, Yongin 449-728, Republic of Koreab Environmental Materials Education Center, Seoul National University of Technology, Seoul, Republic of Koreac Dept. of Civil and Environmental System Engineering, Konkuk University, Seoul, Republic of Korea

a r t i c l e i n f o a b s t r a c t

Article history:Received 3 April 2010Received in revised form 24 June 2010Accepted 26 June 2010

Keywords:Ammonia removalMicroalgaeAnaerobic digestion effluentAlkalinitySemi-continuous operation

0960-8524/$ - see front matter � 2010 Elsevier Ltd. Adoi:10.1016/j.biortech.2010.06.142

* Corresponding author. Tel.: +82 31 330 6689; faxE-mail address: kisay@mju.ac.kr (K. Lee).

The green alga Scenedesmus was investigated for its ability to remove nitrogen from anaerobic digestioneffluent possessing high ammonium content and alkalinity in addition to its growth characteristics.Nitrate and ammonium were indistinguishable as a nitrogen source when the ammonium concentrationwas at normal cultivation levels. Ammonium up to 100 ppm NH4–N did not inhibit cell growth, but diddecrease final cell density by up to 70% at a concentration of 200–500 ppm NH4–N. Inorganic carbon ofalkalinity in the form of bicarbonate was consumed rapidly, in turn causing the attenuation of cellgrowth. Therefore, maintaining a certain level of inorganic carbon is necessary in order to prolong ammo-nia removal. A moderate degree of aeration was beneficial to ammonia removal, not only due to the strip-ping of ammonium to ammonia gas but also due to the stripping of oxygen, which is an inhibitor ofregular photosynthesis. Magnesium is easily consumed compared to other metallic components andtherefore requires periodic supplementation. Maintaining appropriate levels of alkalinity, Mg, aerationalong with optimal an initial NH4

+/cell ratio were all necessary for long-term semi-continuous ammo-nium removal and cell growth.

� 2010 Elsevier Ltd. All rights reserved.

1. Introduction

Livestock wastewater contains a high concentration of nitrogenthat causes eutrophication when discharged without proper treat-ment. Livestock waste is often subjected to anaerobic digestionusing mesophilic or thermophilic bacteria. The main result ofanaerobic digestion is the reduction of organic matter and wastevolume through the fermentative degradation of organicconstituents.

Even so, the levels of nutrients such as ammonia are not com-pletely reduced during anaerobic digestion because the microor-ganisms employed generally lack sufficient autotrophicmetabolism of inorganic nitrogen (Cheng and Liu, 2002; Noikeet al., 2004; Ulundag-Demirer et al., 2008). Traditional bacterialnitrification–denitrification can be used to remove ammonia, butit requires the assimilation of extra organic carbon as a carbonsource. Another important feature of anaerobic digestion is thehigh concentration of alkalinity involved (mostly HCO�3 or CO2�

3 ,1000–5000 ppm as CaCO3). High alkalinity also makes it difficultto apply advanced oxidation processes using ozone or peroxidesbecause bicarbonate acts as a radical scavenger (Ma and Graham,2000; Currie et al., 2003).

ll rights reserved.

: +82 31 336 6336.

Many microalgae species such as green algae and cyanobacteriacan assimilate nitrogen and phosphate into their biomass as wellas inorganic carbon for photosynthesis (Yoshihara et al., 1996;Nagase et al., 2001; Jin et al., 2005). A microalgae system can beemployed as an alternative secondary or post-secondary treatmentprocess to remove nutrients from wastewater, especially when asubstantial amount of nitrogen remains following the reductionof organic matter by traditional biological methods (Przytocka-Ju-siak et al., 1984; Berman and Chava, 1999; Olguin, 2003).

Microalgae have been the subject of recent interest due to theirability to increase growth by uptaking various forms of inorganicnitrogen, including NHþ4 ;NO�3 ;NO�2 , or NO etc. (Nolte and Prouix,1988; Tam and Wong, 1990; Hu et al., 2000; Olguin, 2003; Jinet al., 2008; Park et al., 2009). Microalgae also produce potentiallyvaluable biomass, which can be used as an animal feed additiveand biofuel feedstock for biodiesel and hydrogen (Melis and Happe,2001; Travieso et al., 2006; Tran et al., 2009). It has been recentlysuggested that a microalgae system can be utilized for the treat-ment of flue gas through simultaneous CO2 and NOx fixation, fol-lowed by the conversion of harvested algal biomass to biofuel(Chisti, 2007; Jin et al., 2005).

This study attempted to investigate the characteristics of micro-algal growth along with the removal of ammonia from anaerobicdigestion effluent containing high levels of ammonium and alkalin-ity using the green alga Scenedesmus. The effects of ammonium

8650 J. Park et al. / Bioresource Technology 101 (2010) 8649–8657

content, alkalinity and aeration on cell growth and ammonia re-moval are discussed, and a strategy for long-term semi-continuousoperation is introduced.

2. Methods

2.1. Microalga strain

The green alga Scenedesmus accuminatus (KCTC AG10316) wasobtained from the Biological Resources Center (BRC) of the KoreaResearch Institute of Bioscience and Biotechnology (KRIBB). Cellswere grown at pH 7.5 in modified Bristol medium (NaNO3,250 mg/L; K2HPO4, 75 mg/L; KH2PO4, 175 mg/L; CaCl2, 25 mg/L;NaCl, 25 mg/L; H3BO3, 0.2 mg/L; MgSO4�7H2O, 75 mg/L; FeCl3,0.3 mg/L; MnSO4�4H2O, 0.3 mg/L; ZnSO4�7H2O, 0.2 mg/L; Cu-SO4�5H2O, 0.06 mg/L). 250 mg/L of NaNO3 as a nitrogen source isequivalent to 41 mg-N/L. In order to investigate the effect ofammonium, 157 mg/L of NH4Cl (41 mg-N/L) was used instead.

2.2. Cultivation

Scenedesmus cells were cultivated in a cylindrical glass reactorwith a 1-L working volume. External illuminations were made byfluorescent lamps so that the light intensity at the center of themedium-filled reactor was around 200 lmol/m2/s (Li-250A,Li-COR Lightmeter) with a 12 h:12 h of light:dark cycle.

Table 1Composition of anaerobic digestion effluent wastewater.

Item Concentration (mg/L)

In raw wastewater After pretreatment

pH 8.5 8.4SS 1040 negligibleCODCr 1980 1042SCODCr 1010 1035TOC 370 195NH4-N 1200 1196TN 1240 1220TP 140 75Alkalinity 5990 5562

Time (

0 5 10

Dry

cel

l wei

gh

t (m

g/L

)

0

200

400

600

800

1000

1200

Fig. 1. Cell growth and nitrogen consumption in the pre

2.3. Wastewater

The wastewater used in this study was composed of an anaero-bic digestion effluent obtained from a local piggery farm in Hwa-sung, Korea. In order to prevent interference from othermicroorganisms, wastewater was filtered through a GF/C filterand then autoclaved. The composition of wastewater before andafter this pretreatment is shown in Table 1. After filtration andautoclave, a fraction of SS, COD, TOC and TP was reduced. Mean-while, SCOD, NH4-N, TN and alkalinity were only slightly changed.Since this study focused ammonia removal and the autotrophicmicroalgal growth was influenced by inorganic carbon (alkalinity)and nitrogen source (NH4-N or TN), not by organic carbon (COD,SCOD or TOC), it was considered that the obtained results wouldbe applicable to the raw wastewater used in this study. Ammoniais subject to easily evaporate during autoclave if free ammonium isdominant in water body like domestic wastewater. However, theammonium content is almost invariant in Table 1, because ammo-nium ions exist in salt forms stably with bicarbonate or carbonatedue to high alkalinity in the wastewater used in this study.

Cells were initially cultured in Bristol medium, but wastewaterwas added to the culture suspension when the nitrogen concentra-tion was decreased nearly to zero and the cell density reached adesired value (such as 1.0 g/L). Wastewater was diluted 1/10 uponaddition to the culture, making the nitrogen concentration approx-imately 100 mg/L. Nitrogen removal was performed in batch modeor semi-continuous mode. In semi-continuous mode, new waste-water was added to the culture to increase the nitrogen concentra-tion that had dropped to below 10 ppm back to around 100 ppm.

In order to investigate the influence of metallic components oncell growth, boron, iron, manganese, zinc, copper and magnesiumwere introduced one by one to the ongoing culture when the cellgrowth rate was markedly attenuated.

2.4. Analyses

The algal cell density was expressed as dry cell weight (DCW)per liter of culture suspension. The DCW was evaluated by dryingcells at 85 oC for 24 h after filtration through a GF/C filter. Otheranalyses were performed basically according to standard methods(APHA, 1995). HACH DR4000U was used in determination ofCODCr, SCODCr, TN and TP. Shimadzu 5000A was used for total

day)

15 20

Nit

rog

en (

mg

/L)

0

10

20

30

40

50

Cell / NO3-

Cell / NH4+

NO3-N

NH4-N

sence of nitrate or ammonium as a nitrogen source.

J. Park et al. / Bioresource Technology 101 (2010) 8649–8657 8651

organic carbon (TOC) and inorganic carbon (IC). For NH4-N, HACHDR4000U and Orion electrode were utilized. The Brucine methodwas used to determine NO3-N. Alkalinity was measured throughtitrimetric method. Light intensity was analyses using Li-250Alight meter (Li-COR).

3. Results and discussion

3.1. Preference of nitrogen sources

In order to examine which nitrogen source the Scenedesmus cul-ture prefers, cell growth and nitrogen consumption were comparedin the presence of 45 mg/L NO3-N or NH4-N. Bristol medium con-tained either nitrate or ammonium supplied in the form of NaNO3

or NH4Cl, respectively. Fig. 1 shows that the nitrogen concentrationprofiles were similar regardless of whether nitrate or ammoniumwas used. This result implies that Scenedesmus cells do not differ-entiate between nitrate and ammonium as a nitrogen source, andcan be utilized for nitrogen removal in the treatment of ammo-nia-rich wastewater.

The uptake of ammonium and nitrate is important in microalgalnitrogen removal because nitrogen often exists as ammonium inwastewater (Table 1), especially for livestock wastewater andanaerobically digested wastewater. In fact, the rate of ammoniumuptake is usually inferior to that of nitrate for many microalgaespecies due to the toxicity of ammonia (Przytocka-Jusiak, 1976;Azov and Goldman, 1982; Tam and Wong, 1996). In addition to ni-trate, the Scenedesmus species used in this study was capable ofutilizing ammonium when the ammonium concentration was at

Table 2Cell growth rate and ammonium removal rate depending upon seeding cellconcentrations in Fig. 3.

Seeding celldensity (g/L)

Net growth rate(mg/L/day)

Specific growthrate (day�1)

Ammonium removalrate (mg/L/day)

0.5 45.8 0.091 5.200.75 48.0 0.064 5.871.0 49.4 0.049 6.141.25 53.8 0.043 6.211.5 56.6 0.038 6.46

*Specific growth rate is based upon initial seeding cell concentration.

Tim

0 2 4 6

DC

W (

mg

/L)

0

200

400

600

800

1000

w/o NH4-N

100 ppm 200 ppm 400 ppm 500 ppm 800 ppm 1000 ppm

Fig. 2. Influence of ammonium concentr

normal cultivation levels, resulting in comparable cell growth rates(Fig. 1).

3.2. Growth inhibition by ammonium

Fig. 2 shows the changes in cell density at various initial ammo-nium concentrations ranging from 100 to 1000 ppm NH4-N. Theculture with 100 ppm NH4-N resulted in the best cell growth,which was slightly better than the culture containing 100 ppmNO3-N. The initial growth rates of the culture containing 200–500 ppm NH4-N were similar to those of 100 ppm NH4-N but lev-eled off after 7–8 days to a final cell density about 70% as large.This implied that the levels of inhibition were similar from 200to 500 ppm. The inhibition of cell growth caused by ammoniumbecame severe when the NH4-N content reached 800 ppm, withonly 35% of the final cell density attained in the presence of1000 ppm NH4-N.

It is known that microalgae cells are inhibited at high ammoniaconcentrations, although at low concentrations they can still up-take ammonia (Przytocka-Jusiak, 1976; Azov and Goldman, 1982;Tam and Wong, 1996). Scenedesmus cells used in the current studygrew without any sign of inhibition or toxic influence at 100 ppmNH4-N. The final cell mass was reduced by 30% for 200 to500 ppm NH4-N, although the initial cell growth rate was notaffected.

3.3. Effect of seeding cell concentration

Fig. 3 shows the changes in ammonium content and cell densityat different seeding concentration of Scenedesmus at the beginningof cultivation. This set of experiments was performed using dilutedreal piggery wastewater containing 120 ppm NH4-N. The seedingcell concentration varied from 0.5 g/L to 1.5 g/L DCW. The quanti-tative values of growth rate and ammonium removal rate weresummarized in Table 2. The net cell growth rate for 10 days of cul-tivation was 45.8 mg/L/d at 0.5 g/L seeding and was increased up to55.6 mg/L/d at 1.5 g/L seeding (Fig. 3a). In terms of specific growthrate (Fig. 3b), cells grew relatively rapidly and showed the highestspecific growth rate at the low seeding concentration of 0.5 g/L.The specific growth rate became smaller as the seeding concentra-tion was increased.

Variation in seeding concentration influenced ammonia re-moval during cell cultivation (Fig. 3c and Table 2). Although the

e (day)

8 10 12 14

ation on the growth of Scenedesmus.

Cultivation time (day)

0 2 4 6 8 10 12

Cel

l den

sity

(g

/L)

0.0

0.5

1.0

1.5

2.0

2.5a

c

b

0.5 g/L 0.75 g/L 1.0 g/L 1.25 g/L 1.5 g/L

Cultivation Time (day)

0 2 4 6 8 10 12

Sp

ecif

ic c

ell d

ensi

ty (

C/C

0)

0.8

1.0

1.2

1.4

1.6

1.8

2.0

2.2

2.4

0.5 g/L 0.75 g/L 1.0 g/L 1.25 g/L 1.5 g/L

Cultivation Time (day)

0 2 4 6 8 10 12

NH

4-N

(C

/C0)

0.4

0.5

0.6

0.7

0.8

0.9

1.0

1.1

0.5 g/L 0.75 g/L 1.0 g/L 1.25 g/L 1.5 g/L

Fig. 3. Influence of seeding cell concentration on (a) net cell growth, (b) specific cell growth and (c) ammonium removal.

8652 J. Park et al. / Bioresource Technology 101 (2010) 8649–8657

culture with the lowest seeding concentration, 0.5 g/L, resulted inthe highest ammonia removal rate temporarily during the first

4 days of cultivation, its removal rate was lowered as cultivationtime passed and so the overall removal rate became the lowest.

J. Park et al. / Bioresource Technology 101 (2010) 8649–8657 8653

The ammonium removal rate in cultures of larger seeding size wassteadily maintained during 10 days.

Time

0 2 4

Ino

rgan

ic c

arb

on

(m

g/L

)

0

20

40

60

80

100a

b

c

Time

0 2 4

Cel

l den

sity

(C

/C0)

0.9

1.0

1.1

1.2

1.3

1.4

1.5

1.6

Tim

0 2 4

NH

4-N

(C

/C0)

0.5

0.6

0.7

0.8

0.9

1.0

1.1

No additionBicarbonateCarbonate

Fig. 4. The intermittent addition of inorganic carbon of alkalinity to Scen

Table 2 and Fig. 3 showed that the ammonium removal rate hada parallel relationship with the net growth rate. The overall

(day)

6 8 10

No additionBicarbonateCarbonate

(day)

6 8 10

No additionBicarbonateCarbonate

e (day)

6 8 10

edesmus culture. (a) Inorganic carbon, (b) cell growth and (c) NH4-N.

8654 J. Park et al. / Bioresource Technology 101 (2010) 8649–8657

ammonium removal rate was increased as the seeding concentra-tion was increasing; from 5.20 mg/L/d at 0.5 g/L seeding to6.46 mg/L/d at 1.5 g/L seeding. This observation indicated thatthe ammonium removal by the present Scenedesmus species fol-lowed the substrate utilization kinetics in a batch reactor whereno additional substrate (ammonium as nitrogen source here) issupplied and the product is the increment of biomass itself. Be-cause the final ammonium content reached a similar level after10 days when the seeding concentration ranged from 1.0 to1.5 g/L, the 1.0 g/L seeding condition was routinely used thereafterfor the long-term removal of nitrogen from wastewater.

3.4. Source of alkalinity

Autotrophic algal cells require inorganic carbon for growth;therefore CO2 gas is usually used in many microalgal photobioreac-tors (Keffer and Kleinheinz, 2002; Yamasaki, 2003). Since no CO2

was supplied to the culture in this study, algal cells were forcedto utilize dissolved inorganic carbons in the wastewater. Anaerobi-cally digested wastewater normally contains high level of alkalin-ity (Hill and Bolte, 2000; Bjornsson et al., 2001) of which major

Tim

0 2 4

NH

4-N

(C

/C0)

0.0

0.2

0.4

0.6

0.8

1.0

1.2a

b

Tim

0 2 4

Cel

l den

sity

(C

/C0)

0.9

1.0

1.1

1.2

1.3

1.4

1.5

1.6

Fig. 5. Influence of aeration on amm

constituent is bicarbonate HCO�3 ion. The wastewater used in thisstudy was a digestion effluent containing 5900 – 7500 ppm of alka-linity as CaCO3 (Table 1). The cultivation started with a concentra-tion of 1.0 g of cells per L of diluted wastewater, where inorganiccarbon ranged from 80 to 90 mg/L as C (equivalent to 650–750 mg/L of alkalinity as CaCO3).

Fig. 4 shows the changes in inorganic carbon (alkalinity), celldensity and ammonium concentration during cultivation in thepresence of 120 ppm NH4-N. The inorganic carbon, mostly bicar-bonate, was consumed rapidly to a level near zero within 2 days(Fig. 4a). Along with the exhaustion of inorganic carbon, the cellgrowth was attenuated, resulting in an early approach to station-ary phase (Fig. 4b). Ammonia removal was also decreased and thusonly 13% was removed during 7 days (Fig. 4c).

In order to examine the influence of alkalinity, NaHCO3 orNa2CO3 was supplied intermittently as an external source of inor-ganic carbon (IC) when the level dropped below 5 mg/L on day 2.The level of IC was increased to 85 ppm upon the addition of bicar-bonate on day 2, but this was followed by another drop in IC toaround 10 mg/L after day 5.5. Fig. 4b shows that the growth pat-tern until day 3 was similar to that without bicarbonate. However,

e (day)

6 8 10

without aerationwith aeration

e (day)

6 8 10

without aerationwith aeration

onia removal and cell growth.

J. Park et al. / Bioresource Technology 101 (2010) 8649–8657 8655

from day 3 the cell density was steadily increased without attenu-ation while the ammonium removal rate was maintained highly asthe period of 0 – 2 days.

The addition of carbonate affected differently compared tobicarbonate addition. Growth rate was not improved by carbonateaddition, showing similar growth with no-addition case (Fig. 4b).Ammonium removal rate was a little recovered compared to no-addition case, but its extent was much smaller than that of bicar-bonate addition (Fig. 4c). The reason why carbonate addition didnot improve growth rate is not clear yet and more investigationis required; however, it is suspected that carbonate addition raisedthe culture pH up to 10, which did not favor normal cell growthand carbon uptake. The pH value of the culture decreased to 7.5from the initial value of pH 8.6 upon depletion of inorganic carbon(day 2) and then rose to 8.5 by adding bicarbonate, which meansthat pH was maintained in the range of 7.5–8.6 during ammoniaremoval. Meanwhile, in the case carbonate addition, pH increasedup to 10.3 which was beyond the range for normal growthalthough microalgal cells prefer weak alkaline pH for their growth.After carbonate addition (day 2), the carbon uptake rate was actu-ally reduced to nearly half of the bicarbonate addition case(Fig. 4a). The high pH also favors ammonia stripping by shifting

Tim

0 5 10

Cel

l den

sity

(g

/L)

0.8

1.0

1.2

1.4

1.6

1.8

2.0

2.2

2.4

2.6a

b

(I) (II) (

B F N

Tim

0 5 10

NH

4-N

(p

pm

)

0

20

40

60

80

100

120

Fig. 6. Profiles of cell density and ammonium concentration in semi-continuous treZ = ZnSO4; C = CuSO4; M = MgSO4.

equilibrium from ammonium NHþ4 to ammonia (NH3), which couldbe a reason that the extent of ammonia removal was much greaterthan that of cell growth in the case of carbonate addition.

These results demonstrate that inorganic carbon in the form ofbicarbonate is actively consumed during algal cell growth andammonium removal, and therefore that maintaining certain levelof inorganic carbon is necessary to prolong ammonia removal.The addition of bicarbonate had more positive effects on cultiva-tion than did the addition of carbonate. The bicarbonate ion, whichincreases the alkalinity for anaerobic digestion, was successfullyutilized as a carbon source by autotrophic microalgae grown inammonium-rich wastewater.

Even though the addition of bicarbonate promoted the continu-ous growth of cells, the growth rate during the second period (days2 – 5) looked somewhat inferior to the initial activity (days 0 – 2).Firstly, the declining level of inorganic carbon (Fig. 4a) during days2 – 5 was lower than the rate obtained during days 0 – 2. The cellgrowth rate during days 2 – 5 was also smaller than that duringdays 0 – 2, even though the amount of added bicarbonate wasidentical to the initial level. This difference is perhaps due to thedepletion of other nutritional components such as trace metalsduring the second period (days 2 – 5).

e (day)

15 20 25 30

III) (IV) (V)

Z C M

MgMg

e (day)

15 20 25 30

atment of wastewater using Scenedesmus cells. B = H3BO3; F = FeCl3; N = MnSO4;

8656 J. Park et al. / Bioresource Technology 101 (2010) 8649–8657

3.5. Influence of aeration

In order to determine the influence of aeration on ammoniumremoval and cell growth, sterilized air was supplied at a rate of0.75 vvm (volume to volume per minute). The algal culture wascontrolled at pH 7.5. As seen in Fig. 5, aeration provided significantbenefits to ammonium removal while simultaneously increasingalgal biomass; Ammonium removal rate was 6.3 mg/L/day withoutaeration and 18.4 mg/L/day with aeration. Net cell growth rate was66.8 mg/L/day without aeration and enhanced to 118 mg/L/daywith aeration. This increase in NH4-N removal rate is probablydue to the stripping effect promoted by aeration. It is well knownthat aqueous ammonia can be stripped away from water by aera-tion, a process highly favored in alkaline pH above 9.5.

Nevertheless, ammonia stripping could not be the only causeof ammonium reduction in this study. Besides ammonia, dis-solved oxygen can be stripped away from water, an importantnotion in treating wastewater with photosynthetic microalgae.Molecular oxygen normally evolves as a consequence of photo-synthesis while dissolved oxygen inhibits regular photosynthesisand carbon fixation (Merrett and Armitage, 1982; Markez et al.,1995). The concentration of dissolved oxygen becomes higher inun-aerated algal culture compared to the aerated one, creatingan unfavorable situation for algal cell growth and algal ammoniauptake.

This speculation is supported by the comparison of dissolvedoxygen (DO) values and the enhanced cell density produced inaerated culture (Fig. 5b). The rate of oxygen evolution by microal-gal photosynthesis and the resulting DO increase in the culture isvery fast, and so it is known that the DO value can exceed theambient saturation value (Rubio et al., 1999; Miron et al., 2002).According to separated experiments, the DO value in the culturewas 0.76 mg/L initially and it was maintained without change iflight was not supplied. However, once light illuminated, DO in-creased so rapidly that it exceeded 6.2 mg/L within 18 min and10 mg/L within 35 min. The value 6.2 mg/L was the saturatedDO value that was obtained by continuous bubbling with air atambient condition through the culture containing Scenedesmuscells. DO value could rise even higher than 2.5-folds of saturationvalue after 60 min if inorganic carbon was sufficient. Meanwhile,DO value was kept constant around 6.2 mg/L when aeration wascarried out. Therefore, the use of appropriate aeration was benefi-cial to enhance microalgal ammonium uptake by preventing DOlevel from going up as well as to strip ammonia out of the waste-water. Aeration using inert gases like nitrogen or argon, instead ofcheap air, would achieve further lowered DO value.

3.6. Semi-continuous treatment

Semi-continuous type ammonia removal was performed withrepeated cell withdrawal and wastewater supplement. When theammonium level was dropped to below 10 ppm, a fraction approx-imately 1/10 of the culture broth and cells was withdrawn. Newwastewater was then supplied so the culture could be adjustedto 90–100 ppm NHþ4 , 90 ppm alkalinity and 1 g/L of cell mass(Fig. 6).

Initially, cells harvested from a stock culture were seeded to aconcentration of 1 g/L. The first cycle lasted 5 days during whichthe cell mass more than doubled and 89% of ammonium was suc-cessfully removed. In the second cycle, however, the cell growthrate was decreased to half and the extent of ammonium removalwas also reduced. Therefore, fresh wastewater was supplied atday 11, but no sign of performance recovery was observed. Follow-ing this, metallic components (B, Fe, Mn, Zn, Cu or Mg) were addedindividually with 1-day intervals to the culture medium. Fig. 6shows that none of the elements supplied to the culture produced

any positive effect, except for Mg which increased the cell density.Therefore, Mg was supplied repeatedly at the beginning of the 4thand 5th cycles and caused normal cell growth and ammonium re-moval just like that observed in the 1st cycle. The averaged valuesof cell growth rate and ammonia removal rate in the 1st, 4th and5th cycles were 213 and 19.2 mg/L/day, respectively.

It is speculated that the other metallic elements tested besidesMg were not limiting because they were supplied in sufficientamounts when new wastewater was added. Mg is an importantcomponent of photosynthetic pigments in many algal species (Fin-kel and Appleman, 1963; Walker, 1994). Therefore, it can be con-cluded that the addition of Mg along with the maintenance ofalkalinity and optimal initial NHþ4 /cell ratio are critical for stablelong-term ammonia removal during semi-continuous treatmentusing microalgae.

4. Conclusions

The bicarbonate ion, which is constituent of the alkalinity foranaerobic digestion, was actively consumed as a carbon source byautotrophic microalgae grown in ammonium-rich wastewater.Therefore, maintaining a certain level of inorganic carbon is neces-sary to prolong ammonia removal. In a long-term semi-continuousprocess, Mg is easily consumed compared to other metallic compo-nents and thus requires periodic supplementation. The concertedmaintenance of alkalinity, Mg levels, optimal initial NHþ4 /cell ratioand moderate aeration is required in order to achieve stable long-term ammonia removal in a semi-continuous process usingmicroalgae.

Acknowledgements

This study was financially supported by Korea Research Foun-dation Grant funded by the Korean Government (MOEHRD), KRF-2006-D00131. Jongmin Park is also thankful to scholarships fromthe BK21 Program of the Ministry of Education, Korea.

References

APHA, 1995. Standard Methods for the Examination of Water and Wastewater, 19thed., APHA, Washington, DC.

Azov, Y., Goldman, J.C., 1982. Free ammonia inhibition of algal photosynthesis inintensive cultures. Applied and Environmental Microbiology 43, 735–739.

Berman, T., Chava, S., 1999. Algal growth on organic compounds as nitrogensources. Journal of Plankton Research 21, 1423–1437.

Bjornsson, L., Murto, M., Jantsch, T.G., Mattiasson, B., 2001. Evaluation of newmethods for the monitoring of alkalinity, dissolved hydrogen and the microbialcommunity in anaerobic digestion. Water Research 35, 2833–2840.

Cheng, J., Liu, B., 2002. Swine wastewater treatment in anaerobic digesters withfloating medium. Transactions of the American Society of Agriculture andBiological Engineers 45, 799–805.

Chisti, Y., 2007. Biodiesel from microalgae. Biotechnology Advances 25, 294–306.Currie, M., Graham, N., Hall, T., Lambert, S., 2003. The effect of bicarbonate on

ozone-enhanced particle removal in water treatment. Ozone Science andEngineering 25, 285–293.

Finkel, B.J., Appleman, D., 1963. The effect of magnesium concentration on growthof Chlorella. Plant Physiology 28, 664–673.

Hill, D.T., Bolte, J.P., 2000. Methane production from low solid concentration liquidswine waste using conventional anaerobic fermentation. BioresourceTechnology 74, 241–247.

Hu, Q., Westerhoff, P., Vermaas, W., 2000. Removal of nitrate from groundwater bycyanobacteria: quantitative assessment of factors influencing nitrate uptake.Applied and Environmental Microbiology 66, 133–139.

Jin, H.F., Santiago, D., Park, J., Lee, K., 2008. Enhancement of nitric oxide solubilityusing Fe(II)EDTA and its removal by green algae Scenedesmus sp. Biotechnologyand Bioprocess Engineering 13, 48–52.

Jin, Y., Veiga, M.C., Kennes, C., 2005. Bioprocesses for the removal of nitrogen oxidesfrom polluted air. Journal of Chemical Technology and Biotechnology 80, 483–494.

Keffer, J.E., Kleinheinz, G.T., 2002. Use of Chlorella vulgaris for CO2 mitigation in aphotobioreactor. Journal of Industrial Microbiology and Biotechnology 29, 275–280.

J. Park et al. / Bioresource Technology 101 (2010) 8649–8657 8657

Markez, F.J., Sasaki, K., Nishio, N., Nagai, S., 1995. Inhibitory effect of oxygenaccumulation on the growth of Spirulina platensis. Biotechnology Letters 17,222–228.

Ma, J., Graham, N., 2000. Degradation of atrazine by manganese-catalysedozonation – influence of radical scavengers. Water Research 34, 3822–3828.

Melis, A., Happe, T., 2001. Hydrogen production: green algae as a source of energy.Plant Physiology 127, 740–748.

Merrett, M.J., Armitage, T.I., 1982. The effect of oxygen concentration onphotosynthetic biomass production by algae. Planta 155, 95–96.

Miron, A.S., Garciam, M.-C.C., Camacho, F.G., Grima, E.M., Chisti, Y., 2002. Growthand biochemical characterization of microalgal biomass produced in bubblecolumn and airlift photobioreactors: studies in fed-batch culture. EnzymeMicrob. Technol. 31, 1015–1023.

Nagase, H., Yoshihara, K.I., Eguchi, K., Okamoto, Y., Murasaki, S., Yamashita, R.,Hirata, K., Miyamoto, K., 2001. Uptake pathway and continuous removal ofnitric oxide from flue gas using microalgae. Biochemical Engineering Journal 7,241–246.

Noike, T., Goo, I.S., Matsumoto, H., Miyahara, T., 2004. Development of a new type ofanaerobic digestion equipped with the function of nitrogen removal. WaterScience and Technology 49, 173–179.

Nolte, J., Prouix, D., 1988. Biological tertiary treatment of urban wastewaters withchitosan-immobilized Phormidium. Applied Microbiology and Biotechnology 29,292–297.

Olguin, E.J., 2003. Phycoremediation: Key issues for cost-effective nutrient removalprocesses. Biotechnology Advances 22, 81–92.

Park, K.Y., Lim, B.R., Lee, K., 2009. Growth of microalgae in diluted process water ofthe animal wastewater treatment plant. Water Science and Technology 59,2111–2116.

Przytocka-Jusiak, M., 1976. Growth and survival of Chlorella vulgaris in highconcentrations of nitrogen. Acta Microbiolica Polonica 25, 287–289.

Przytocka-Jusiak, M., Duszota, M., Matusiak, K., Mycielski, R., 1984. Intensive cultureof Chlorella vulgaris as the second stage of biological purification of nitrogenindustry wastewaters. Water Research 18, 1–7.

Rubio, F.C., Fernandez, F.G.A., Perez, J.A.S., Camacho, F.G., Grima, E.M., 1999. Carbondioxide concentration profiles in tubular photobioreactors for microalgalculture. Biotechnol. Bioeng. 62, 71–86.

Tam, N.F.Y., Wong, Y.S., 1990. The comparison of growth and nutrient removalefficiency of Chlorella pyrenoidosa in settled and activated sewages. EnvironmentalPollution 65, 93–108.

Tam, N.F.Y., Wong, Y.S., 1996. Effect of ammonia concentrations on growth ofChlorella vulgaris and nitrogen removal from media. Bioresource Technology 57,45–50.

Tran, H.L., Hong, S.J., Lee, C.G., 2009. Evaluation of extraction methods for recoveryof fatty acids from Botryococcus braunii LB 572 and Synechocystis sp. PCC 6803.Biotechnology and Bioprocess Engineering 14, 187–192.

Travieso, L., Benitez, F., Sanchez, E., Borja, R., Martin, A., Colmenarejo, M.F., 2006.Batch mixed culture of Chlorella vulgaris using settled and diluted piggerywaste. Ecological Engineering 28, 158–162.

Ulundag-Demirer, S., Demirer, G.N., Frear, C., Chen, S., 2008. Anaerobic digestion ofdairy manure with enhanced ammonia removal. Journal of EnvironmentManagement 86, 193–200.

Walker, G.M., 1994. The role of magnesium in biotechnology. Critical Reviews inBiotechnology 14, 311–354.

Yamasaki, A., 2003. An overview of CO2 mitigation options for global warming-emphasizing CO2 sequestration options. Journal of Chemical Engineering Japan36, 361–375.

Yoshihara, K.I., Nagase, H., Eguchi, K., Hirata, K., Miyamoto, K., 1996. Biologicalelimination of nitric oxide and carbon dioxide from flue gas by marinemicroalga NOA-113 cultivated in a long tubular photobioreactor. Journal ofFermentation and Bioengineering 82, 351–354.