Factors affecting biological denitrifying ...

J ou rna l of Ha rbin Institu te of Technology ( N ew S eries) , V ol. 14, N o. 4, 2007

Factors affecting biological denitrifying dephosphatation in anaerob ic /anox ic /aerob ic sequenc in g ba tch reactor L IN Yan , WANG Xin 2ze , YUAN L in 2jiang , WANG Zhi2ying , KONG Hai2nan 1

1

2

2

1

林　燕 , 　　王欣泽 , 　　　袁林江 , 　　　王志盈 , 　　　　孔海南　　 ( 1. School of Environmental Science and Engineering, Shanghai J iaotong University, Shanghai 200240, China, E 2mail: hainankong@126. com; 2. School of Environmental and Municipal Engineering, Xi’an University of A rchitecture & Technology, Xi’an 710055, China)

Abstract: This study was conducted to verify and discuss the denitrifying dephosphatation under different levels of nitrate concentration and retention tim e of anoxic / aerobic p rocess in a Sequencing Batch Reactor ( SBR ) . The results of tests demonstrated that there were two kinds of phosphorus2accumulating organism s ( PAO s) in the biological excess phosphorus removal (BEPR ) system. One was non 2DNPAO s that could only use oxygen as term inal electron accep tors, the other was denitrifying PAO s (DNPAO s) that could use both nitrate and oxygen as term inal electron accep tors. Phosphorus up take efficiency could be attained under anoxic period ranging from 2817% - 9617% in an anaerobic / anoxic / aerobic system. Experim ental results showed that nitrate concentra2 tion and retention tim e of anoxic / aerobic p rocess were the key factors affecting the course of denitrifying dephos2 phatation. Key words: biological wastewater treatm ent; denitrifying; dephosphatation; biological phosphorus removal; se2 quencing batch reactor CLC num ber: X70311 　　　D ocum en t code: A 　　　　Article ID : 1005 29113 ( 2007 ) 04 20465 205

　　O rganic substrate is separately necessary for phos2 phorus and nitrogen removal in conventional system s. This causes the competition for COD betw een phosphor2 us2accumulating organism s and denitrifers, since organ2 ic substrates in municipal wastewater are often lim ited [1 - 3] for extensive phosphorus and nitrogen removal . However, if PAO s ( or part of them ) are able to accu2 mulate high amounts of polyphosphate by using nitrate as electron accep tor instead of oxygen, there would be less competition for organic substrate. The app lication of denitrifying phosphorus accumulating organism s (DNPAO s) in biological phosphorus removal ( BPR ) system would be advantageous for the saving of organic substrates and energy ( aeration ) , in addition, the an2 oxic phosphate removal occurs sim ultaneously w ith [4] denitrification in the same reactor . In a wastewater treatm ent system , the study of m i2 croorganism ecological system of activated sludge has demonstrated that changes of the environm ental condi2 tions, even slight changes, could lead to changes in [ 5, 6 ] m icrobial population in the system . It is commonly thought that the activity of excess up take of phosphorus by PAO s is the result of continuously recirculating them [7 - 9] betw een anaerobic and aerobic environments . Mo2 reover, most of aerobic heterotrophs can use nitrate as term inal electron accep tor instead of oxygen to generate energy. Therefore, the metabolism method of denitrif2

ying dephosphatationn could be induced by the stim ula2 tion of environment. The objectives of this study were to p rovide an ad2 ditional evidence of denitrifying dephosphatation in one system and to investigate the effects of nitrate concen2 tration and retention tim e of an anaerobic / anoxic / aero2 bic p rocess on denitrification and phosphorus up take. 1 　M a ter ia ls and M ethods 1. 1 　Ba tch Tests The experim ental set2up is illustrated in Fig. 1. It consisted of one 5 2liter Plexiglas cylindrical batch reac2 tor. The reactor was sealed to m aintain anaerobic and anoxic conditions. Aerobic periods were initiated by sparging comp ressed air through a diffuser at the bottom of the reactor. The pH was manually controlled at 712 - 810 via addition of 1 mol/L HC l or NaOH. The tem 2 perature of the m ixed liquid remained at 25 ±1 ℃. The sludge concentration was diluted app roxim ately 215 g (MLSS) /L. Chem ical addition was performed w ith a calibrated peristaltic pump. 112 　Syn thetic Sewage and Seed Sludge The composition of the synthetic sewage was as follow s: NaAC ・ 3H2 O ( Xi ’an Chem ical Reagent Co. , L td. Xi ’an , China , sam e as the follow s excep t

Received 2004 - 08 - 30. Sponsored by the National Natural Science Foundation of China ( Grant No. 50008014) .

・465・


otherwise described ) 0152 g/L , KH2 PO4 01045 g/L , K2 SO4 01002 g/L , MgCl2 ・6H2 O 01008 g/L , MnSO4 ・ 7H2 O 01003 g/L , FeSO4 ・7H2 O 010002 g/L , CaCl2 01003 g /L , NH4 Cl 01115 g/L , NaHCO3 01420 g/L. The influent characteristics are described in Tab. 1.

-

W hen 5 mg /L NO3 - N was added, the m axim um phosphorus up take rate was 8153 m gP / ( gSS ・ h ) ; while the NO3 - N concentration was 10 m g /L , the maxim um phosphorus up take rate was 9101 m gP / ( gSS・h ) corresponding to a higher anoxic phosphorus removal efficiency ( 4210 % ) . And almost all nitrite and nitrate was consumed during this period. W hen the nitrate concentration increased to 20 mg /L , the m axi2 mum phosphorus up take rate was 9168 mgP / ( gSS ・ h ) . This means that it is necessary to supp ly sufficient nitrate as the electron accep tor to the system , and the denitrification and dephosphatation could be thereby imp roved.

F ig. 1 　Sketch of the exper im en ta l set2up Tab. 1 　The average syn thetic sewage Component

Concentration

ρ ( CODC r ) / (mg・L ρ (NH4

-1

350 ±10

)

- N ) / (mg・L

-1

)

30 ±3

ρ( NO x - N ) / (mg・L

-1

)

0 ±015

+ -

ρ ( TN ) / (mg・L - 1 ) ρ ( PO 4

3-

- P) / (mg・L

30 ±3 -1

10 ±1

)

ρ ( TP) / (mg・L - 1 ) ρ ( SS) / (mg・L

-1

10 ±1 10 ±015

)

A lkalinity ( CaCO 3 ) / (mg・L

-1

350 ±10

)

715 ±015

pH

　　3 NO x - N : NO 3 -

-

- N +NO 2

-

-N

　　The seed activated sludge was taken from oxidation ditch of Bei Shi Q iao W astewater Treatm ent Plant loca2 ted in Xi ’an City, Shaanxi Province, China. It was cultivated for two weeks in aerobic and phosphorus con2 taining medium before feeding of the synthetic sewage. 113 　Ana lytica l M ethods Chem ical oxygen demand ( COD ) , total phosphor2 3+ ( us TP) , total nitrogen ( TN ) , PO4 - P, NH4 - N, NO2 - N , NO3 - N and pH, were measured accord2 [ 10 ] ing to the standard methods as described in APHA . 2 　Results and D iscussion 211 　 Effect of N itra te Concen tra tion on D en itr if2 y in g D ephospha ta tion The experim ental data shown in Fig. 2 illustrate the typ ical behavior of the changes in phosphorus up 2 take rates. The maxim um phosphorus up take rate in2 creased w ith the increasing of the nitrate concentration.

・466・

F ig. 2 　Change of phosphorus uptake ra te in one c ircle un 2 der d ifferen t n itra te concen tra tion ( anaerob ic( 2 h) - anox ic( 1 h) - aerob ic( 2 h) ) -

　　Moreover, when the NO3 - N concentrations were 5 m g /L and 10 m g /L , the anoxic phosphorus up 2 take rates slowed down over tim e. A t the NO3 - N concentrations of 5 mg /L and 10 mg /L , the phosphor2 us up take rates at the end of the anoxic periods had de2 creased to nearly 0 m gP / ( gSS ・ h ) , and meanwhile the NO3 - N concentration fell to 0 m g /L ( Fig. 3 ) . The most likely exp lanation for this phenomenon m ight be that the lack of nitrate blocked the oxidization of PHB in the cell and thus p revented the bacteria from obtaining sufficient energy to up take large quantities of phosphorus into the cell. Additionally, the p roportion of the DNPAO s among the PAO s and the quantity of PHB stored in the DNPAO s were also the key factors. After aeration was started in reactor, phosphorus up 2 take rate sharp ly increased, since the rem nant non 2 DNPAO s could utilize PHB stored in cells. Of course, the DNPAO s m ight also contribute to the phosphorus up take under aerobic condition if there were still some PHB in cells. A ll the experim ental data indicate that there exis2 ted two populations of PAO s: DNPAO s which could use either oxygen or nitrate as the electron accep tor and non 2DNPAO which could utilize only oxygen. DNPAO s p layed an im portant role in phosphorus removal during anoxic phosphorus up take. Under the ideal nutrient en2 vironment the capability of denitrifying dephosphatation


would be enhanced significantly w ith the increasing of the quantity and activity of the DNPAO s. Figs. 3 and 4 show the changes of NO3 - N and phosphorus concentrations on the course of denitrifying dephosphatation under different nitrate concentrations. During this course, about half of the phosphorus was accumulated by the DNPAO s under anoxic condition, and the anoxic phosphorus removal efficiency increased as the nitrate concentration increased. W hen 5 m g /L NO3 - N was added, the anoxic phosphorus removal efficiency was 2817 % ; while the NO3 - N concen2 tration was 10 mg /L , a higher anoxic phosphorus re2 moval efficiency of 4210 % appeared. W hen the NO3 - N concentration was increased to 20 m g /L , the an2 oxic phosphorus removal efficiency was 4817 %. Mo2 reover, when 20 m g /L NO3 - N was added, there was still some nitrate left at the end of the anoxic peri2 od because of the shorter retention tim e of anoxic peri2 od, which indicated that the aerobic 2anoxic retention tim e also had effect on the course of the denitrifying dephosphatation.

F ig. 3 　Change of NO x - N in one c ircle under d ifferen t n i2 tra te concen tra tion ( anaerob ic ( 2 h) - anox ic( 1 h) - aerob ic( 2 h) ) -

　　 The batch test was performed in the anaerobic ( 2 h / circle ) 2anoxic ( 215 h / circle ) 2aerobic ( 015 h / circle ) system , and three concentrations of the nitrate were app lied. A s indicated in Figs. 5 and 6, the sys2 tem got the best performance of denitrifying dephos2 phatation when the NO3 - N concentration was 20 mg /L. In other conditions, nevertheless, the PHB in the DNPAO s was run out by a large amount of the e2 lectron accep tors. Once the nitrate retained in the sys2 tem , it would consume part of the organic substrate in the influent. W hen the NO3 - N concentration was 20 mg /L , 30 mg /L and 40 mg /L , the residual nitrite p lus nitrate nitrogen at the end of the anoxic period was 0110 m g /L , 8100 mg /L and 23164 m g /L respective2 ly, as shown in Fig. 6. This is disadvantageous to the PAO s, since there w ill not be sufficient organic sub2 strate for them to utilize in the aerobic or anoxic peri2 od.

F ig. 5 　Change of phosphorus in one c ircle under d ifferen t n itra te concen tra tion ( anaerob ic ( 2 h ) - anox ic ( 2. 5 h) - aerob ic( 0. 5 h) )

　　 W hen the nitrate concentration increased to 30 m g /L and 40 mg /L , the p rofiles changed as illus2 trated in Figs. 5 and 6.

F ig. 6 　Change of NO x - N in one c ircle under d ifferen t n i2 tra te concen tra tion ( anaerob ic ( 2 h ) - anox ic ( 2. 5 h) - aerob ic( 0. 5 h) ) -

F ig. 4 　Change of phosphorus in one c ircle under d ifferen t n itra te concen tra tion ( anaerob ic ( 2 h ) - anox ic ( 1 h) - aerob ic( 2 h) )

　　The change of phosphorus up take rate in one cir2 cle was shown in Fig. 7. W hen the aeration started, the phosphorus up take rate did not enhance, but kep t decline. This indicates that nearly all the PHB in the DNPAO s was consumed. Even the bacteria were in the aerobic condition, they would not be able to accum u2 late phosphorus for lack of the PHB. This also show s ・467・


that the DNPAO s in the system p roliferated through in2 ducement.

from Figs. 10 and 11 that when the duration of aeration was shortened to 10 m inutes, denitrifying dephosphata2 tion was enhanced and the phosphorus up take rate in2 creased slow ly. This also indicates that reduction of the aeration tim e could p romote denitrifying dephosphata2 tion.

F ig. 7 　Change of phosphorus uptake ra te in one c ircle un 2 der d ifferen t n itra te concen tra tion ( anaerob ic( 2 h) - anox ic( 2. 5 h) - aerob ic( 0. 5 h) )

212 　 Effect of the Reten tion T im e of an Anox ic 2 aerob ic Process on D en itr ify in g D ephos2 pha ta tion A s shown in Fig. 3 the retention tim e of an anox2 ic 2aerobic p rocess also had an effect on the course of denitrifying dephosphatation. A longer aerobic period could lower the capacity of denitrifying dephosphatation since phosphorus up take mainly happened in aerobic period. Moreover, it is important for cost saving to shorten aerobic period. W hen the nitrate nitrogen concentration was 20 m g /L , it could imp rove the denitrifying dephos2 phatation ( Fig. 3 ) . The anoxic phosphorus up take rate was high, but there was still som e nitrite and ni2 trate left at the end of the circle. This would bring the disadvantage to the follow ing circle. In order to con2 sume all added nitrate, the batch test was adjusted to anaerobic ( 2 h / circle ) 2anoxic ( 215 h / circle ) 2aerobic ( 015 h / circle ) . Fig. 8 disp lays the characteristics of the changes of the durations of the anoxic and aerobic periods. The efficiency of denitrifying dephosphatation was enhanced when the durations of the aerobic periods reduced. The anoxic phosphorus removal efficiency increased from 4817% ( 2 hours’aeration ) to 9617% ( 015 hour’s aeration ) . Fig. 9 show s the changes in the concentra2 tion of nitrogen in the system. A longer aeration ( 2 h ) would lead to a higher concentration of nitrite and ni2 trate at the end of the circle due to nitrification. A t the same tim e, the nitrite and nitrate still remained at the end of the anoxic period. On the contrast, a shorter aeration period ( 015 h ) would not result in much ni2 trate to the next circle. Therefore denitrifying dephos2 phatation would be little affected by nitrate. 　　 W hen the added nitrate concentration was in2 creased to 40 mg /L , the changes of the duration of the aeration also affected the whole running. It can be seen

・468・

F ig. 8 　Change of phosphorus in one c ircle under d ifferen t dura tion of aera tion ( [ NO 3 - N] = 20 m g /L )

-

F ig. 9 　 Change of NO x - N in one c ircle under d ifferen t dura tion of aera tion

F ig. 10 　Change of phosphorus in one c ircle under d ifferen t dura tion of aera tion( [ NO 3 - N] = 40 m g /L )

　　 It should be noted that Fig. 11 disp lays the per2 form ance on the third day. In the follow ing days the


system did not reach to a stable situation, as shown in Fig. 12.

-

F ig. 11 　Change of NO x - N in one c ircle under d ifferen t dura tion of aera tion( NO 3 - N] = 40 m g /L )

trifying dephosphatation couldn ’t be im p roved any2 more. The op tim al result could be obtained when influ2 ent COD concentration was 350 mg /L , SRT was 15 days, the durations for anaerobic, anoxic and aerobic were 2 h, 215 h and 015 h, respectively, and the NO3 - N concentration in the anoxic zone was 20 mg /L. 3 ) Shorter aeration tim e could be helpful for accu2 mulation of DNPAO s. However, an aerobic phase was necessary after an anoxic phase. W hen influent COD concentration was 350 mg /L , SRT was 15 days, and the NO3 - N concentration in the anoxic zone was 20 mg /L , the op tim al ratio of durations for anaerobic, anoxic and aerobic was 2 ∶215 ∶015. 4 ) There were two populations of PAO s, namely DNPAO s and non - DNPAO s. DNPAO s could use either oxygen or nitrate as electron accep tor. Non 2DNPAO s could use only oxygen. DNPAO s p layed an important role in phosphorus removal under anoxic condition. References:

F ig. 12 　Change in the concen tra tion of phosphorus in the long 2term opera tion

　　 It illustrates that shorter aeration could save ener2 gy and intensify the function of denitrifying dephos2 phatation, but there should be a certain course of aera2 tion as a supp lement to recover metabolic activity of bacteria and get better settling ability so as to im p rove the removal effect of the system. The duration of aera2 tion should be longer than 10 m inutes; otherw ise the effect of denitrifying dephosphatation would decrease significantly. 3 　Conclusion s 1 ) In anaerobic 2anoxic 2aerobic tests, the phe2 nomenon of denitrifying dephosphatation was observed. The phosphorus up take efficiency attained under anoxic period was ranged from 2817% - 9617%. 2 ) NO3 - N concentration was one of the key factors affecting the denitrifying dephosphatation. Gen2 erally, a higher NO3 - N concentration is p rofitable. However, when the NO3 - N concentration was in2 creased to higher than 20 mg /L , the efficiency of deni2

[ 1 ] Feng S H. Investigation of biological nitrogen and phos2 phorus removal. W ater and W astewater, 1994, 20 ( 2 ) : 18 - 21. [ 2 ] Carucci A , Kuhni M , B run R. M icrobial competition for the organic substrates and its impact on Ebp r system under conditions of changing carbon feed. W ater Science & Tech2 nology, 1999, 39 ( 1 ) : 75 - 85. [ 3 ] L iu J X, VAN Groenestijin J W. Op tim um operation condi2 tions of nitrogen and phosphorus removal by a biofilm 2acti2 vated 2sludge system. Journal of Environmental Sciences, 2003, 15 (1) : 25 - 30. [ 4 ] M einhold J, A rnold E, Isaacs S. Effect of nitrite on anoxic phosphate up take in biological phosphorus removal activa2 ted sludge. W ater Research, 1999, 33 ( 8 ) : 1871 1883. [ 5 ] W ang J C, Park J K. Effect of wastewater composition on m icrobial populations in biological phosphorus removal p rocesses. W ater Science & Technology, 1998, 38 ( 1 ) : 159 - 166. [ 6 ] W itten G Q , R ichardson F D. Competition of three aggre2 gated m icrobial species for four substrates in the rumen. Ecological Modelling, 2003, 164: 121 - 135. [ 7 ] Ubukata Y, Takii S. Induction method of excess phosphate accumulation for phosphate removing bacteria isolated from anaerobic / aerobic activated sludge. W ater Science & Technology, 1994, 30 ( 6 ) : 221 - 227. [ 8 ] Shahnaz D , O leszkiew icz J A. U se of a new anaerobic 2aer2 obic sequencing batch reactor system to enhance biological phosphorus removal. W ater Science & Technology, 1997, 35 ( 1 ) : 137 - 144. [ 9 ] Sang Y L , Choi J. Production and degradation of poly2 hydroxyalkanoates in waste environment. W aste M anage2 ment, 1999, 19: 133 - 139. [ 10 ]APHA 2AWWA 2W EF. Standard M ethods for the Exam ina2 th tion of W ater and W astewater. 19 ed. W ashington DC: American Public Health A ssociation, 1995.

・469・