Biological nitrogen removal from landfill leachate using anaerobic–aerobic process Denitritation

Biological nitrogen removal from land?ll leachate using anaerobic–aerobic process:Denitritation via organics in raw leachate and intracellular storage polymers

"

Endogenous DNR changed with type alteration of internal carbon source used.

a r t i c l e i n f o Article history:

Received 7August 2012

Received in revised form 13October 2012Accepted 13October 2012

Available online 23October 2012Keywords:ASBR

Pulsed SBR

Endogenous denitritation Land?ll leachate

a b s t r a c t

A system which combined ASBR with pulsed SBR (PSBR)was introduced to enhance COD and nitrogen removal from the real land?ll leachate.ASBR was used to degrade the organics from raw leachate mainly.Three equal feeds mode was applied in PSBR operation.The results obtained from the joint operation per-iod (157days)show that the COD removal rate of ASBR was 83–88%under the speci?c loading rate of 0.43–0.62gCOD gVSS à1day à1.PSBR’s operation can be divided into four phases according to the different in?uent NH 4+-N which increased to 800–1000mg L à1?nally,and total nitrogen (TN)removal rate of more than 90%with the ef?uent TN of less than 40mg L à1was obtained.PH

B and glycogen can act as electron donor for endogenous denitritation orderly with the hypothetical function from DNGAOs.Con-sequently,the system achieved COD and TN removal rate of 89.61–96.73%and 97.03–98.87%,respec-tively,without any extra carbon source addition.

ó2012Elsevier Ltd.All rights reserved.

1.

Introduction

Sanitary land?ll is widely used in municipal solid waste (MSW)treatment throughout the world because of the convenience and low capital cost.As a result,the rainwater percolates through the waste material and the biodegradation of the MSW

organic frac-tion generates a severely contaminated leachate,which is charac-terized by high concentration of organics,nitrogen,inorganic salts and heavy metals.If the land?ll leachate has not been collected cautiously and discharge safely,it could be the potential pollution source that contaminates soil,surface water and ground-water (Nehrenheim et al.,2008).

At present,the chief methods to treat sanitary land?ll leachate are physico-chemical and biological processes.Some researches (Uygur and Kargi,2004;Park et al.,2001;Chiang et al.,2001)found

0960-8524/$-see front matter ó2012Elsevier Ltd.All rights reserved.https://www.sodocs.net/doc/2e1455230.html,/10.1016/j.biortech.2012.10.063

Corresponding authors.Tel./fax:+861067392627.

E-mail addresses:wsy@https://www.sodocs.net/doc/2e1455230.html, (S.Wang),pyz@https://www.sodocs.net/doc/2e1455230.html, (Y.Peng).

that the high concentration of ammonium is the main reason for causing a low ef?ciency in biological treatment of land?ll leachate and physico-chemical process is a necessary pretreatment.The most widely used physico-chemical processes are chemical precip-itation(Ozturk et al.,2003),ammonium stripping(Castrillon et al., 2010)and reverse osmosis(Hasar et al.,2009).Although the phys-ico-chemical processes have advantages of simplicity,ef?cient, etc.,their bene?ts are counteracted by drawbacks like high opera-tional costs and energy consumption.Consequently,the biological processes,such as ANAMMOX(Tao et al.,2012)and anaerobic–aerobic system(Peng et al.,2008),become the feasible technolo-gies for removing organics and nitrogen from land?ll leachate. Between them,anaerobic–aerobic system is more suitable for immature land?ll leachate treatment.The results from previous studies(Peng et al.,2008;Huo et al.,2008;Hoilijoki et al.,2000) show that most COD and NH4+-N in the raw leachate can be removed via methanogenesis and nitri?cation,however,TN removal becomes an intractable problem because of the shortage of available carbon source.

In terms of immature leachate,the COD concentration could reach above10,000mg Là1,and the COD/NH4+-N ratio could ex-ceed10:1.In the previous studies,anaerobic technique was used to convert the organics into CO2and CH4via methanogenesis, and aerobic/anoxic reactor was utilized to realize nitri?cation/ denitri?cation.Consequently,the in?uent of aerobic/anoxic reactor is always with low COD/NH4+-N ratio because of the previous methanogenesis.As a result,organics in raw leachate were wasted and extra carbon source must be added in conventional anaerobic–aerobic system in order to obtain satisfying TN removal ef?ciency, otherwise only NH4+-N could be oxidized and NO xà-N could not be denitri?ed thoroughly.

Biomass can store excess organics as intracellular storage poly-mers under the dynamic conditions,in which the stored polymers like polyhydroxybutyrate(PHB)and glycogen can act as electron donor for denitri?cation(endogenous denitri?cation)when there is no available external carbon source presenting in wastewater (Beun et al.,2000;Majone et al.,1998;Vocks et al.,2005).For the purpose of realization of advanced nitrogen removal without extra carbon source addition,taking full advantage of organics in raw leachate and intracellular storage polymers of microorganism would be necessary.To authors’knowledge,advanced nitrogen re-moval form leachate through endogenous denitri?cation was never reported,as well as the exact types of internal carbon sources used in that.

This work presents a lab-scale anaerobic sequencing batch reac-tor(ASBR)combining pulsed sequencing batch reactor(PSBR)bio-logical system adapting the removal of organics and nitrogen as a feasible process for the treatment of land?ll leachate.The objective of this study is to investigate the possibility of achieving advanced nitrogen removal from land?ll leachate without extra carbon source addition by utilizing the organics in raw leachate and intra-cellular storage polymers of https://www.sodocs.net/doc/2e1455230.html,anics and nitrogen removal performance of this biological system was monitored,and variation of nitrogen and intracellular storage polymers during the typical cycle was also investigated.Meanwhile a hypothesis was made to explain which group of microorganism is responsible for nitrogen removal using intracellular storage polymers.

2.Methods

2.1.Experimental lab-scale reactor

Fig.1shows the experimental system,which includes an ASBR and a pulsed SBR.

The raw leachate stored in the feed tank was used as the in?uent of ASBR.And leachate in equalization tank utilized as the in?uent of PSBR was a mix of ef?uent from ASBR,raw leachate and ef?uent from PSBR with a proper ratio in order to adjust the COD/NH4+-N ratio from3to5,of these,the raw leachate and ef?u-ent of PSBR fed into the equalization tank through the bypass pipe and recirculation pipe,respectively.The working volume of feed tank and equalization tank was9L and10L for ASBR and PSBR which were made from polymethyl methacrylate.A pH meter and mechanical stirrer were installed inside the ASBR.For PSBR, a pH meter,DO meter,ORP meter,mechanical stirrer and three air diffusers were set up.Temperature adjusting was realized through temperature controller and heating band both for ASBR and PSBR.

2.2.Inoculums

The anaerobic activated sludge from Liulitun Land?ll Leachate Treatment Plant(Beijing,China)was inoculated in the ASBR with initial MLSS and MLVSS of19,780mg Là1and8775mg Là1,respec-tively.On the other hand,inoculums for PSBR was aerobic acti-vated sludge from a SBR mainly treating the mature leachate, whose initial MLSS,MLVSS and SV were5365mg Là1,4237mg Là1 and27%,respectively.

2.3.Raw leachate

Raw leachate for this experiment was supplied from the Liuli-tun MSW Sanitation Land?ll Site(Beijing,China),and it was con-served at4°C to prevent the natural degradation of organics.The raw leachate,whose BOD5/COD ratio and COD/NH4+-N ratio reached0.5–0.7and7–9,respectively,with the comparably high concentration of COD,can be characterized as immature one.The average values of the principal chemical compounds concentration are summarized in Table1.

2.4.Experimental procedure

The experiment was divided into start-up and joint operation period.In start-up period,ASBR and PSBR operated separately,then in the joint operation period,the system operated with ASBR link-ing to the PSBR.The operational modes of ASBR during these two periods were same.One ASBR standard cycle consisted of a rapid ?ll phase(10min),an anaerobic phase(22.5h),a settle phase (30min),an ef?uent withdrawal phase(20min),an idle phase (30min),in which case5L of ef?uent was discharged,resulting in a hydraulic retention time(HRT)of2days.The MLVSS of ASBR was kept at8000–9000mg Là1under the sludge retention time (SRT)of27days approximately.Temperature was maintained at 35°C.During the joint operation period,the ef?uent of ASBR, raw leachate and ef?uent of PSBR went into the equalization tank under a proper volumetric ratio in order to obtain a demanded COD/NH4+-N(3–5),consequently,the ef?uent of ASBR would be exceeding,so it was used to do other researches in the lab.

In joint operation period,three equal feeds mode(i.e.three times of in?uent?lling with the same volume)was applied for PSBR.Discharging of3L ef?uent every cycle led to1L in?uent ?lled for3times under three equal feeds mode.Fig.2shows the operational mode of PSBR.One standard cycle of PSBR could be concluded as follows:An1?O1?An2?O2?An3?O3? An4?sedimentation?withdrawal as described.Every day one standard cycle was implemented.No extra carbon source was added at the end of O3,which led to implementation of denitrita-tion in An4with the internal carbon source(endogenous denitrita-tion).All the three anoxic stages(except An4)lasted for1h,while the time of rapid?ll phase,settle phase and ef?uent withdrawal phase was2min,30min and10min,respectively.However,the time of idle phase was not?xed.The MLSS and MLVSS of PSBR

402R.Zhu et al./Bioresource Technology128(2013)401–408

was maintained at 4500–5500mg L à1and 4200–4800mg L à1,respectively,under the SRT of around 18days,and the operational temperature maintained at 25°C.Air ?ow rate kept at 100L h à1using gas-?owmeter,which resulted in the DO concentration of less than 2.5mg L à1.

Meanwhile,operational mode of PSBR during start-up period was the same as the one in joint operation period,with the excep-tion of the addition of 2.5g acetate into the reactor at the end of O3.pH variation was applied to indicate the completion of nitrita-tion during all oxic stages,while the terminal point of An4was indicated with ORP variation.

2.5.Typical cycle tests of PSBR in the joint operation period

Dividing the joint operation period of PSBR into phase I,phase II,phase III and phase IV (see the performance of PSBR,below),typical cycle tests were conducted in each phase,in which case principal chemical compound concentrations like COD,NH 4+-N,NO 3à-N,NO 2à-N,TP,PHB and glycogen were monitored.Every typical cycle test of each phase was performed three times with the same leach-ate from the equalization tank.The experimental values of the typ-ical cycle tests were the average of three samples,and the results also showed over 95%con?dence.2.6.Analytical methods

The ammonia nitrogen (NH 4+-N),nitrate (NO 3à-N),nitrite (NO 2à-N),chemical oxygen demand (COD),total phosphorus (TP),MLSS/MLVSS and alkalinity were measured according to the

standard methods (APHA,1995).Total nitrogen (TN)was analyzed using TN/TOC analyzer (Multi N/C3000,AnanltikjenaAG,Germany).BOD 5was measured by Oxitop Control WTW.DO,pH and ORP were monitored using pH/Oxi 340i analyzer (WTW Company,Germany).

Glycogen was extracted and hydrolyzed from activated sludge by heating a sample of a known amount of freeze-dried activated sludge (typically 20mg)in a known volume of 0.6M HCl (typically 5mL)for 5h at 100°C.The glucose content in the supernatant was determined by HPLC.

PHB was determined by the method of Zeng et al.(2003).2.7.Calculations

The ef?ciency of NH 4+-N,NO 3à-N,NO 2à-N,COD and TN removal was estimated according to:

Efficiency e%T?

C inf emg L à1TàC eff emg L à1T

C inf emg L T

?100e1T

where C inf is the NH 4+-N (or NH 4+-N,NO 3—N,etc.)concentration in the in?uent and C eff is the one in the ef?uent.

The PHB fraction of active biomass (f PHB )in Cmol per total Cmol can be calculated as:

f PHB eCmol =Cmol T?

PHB eCmol T

e2T

Calculation by this method considers the slightly different com-position and molecular weight of PHB (CH 1.5O 0.5,21.5g Cmol à1)and net biomass (CH 1.8O 0.5N 0.2,24.6g Cmol à1),where the Xnet can be calculated from:

X net eg L à1T?DW eg L à1TàPHB eg L à1T

e3T

where DW is the dry weight of biomass.

The glycogen fraction of active biomass (f Gly )can be calculated as the same of f PHB :

f Gly eCmol =Cmol T?

Glycogen eCmol T

X net

eCmol TtGlycogen eCmol T

e4T

where the molecular weight

of

glycogen

is

27g Cmol à1

(CH 1.67O 0.83

).

of the ASBR +PSBR system:(1)feed tank,(2)peristaltic pump,(3)mechanical stirrer,(4)alkali liquor collection outlet,(9)equalization tank,(10)temperature controller,(11)air compressor,(12)air diffuser,(13)pH meter,Table 1

Characteristics of the raw leachate from the Liulitun MSW Sanitation Land?ll Site of Beijing City.Values are in mg L à1,except the https://www.sodocs.net/doc/2e1455230.html,pound Mean Max Min SD a COD 852810,4487341752BOD 5566969523758432NH 4+-N 11541305104966NO 3à-N 2.21 3.460.110.31NO 2à-N 1.23 1.920.060.18TN

13841578127965Alkalinity 10,23011,32082101020pH 8.08.27.70.2TP

5.2

9.6

3.3

0.8

a

Standard deviation.

The theoretical denitritation rate (TDNR)of pulsed SBR’s An4can be expressed as:

TDNR emgN h

à1

gVSS à1

T?

NO à2-N ini emg L à1TàNO à2-N end emg L à1

T

t eh TáMLVSS eg L T

e5T

where NO 2à-N ini is the initial nitrite concentration at the start of

An4,NO x à-N end is the nitrite concentration at the end of An4,t is the reaction time of An4,MLVSS is the mixed liquor volatile sus-pended solid concentration of PSBR.

The anoxic substrate (glycogen)utilization rate r s is calculated by:

r s emgCOD h

à1

gVSS à1T?

1:71?DNR Gly emgN h à1

gVSS à1T

1àY H emgCOD =mgCOD T

e6T

where the DNR Gly is the DNR based on glycogen and Y H is the bio-mass yield coef?cient.3.Results and discussion

3.1.Overview of system performance

As shown in Fig.3,during the whole joint operation period last-ing for 157days,the in?uent COD of system was 7341–10,448mg L à1,when ef?uent COD was less than 910mg L à1,which resulted in COD removal rate of 89.61–96.73%.On the other hand,the in?uent TN of system reached 1279–1578mg L à1,while ef?u-ent TN was below 40mg L à1,which led to the TN removal rate of

nitri?cation.The results from Klimiuk and Kulikowaka (2004),Tsilogeorgis et al.(2008)and Spagni and Marsili-Libelli (2010)(Table 2)ensured that extra carbon source addition seems to be necessary in the anoxic stage,in order to reduce the TN concentra-tion in the ef?uent.Apparently,this would raise the operational cost drastically for the real land?ll leachate treatment plant.On the other hand,although Peng et al.(2008)realized relatively out-standing nitrogen removal without extra carbon source addition,it still could not meet the stricter and stricter TN concentration discharge standard for leachate around the world,like China (<40mg L à1).In the case of land?ll leachate treatment,this work for the ?rst time shows that advanced nitrogen removal with the ef?uent TN of less than 40mg L à1can be obtained via the biologi-cal process,in which no carbon source addition happened.3.2.The performance of ASBR

As Fig.4a presents,the start-up period of ASBR can be divided into two parts (part I and part II).In part I,the raw leachate mixed with the tap water (1:1)was used as in?uent,which resulted in the in?uent COD of 3500–5500mg L à1corresponding to the COD spe-ci?c loading rate (SLR)of 0.20–0.32gCOD gVSS à1day à1,and the ef?uent COD reached 600–800mg L à1after 10days of acclimatiza-tion.In part II which lasted for 10days,too,the in?uent was the raw leachate with the in?uent COD of 7000–11,000mg L à1,which led to the SLR of 0.41–0.64gCOD gVSS à1day à1.For the last 5days of part II,the performance of ASBR tended to be stabilized with the ef?uent COD of 1000–1200mg L à1,which meant the completion of acclimatization.Due to the well adaption for inoculums to the leachate before inoculation,the duration of acclimatization was surprisingly short.

After the success of the acclimatization,ASBR operated combining with the PSBR (i.e.joint operation period)for 157days with the raw leachate as in?uent.Fig.4a shows that the in?uent COD was 7341–10,448mg L à1corresponding to the SLR of 0.43–0.62gCOD gVSS à1d à1,while ef?uent COD varied between 911and 1355mg L à1,which resulted in the COD removal rate of more than 83%.However,the ef?uent NH 4+-N was always higher than the in?uent NH 4+-N,this phenomenon might ascribe to the break-down of organic nitrogen to ammonium because of the relatively high concentration of organic nitrogen in leachate.3.3.The performance of pulsed SBR(PSBR)

During the start-up period,the mixture with a proper ratio of raw leachate,ef?uent of ASBR and tap water was applied as the in?uent of PSBR,whose NH 4+-N and COD concentrations were 248.1–252.0mg L à1and 893–1032mg L à1,respectively,according to Fig.4b.In the 20days of inoculation (with extra carbon source addition at the beginning of An4),the ef?uent NH 4+-N,TN and COD reached less than 10,20and 250mg L à1,respectively.On the other hand,at the end of each oxic stage,nitrite accumulated over 90%(not shown in Fig.4b)at the very start,this might be attributed to excellent nitritation performance of the inoculums (Zheng et al.,2012).The nitritation and denitritation performance

Fig.2.Operational mode of the PSBR in joint operation period.

Fig.3.System performance in the joint operation period.

went well and to be stable(the removal rate difference of each ef?uent index was less than10%within5cycles).

In the joint operation period(with no extra carbon source addi-tion),the ratio of raw leachate,ef?uent of ASBR and PSBR was ad-justed to obtain the demanded concentration of NH4+-N and COD/ NH4+-N ratio.The mixed leachate was used as the PSBR’s in?uent whose NH4+-N concentration increased step by step for4phases (phase I,phase II,phase III,phase IV),while the in?uent COD/ NH4+-N ratio was always kept within the range of3–5.Table3 shows the parameters controlled in different phases.

As shown in Fig.4b,the in?uent TN of PSBR changed from 301.88–365.80mg Là1in phase I to920.52–1105.54mg Là1in phase IV,and the ef?uent TN of all phases kept beneath40mg Là1, which meant the TN removal ef?ciency of more than90%.Ad-vanced nitrogen was achieved without extra carbon source addi-tion because the2nd and3rd?lling of in?uent provided enough organics for denitritation in An2and An3,respectively,through the three equal feeds mode.

The in?uent NH4+-N,which reached829.30–1000.66mg Là1in phase IV(data not shown),improved with the TN concentration increasing,while the ef?uent NH4+-N of each phase was less than 20mg Là1,which resulted in realization of NH4+-N removal rate over93%.

The above results about the variations of TN and NH4+-N re-moval ef?ciencies were quite opposite to these from Xie et al. (2012),which demonstrated that the system used in this study had better resistance when nitrogen loading improved.

With ensuring the proper COD/NH4+-N ratio of in?uent,the in?uent COD also increased along with the rise of NH4+-N concen-tration.Different form the relative constancy of ef?uent TN and NH4+-N during all the phases,the ef?uent COD went up from 298.0–402.5mgáLà1in phase I to771.74–901.48mg Là1in phase IV.That might be attributed to the existence of nondegradable organics in leachate.

3.4.Evolution of the nitrogen during the oxic stages over the typical PSBR cycle

It is found that the changes in concentration of NH4+-N,NO2à-N, COD,PHB,glycogen,etc.over a PSBR cycle had almost the same disciplines for all the phases,so the cycle pro?le of phase IV (Fig.5)was analyzed as an example for the purpose of studying more details in the PSBR process.According to Fig.5a,under the process of nitritation,NH4+-N concentration at the end of each oxic stage was below14mg Là1.The outstanding NH4+-N removal was attributed to the three equal feeds mode,which resulted in the highest NH4+-N concentration occurring at the start of the?rst?ll-ing just5/12of the highest NH4+-N concentration theoretically when one feed mode was applied in SBR.As a result,lower concen-tration of free ammonia(FA)could inhibit the nitritation.On the other hand,denitritation in An2and An3provided extra alkalinity for nitritation in O2and O3.

Fig.5a shows that the duration of O1was obviously longer than the one of O2and O3,whose reasons can be explained as follows: (1)Most COD of the2nd and3rd in?uent?lling was used as the carbon source for denitritation in An2and An3,which led to the lit-tle in?uence of COD on nitritation in O2and O3.As a result,nitri-tation started at the beginning of O2and O3,but it proceeded after the organics degradation?nished in O1because the autotrophic bacteria were at competitive disadvantage compared with hetero-trophic bacteria.(2)The inhibition of FA to nitritation reduced at the beginning of O2and O3because of the dilution.

As Fig.5b presents,NO2à-N accumulated above90%at the end of O1,O2and O3,which meant the achievement of partial nitri?-cation(nitritation).

3.5.Evolution of the nitrogen during the anoxic stages over the typical PSBR cycle

As shown in Fig.5b,there was no NO2à-N left in An1after last cy-cle?nished,and NO2à-N concentration kept below1mg Là1when An2and An3completed,which led to the denitritation rate(DNR) in An2and An3of38.8mgN hà1gVSSà1and23.7mgN hà1gVSSà1 (MLVSS=4.5g Là1),respectively.The reasons for excellent denitri-tation performance were concluded as follows:the in?uent COD and NH4+-N of PSBR was controlled under a proper COD/NH4+-N ratio(3–5),on the other hand,the raw leachate had relatively high B/C ratio.Consequently,the2nd and3rd in?uent?lling offered suf?cient degradable organics for dentritation in An2and An3, respectively.The dentritation rate could be surprisingly high via the carbon source in raw leachate.

In this research,at least2/3of the in?uent TN was denitri?ed under the condition with enough external carbon source,which not only improved dentritation rate but also took full advantage of organics in the raw leachate so that it reduced the negative ef-fects on the next nitritation.

Fig.5a shows that there was no apparent variation of COD in An4,which illustrated the lack of external carbon source for dent-ritation in An4,as a result,the removal of NO2à-N realized via endogenous denitritation.The DNR of An4lasting for462min was obviously divided into a fast one(DNR1:2.97mgN hà1gVSSà1) and a slow one(DNR2:1.22mgN hà1gVSSà1),while the change point happened when endogenous dentritation had run for 210min according to Fig.5b.This?nd is in correspondence with the work of Vocks et al.(2005),that found the endogenous DNR would change with the type alteration of internal carbon source used in denitri?cation(see Section3.6,below).

3.6.Evolution of the intracellular storage polymers over the typical PSBR cycle

Fig.5c shows the variation of PHB and glycogen concentration of microorganisms over the PSBR cycle.There was no NO xà-N at the beginning of An1,but the COD concentration kept decreasing

Table2

Comparison of COD and nitrogen removal from land?ll leachate treatment reported in the literatures using biological process.

Literature Process In?uent COD

(mg Là1)Ef?uent COD

(mg Là1)

In?uent

TN(mg Là1)

Ef?uent TN

(mg Là1)

Extra carbon source

addition

(mgCOD mgNO xà1)

Klimiuk and Kulikowaka(2004)Two-stage SBRs757a348a397a<20 3.6a

Tsilogeorgis et al.(2008)MSBR2456a1228a375a<50 5.2a

Spagni and Marsili-Libelli(2010)SBR2055a1335a1319a<70 2.2a

Peng et al.(2008)Two-stage UASB-A/O2800–6200800–14001450–2450170–250–

This study ASBR–PSBR7341–10,448<9001279–1578<40–

a Values are on average.

R.Zhu et al./Bioresource Technology128(2013)401–408405

during An1according to Fig.5a and b.This can be explained as fol-lows:assuming 2/3of the removed glycogen is transformed into PHB (Smolders et al.,1994a ),so the D (PHB-2/3glycogen)was 0.0414Cmol under the reduced glycogen and increased PHB of 0.0253Cmol and 0.0583Cmol,respectively,which showed the conversion from the organics into PHB.

TN loss derived from simultaneous nitritation and denitritation (SND)happened in oxic stages (Table 4),however,glycogen cannot be used as electron donor for SND because of the occurrence of synthesis in these stages.In contrast with glycogen,PHB was avail-able as internal carbon source in O2,O3and O1(famine)for SND,while organics in leachate can do the same way in O1(feast).According to Table 4,PHB oxidation rates of oxic stages where no degradable organics presented were at least 5.14times slower than the organics oxidation rate of O1(feast).This shows that PHB stor-age distinctly reduced the rate of carbon oxidation and prevented its rapid conversion into CO 2,which made PHB available as the electron donor for SND throughout all the oxic stages.As Fig.5c presents,PHB concentration decreased initially in An4,but the glycogen concentration stayed at a stable level until the PHB concentration stopped declining.We de?ned the points where the concentration of PHB stopped decreasing and glycogen started decreasing as the change points.These two change points matched very well with the change point in Fig.5b at which the endogenous DNR changed from fast to slow.This phenomenon indicated that the reason why endogenous DNR would vary was the type alteration of stored polymers used as the electron donor for denitritation.Furthermore,PHB was a preferable internal car-bon source for denitritation than glycogen.

Storage of PHB can occur in nitritation and denitritation system where microbial population subjecting dynamic conditions with respect to the availability of external substrate (Majone et al.,1998).Then in the anaerobic period,when there is no external sub-strate,they use it as internal energy and carbon source for growth and denitritation.The PHB loss in An4corresponded to 981.72mgCOD (determined that 1Cmol PHB re?ects 36gCOD),while the NO 2à-N loss was 465.85mgN before the change point.It is clear that the COD derived from PHB was not suf?cient for the denitritation during the DNR 1stage.More studies need to be done to identify whether other stored polymers exist for denitrita-tion during the DNR 1stage.The PHB concentration maintained constant after the change point,which was corresponding with the work of Lei et al.(2005)who found microorganisms would preferentially utilize PHB for cell functions rather than for denitri-tation when the PHB reservoir decreased to below a critical level,resulting in the little degradation of PHB.

When the PHB stopped degrading,glycogen concentration started to decline with respect to the change of DNR into DNR 2.Vocks et al.(2005)found glycogen can also be used as carbon source for denitritation.The anoxic substrate utilization rate r s is calculated by Eq.(6).DNR gly was equal to DNR 2.Y H was determined as 0.45mg biomass COD/mg carbohydrate COD for glycogen con-sumption (Goel et al.,1998).With these ?gures,a theoretical gly-cogen consumption of 3.79mgCOD h à1gVSS à1was calculated,this matched well with the measured 3.719mgCOD h à1gVSS à1(determined that 1Cmol glycogen re?ects 32gCOD).This phenom-enon indicated that the glycogen could also be used as a carbon source for endogenous denitritation.

PHB and glycogen can be applied as the internal carbon source in SND or endogenous denitritation when no external carbon source is available.More details need to be determined about the endogenous denitritation in the further studies.

3.7.Endogenous denitritation rate of An4in PSBR’s joint operation period

Fig.6indicates the variation of TDNR (Eq.(5))of An4in PSBR’s joint operation period,which maintained at a relatively steady level (Std.Dev:0.139)with the mean value (TDNR m )of 1.903mgN h à1gVSS à1.Although the in?uent NH 4+-N of PSBR in-creased from phase I to phase IV,the TDNR did not decrease,which might be due to the constant in?uent COD/NH 4+-N ratio of 3–5.This result is consistent with the study of Bernat et al.(2008),whose result indicated that the in?uent C/N ratio would in?uence the storage of intracellular polymers,subsequently,it could impact the endogenous DNR.Thus the constant in?uent COD/NH 4+-N ratio would lead to a little difference of TDNR in An4during the whole joint operation period.

The max endogenous DNR obtained from Henze (1991),Kujawa et al.(1997),and Kujawa and Klapwijk (1999)was 0.5mgN h à1gVSS à1,0.8mgN h à1gVSS à1and 0.6mgN h à1gVSS à1,respectively,all of them were below 1mgN h à1gVSS à1.That might because these values are normally found in the low-loaded activated sludge systems compared with this study,which indicated that

the

Fig.4.COD and nitrogen variations of ASBR and PSBR during the experimental period:(a)COD and NH 4+-N variations in ASBR;(b)COD and TN variations in PSBR.

Table 3

Parameters of PSBR controlled in the different phases.Item Time (days)In?uent NH 4+-N (mg L à1)Phase I 20–57200–400Phase II 58–97400–600Phase III 98–137600–800Phase

IV

138–177

800–1000

128(2013)401–408

organic loading rate might in?uence the gross of intracellular poly-mers.On the other hand,the value got from Third et al.(2003)was found in a system which had a considerable amount of PHB stor-age,and 2.21mgN h à1gVSS à1was close to 2.97mgN h à1gVSS à1found in this study,that explained the operation mode of PSBR treating immature land?ll leachate has an advantage in transform-ing organics in raw leachate into stored polymers,which provided the possibility of thorough nitrogen removal without extra carbon source addition.

3.8.Microorganisms participating in the endogenous denitritation This work for the ?rst time shows the phenomenon that PHB and glycogen can be used as electron donors orderly for

Fig.5.Variations of COD,nitrogen,phosphorus and intracellular storage polymers over the typical PSBR cycle in phase IV:(a)COD,NH 4+-N and phosphorus variations;(b)NO 3à-N and NO 2à-N variations;(c)PHB and glycogen variations.

Fig.6.Variations of TDNR of An4in PSBR’s joint operation period.

ability of denitrifying within GAOs,denitrifying glycogen-accumulating organisms(DNGAOs),rather than DNPAOs,were responsible for the denitri?cation/denitritation activity.Although no evidence was presented in this work,the existence of DNGAOs was postulated,which provided DNGAOs’possibility of playing an important role in endogenous denitritation.

4.Conclusions

ASBR–PSBR system was suitable for biological treatment of immature land?ll leachate,which realized COD and TN removal rate of above89%and97%,respectively,with respect to the ef?u-ent TN of less than40mg Là1.Two reasons contributed to out-standing system performance.Firstly,due to the three equal feeds mode applied in the PSBR,most of the organics in the raw leachate was used as the carbon source for denitritation.Secondly, PHB and glycogen acted as internal carbon source orderly for endogenous denitritation,which con?rmed the possibility of ad-vanced nitrogen removal without extra carbon source addition. Acknowledgements

The authors gratefully acknowledge the?nancial support from National Natural Science Foundation of China(51178007).This work was also supported by the Project of Scienti?c Research Base and Scienti?c Innovation Platform of Beijing Municipal Education Commission.

References

Ahn,J.,Daidou,T.,Tsuneda,S.,Hirata,A.,2001.Metabolic behavior of denitrifying phosphate-accumulating organisms under nitrate and nitrite electron acceptor conditions.Journal of Bioscience and Bioengineering92,442–446.

APHA,1995.Standard Methods for the Examination of Water and Wastewater.

American Public Health Association,New York.

Bernat,K.,Wojnowska-Baryla,L.,Dobrzynska, A.,2008.Denitri?cation with endogenous carbon source at low C/N and its effect on P(3HB)accumulation.

Bioresource Technology99,2410–2418.

Beun,J.J.,Verhoef,E.V.,Van Loosdrecht,M.C.M.,Heijnen,J.J.,2000.Stoichiometry and kinetics of poly b-hydroxybutyrate metabolism under denitrifying conditions in activated sludge cultures.Biotechnology and Bioengineering68, 496–507.

Castrillon,L.,Fernandez-Nava,Y.,Ulmanu,M.,Anger,I.,Maranon,E.,2010.Physico-chemical and biological treatment of MSW land?ll leachate.Waste Management30,228–235.

Chiang,L.,Chang,J.,Chung,C.,2001.Electrochemical oxidation combined with physical–chemical pretreatment processes for the treatment of refractory land?ll leachate.Environmental Engineering Science18,369–378.

Goel,R.,Mino,T.,Satoh,H.,Matsuo,T.,1998.Intracellular storage compounds, oxygen uptake rates and biomass yield with readily and slowly degradable substrates.Water Science and Technology38,85–903.

Hasar,H.,Unsal,S.A.,Ipek,U.,Karatas,S.,Cinar,O.,Yaman,C.,Kinaci,C.,2009.

Stripping/?occulation/membrane bioreactor/reverse osmosis treatment of municipal land?ll leachate.Journal of Hazardous Materials171,309–317. Henze,M.,1991.Capabilities of biological nitrogen removal processes from wastewater.Water Science and Technology23,669–679.

Hoilijoki,T.H.,Kettunen,R.H.,Rintala,J.A.,2000.Nitri?cation of anaerobically pretreated municipal land?ll leachate at low temperature.Water Research34, 1435–1446.Huo,S.L.,Xi,B.D.,Yu,H.C.,Fan,S.L.,Su,J.,Liu,H.L.,2008.In situ simultaneous organics and nitrogen removal from recycled land?ll leachate using an anaerobic–aerobic process.Bioresource Technology99,6456–6463.

Klimiuk,E.,Kulikowaka,D.,2004.Effective of organics and nitrogen removal from municipal land?ll leachate in single and two-stage SBR systems.Polish Journal of Environmental Studies13,525–532.

Kuba,T.,Van loosdrecht,M.C.M.,Heijnen,J.J.,1997.An integrated metabolic model for the aerobic and denitrifying biological phosphorus removal.Biotechnology and Bioengineering54,434–450.

Kujawa,K.,Klapwijk,B.,Maas,P.,Majoor,E.,1997.Estimation of nitrogen removal potential on WWTP Assen.Report of the Agricultural University Wageningen. Kujawa,K.,Klapwijk,B.,1999.A method to estimate denitri?cation potential for predenitri?cation systems using NUR batch test.Water Research33,2291–2300.

Lei,Q.,Yu,L.,Joo-Hwa,T.,2005.Denitri?cation on poly-b-hydroxybutyrate in microbial granular sludge sequencing batch reactor.Water Research39,1503–1510.

Majone,M.,Massanisso,P.,Ramadori,R.,https://www.sodocs.net/doc/2e1455230.html,parison of carbon storage under aerobic and anoxic conditions.Water Science and Technology38,77–84. Nehrenheim,E.,Waara,S.,Westholm,J.L.,2008.Metal retention on pine bark and blast furnace slag–on-site experiment for treatment of low strength land?ll leachate.Bioresource Technology99,998–1005.

Ozturk,I.,Altinbas,M.,Koyuncu,I.,Arikan,O.,Gomec-Yangin,C.,2003.Advanced physico-chemical treatment experiences on young municipal land?ll leachates.

Waste Management23,441–446.

Park,S.,Choi,K.S.,Joe,K.S.,Kim,W.H.,Kim,H.S.,2001.Variations of land?ll leachate properties in conjunction with the treatment process.Environmental Technology22,639–645.

Peng,Y.Z.,Zhang,S.J.,Zeng,W.,Zheng,S.W.,Mino,T.,Satoh,H.,https://www.sodocs.net/doc/2e1455230.html,anic removal by denitritation and methanogenesis and nitrogen removal by nitritation from land?ll leachate.Water Research42,883–892.

Pijuan,M.,Saunders,A.M.,Guisasola,A.,Baeza,J.A.,Casas,C.,Blackall,L.L.,2004.

Enhanced biological phosphorus removal in a sequencing batch reactor using propionate as the sole carbon source.Biotechnology and Bioengineering85,56–

67.

Smolders,G.J.F.,Vandermeij,J.,Van loosdrecht,M.C.M.,Heijnen,J.J.,1994a.Model of the anaerobic metabolism of the biological phosphorus removal process: stoichiometry and pH in?uence.Biotechnology and Bioengineering43,461–470.

Smolders,G.J.F.,Vandermeij,J.,Van loosdrecht,M.C.M.,Heijnen,J.J.,1994b.

Stoichiometric model of the aerobic metabolism of the biological phosphorus removal process.Biotechnology and Bioengineering44,837–848.

Spagni,A.,Marsili-Libelli,S.,2010.Nitrogen removal via nitrite in a sequencing batch reactor treating sanitary land?ll leachate.Bioresource Technology100, 609–614.

Tao,Y.,Gao,D.W.,Fu,Y.,Wu,W.M.,Ren,N.Q.,2012.Impact of reactor con?guration on anammox process start-up:MBR versus SBR.Bioresource Technology104, 73–80.

Third,K.A.,Burnett,N.,Cord-Ruwisch,R.,2003.Simultaneous nitri?cation and denitri?cation using stored substrate(PHB)as the electron donor in an SBR.

Biotechnology and Bioengineering83,706–720.

Tsilogeorgis,J.,Zouboulis,A.,Samaras,P.,Zamboulis,D.,2008.Application of a membrane sequencing batch reactor for land?ll leachate treatment.

Desalination221,483–493.

Uygur, A.,Karg?, F.,2004.Biological nutrient removal from pretreated land?ll leachate in a sequencing batch reactor.Journal of Environmental Management 71,9–14.

Vocks,M.,Adam, C.,Lesjean, B.,Gnirss,R.,Kraume,M.,2005.Enhanced post-denitri?cation without addition of an external carbon source in membrane bioreactors.Water Research39,3360–3368.

Xie,B.,Xiong,S.Z.,Liang,S.B.,Hu,C.,Zhang,X.J.,Lu,J.,2012.Performance and bacterial compositions of aged refuse reactors treating mature land?ll leachate.

Bioresource Technology103,71–77.

Zeng,R.J.,Lemaire,R.,Yuan,Z.G.,Keller,J.,2003.Simultaneous nitri?cation, denitri?cation,and phosphorus removal in a lab-scale sequencing batch reactor.Biotechnology and Bioengineering84,170–178.

Zheng,D.,Deng,L.W.,Fan,Z.H.,Liu,G.J.,Chen,C.,Yang,H.,2012.In?uence of sand layer depth on partial nitritation as pretreatment of anaerobically digested swine wastewater prior to anammox.Biotechnology and Bioengineering104, 274–279.

408R.Zhu et al./Bioresource Technology128(2013)401–408