Author's personal copy Advancing post-anoxic denitrification for biological nutrient removal Matt Winkler, Erik R. Coats*, Cynthia K. Brinkman Department of Civil Engineering, University of Idaho, PO Box 441022, Moscow, ID 83844-1022, USA article info Article history: Received 21 April 2011 Received in revised form 2 August 2011 Accepted 3 September 2011 Available online 14 September 2011 Keywords: Post-anoxic denitrification Biological nutrient removal (BNR) Enhanced biological phosphorus removal (EBPR) Surface oxygen transfer qPCR Polyphosphate accumulating organisms (PAOs) Glycogen accumulating organisms (GAOs) Secondary phosphorus release abstract The objective of this research was to advance a fundamental understanding of a unique post-anoxic denitrification process for achieving biological nutrient removal (BNR), with an emphasis on elucidating the impacts of surface oxygen transfer (SOT), variable process loadings, and bioreactor operational conditions on nitrogen and phosphorus removal. Two sequencing batch reactors (SBRs) were operated in an anaerobic/aerobic/anoxic mode for over 250 days and fed real municipal wastewater. One SBR was operated with a headspace open to the atmosphere, while the other had a covered liquid surface to prevent surface oxygen transfer. Process performance was assessed for mixed volatile fatty acid (VFA) and acetate-dominated substrate, as well as daily/seasonal variance in influent phosphorus and ammonia loadings. Results demonstrated that post-anoxic BNR can achieve near- complete (>99%) inorganic nitrogen and phosphorus removal, with soluble effluent concentrations less than 1.0 mgN L 1 and 0.14 mgP L 1 . Observed specific denitrification rates were in excess of typical endogenous values and exhibited a linear dependence on the glycogen concentration in the biomass. Preventing SOT improved nitrogen removal but had little impact on phosphorus removal under normal loading conditions. However, during periods of low influent ammonia, the covered reactor maintained phosphorus removal performance and showed a greater relative abundance of polyphosphate accu- mulating organisms (PAOs) as evidenced by quantitative real-time PCR (qPCR). While GAOs were detected in both reactors under all operational conditions, BNR performance was not adversely impacted. Finally, secondary phosphorus release during the post-anoxic period was minimal and only occurred if nitrate/nitrite were depleted post-anoxically. ª 2011 Elsevier Ltd. All rights reserved. 1. Introduction Phosphorus (P) and nitrogen (N) are nutrients of primary concern in regard to accelerated surface water eutrophication, and many wastewater treatment plants (WWTPs) are facing increasingly stringent effluent limitations for both nutrients. P and N can be readily removed biologically, with P removal achieved using an engineered process known as enhanced biological P removal (EBPR). EBPR is accomplished by exposing microbes to cyclical anaerobic/aerobic and/or anoxic condi- tions, with influent wastewater first directed to the anaerobic zone. The prescriptive EBPR configuration provides a selective advantage to organisms capable of storing volatile fatty acids (VFAs) anaerobically as polyhydroxyalkanoates (PHAs), such as polyphosphate accumulating organisms (PAOs), which remove and store excess P as intracellular polyphosphate (poly-P) and are the putative organisms responsible for EBPR. EBPR can also enrich for glycogen accumulating organisms * Corresponding author. Tel.: þ1 208 885 7559; fax: þ1 208 885 6608. E-mail address: [email protected](E.R. Coats). Available online at www.sciencedirect.com journal homepage: www.elsevier.com/locate/watres water research 45 (2011) 6119 e6130 0043-1354/$ e see front matter ª 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.watres.2011.09.006
12
Embed
Advancing post-anoxic denitriÞcation for biological ... · post-anoxic denitriÞcation process for achieving biological nutrient removal (BNR), with an emphasis on elucidating the
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Author's personal copy
Advancing post-anoxic denitrification for biological nutrientremoval
Matt Winkler, Erik R. Coats*, Cynthia K. Brinkman
Department of Civil Engineering, University of Idaho, PO Box 441022, Moscow, ID 83844-1022, USA
a r t i c l e i n f o
Article history:
Received 21 April 2011
Received in revised form
2 August 2011
Accepted 3 September 2011
Available online 14 September 2011
Keywords:
Post-anoxic denitrification
Biological nutrient removal (BNR)
Enhanced biological phosphorus
removal (EBPR)
Surface oxygen transfer
qPCR
Polyphosphate accumulating
organisms (PAOs)
Glycogen accumulating organisms
(GAOs)
Secondary phosphorus release
a b s t r a c t
The objective of this research was to advance a fundamental understanding of a unique
post-anoxic denitrification process for achieving biological nutrient removal (BNR), with an
emphasis on elucidating the impacts of surface oxygen transfer (SOT), variable process
loadings, and bioreactor operational conditions on nitrogen and phosphorus removal. Two
sequencing batch reactors (SBRs) were operated in an anaerobic/aerobic/anoxic mode for
over 250 days and fed real municipal wastewater. One SBR was operated with a headspace
open to the atmosphere, while the other had a covered liquid surface to prevent surface
oxygen transfer. Process performance was assessed for mixed volatile fatty acid (VFA) and
acetate-dominated substrate, as well as daily/seasonal variance in influent phosphorus
and ammonia loadings. Results demonstrated that post-anoxic BNR can achieve near-
complete (>99%) inorganic nitrogen and phosphorus removal, with soluble effluent
concentrations less than 1.0 mgN L�1 and 0.14 mgP L�1. Observed specific denitrification
rates were in excess of typical endogenous values and exhibited a linear dependence on
the glycogen concentration in the biomass. Preventing SOT improved nitrogen removal but
had little impact on phosphorus removal under normal loading conditions. However,
during periods of low influent ammonia, the covered reactor maintained phosphorus
removal performance and showed a greater relative abundance of polyphosphate accu-
mulating organisms (PAOs) as evidenced by quantitative real-time PCR (qPCR). While GAOs
were detected in both reactors under all operational conditions, BNR performance was not
adversely impacted. Finally, secondary phosphorus release during the post-anoxic period
was minimal and only occurred if nitrate/nitrite were depleted post-anoxically.
ª 2011 Elsevier Ltd. All rights reserved.
1. Introduction
Phosphorus (P) and nitrogen (N) are nutrients of primary
concern in regard to accelerated surfacewater eutrophication,
and many wastewater treatment plants (WWTPs) are facing
increasingly stringent effluent limitations for both nutrients. P
and N can be readily removed biologically, with P removal
achieved using an engineered process known as enhanced
biological P removal (EBPR). EBPR is accomplished by exposing
microbes to cyclical anaerobic/aerobic and/or anoxic condi-
tions, with influent wastewater first directed to the anaerobic
zone. The prescriptive EBPR configuration provides a selective
advantage to organisms capable of storing volatile fatty acids
(VFAs) anaerobically as polyhydroxyalkanoates (PHAs), such
as polyphosphate accumulating organisms (PAOs), which
remove and store excess P as intracellular polyphosphate
(poly-P) and are the putative organisms responsible for EBPR.
EBPR can also enrich for glycogen accumulating organisms
butyrate (HBu), and 5 � 1% valerate (HVa) on a Cmol basis;
nominal amounts of iso-HBu and iso-HVa were observed as
well. Influent P varied from 3.5 to 6.0 mgP L�1, and influent
ammonia concentrations ranged from 14 to 50 mgN L�1. The
reactors received the 90:10 substrate for a majority of the
study (Fig. 1), and an example of the daily variations in the
90:10 characteristics are shown in Fig. 2 for about 30 days prior
to the first sampling run. During this time period, total VFAs
ranged from 200 to 275 mgCOD L�1 which yielded an influent
VFA:P ratio of 35e50 mgCOD mgP�1, which was theoretically
sufficient to be favorable for PAOs (Oehmen et al., 2007).
Reactors were allowed to stabilize for about 3 SRTs between
operational changes before performance was assessed.
Table 1 provides additional information about the influent
wastewater characteristics for each sampling event.
2.3. Stoichiometric calculations
Due to the SBR configuration and post-anoxic operational
mode, residual NOx carryover from the anoxic period to the
subsequent anaerobic cycle occurred. While the amount of
carryover was insufficient to upset EBPR performance, some
of the influent VFAs would have been consumed to reduce the
residual NOx (mainly nitrate). Much debate exists regarding
how much influent COD would be consumed for nitrate
reduction, and others have commonly assumed a ratio of
8.6 mgCOD mgNO3eN�1 (Henze et al., 2008). However, this
ratio likely overestimates COD utilization because it assumes
an anoxic yield equal to the typical aerobic heterotrophic yield
value of 0.666 mgBiomassCOD (mgCODutilized)�1. Others
have suggested that the anoxic yield is 60e70% of the aerobic
yield and also dependent on the substrate provided (Copp and
Dold, 1998). In this study, residual nitrate at the beginning of
the anaerobic period was low and available for less than
5 min, which would also limit the anoxic yield. Assuming that
acetate was the preferred VFA for denitrification (Elefsiniotis
and Wareham, 2007) and that the anoxic yield on acetate
was 0.192 mgBiomassCOD (mgCODutilized)�1 (Copp and Dold,
1998), a ratio of 3.54 mgCOD mgNO3eN�1 was obtained.
Accordingly, all EBPR stoichiometric VFA ratios were calcu-
lated assuming that 3.54 mgCOD mgNO3eN�1 and
1.71 mgCOD mgNO2eN�1 were utilized for nitrate and nitrite
reduction, respectively (Tchobanoglous et al., 2003). Given
that the NOx residual was small in comparison to the influent
VFAs, this assumption causedminimal variation in the ratios.
2.4. Analysis
2.4.1. Chemical analysesAll soluble constituents were filtered through a 0.22 mmMillex
GP syringe-driven filter (Millipore, MA, USA). Phosphate
(PO4eP) and nitrate (NO3eN)were determined colorimetrically
0
10
20
30
40
50
60
0
3
6
9
12
15
7/26 7/31 8/5 8/10 8/15 8/20 8/25 8/30
NH
3-N
(m
g L
-1)
PO
4-P
(m
g L
-1)
Date
Feed P Feed NH3-Na
0
10
20
30
40
50
60
0
100
200
300
400
7/26 7/31 8/5 8/10 8/15 8/20 8/25 8/30
VF
A:P
(m
gC
OD
m
gP
-1)
VF
As
(m
gC
OD
L
-1)
Date
Feed VFAs VFA:Pb
Fig. 2 e Typical influent characteristics for (a) P and
ammonia, and (b) VFAs and VFA:P ratio at the beginning of
this study. The dashed lines show each time that a new
batch of wastewater was collected.
wat e r r e s e a r c h 4 5 ( 2 0 1 1 ) 6 1 1 9e6 1 3 0 6121
Author's personal copy
(Coats et al., 2011b). Ammonia (NH3eN) and nitrite (NO2eN)
were measured in accordance with Hach (Loveland, CO, USA)
methods 10031 and 10019, respectively. MLSS andMLVSSwere
analyzed according to the standard methods (APHA, 1998). pH
was monitored using American Marine (Ridgefield, CT, USA)
Pinpoint� pH controllers. VFAs were measured using a gas
chromatograph (GC) equipped with a flame ionization
detector (Coats et al., 2011a). Glycogen was determined with
dried biomass samples as described by Parrou and Francois
(1997); biomass samples were washed with a 1% NaCl solu-
tion prior to analysis to minimize potential interference with
exopolysaccharide (EPS), the latter being a source of slowly
biodegradable carbon for bacteria. Intracellular PHA was
quantified using a GC equipped with a mass spectrometer
(Coats et al., 2011b). Surface oxygenmass transfer coefficients
(KLaSUR) were determined using room temperature tap water
according to the dynamic gassing out method of Van’t Riet
(1979), where saturation DO concentrations were estimated
using temperature and pressure correction equations (APHA,
1998). KLaSUR values were calculated by minimizing the total
sum of square errors between the predicted DO from themass
transfer model and the actual DO measurements. Both DO
and temperature were recorded simultaneously using a Hach
HQ30d Meter and LDO101 DO Probe. The volumetric oxygen
mass transfer coefficient, KLaSUR, in reactor O was measured
at 0.54 h�1.
2.4.2. Microbial population analysesGenomic DNA was extracted from each reactor on the dates
shown (Fig. 1) according to the procedure outlined in Coats
et al. (2011c). Quantitative real-time PCR (qPCR) was used to
quantify 16S rDNA genes from total bacteria, Accumulibacter
(the model PAO), and GAOs to provide an estimation of rela-
tive abundance. qPCR was conducted on a StepOne Plus�Real-Time PCR system (Applied Biosystems, Foster City, CA)
using iTaq� SYBR� Green Supermix w/ROX (Bio-Rad Labora-
tories, Inc., Hercules, CA, USA) with a total reaction volume of
25 ml. Total bacterial and total Accumulibacter 16S rDNA genes
were quantified with primer sets 341f/534r and 518f/846r,
respectively (He et al., 2007). GAOs were quantified using
primer set GAOQ431f/GAOQ989r (specifically designed to
target Candidatus Competibacter phosphatis, which is a puta-
tive model GAO (Crocetti et al., 2002)) and the total bacteria
primer set. In addition, a primer set targeting the GB lineage
(specifically GB612f/GAOQ989r (Kong et al., 2002), coupled
with the total bacteria primer set) was employed to quantify
Gammaproteobacteria. The GB lineage, also referred to as the
Competibacter lineage, is proposed to capture the predominant
GAOs within the class Gammaproteobacteria that would be
present in EBPR WWTPs (Kong et al., 2006; Oehmen et al.,
2007). qPCR conditions were as follows: 3 min at 95 �C, 45cycles of 30 s at 95 �C, 45 s at 60 �C, and 30 s at 72 �C. Allunknown samples were assessed in triplicate with 5 ng of
total genomic DNA per reaction. Amplification efficiency was
estimated for each primer set using baseline-corrected fluo-
rescence data (from StepOne Software v2.0) with LinRegPCR
(Ramakers et al., 2003). For PAO quantification, mean ampli-
fication efficiencies for the total bacterial and PAO primer sets
were 96.8 � 1.8% (n ¼ 37) and 93.9 � 1.5% (n ¼ 36), respectively.
For GAO and GB lineage quantification, mean amplification
efficiencies for the respective primer sets were 85.2 � 3.1%
(GAOs; n ¼ 76), 93.6� 3.4% (GB lineage; n ¼ 76), and 93.9� 3.0%
(n¼ 76). The cycle threshold was set at a constant value across
all samples based on location within the log-linear region for
Table 1 e Influent characteristics for all sampling investigations in this study.
Study type Influent characteristicsa
PO4eP (mg L�1) NH3eN (mg L�1) HAc HPr HBu HVa Total VFAs Influent VFA:P Influent VFA:NH3eN
Table 3 e Relative fraction of PAOs (n [ 3) and GAOs (n [ 6; except DNA5 n [ 8) within the bacterial community asestimated by qPCR for the DNA extraction dates shown in the timeline (Fig. 1).
Sample ID PAOs GAOs
GAO primer set GB lineage
Reactor O Reactor C Reactor O Reactor C Reactor O Reactor C
Bortone, G., Saltarelli, R., Alonso, V., Sorm, R., Wanner, J.,Tilche, A., 1996. Biological anoxic phosphorus removal - thedephanox process. Water Sci. Technol. 34, 119e128.
Bracklow, U., Drews, A., Gnirss, R., Klamm, S., Lesjean, B.,Stuber, J., Barjenbruch, M., Kraume, M., 2010. Influence ofsludge loadings and types of substrates on nutrients removalin MBRs. Desalination 250, 734e739.
Carvalho, G., Lemos, P.C., Oehmen, A., Reis, M.A.M., 2007.Denitrifying phosphorus removal: linking the processperformance with the microbial community structure. WaterRes. 41, 4383e4396.
Coats, E., Gregg, M., Crawford, R., 2011a. Effect of organic loadingand retention time on dairy manure fermentation. Bioresour.Technol. 102, 2572e2577.
Coats, E.R., Mockos, A., Loge, F.J., 2011b. Post-anoxicdenitrification driven by PHA and glycogen within enhancedbiological phosphorus removal. Bioresour. Technol. 102,1019e1027.
Coats, E.R., Watkins, D.L., Brinkman, C.K., Loge, F.J., 2011c. Effectof anaerobic HRT on biological phosphorus removal and theenrichment of phosphorus accumulating organisms. WaterEnviron. Res. 83, 461e469.
Copp, J.B., Dold, P.L., 1998. Comparing sludge production underaerobic and anoxic conditions. Water Sci. Technol. 38,285e294.
Cornel, P., Wagner, M., Krause, S., 2003. Investigation of oxygentransfer rates in full scale membrane bioreactors. Water Sci.Technol. 47, 313e319.
Elefsiniotis, P., Wareham, D.G., 2007. Utilization patterns ofvolatile fatty acids in the denitrification reaction. Enzym.Microb. Technol. 41, 92e97.
Erdal, U.G., Erdal, Z.K., Daigger, G.T., Randall, C.W., 2008. It isPAO-GAO competition or metabolic shift in EBPR system?Evidence from an experimental study. Water Sci. Technol. 58,1329e1334.
Flowers, J.J., He, S., Yilmaz, S., Noguera, D.R., McMahon, K.D.,2009. Denitrification capabilities of two biologicalphophorus removal sludges dominated by different’Candidatus Accumulibacter’ clades. Environ. Microbiol.Rep. 1, 583e588.
He, S., Gall, D.L., McMahon, K.D., 2007. CandidatusAccumulibacter" population structure in enhancedbiological phosphorus removal sludges as revealed bypolyphosphate kinase genes. Appl. Environ. Microbiol. 73,5865e5874.
wat e r r e s e a r c h 4 5 ( 2 0 1 1 ) 6 1 1 9e6 1 3 0 6129
Author's personal copy
Henze, M., van Loosdrecht, M.C.M., Ekama, G.A., Brdjanovic, D.,2008. Biological Wastewater Treatment: Principles, Modelling,and Design, First ed. IWA Publishing.
Kong, Y., Ong, S.L., Ng, W.J., Liu, W.-T., 2002. Diversity anddistribution of a deeply branched novel proteobacterial groupfound in anaerobic-aerobic activated sludge processes.Environ. Microbiol. 4, 753e757.
Kong, Y., Xia, Y., Nielsen, J.L., Nielsen, P.H., 2006. Ecophysiology ofa group of uncultured Gammaproteobacterial glycogen-accumulating organisms in full-scale enhanced biologicalphosphorus removal wastewater treatment plants. Environ.Microbiol. 8, 479e489.
Kuba, T., Murnleitner, E., Van Loosdrecht, M.C.M., Heijnen, J.J.,1996a. A metabolic model for the biological phosphorusremoval by denitrifying organisms. Biotechnol. Bioeng. 52,685e695.
Kuba, T., van Loosdrecht, M.C.M., Heijnen, J.J., 1996b. Phosphorusand nitrogen removal with minimal COD requirement byintegration of denitrifying dephosphatation and nitrificationin a two-sludge system. Water Res. 30, 1702e1710.
Kuba, T., van Loosdrecht, M.C.M., Murnleitner, E., Heijnen, J.J.,1994. Effect of nitrate on phosphorus release in biologicalphosphorus removal systems. Water Sci. Technol. 30,263e269.
Kujawa, K., Klapwijk, B., 1999. A method to estimatedenitrification potential for predenitrification systems usingNUR batch test. Water Res. 33, 2291e2300.
Liu, W.T., Mino, T., Nakamura, K., Matsuo, T., 1994. Role ofglycogen in acetate uptake and polyhydroxyalkanoatesynthesis in anaerobic-aerobic activated sludge witha minimized polyphosphate content. J. Fermentation Bioeng.77, 535e540.
Lopez-Vazquez, C.M., Oehmen, A., Hooijmans, C.M.,Brdjanovic, D., Gijzen, H.J., Yuan, Z., van Loosdrecht, M.C.M.,2009. Modeling the PAO-GAO competition: effects of carbonsource, pH and temperature. Water Res. 43, 450e462.
Lopez, C., Pons, M.N., Morgenroth, E., 2006. Endogenous processesduring long-term starvation in activated sludge performingenhanced biological phosphorus removal. Water Res. 40,1519e1530.
Lu, H., Keller, J., Yuan, Z., 2007. Endogenous metabolism ofCandidatus Accumulibacter phosphatis under variousstarvation conditions. Water Res. 41, 4646e4656.
Martins, A., Heijnen, J.J., van Loosdrecht, M.C.M., 2004. Bulkingsludge in biological nutrient removal systems. Water Res. 86,125e135.
Oehmen, A., Carvalho, G., Lopez-Vazquez, C.M., vanLoosdrecht, M.C.M., Reis, M., 2010a. Incorporating microbialecology into the metabolic modelling of polyphosphateaccumulating organisms and glycogen accumulatingorganisms. Water Res. 44, 4992e5004.
Oehmen, A., Lemos, P.C., Carvalho, G., Yuan, Z.G., Keller, J.,Blackall, L.L., Reis, M.A.M., 2007. Advances in enhanced
biological phosphorus removal: from micro to macro scale.Water Res. 41, 2271e2300.
Oehmen, A., Lopez-Vazquez, C.M., Carvalho, G., Reis, M., vanLoosdrecht, M.C.M., 2010b. Modelling the populationdynamics and metabolic diversity of organisms relevant inanaerobic/anoxic/aerobic enhanced biological phosphorusremoval processes. Water Res. 44, 4473e4486.
Oehmen, A., Zeng, R.J., Yuan, Z., Keller, J., 2005. Anaerobicmetabolism of propionate by polyphosphate-accumulatingorganisms in enhanced biological phosphorus removalsystems. Biotechnol. Bioeng. 91, 43e53.
Oh, J., Silverstein, J., 1999. Oxygen inhibition of activated sludgedenitrification. Water Res. 33, 1925e1937.
Parrou, J.L., Francois, J., 1997. A simplified procedure for a rapidand reliable assay of both glycogen and trehalose in wholeyeast cells. Anal. Biochem. 248, 186e188.
Plosz, B., Jobbagy, A., Grady Jr., C.P.L., 2003. Factors influencingdeterioration of denitrification by oxygen entering an anoxicreactor through the surface. Water Res. 37, 853e863.
Ramakers, C., Ruijter, J.M., Deprez, R.H.L., Moorman, A.F.M., 2003.Assumption-free analysis of quantitative real-timepolymerase chain reaction (PCR) data. Neurosci. Lett. 339,62e66.
Schuler, A., Jenkins, D., 2003. Enhanced biological phosphorusremoval from wastewater by biomass with differentphosphorus contents, Part I: experimental results andcomparison with metabolic models. Water Environ. Res. 75,485e498.
Smolders, G.J.F., van der Meij, J., van Loosdrecht, M.C.M.,Heijnen, J.J., 1995. A structured metabolic model for anaerobicand aerobic stoichiometry and kinetics of the biologicalphosphorus removal process. Biotechnol. Bioeng. 47, 277e287.
Smolders, G.J.F., van Dermeij, J., van Loosdrecht, M.C.M.,Heijnen, J.J., 1994. Model of the anaerobic metabolism of thebiological phosphorus removal process: stoichiometry and pHinfluence. Biotechnol. Bioeng. 43, 461e470.
Van’t Riet, K., 1979. Review of measuring methods and results innonviscous gas-liquid mass transfer in stirred vessels. Ind.Eng. Chem. Process Dev. 18, 357e364.
Vocks, M., Adam, C., Lesjean, B., Gnirss, R., Kraume, M., 2005.Enhanced post-denitrification without addition of an externalcarbon source in membrane bioreactors. Water Res. 39,3360e3368.
Wentzel, M.C., Dold, P.L., Ekama, G.A., Marais, G.V.R., 1989.Enhanced polyphosphate organism cultures in activatedsludge. Part III: kinetic model. Water SA 15, 89e102.
Zhang, C., Chen, Y., Randall, A.A., Gu, G., 2008. Anaerobicmetabolic models for phosphorus- and glycogen accumulatingorganisms with mixed acetic and propionic acids as carbonsources. Water Res. 42, 3745e3756.
Zhou, Y., Pijuan, M., Oehmen, A., Yuan, Z., 2010. The source ofreducing power in the anaerobic metabolism ofpolyphosphate accumulating organisms (PAOs) - a mini-review. Water Sci. Technol. 61, 1653e1662.
wat e r r e s e a r c h 4 5 ( 2 0 1 1 ) 6 1 1 9e6 1 3 06130