Content uploaded by Joshua Boltz
Author content
All content in this area was uploaded by Joshua Boltz on Mar 13, 2017
Content may be subject to copyright.
Questions and Answers About Integrated Fixed-Film/
Activated Sludge (IFAS) in a BNR Pilot Plant
Hyun-su Kim1, Sarah Hubbell2, Joshua P. Boltz3, Wayne Flournoy2, James Gellner4,
Paul Pitt4, Robert Dodson5, Andrew J. Schuler1,6*
1Duke University, 2Entex Technologies Inc., 3CH2M HILL, Inc., 4Hazen and Sawyer,
5South Durham Water Reclamation Facility, 6University of New Mexico
*Corresponding author: Andrew J. Schuler, Box 90287,
Duke University, Durham, NC, 27708, aschuler@duke.edu
ABSTRACT
Integrated Fixed-film/Activated Sludge (IFAS) is an increasingly popular technology that
promises to greatly improve the performance of activated sludge wastewater treatment systems.
However, many questions remain about the fundamental behaviors and performance of these
hybrid systems. Their future development will require a better understanding of the biokinetic
behaviors and distribution of specific organisms in the IFAS biofilms, including the effects of
media placement, on these parameters. For this reason a team of researchers from Duke
University, Entex Technologies Inc, Ashbrook Simon-Hartley, CH2M Hill and Hazen and
Sawyer are conducting an in-depth pilot study of IFAS at the South Durham Water Reclamation
Facility in North Carolina. This paper provides early data from this study, providing key insights
to how such systems should be designed and the expected benefits.
KEYWORDS
Biokinetics, BioPortz, BioWeb, integrated fixed film activated sludge, nitrification
INTRODUCTION
Integrated fixed film activated sludge (IFAS) systems are hybrid biological wastewater treatment
systems consisting of microbial activity in both suspended phase, as in conventional activated
sludge systems, and a fixed (biofilm) phase. The fixed phase is included through the addition of
solid media, often to conventional activated sludge reactors, which provides a surface for biofilm
growth. The media used is typically some form of extruded, honey-comb like plastic media (e.g.
BioPortzTM, Entex Technologies), or a series of mesh netting screens through which the waste
stream flows (BioWebTM, Entex Technologies).
The addition of IFAS media greatly increases microbial mass in biological reactors, which
accelerates the removal rates of soluble contaminants and provides additional capacity relative to
conventional systems in the same volume. A major motivation for the use of these systems is to
cost-effectively increase plant capacity, particularly with respect to nitrification, which is thought
to be encouraged because the IFAS media provides a relatively stable environment with a long
solids retention time (SRT) for slow growing, nitrifying bacteria.
143
WEFTEC®.07
Copyright 2007 Water Environment Federation. All Rights Reserved
©
In spite of the increasing popularity of IFAS systems, there is still much that remains unknown
with respect to the fundamental behaviors and benefits of these systems, and this limits the
ability to develop rational design criteria and to apply this set of technologies in a manner that
optimizes performance and minimizes costs.
The objectives of this ongoing study are to:
1. Determine and compare the specific nitrification rates of biomass on textile mesh BioWeb
media, on plastic (BioPortz) media, and suspended phase biomass.
2. Determine the effects of media placement location (e.g., in which reactors) on performance.
3. Determine potential benefits of IFAS systems to secondary clarification and post-clarification
membrane filtration.
4. Evaluate the performance of IFAS systems in a pilot scale system with a non-IFAS control.
5. Determine the differences in microbial communities in suspended and fixed phases through
the application of molecular methods.
METHODOLOGY
Systems Studied
Three systems are being studied as part of this project: (1) a single 25 gallon reactor was
installed at the South Durham WRF to investigate biofilm growth on suspended carrier plastic
media (BioPortz) in a simple, single reactor system, (2) BioWeb coupons were placed in
anaerobic, anoxic, and aerobic reactors in the South Durham WRF to provide initial information
about biokinetics of this type of IFAS biofilm relative to suspended growth under identical
conditions, and (3) a 5 GPM IFAS system, with a parallel non-IFAS control, was constructed at
the South Durham WRF to investigate IFAS performance under controlled conditions.
System 1, the single plastic media (BioPortz) reactor, was fed with about 215 gpd primary
effluent and contained a 50% fill fraction of the media. The reactor was operated in a moving
bed biofilm reactor (MBBR) mode, without activated sludge recycle, for a period of at least 90
days prior to the solids and batch nitrification testing. A course bubble diffuser kept the reactor
aerated and the media well mixed.
In System 2, BioWeb “coupons” (1 ft x 1 ft PVC frames with BioWeb media stretched across
them) were placed at points of interest within the South Durham WRF train as indicated in
Figure 2. The purpose of this portion of the study was to determine how media location affected
the performance of the IFAS media. The media was suspended in the full scale reactors at the
locations shown in Figure 2, and they were left in place for approximately 2.5 months (April to
June) before batch testing described below. Well-developed biomass on a BioWeb coupon is
shown in Figure 3.
144
WEFTEC®.07
Copyright 2007 Water Environment Federation. All Rights Reserved
©
Figure 1: System 1 - Single Pilot Reactor containing BioPortz Media at the South
Durham WRF, and a close up of media with biofilm.
Figure 2: Full scale South Durham WRF with BioWeb coupon locations (System
2).
Return Activated Sludge
Mixed Liquor Recycle
Final
clarifier
Anaerobic
Aerobic
(Ae)
Anoxic
(An)
Primary
clarifier
AnAe
: Coupon locations
12
145
WEFTEC®.07
Copyright 2007 Water Environment Federation. All Rights Reserved
©
Figure 3: Bioweb coupon with well-developed biomass.
System 3, consisting of the pilot scale IFAS train and a control activated sludge system train,
were set up in a pilot trailer at the South Durham WRF (Figure 4). Each train consisted of five
reactors in an A2O configuration for Biological Nutrient Removal. A membrane unit was also
provided for the tail end of each of the two IFAS trains, so that the research team could examine
the effects of pairing IFAS with membrane technology.
Batch Tests
Batch tests were conducted with fixed media coupons, suspended media, and mixed liquor
suspended solids (MLSS) from the reactors of South Durham WRF to study nitrification kinetics
of the suspended growth and attached growth processes. Tests on MLSS were performed by
collecting samples from locations where fixed media coupons were placed. Equal volumes of
MLSS sample and primary effluent (PE) were mixed to make up 3 L, which was aerated for the
duration of experiments. Experiments on media coupons were performed in 13.2 L reactors,
which were made by partitioning a commercially available aquarium for fish with plexiglass
plates. Coupons were placed in the reactors, which were filled with PE and aerated for the
duration of experiment. Nitrification of Bioportz media was investigated by putting
approximately 350 media pieces in a 3 L reactor that contained PE and was aerated.
Analytical Methods
Suspended particles in the samples collected during batch experiments were filtered with glass
fiber filters (25 mm diameter, 0.45 mm pore diameter) before measuring concentrations of nitrate
and ammonia, which were measured with Hach DR 2000 Spectrophotometer. The cadmium
reduction method (Hach method # 8039; concentration range 0.3 ~ 30 mg/L) and salicylate
method (Hach method # 10031; concentration range 0.4 ~ 50 mg/L) were used for measurement
of nitrate and ammonia concentrations, respectively. Dissolved oxygen (DO), pH and
temperature were measured at each time point when the samples were collected. No adjustment
in alkalinity was made as initial alkalinity measured (Standard Method 2320 B; Clesceri et al.,
1998) was higher than 120 mg/L in all reactors and pH during the experiments remained fairly
constant between 7 and 7.3.
146
WEFTEC®.07
Copyright 2007 Water Environment Federation. All Rights Reserved
©
Figure 4: BNR Flow Schematic of IFAS Pilot Plant and photograph looking at
aerobic reactors from the direction of the clarifier. A parallel system without IFAS
media is being operated as a control (left train in lower photo).
Biomass in MLSS and media was quantified by measuring total suspended solids (TSS) and
volatile suspended solids (VSS) in accordance with Standard Methods (Method 2540 D and 2540
E, respectively; Clesceri et al., 1998). Biomass on the Bioweb media was removed by high
pressure water spray from syringe. The removal of biomass was checked by comparing the
weight of the media from coupon and that of fresh media of the same length after drying at 105o
C for 12 hours. Comparison of triplicate measurements showed that the difference in weight was
less than 5%. TSS and VSS of solids detached from Bioweb and suspended in known volume of
water were measured. Biomass in Bioportz media could not be removed by this method, and
instead 100 media (with biofilm) were dried in the oven (105o C) for 12 hrs and the dry weight
was measured. Biomass on the media was then removed with commercial bleach after which the
media were dried again for 12 hour and media weight was measured. The difference between
these two weights was noted as TSS. The measured amount of total solids in the reactor was
divided by reactor volume to have concentration in the reactor (mg/L). TSS and VSS in MLSS
were measured following the standard methods as described above.
Anaerobic Anoxic Aerobic
Clarifier Membrane Unit
Nitrate recycle
RAS
RAS
147
WEFTEC®.07
Copyright 2007 Water Environment Federation. All Rights Reserved
©
RESULTS AND DISCUSSION
Relatively high rates of nitrification were found in System 1, the flow-through BioPortz media
reactor fed with primary effluent. The changes in ammonia and nitrate-during a batch test of this
media are shown in Figure 5. The average biomass attached to the BioPortz media in two batch
tests was 31.4 g/ft3 of media (1.11 g/L), and ranged from 30.1 to 32.8 g/ft3 (1.06 to 1.16 g/L).
The measured nitrification varied to a greater degree, however, ranging from 20.9 to 50.8 lb
NH3-N/(1000 ft3 media*d), or 0.335 to 0.816 kg NH3-N removed/(m3 media*d). Nitrification
rates based on NH3 consumption were about 20% higher than when based on nitrate production.
The reason for the variability in nitrification rates is not certain, but the lower rates could have
been related to ongoing recovery of the reactor after a pump malfunction in the weeks prior to
sampling. These rates compare favorably with published and other suggested values, as shown in
Table 1. These results demonstrated that even a barely visible biofilm on the media (about 4
mg/media piece) can contain a substantial population of autotrophic organisms. The rates
calculated per attached biomass are discussed below in relation to measurements of suspended
growth and BioWeb batch tests.
Figure 5: System 1 - Ammonia consumption and nitrate production (nitrification)
in batch tests of BioPortz media.
Table 1: Comparison of observed nitrification rates in BioPortz media with published
values (rates calculated based on NO3- production).
Nitrification rate
lb NH3-N/(1000 ft3 media*d) Nitrification rate
kg NH3-N /(m3 media*d)
Reference
14.4 to 28.7 0.230 to 0.461 Sen et al., 2000
17.9 0.288 Sen et al., 2005
20.9 to 50.8 0.335 to 0.816 This study
0
5
10
15
20
25
0 30 60 90 120
Time, min.
NH3
NO3
148
WEFTEC®.07
Copyright 2007 Water Environment Federation. All Rights Reserved
©
A series of batch nitrification tests were also conducted on BioWeb coupons installed at different
locations in the full scale South Durham WRF (System 2) to gain a better understanding of the
effects of media location on performance. As noted, coupons were installed in the anaerobic
reactor, upstream anoxic reactor, and the upstream (aerobic 1) and downstream (aerobic 2) ends
of an aerobic reactor (Figure 2).
Figure 6 shows the nitrification rates and ammonia uptake rates for the anaerobic and aerobic
reactors, for both the suspended (activated sludge) and the attached biomass. The nitrate
production and ammonia consumption rates were calculated for the first 60 minutes of each
batch test, and these are shown in Figure 7. Figure 7a indicates that the nitrification rate in terms
of nitrate production was higher in the attached phase than in the suspended phase for the aerobic
reactors, indicating that the attached biomass contained a higher level of nitrifying activity than
did the suspended phase, which likely indicated a larger nitrifying fraction of the population. For
comparison with Figure 7a, the BioPortz batch tests yielded even higher rates of nitrate
production per biomass, at 10.3 to 25.8 mg NO3--N/(g TSS*h).
In contrast, the anaerobic and anoxic reactor data suggested lower nitrification rates in the
attached growth than the suspended phase, when express per biomass (Figure 7a). This is
consistent with expectations, since oxygen concentrations were low in these reactors, creating
conditions unfavorable for nitrifier growth in the biofilm fixed in these reactors.
In theory, the suspended phase nitrification rate should be nearly constant through all reactors,
since the suspended populations should be nearly the same in suspended growth systems with
high SRT/HRT ratios. This was generally true, although the anaerobic phase rate was somewhat
lower than the other phases (Figure 7a; this was confirmed in a separate set of batch tests). The
reason for this is not known at this time, but it can be speculated that nitrifiers may be less active
after even relatively short exposure to anaerobic conditions, relative to when they are taken
directly from aerobic reactors.
Ammonia consumption rates were also calculated for the suspended and attached phases in the
BioWeb batch tests (Figure 7b). Ammonia consumption occurs primarily because of nitrification
and biomass growth. In each case, ammonia consumption rates were greater than the nitrate
production rates, indicating uptake of ammonia for growth, simultaneous denitrification, or
nitrite production. Ammonia consumption rates in the aerobic reactors were similar between the
attached and suspended phases, which was in contrast to the nitrate production rates (Figure 7b).
149
WEFTEC®.07
Copyright 2007 Water Environment Federation. All Rights Reserved
©
Figure 6: Ammonia consumption and nitrate production (nitrification) in batch
tests of suspended growth samples and BioWeb coupons at three locations in
the South Durham WRF.
0
5
10
15
20
25
0 30 60 90 120
Time, min.
C. Suspended Aerobic 1
0
5
10
15
20
25
0 30 60 90 120
Time, min.
E. Suspended Aerobic 2
0
5
10
15
20
25
0 30 60 90 120
Time, min.
D. BioWeb Aerobic 1
0
5
10
15
20
25
0 30 60 90 120
Time, min.
F. BioWeb Aerobic 2
0
5
10
15
20
25
0 30 60 90 120
Time, min.
NH4
NO3
A
. Suspended Anaerobic
0
5
10
15
20
25
0 30 60 90 120
Time, min.
B. BioWeb Anaerobic
150
WEFTEC®.07
Copyright 2007 Water Environment Federation. All Rights Reserved
©
0
1
2
3
4
5
6
1. Anaerobic 2. Anoxic 3. Aerobic 1 4. Aerobic 2
mg NO3-N/(g TSS*h)
Suspended phase
Attached phase
Figure 7a: Nitrate production rates in batch tests of suspended phase
and attached phase (BioWeb) samples taken from 4 reactors along the
full scale treatment train.
0
1
2
3
4
5
6
7
1. Anaerobic 2. Anoxic 3. Aerobic 1 4. Aerobic 2
mg NH3-N/(g TSS*h)
Suspended phase
Attached phase
Figure 7b: Ammonia consumption rates in batch tests of suspended
phase and attached phase (BioWeb) samples taken from 4 reactors
along the full scale treatment train.
151
WEFTEC®.07
Copyright 2007 Water Environment Federation. All Rights Reserved
©
Figure 8 shows the nitrification rates per coupon surface area (that is, a 1 ft x 1 ft coupon would
have a 1 ft2 surface area, rather than using the surface area of the textile media itself). Because
the coupons had similar quantities of attached biomass, the trends shown in Figure 8 are similar
to those in Figure 7b, with much higher rates for the aerobic coupons than the anaerobic and
anoxic coupons. The aerobic reactor rates, at 3.6 to 4.3 lb NH3-N/(1000 ft2*d), were somewhat
higher than those previously reported for BioWeb media, at 2.5 lb NH3-N/(1000 ft2*d) (Hubbell
and Krichten, 2004). This information, and data from ongoing batch tests, may be used to
improve the rational design of BioWeb systems.
In spite of the differences in behaviors of the different BioWeb biofilms, the biofilm quantity on
each coupon was remarkably similar, at 49.0 ± 1.6 mg/in (125 ± 3 mg/cm). Depending on the
activity of this biomass, there may be benefits to installing IFAS media in non-aerobic reactors
as well. These results also indicate that significant biomass can accumulate on fixed media, even
under non-aerobic conditions. Depending on the activity of this biomass, there may also be
benefits to IFAS denitrification in anoxic environments.
With respect to the effects of the location, this data set suggests that, as expected, placement of
media in aerobic reactors is more effective for increasing nitrifier populations than in anaerobic
or anoxic reactors. This data also tentatively suggests that there is a larger nitrification benefit
gained by placing media at the downstream end of aerobic reactors, since these provided the
highest rate of nitrification, but this needs to be verified and compared with previous studies.
This result does make some intuitive sense, since lower organic carbon availability in
downstream locations may decrease heterotrophic growth in the biofilm, which could decrease
competition with autotrophs for space within the biofilm.
The next phase of this study will include the evaluation of the pilot scale System 3 (currently in
the startup phase), in terms of the effects of media placement in the anaerobic and aerobic
reactors on nitrification and overall system performance, the effects of substrate concentrations
on biofilm quality and activity, and the biofilm populations as measured through the application
of oligonucleotide probes. The study will initially focus on the effects of the BioPortz media, and
will compare with BioWeb media at a later date, when the control system is planned to be
converted to a BioWeb system. An important component of the study will be the evaluation of
IFAS effects on fouling rates on post-clarification membranes.
152
WEFTEC®.07
Copyright 2007 Water Environment Federation. All Rights Reserved
©
CONCLUSIONS
Data from a pilot plastic media IFAS system and mesh textile (BioWeb) media installed in full
scale systems provided both confirmation of previous biokinetic data ranges, as well as insights
to the effects of location on IFAS biofilm activity. Measurements of nitrate production and
ammonia consumption rates indicated that activities on plastic media confirmed previous results
in terms of ammonia consumption rates expressed per volume of media. Placement of BioWeb
media coupons in aerobic reactors demonstrated the biofilm populations were enriched with
nitrifiers relative to the suspended growth populations, and substantial biofilm development in
anaerobic and anoxic reactors suggested potential benefits for other functions, such as anaerobic
carbon consumption and denitrification, although further work is necessary on this point.
ACKNOWLEDGEMENTS
The authors are grateful to the South Durham WRF personnel for their assistance with sampling
and setup of the pilot facilities. We also thank Jeff Devine, Ben Gould, and Xia Yongming
(Simon-Ashbrook), Jim Mcquarrie and Bruce Johnson (CH2M HILL), and Katya Bilyk, Bob
Fergen, Joe Rohrbacher, Ron Taylor, and Andre van Niekirk (Hazen and Sawyer) for their
invaluable technical and material support.
REFERENCES
Clesceri, L.S.; Greenberg, A.E.; Eaton, A.D., Eds. (1998). Standard Methods for the
Examination of Water and Wastewater. Washington, D.C., American Public Health
Association, American Water Works Association, Water Pollution Control Federation.
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
1. Anaerobic 2. Anoxic 3. Aerobic 1 4. Aerobic 2
lb NH3-N/(1000 ft2*d)
Figure 8: Nitrification rates (based on NH3 production) in batch tests
of attached phase (BioWeb) samples taken from 4 reactors along the
full scale treatment train, expressed per ft2 of BioWeb coupon.
153
WEFTEC®.07
Copyright 2007 Water Environment Federation. All Rights Reserved
©
Hubbell, S.B.; Krichten, D.J. (2004). Demonstration and full scale results of a plant upgrade for
BNR using integrated fixed-film activated sludge (IFAS) Technology. Proceedings of the
Water Environment Federation Annual Conference (WEFTEC), New Orleans.
Sen, D.; Copithorn, R.; Randall, C.W. (2005). Operating Threshholds for Single Stage
Nitrification In Municipal IFAS and MBBR Systems as Measured In Terms of Minimum
Hydraulic Retention Times and Mixed Liquor MCRT. WEFTEC 2005, Washington, D.C.
Sen, D.; Copithorn, R.; Randall, C.W.; Jones, R.; Phago, D.; Rusten, B. (2000). Investigation of
Hybrid Systems for Enhanced Nutrient Control, Water Environment Federation.
154
WEFTEC®.07
Copyright 2007 Water Environment Federation. All Rights Reserved
©
