Bioresource Technology 232 (2017) 380–388

Contents lists available at ScienceDirect

Bioresource Technology
journal homepage: www.elsevier.com/locate/biortech

Enhancement of microbial density and methane production in advanced

anaerobic digestion of secondary sewage sludge by continuous removal
of ammonia
Bing Tao a, Joanne Donnelly a, Ivo Oliveira a, Ruth Anthony a,b, Victoria Wilson b, Sandra R. Esteves a,⇑
a
Wales Centre of Excellence for Anaerobic Digestion, Sustainable Environment Research Centre, Faculty of Computing, Engineering and Science, University of South Wales,
Pontypridd CF37 1DL, UK
b
Welsh Water, Nelson, Treharris CF46 6LY, UK

h i g h l i g h t s g r a p h i c a l a b s t r a c t

NH3 inhibition was mitigated by

continuous NH4+ removal using ion-
exchangers.

CH4 yields improved by 54% when
NH3 reduced from 630 to 92 mg/L.

Population density of MSC and MBT
increased by over 6 and 5 times.

Total bacterial density increased by
over 2 times facilitated by ammonia
reduction.

Enhanced carbohydrates and proteins
hydrolysis was achieved with
ammonia removal.

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

Article history: Ammonia inhibition mitigation in anaerobic digestion of high solids content of thermally hydrolysed sec-
Received 6 January 2017 ondary sewage sludge by the NH+4 affinitive clinoptilolite and a strong acid type ion-exchange resin S957
Received in revised form 13 February 2017 was investigated. Continuous NH+4-N removal was achieved through ion-exchanging at both tempera-
Accepted 15 February 2017
tures with average removals of 50 and 70% for the clinoptilolite and resin dosed reactors, respectively.
Available online 20 February 2017
Approximate 0.2–0.5 unit of pH reduction was also observed in the dosed reactors. The synergy of
NH+4-N removal and pH reduction exponentially decreased free NH3 concentration, from 600 to 90 mg/
Keywords:
L at 43 °C, which mitigated ammonia inhibition and improved methane yields by approximately 54%.
Advanced anaerobic digestion
Ammonia inhibition
Microbial community profiling suggested that facilitated by ammonia removal, the improvement in
Ion-exchange resin methane production was mainly achieved through the doubling in bacterial density and a 6-fold increase
Zeolite in population of the Methanosarcinaceae family, which in turn improved the degradation of residual
Thermal hydrolysed secondary sewage volatile fatty acids, proteins and carbohydrates.
sludge Ó 2017 Elsevier Ltd. All rights reserved.
Bacteria and methanogens gene abundance

1. Introduction of sewage sludge is treated via AD in the UK and 90% in Germany

Anaerobic digestion (AD) is a well-established process for
sewage sludge treatment in Europe and it is estimated that 66%
⇑ Corresponding author.
E-mail address: sandra.esteves@southwales.ac.uk (S.R. Esteves).
(European Biogas Association, 2014). As secondary sewage sludge
is particularly difficult to treat, a number of pre-treatment pro-
cesses have been investigated and thermal hydrolysis is currently
being implemented on over 55 full scale plants across the world
and deployment continues to grow.

http://dx.doi.org/10.1016/j.biortech.2017.02.066
0960-8524/Ó 2017 Elsevier Ltd. All rights reserved.
 B. Tao et al. / Bioresource Technology 232 (2017) 380–388
Despite the attractive benefits offered by AD, a number of diges-
ters have often suffered inhibition imposed by compounds, among
which ammonia is the most common one. It has been proposed
that ammonia inhibits AD mainly through changing intracellular
pH of methanogens, increasing maintenance energy requirements
and suppressing a specific enzyme reaction (Chen et al., 2008;
Rajagopal et al., 2013), although knowledge of how ammonia tox-
icity occurs is still limited. Free ammonia nitrogen (FAN, NH3) is
the main cause of inhibition as it can freely permeate cell mem-
branes. It is recognized that when total ammonia nitrogen (TAN,
i.e.NH3-N + NH+4-N) exceeds 3000 mg/L (McCarthy, 1964), the AD
process is inhibited at any operating pH and severe methanogene-
sis inhibition occurs when TAN exceeds 4000 mg/L (Procházka
et al., 2012).
Typically, to counteract ammonia inhibition, anaerobic diges-
ters have been operated using co-digestion substrates with high
carbon content or using water for dilution. Alternatively, digesters
have been operated sub-optimally at lower loading rates and
reduced conversions with a high level of volatile fatty acids (VFAs)
accumulation (Banks et al., 2011) facilitated by a high buffering
from the level of ammonium. Other researchers on the contrary
have identified that after adaptation of the microbial community
to high levels of ammonia, biogas production can, however, occur
without instability or reduced performance (Kovács et al., 2015)
and particularly, Methanosarcina sp. have been reviewed as highly
tolerant to high ammonia levels (De Vrieze et al., 2012). Ammonia
impact on digestion is therefore a topic of argument among the
academic and industrial communities.
Some research on ammonia removal aiming to reduce inhibi-
tion or recover the nutrient has also been conducted. There are a
number of reports applying different methods to remove/recover
ammonia from anaerobic digesters. Ammonia stripping requires
increasing pH of effluent to 10–11 with addition of alkali and also
demands heat and/or constant gas flow to effectively remove
ammonia (Oliveira et al., 2013; Park and Kim, 2015; Tao and
Ukwuani, 2015), hence, it is normally operated in batch mode
using anaerobic digestate instead of using an online ammonia
removal method. Similarly, struvite precipitation requires adjust-
ment of pH and addition of chemicals, and therefore it has been
used offline as well (Cerrillo et al., 2015; Romero-Güiza et al.,
2015). With recent advancements in membrane technology, coun-
teracting ammonia inhibition has been investigated using both
pressure- and electrical-driven membrane technologies. Lau-
terböck and colleagues reported that continuous removal of
ammonia, which resulted in a drop of pH and VFA levels, has led
to a higher gas production (Lauterböck et al., 2012, 2014), however
microbial communities and detailed degradation rates were not
investigated. Desloover and co-workers investigated the feasibility
of applying electrical-driven cation exchange membrane to extract
ammonia from the effluent of an upflow anaerobic sludge blanket
reactor fed by diluted molasses solution (Desloover et al., 2012,
2015). The simultaneous ammonium extraction and pH control
achieved by the electrochemical unit improved the methane pro-
duction by 4.5 times in comparison to the control reactor. Despite
the advantages, the membrane unit suffered fouling from sus-
pended solids and that is why most studies involving membranes
focused on the treatment of either a model solution or low suspend
solids wastewater effluents only (Desloover et al., 2015;
Lauterböck et al., 2012).
Ammonium removal by ion-exchange has been reported by
many researchers. Natural zeolites have also been used to mitigate
ammonia inhibition of AD (Wang et al., 2011; Zheng et al., 2015).
The adsorption of ammonium ion by natural zeolites has shown
improved stability of digesters and better methane production
(Ho and Ho, 2012; Zheng et al., 2015). However, the ion-
exchanging capacities of zeolites are usually quite low, which leads
381
to the need for quite high zeolite dosage. For example, it has been
demonstrated that the methane production was improved by only
26% at a dosage of 10 g/L, while it was improved by 60% when
dosage was increased to 20 g/L (Ho and Ho, 2012). Also, natural
zeolite mostly exchanges non-proton cations (e.g. Ca2+, Mg2+) for
ammonium ion; thus, there is usually little effect in pH reduction
by natural zeolites. To the authors’ knowledge, there is little infor-
mation available on how the microbial community is affected dur-
ing the ammonium removal by ion-exchanging materials.
Therefore, in this work, continuous ammonium removal using
ion-exchanging materials was investigated, with the objective of
understanding how simultaneous ammonium removal and pH
reduction impact on the performance of digesters and especially
the microbial community characteristics.
2. Materials and methods
2.1. Ion-exchanging materials
Two ion-exchanging materials, including clinoptilolite and a
strong acid type ion-exchange resin S957, were evaluated for
ammonium removal. Clinoptilolite with particle size of 1–2 mm
was obtained from Rota Mining Corporation, Turkey. The Sulfonic
and phosphonic acid functionalized cation exchange resin S957
was provided by PurolitÒ (Llantrisant, UK). Zeolite refers to clinop-
tilolite and resin refers to S957 in this work, when no other speci-
fication is given. The typical characteristics of the used zeolite and
resin are summarized in Table 1.
2.2. Setup of the anaerobic digesters
Twelve lab scale continuously stirred-tank reactors (CSTR) (1 L
size) were set up and operated at 37 and 43 °C (6 reactors on each
temperature). The reactor configurations comprised of 3 types:
control, zeolite and resin dosed reactors. Each reactor condition
was run in duplicate. The temperatures were selected to simulate
the full scale mesophilic advanced anaerobic digesters at Cardiff
wastewater treatment works (WWTWs) treating thermally hydrol-
ysed secondary sewage sludge with an incoming feed of 8–9% total
solids (TS). As sludges are pre-treated at 165 °C and 6 bar, during
hot weather conditions and due to sub-optimal heat exchanger
performance, the actual temperature of the digesters are at the
higher end of mesophilic temperatures (42–43 °C), hence 43 °C
was selected as well.
Inoculum for seeding the reactors were obtained from the full
scale advanced anaerobic digesters from Cardiff WWTWs (Cardiff,
UK). The laboratory reactors were fed with thermally hydrolysed
waste activated sludge (THWAS) sourced from Cardiff WWTW.
Typical characteristics of THWAS is summarized in Table 2. Each
reactor was initially seeded with 800 g of digestate inoculum,
heated to 37 or 43 °C and then fed with thermally hydrolysed
sludge and allowed to acclimatise for 14 days to allow for a similar
Table 1
Summary of the characteristics of the used zeolite and resin.
Clinoptilolite Resin S957
Structure Monoclinic platy Polystyrene cross-linked
crystals divinylbenzene
Ionic form Ca2+, Mg2+, K+, Na+ H+
Physical Angular granules Spherical beads
appearance
Particle size (mm) 1000–2000 425–825
Pore size type Microporous Macroporous
Water content <3% 45–55%
Acidic capacity 1.5–1.8 11.25
(mmol/g)
382 B. Tao et al. / Bioresource Technology 232 (2017) 380–388

Table 2
Typical characteristics of the THWAS fed to the reactors.

Parameters Values Parameters Values Parameters Values

pH 5.82 VS (%TS) 74.87 SO2
4 (g/L) 0.16
tCOD (g/L) 88.21 NH+4 (g/L) 1.46 PO3
4 (g/L) 0.46
2+
sCOD (g/L) 32.54 TKN (g/L) 4.84 Mg (g/L) 0.11
2+
TS (%) 7.81 tVFA (g/L) 4.38 Ca (g/L) 0.21

operation and performance amongst all the reactors running at the
same temperature before the ion-exchangers were dosed. The
reactors were operated within a rotating incubator, which main-
tained them at the defined temperatures with a stirring of
100 rpm. The reactors were operated at a hydraulic retention time
(HRT) of 14 days and an organic loading rate (OLR) of 4 g volatile
solids (VS) per L per day and were fed semi-continuously once a
day during week days only. After 1 HRT acclimatisation, dosage
of ion-exchange resin and zeolite started (Fig. 1). The dosages of
ion-exchangers were 30 and 5 g each time for the zeolite and the
resin, respectively. The dosages were determined based on the typ-
ical cation exchange capacities of the zeolite (1.5–1.8 mmol/g) and
resin (11.25 mmol/g), and with the consideration of the typical
ammonium level found in the thermally hydrolysed sludge. The
zeolite and resin were retained loosely by a porous nylon mesh
purchased from a local shop and replaced every three days. The
biogas produced passed through a NaOH solution (6 mol/L) to
remove CO2 before it was measured continuously by wet tip gas
meters that were assembled and calibrated in-house, with the
maximum flow being 2 mL/min. Samples were taken every work-
ing day and the characteristics of the digestate were monitored.
2.3. Analytical methods
Determinations of TS and VS contents were carried out accord-
ing to the standard methods from American Public Health Associ-
ation (APHA, 2005). Concentrations of VFA levels in the digestate
were determined by gas chromatography (Perkin Elmer Clarus
500) using the static headspace injection method (Perkin Elmer
HS40), as described elsewhere (Cruwys et al., 2002). Ammonium
2.4. Microbial profiling
In order to evaluate the effect of ammonia removal on the abun-
dance of bacteria and diversity of methanogenic communities as
well as population dynamics, samples were subjected to quantita-
tive polymerase chain reaction (qPCR) analysis, as described else-
where (Williams et al., 2013). Samples were extracted in
triplicate and analyses were performed in duplicate. Total genomic
DNA was extracted using a PowerSoil DNA Isolation kit (Mo Bio
Laboratories Inc.) and the concentration of extracted DNA was
quantified based on absorbance at 260 nm using a NanoDrop
1000 spectrophotometer (Thermo Scientific). Two families of ace-
toclastic methanogens (Methanosarcinaceae (MSC) and
Methanosaetaceae (MST)) and three orders of hydrogenotrophic
methanogens (Methanobacteriales (MBT), Methanomicrobiales
(MMB) and Methanococcales (MCC)) were analysed. Quantitative
calibration curves were constructed using the representative
strains corresponding to each primer and probes set targeting the
above mentioned methanogens. Strains were obtained from the
Deutsche Sammlung von Mikrooganismen and Zellkulturen,
Braunschweig, Germany and real-time PCR was performed on a
Biorad iQ5 system from Bio Rad Laboratories.
3. Results and discussion
3.1. Ion-exchange process and pH reduction
The clinoptilolite has an empirical formula of (Ca, Mg, K, Na)3–
6Si30Al6O7224H2O, meaning that the exchangeable ions available
could be Ca , Mg , K or Na+. In order to determine the species
2+ 2+ +

magnesium, potassium and calcium ion concentrations were anal-
ysed by ion chromatography using (Dionex ICS3000) using an Ion-
Pac CS12A as the separation column. The eluent was 20 mmol/L
methansulfonic acid and suppressed conductivity was used as
the detection method. Total polysaccharide levels were deter-
mined using anthrone-sulfuric acid method with glucose as stan-
dard (Frølund et al., 1996). Total protein levels were measured
using the Lowry-Folin method using bovine serum albumin (BSA)
as standard (Frølund et al., 1996).
of exchangeable ions, a preliminary batch test was carried out.
Clinoptilolite and resin S957 were added to the reactor contents.
Then the vessel was placed in a shaking incubator and samples
were taken at predetermined time intervals. The cation concentra-
tions evolution over time and the initial concentrations were
determined by ion chromatography. For the zeolite, it is shown
that the concentration of ammonium and potassium decreased,
while the level of calcium and magnesium ion increased, indicating
the main exchangeable ions in clinoptilolite were Ca2+ and Mg2+. In

Fig. 1. Schematic illustration of the experimental setup.

 B. Tao et al. / Bioresource Technology 232 (2017) 380–388
the case of the resin, since the functionality is sulfonic and phos-
phonic acid, the exchangeable ion is proton, where the results
showed the levels of all cations decreased over time and pH
dropped.
Fig. 2 shows the pH values of the tested reactors at both tem-
peratures. The pH values of the control reactors at 37 °C were sta-
bilized in the range of 7.5–7.6, while a lower pH was observed for
the zeolite and resin dosed digesters, where the pH values in the
zeolite and resin dosed reactors gradually dropped from 7.5 to
7.3. More significant pH reductions for the ion-exchangers dosed
reactors were observed at 43 °C, the pH decreased from 8.0 to a
final value of 7.4 with zeolite and 7.5 with the resin. A previously
reported detailed study of the interaction of clinoptilolite with
the aqueous solution showed that clinoptilolite increased the pH
of solution in the weakly acid region while it decreased the pH of
solution in the basic region, proposing that it is an amphoteric sub-
stance that tends to neutralize the aqueous medium acting as
either proton acceptor or donor (Rivera et al., 2000). The pH reduc-
tion in the reactors with addition of clinoptilolite could be caused
by this property of clinoptilolite. In this study, 0.2 unit of pH reduc-
tion was achieved by the zeolite at 37 °C and 0.6 unit at 43 °C.
Therefore, the ion-exchange process that occurred in the zeolite
dosed reactors could be presented as in following equation.
ðCa=MgÞ3 Si30 Al6 O72  24H2 O þ 6NHþ4 NHþ4 6 Si30 Al6 O72  24H2 O
þ 3ðCa=MgÞ 2þ
ð1Þ
The pH reduction for the resin dosed reactors was caused by the
release of H+ from the strong proton type resin. Considering the
functionality of the resin, the ion-exchange process that occurred
in the reactors dosed by resin could be presented as follow.
R  SO3 H þ NHþ4 R  SO3 NH4 þ H þ
ð2Þ
3.2. Ammonium removal and residual VFA levels
The concentrations of ammonium in the control and dosed
reactors as a function of time are shown in the top part of Fig. 3
in solid symbols. It is clearly seen that the ammonium concentra-
tions were continuously decreased with the addition of ion-
exchanging materials. At 37 °C, the ammonium level was continu-
ously lowered by the presence of zeolite and resin. The ammonium
concentration was decreased from the initially approximately 4.0
to 2.3 g/L with addition of zeolite to the reactors. A significant low-
ering of ammonium concentration was achieved when the resin
was added into reactor, where the ammonium level was found to
decrease from approximately 4.0 to 2.0 g/L. At 43 °C, similar trends
Fig. 2. pH values of the control and ion-exchanger dosed reactors at 37 °C (a) and 43 °C (b).
383
were observed in terms of ammonium concentration, where with
the addition of zeolite and resin, ammonium levels decreased from
initially 4.1 down to 1.6 and 1.4 g/L, respectively.
Ammonium ion (NH+4) and free ammonia (NH3) are the principal
two forms of ammonia nitrogen existing in anaerobic reactors.
Although exact knowledge of how ammonia nitrogen inhibits AD
is limited, it is generally suggested that NH3 is the main toxicant
between the two forms because of its high permeability in bacte-
rial cell membranes (Müller et al., 2006). It is also proposed that
the inhibitory effect of NH3 is mainly associated with enzymes in
methanogenic populations particularly in acetoclastic species.
Fractions of NH3 enter the cell and combine with proton to form
NH+4; besides the direct inhibitory effect of NH+4 to methane pro-
ducing enzymes (Gallert et al., 1998), as it also causes proton
imbalance and pH change in the cells. Then, the cells must con-
sume more energy to maintain proton balance and high intracellu-
lar pH, which potentially causes inhibition to specific enzymatic
reaction (Wittmann et al., 1995). Therefore, it is important, besides
the ammonium level, to calculate the free ammonia level in the
reactors.
Given that the pH and NH+4 concentration as well as tempera-
tures, the concentration of NH3 can be calculated by the following
equations, as proposed by (Anthonisen et al., 1976).
2755:16
pK NHþ ¼ ð3Þ
4 T
½NHþ4 
½NH3  ¼ pK NHþ pH
ð4Þ
10 4 þ1
where [NH3] and [NH+4] denote the concentrations of NH3 and NH+4;
pK NHþ is the dissociation constant of NH+4. The calculated free
4
ammonia concentration evolution as a function of time is presented
in the bottom part of Fig. 3 in open symbols. It is clearly shown that,
with the addition of ion-exchanging materials, the free ammonia
concentrations of the reactors were significantly reduced as well
at both temperatures. At 37 °C, FA concentration decreased from
148 mg/L to 50 mg/L with addition of zeolite and resin; and at
43 °C, FA levels were reduced from 630 mg/L to 90 mg/L within
dosed reactors.
According to Eqs. (3) and (4), the concentration of free ammonia
is mainly determined by the operating temperature and the con-
centration of NH+4. It has been reported that the initial inhibition
occurs for unadapted microbial cultures when free ammonia level
exceeds 100 mg/L (Hansen et al., 1998). The NH3 contents in the
reference reactors averaged at 160 mg/L at 37 °C and 600 mg/L at
43 °C during the operating period, indicating that the reference
384 B. Tao et al. / Bioresource Technology 232 (2017) 380–388

Fig. 3. Ammonium and free ammonia concentration evolutions as a function of time in control reactors and zeolite/resin dosed reactors at 37 °C (a) and 43 °C (b), where the
solid symbol data denotes NH+4 concentration and the open symbol denotes NH3.

reactors were probably at inhibited steady state or high ammonia the activities of microflora in anaerobic reactors, by providing a
level acclimatised steady state. This also suggested that ammonia more suitable growth and functional environment. The pH was
inhibition was probably more significant at 43 than 37 °C, which decreased from 7.5 to 7.3 for reactors at 37 °C and from 8.0 to
agrees with many other similar anaerobic digestion systems. For 7.5 at 43 °C, where the final pH are very close to the upper end
example, Hansen and Angelidaki and co-workers have evaluated of the reported optimal pH range for acetogens and methanogens
the anaerobic digestion of swine manure with high ammonia nitro- (Angelidaki et al., 2003).
gen content (6 g/L) at different temperatures. The results clearly
demonstrated that 37 °C was the best operating temperature 3.3. Methane production
among 37, 45, 55 and 60 °C, where the methane yields were 188,
141, 67 and 22 mL-CH4/g-VS at 37, 45, 55 and 60 °C, respectively The methane production profiles of the 37 and 43 °C operated
(Hansen et al., 1998). These methane yields were much lower than reactors are presented in Fig. 4. At both temperatures, reactors
the potential methane yield (300 mL-CH4/g-VS), indicating ammo- with the addition of zeolite and resin showed improved methane
nia inhibition at all tested temperatures and more significant inhi- yields than the control ones, with the ones with addition of resin
bition at higher operating temperature. The additions of zeolite
and resin have, however, demonstrated working effectively at low-
ering free ammonia levels at both temperatures. As shown in Fig. 3,
the NH3 level within zeolite and resin modified digesters were
decreased substantially. The significant decrease of ammonia level
could be explained by NH+4-N removal by the ion-exchanging pro-
cess and the pH reduction effect imposed by zeolite and resin, as
indicated in Eqs. (3) and (4).
It has been reported that zeolite could effectively mitigate
ammonium inhibition of anaerobic digestion and the synergic
effect of ammonia adsorption and microorganism immobilization
by presence of zeolite have enhanced the AD efficiency (Wang
et al., 2011; Zheng et al., 2015). Tada and co-workers have investi-
gated the effect of various zeolites on the methane yields of anaer-
obic digestion and proposed that the mechanism of zeolite
improving anaerobic digestion performance could be attributed
to the synergistic effect of Ca2+ supply and ammonium ion removal
achieved by Ca2+/NH+4 exchange (Tada et al., 2005). However, these
studies only evaluated the effect of zeolite addition in batch mode
and the pH was manually controlled; thus, the effect of pH reduc-
tion of the zeolite addition was not clear. According to Eq. (4), the
pH value has a vital effect on free ammonia concentration and
other research has also demonstrated that pH reduction had the
greatest effect on mitigating ammonia inhibition among the inves-
tigated methods, including pH reduction, zeolite addition, biomass
addition and humic acid addition (Ho and Ho, 2012). Therefore, the
benefit of continuously using zeolite and resin in this study is not
only attributed to the removal of NH+4-N through ion-exchanging,
but also to the pH reduction effect of the materials. The reduction
of pH decreased the free ammonia concentration exponentially, Fig. 4. Cumulative methane production in control reactors, together with zeolite
according to Eq. (4). Most importantly, pH reduction also affected and resin dosed ones at a) 37 °C and b) 43 °C.
 B. Tao et al. / Bioresource Technology 232 (2017) 380–388
achieving the highest methane yields. At 37 °C, approximately
7.95 L of cumulative methane gas was produced for the control
reactors at the end of experiment, while 8.62 and 9.58 L of
methane were obtained from the zeolite and the resin dosed reac-
tors, respectively. The methane production yields were improved
7.7% with addition of zeolite and 18.9% with resin at 37 °C. The
improved methane production for the ion-exchanger modified
reactors suggested that the control reactors were being operated
at a suboptimal stable condition due to ammonia inhibition.
Improvement of methane yields also indicated that continuous
ammonia removal using ion-exchanging materials could be
applied as an effective measure to mitigate ammonia inhibition.
Similar trends were observed for 43 °C operated reactors. At
43 °C, only 6.17 L of methane was obtained from control reactors,
while from zeolite and resin dosed ones 7.57 and 9.52 L of methane
were obtained, respectively. The methane yields were improved by
22.7% with the addition of zeolite and 54.3% with resin for 43 °C
operated digesters.
Some observations on the effect of operating temperature on
anaerobic digestion of high nitrogen content sewage sludge were
also made from the methane yields of the control reactors at 37
and 43 °C. For the control reactors at 37 °C, methane production
was 7.95 L, much higher than 6.17 L at 43 °C. This is in agreement
with the observed ammonia concentration results of the control
reactors, where the free ammonia concentration was 160 mg/L at
37 °C and 600 mg/L at 43 °C; and this could be explained by the
fact that at a higher temperature, more free ammonia nitrogen
was presented due to the equilibrium of NH3 + H+  NH+4 (Guo
et al., 2013; Rajagopal et al., 2013). Similar trends were observed
for zeolite and resin dosed reactors at 37 and 43 °C, where the
methane yield improvement was more significant at 43 °C,
Fig. 5. Variability of acetic and propionic acid concentrations within the studied reactors; a) acetic acid levels at 37 °C operated reactors; b) propionic acid levels at 37 °C; c)
acetic acid levels at 43 °C; and d) propionic acid levels at 43 °C.
385
confirming a more significant ammonia inhibition in anaerobic
digestion of high ammonia nitrogen content sewage sludge.
In order to investigate where the carbon source came from
for the extra methane production within zeolite and resin mod-
ified reactors, the residual VFA levels of the studied reactors
were analysed. GC analysis of the digestate samples showed that
the main residual VFA species were acetic and propionic acids.
The concentration of acetic and propionic acids in the studied
reactors as a function of time are shown in Fig. 5. As shown in
the graph, in the case of control reactors, the acetic acid concen-
trations were in the range of 350–400 mg/L at 37 °C and 450–
500 mg/L at 43 °C, respectively; and the propionic acid concen-
trations were in the range of 340–420 mg/L at 37 °C and 300–
400 mg/L at 43 °C, respectively. The VFA concentration levels
within the reactors were consistent with the ammonia profile
and methane production trends. The acetic and propionic resid-
ual levels within the digestate were high, when considering
digestion of THWAS. This could possibly be attributed to the
quite high content of free ammonia presented in the reactors
because when ammonia levels reduced, the concentrations of
acetic and propionic acid concentrations within the modified
reactors were much lower than those within the control reac-
tors. For the reactors with zeolite addition, the acetate levels
were gradually decreased from 375 to 247 mg/L at 37 °C and
from 472 to 119 mg/L at 43 °C. In regard to resin dosed reactors,
the acetic acid concentrations were lowered from 356 to
215 mg/L at 37 °C and from 426 to 126 mg/L at 43 °C.
The improved degradation of VFAs in the dosed reactors indi-
cated that the addition of ion-exchanging materials improved the
degradation of residual VFAs and was also indicative that the extra
amount of methane produced within the zeolite and resin modified
386 B. Tao et al. / Bioresource Technology 232 (2017) 380–388

Fig. 6. Concentrations of total carbohydrates in control and zeolite/resin dosed reactors at 37 °C (a) and 43 °C (b); and concentrations of total proteins in control and dosed
reactors at 37 °C (c) and 43 °C (d).

Table 3
Quantification of gene copies for the methanogenic communities in the control and resin dosed reactors at 43 °C.

Control (copies/mL) Resin (copies/mL)

Day 5 Day 10 Day 19 Day 5 Day 10 Day 19
MSC 4.1  108 4.4  108 5.4  108 1.8  108 8.2  108 3.1  109
MST 2.4  105 1.2  105 1.6  105 3.9  104 9.0  104 4.1  105
MMB n.d. n.d. n.d. 2.7  102 2.4  103 n.d.
MBT 3.2  106 5.0  106 5.4  106 2.3  106 1.0  107 1.2  107
MCC n.d. n.d. n.d. n.d. n.d. n.d.

n.d. denotes concentration lower than detection limit of 100 copies/mL.

Table 4 of carbohydrate degradation was observed at 43 °C than at 37 °C,

The gene abundance of the total bacteria within the control and dosed reactors at the
end of the experiment and at both studied temperatures.
Total bacteria at 37 °C Total bacteria at 43 °C
Control 2.8  107 copies/mL 4.0  109 copies/mL
Resin 3.4  107 copies/mL 9.2  109 copies/mL
reactors were at least partially attributed to the further degrada-
tion of residual VFAs.
Other possible carbon sources for the extra methane production
were the carbohydrates, fats and proteins. Hence, the residual con-
centration of total carbohydrates, fats and proteins were analysed
and significant differences were observed for total carbohydrate
degradations in the control and dosed reactors in particular.
Fig. 6 shows the concentration evolution of the total carbohydrates
and proteins in control digesters and the ones dosed with the zeo-
lite and the resin at both temperatures. It is clearly seen that more
polysaccharides were digested in the zeolite and resin dosed reac-
tors than those in the control reactors. Also, a larger improvement
which could be attributed to the ammonia inhibition alleviation
within the reactors at 43 °C. Similar trends occurred in terms of
the total protein contents within the reactors (as shown in
Fig. 6c and d). Further degradation of proteins was observed when
reactors were dosed with zeolite and resin. This improved degrada-
tion of total carbohydrates and proteins indicated that the zeolite
and the resin had an enhancing effect on the anaerobic digestion
efficiency of the THWAS and that the effect was greater at the
higher operating temperature. The improved degradation of carbo-
hydrates and proteins could be also related to the interactions
between multivalent ions and extracellular polymers (EPS), the
category which carbohydrates and proteins fall into. It has been
reported that the EPS and bivalent cations are combined together,
forming a three-dimensional matrix, which holds the sludge floc
together (Rudd et al., 1983). Hence, in the ion-exchanging materials
dosed reactors, the presence of zeolite and resin may have caused a
deflocculation of the sludge flocs by the continuous ion-exchange
step between the zeolite/resin and the three-dimensional matrix,
resulting in the release of EPS (e.g. carbohydrates and protein),
 B. Tao et al. / Bioresource Technology 232 (2017) 380–388
which then were available to microorganisms to degrade (Dai
et al., 2015). This is also the mechanism behind the widely
accepted method of extracting EPS from sludge by cation exchange
resin, as developed by Frølund and co-workers (Frølund et al.,
1996).
3.4. Microbial profiling
In order to understand the ammonia mitigation effect on the
microbial abundance and profile, the qPCR analysis of acetoclastic
and hydrogenotrophic methanogens as well as total bacterial pro-
filing in the control and the resin dosed digesters at 37 and at 43 °C
were performed. The microbial community profiles of the control
and resin modified reactors at 43 °C are summarized in Table 3.
It is clearly shown in Table 3 that the methanogenic communities
in the control and the resin dosed reactors were dominated by MSC
species throughout the experimental period, with over 99% of gene
copies being from MSC. These reactors would be classed as operat-
ing under adapted microbial cultures to high levels of ammonia at
the start of the experiment. The dominance of methanogenic com-
munity by MSC in these conditions is likely to be related to a higher
levels of acetate and the possibly inhibition of MST species due to
the high level of ammonia (Conklin et al., 2006; Karakashev
et al., 2005) within the reactors, since as reviewed by De Vrieze
et al. (2012), MSC methanogens have a much higher tolerance to
ammonia inhibition.
Comparing the MSC populations within control and dosed reac-
tors at 43 °C, the gene abundance of MSC methanogens in the con-
trol reactor maintained a quite stable level, with a range of 4.1–
5.4  108 gene copies/mL, as shown in Table 3. However, the gene
abundance of MSC within the resin dosed digesters, have experi-
enced a stable increase with the addition of resin, from 1.8  108 -
copies/mL on day 5 to 3.1  109 copies/mL on day 19,
approximately 6 times higher than those within the control reac-
tor. This surge of gene abundance (growth and activity) of MSC in
the resin modified reactor was in synchronization with the
decrease of free ammonia level as shown in Fig. 3b, where free
ammonia level was reduced by 6 times from 600 mg/L in the con-
trol digesters to 90 mg/L in the resin modified digesters. As shown
in Table 3, similar trends were observed regarding the gene abun-
dance of MST, the other family of acetoclastic methanogens, the
gene copies stayed at quite stable levels in the control reactors
(1.2–2.4  105 copies/mL), while a steady increase was registered
in the resin modified reactors. This continual increase would be
further expected with the continuation of the experiment as MST
species require a number of days to double.
In terms of hydrogenotrophic methanogens, the most abundant
species among the three orders of hydrogenotrophic microbes was
MBT (Table 2). The same gene abundance variation trends as for
acetoclastic methanogens were observed for hydrogenotrophic
methanogens, where in control reactors the gene abundance main-
tained at a relative steady level and in the resin modified reactors,
the gene abundance experienced a gradual increase with the addi-
tion of resin. The increase of the population of MBT within the resin
dosed reactors suggested that the hydrogenotrophic methanogens
in control reactors were probably inhibited due to the quite high
amount free ammonia presented, since ammonia is a strong inhibi-
tor of the formation of methane from H2 and CO2 (Wiegant and
Zeeman, 1986). As observed for MSC, the alleviation effect of
ammonia inhibition by the resin presence increased the MBT pop-
ulation density and probably ultimately boosted the methane pro-
duction and by lowering the hydrogen partial pressure a more
efficient conversion of propionic acid was also achieved.
Besides the methanogenic populations, another important find-
ing of this study is the total bacterial population density within the
reactors at the studied temperatures. Table 4 summarises the gene
387
abundance of total bacterial in the control and resin modified reac-
tor at the end of experimental trial and studied temperatures. As
shown in Table 4 the gene abundance of the total bacteria within
both control and resin dosed reactors operated at 43 °C were gen-
erally two orders of magnitude higher than those of at 37 °C.
Higher bacterial densities could indicate more fermentative
microbes available to degrade organic matter and ultimately help
to make secondary sludge more available during digestion and
improve methane production and many available technologies
have taken advantage of this. For example, Monsal Enzymic Pro-
cess (Monsal, GE) applies a higher operating temperature (e.g.
42 °C) for the hydrolysis step of anaerobic digestion since a higher
temperature is beneficial to enzymatic hydrolysis of sewage sludge
(Bungay and Abdelwahab, 2006) comparing to 37 °C. Although a
higher temperature is beneficial to enzymatic reactors, methano-
genic systems with high ammonia content are generally operated
at 37 °C, due to the severer ammonia inhibition at higher temper-
ature. However, from the methane production data and microbial
profiling of the reactors, it is possible that a higher operating tem-
perature could be adopted for anaerobic digestion of high ammo-
nia content feedstock to boost hydrolysis and methanogenesis if
the ammonia inhibition was mitigated by a technology, such as
ion-exchange as investigated in this work.
3.5. Research implications to the AD industry
The findings from this work suggest that mitigation of ammonia
inhibition using ion-exchanging materials is a promising way to
improve the methane yields and enhance anaerobic degradation
of feedstocks with high ammonia content at high temperatures
(>37 °C). The positive impact would likely be much broader than
just treating high solids THWAS feedstock. For example, in recent
years, anaerobic digestion of food waste has received significant
attention from academia and industry, but the high content of
ammonia in food waste has created some barriers for this technol-
ogy, such as the limitation of operation at mesophilic temperatures
only and quite high organic residual content in digestates. The high
residual organic matter of digestates would likely increase
methane losses during digestate storage and land disposal; and
more importantly, may potentially be responsible for difficulties
in digestate dewatering. This work has indicated that continual
removal of ammonium using ion-exchanging materials would
enable higher operating temperatures to be used, which could
improve methane production and lead to a further degradation of
organic matter because of enhanced microbial density and activity
within the reactors.
In the case of digesters without ammonia removal, this study
have led to the recommendation that reactors with high ammonia
levels should avoid operating at the top end of the mesophilic tem-
peratures, because of the ammonia inhibition to methanogens,
although higher operating temperature is beneficial to fermentative
microbes. Removal of ammonia allows higher operating temperature
(43 °C), which promoted a significant growth of all the microbial
communities. This is particularly important in AD plants with related
thermal hydrolysis pretreatments, because the sizing and specifica-
tion of heat exchangers in AD plants could be controlled within a
very narrow range of temperatures and constantly throughout the
year despite environmental temperature variations.
The research also demonstrated that both the clinoptilolite and
the strong cation exchange resin are promising in ammonia inhibi-
tion mitigation and improving of methane yields. However, before
this technology becomes more applied in industrial settings, some
factors will require further investigations. For the clinoptilolite, a
high dosage requirement is one of the obstacles, despite the quite
low price of clinoptilolite. For instance, Ho and co-worker investi-
gated the effect of zeolite dosage and the results showed that the
388 B. Tao et al. / Bioresource Technology 232 (2017) 380–388
highest dosage (20 g/L) resulted in the best performance, hence
suggested that even a higher dosage would be required to deal
with substrates with ammonium levels higher than 1740 mg/L
(Ho and Ho, 2012). Tada and co-workers reported that the best per-
formance was achieved with a dosage of 100 g/L for a sludge sub-
strate with an ammonium level of 4500 mg/L (Tada et al., 2005). In
this study, the dosage for clinoptilolite was 37.5 g/L and occupied
part of the reactor volume. For the resin, the performance was very
promising and the dosage requirement was quite low (6 g/L) in
comparison to the zeolite, whist the price of resin S957 (£20/kg)
is much higher than the clinoptilolite currently, because the speci-
fic resin was originally designated to high value-added applica-
tions. The material cost for the resin could be significantly
reduced in a near future if common ion-exchange resins (e.g. the
ones for water softening, £3/kg) could deliver similar performance,
although would require further investigation including ability for
multiple times regeneration.
Other methodologies for ammonia removal may also be estab-
lished and this study for the first time identified that even for
ammonia adapted methanogenic cultures such as the ones abun-
dant in MSC species and already classed as high ammonia tolerant,
microbial growth could be significant enhanced if ammonia was
removed to avoid growth/activity inhibition.
4. Conclusions
Ammonia inhibition of anaerobic digestion has been mitigated
using ion-exchanging technology, where concentration of NH+4-N
was decreased from 4.0 g/L to as low as 2.0 g/L at 37 °C and down
to 1.5 g/L at 43 °C. The synergic effect of NH+4-N removal and pH
reduction of the reactors contributed to the significant decrease
of free ammonia level within the ion-exchanging material modified
digesters (from 600 mg/L to 90 mg/L), which alleviated the ammo-
nia inhibition and improved the methane yields by 54%. The micro-
bial profiling confirmed that the populations of Methanosarcinaceae
and Methanobacteriales methanogens were increased by 6 and 5
times, respectively.
Acknowledgements
The authors would like to acknowledge the Welsh Government
and the European Regional Development Fund (ERDF) for the finan-
cial support for the A4B SuPER CIRP and the KTC AAPBS projects.
References
Angelidaki, I., Ellegaard, L., Ahring, B.K. 2003. Applications of the anaerobic
digestion process. in: Biomethanation II, Springer, pp. 1–33.
Anthonisen, A., Loehr, R., Prakasam, T., Srinath, E., 1976. Inhibition of nitrification by
ammonia and nitrous acid. J. Water Pollut. Control. Fed., 835–852
APHA, 2005. Standard Methods for the Examination of Water and Wastewater.
American Public Health Association (APHA), Washington, DC, USA.
Banks, C.J., Chesshire, M., Heaven, S., Arnold, R., 2011. Anaerobic digestion of source-
segregated domestic food waste: performance assessment by mass and energy
balance. Bioresour. Technol. 102, 612–620.
Bungay, S., Abdelwahab, M., 2006. Monsal enzymic hydrolysis – new developments
and lesson learnt. 13th European Biosolids & Organic Resources Conference &
Workshop. Aqua Enviro, Manchester, UK.
Cerrillo, M., Palatsi, J., Comas, J., Vicens, J., Bonmati, A., 2015. Struvite precipitation
as a technology to be integrated in a manure anaerobic digestion treatment
plant – removal efficiency, crystal characterization and agricultural assessment.
J. Chem. Technol. Biotechnol. 90, 1135–1143.
Chen, Y., Cheng, J.J., Creamer, K.S., 2008. Inhibition of anaerobic digestion process: a
review. Bioresour. Technol. 99, 4044–4064.
Conklin, A., Stensel, H.D., Ferguson, J., 2006. Growth kinetics and competition
between Methanosarcina and Methanosaeta in mesophilic anaerobic digestion.
Water Environ. Res., 486–496
Cruwys, J., Dinsdale, R., Hawkes, F., Hawkes, D., 2002. Development of a static
headspace gas chromatographic procedure for the routine analysis of volatile
fatty acids in wastewaters. J. Chromatogr. A 945, 195–209.
Dai, X., Ye, N., Luo, F., Dong, B., 2015. Enhancement of anaerobic digestive efficiency
by the use of exchange resin to remove cations in sewage sludge. Desalin. Water
Treat., 1–7
De Vrieze, J., Hennebel, T., Boon, N., Verstraete, W., 2012. Methanosarcina: the
rediscovered methanogen for heavy duty biomethanation. Bioresour. Technol.
112, 1–9.
Desloover, J., Abate Woldeyohannis, A., Verstraete, W., Boon, N., Rabaey, K., 2012.
Electrochemical resource recovery from digestate to prevent ammonia toxicity
during anaerobic digestion. Environ. Sci. Technol. 46, 12209–12216.
Desloover, J., De Vrieze, J., Van de Vijver, M., Mortelmans, J., Rozendal, R., Rabaey, K.,
2015. Electrochemical nutrient recovery enables ammonia toxicity control and
biogas desulfurization in anaerobic digestion. Environ. Sci. Technol. 49, 948–955.
European Biogas Association. 2014. Biogas production report in Europe 2014.
Frølund, B., Palmgren, R., Keiding, K., Nielsen, P.H., 1996. Extraction of extracellular
polymers from activated sludge using a cation exchange resin. Water Res. 30,
1749–1758.
Gallert, C., Bauer, S., Winter, J., 1998. Effect of ammonia on the anaerobic
degradation of protein by a mesophilic and thermophilic biowaste
population. Appl. Microbiol. Biotechnol. 50, 495–501.
Guo, J., Dong, R., Clemens, J., Wang, W., 2013. Kinetics evaluation of a semi-
continuously fed anaerobic digester treating pig manure at two mesophilic
temperatures. Water Res. 47, 5743–5750.
Hansen, K.H., Angelidaki, I., Ahring, B.K., 1998. Anaerobic digestion of swine
manure: inhibition by ammonia. Water Res. 32, 5–12.
Ho, L., Ho, G., 2012. Mitigating ammonia inhibition of thermophilic anaerobic
treatment of digested piggery wastewater: use of pH reduction, zeolite, biomass
and humic acid. Water Res. 46, 4339–4350.
Karakashev, D., Batstone, D.J., Angelidaki, I., 2005. Influence of environmental
conditions on methanogenic compositions in anaerobic biogas reactors. Appl.
Environ. Microbiol. 71, 331–338.
Kovács, E., Wirth, R., Maróti, G., Bagi, Z., Nagy, K., Minárovits, J., Rákhely, G., Kovács,
K.L., 2015. Augmented biogas production from protein-rich substrates and
associated metagenomic changes. Bioresour. Technol. 178, 254–261.
Lauterböck, B., Ortner, M., Haider, R., Fuchs, W., 2012. Counteracting ammonia
inhibition in anaerobic digestion by removal with a hollow fiber membrane
contactor. Water Res. 46, 4861–4869.
Lauterböck, B., Nikolausz, M., Lv, Z., Baumgartner, M., Liebhard, G., Fuchs, W., 2014.
Improvement of anaerobic digestion performance by continuous nitrogen
removal with a membrane contactor treating a substrate rich in ammonia and
sulfide. Bioresour. Technol. 158, 209–216.
McCarthy, P., 1964. Anaerobic waste treatment fundamentals. Part III: toxic
materials and their control. Public Works 95, 91–94.
Müller, T., Walter, B., Wirtz, A., Burkovski, A., 2006. Ammonium toxicity in bacteria.
Curr. Microbiol. 52, 400–406.
Oliveira, I., Hegarty, F., Reed, J., Wilson, V., Esteves, S., 2013. Ammonia stripping
methodologies for pre and post digested thermally hydrolysed waste activated
sludge: Preliminary investigations. 18th European Biosolids & Organic
Resources Conference & Exhibition. Aqua Enviro, Manchester UK.
Park, S., Kim, M., 2015. Innovative ammonia stripping with an electrolyzed water
system as pretreatment of thermally hydrolyzed wasted sludge for anaerobic
digestion. Water Res. 68, 580–588.
Procházka, J., Dolejš, P., Máca, J., Dohányos, M., 2012. Stability and inhibition of
anaerobic processes caused by insufficiency or excess of ammonia nitrogen.
Appl. Microbiol. Biotechnol. 93, 439–447.
Rajagopal, R., Massé, D.I., Singh, G., 2013. A critical review on inhibition of anaerobic
digestion process by excess ammonia. Bioresour. Technol. 143, 632–641.
Rivera, A., Rodrıguez-Fuentes, G., Altshuler, E., 2000. Time evolution of a natural
clinoptilolite in aqueous medium: conductivity and pH experiments.
Microporous Mesoporous Mater. 40, 173–179.
Romero-Güiza, M.S., Astals, S., Mata-Alvarez, J., Chimenos, J.M., 2015. Feasibility of
coupling anaerobic digestion and struvite precipitation in the same reactor:
evaluation of different magnesium sources. Chem. Eng. J. 270, 542–548.
Rudd, T., Sterritt, R.M., Lester, J.N., 1983. Extraction of extracellular polymers from
activated sludge. Biotechnol. Lett. 5, 327–332.
Tada, C., Yang, Y., Hanaoka, T., Sonoda, A., Ooi, K., Sawayama, S., 2005. Effect of
natural zeolite on methane production for anaerobic digestion of ammonium
rich organic sludge. Bioresour. Technol. 96, 459–464.
Tao, W.D., Ukwuani, A.T., 2015. Coupling thermal stripping and acid absorption for
ammonia recovery from dairy manure: Ammonia volatilization kinetics and
effects of temperature, pH and dissolved solids content. Chem. Eng. J. 280, 188–
196.
Wang, Q., Yang, Y., Yu, C., Huang, H., Kim, M., Feng, C., Zhang, Z., 2011. Study on a
fixed zeolite bioreactor for anaerobic digestion of ammonium-rich swine
wastes. Bioresour. Technol. 102, 7064–7068.
Wiegant, W.M., Zeeman, G., 1986. The mechanism of ammonia inhibition in the
thermophilic digestion of livestock wastes. Agric. Wastes 16, 243–253.
Williams, J., Williams, H., Dinsdale, R., Guwy, A., Esteves, S., 2013. Monitoring
methanogenic population dynamics in a full-scale anaerobic digester to
facilitate operational management. Bioresour. Technol. 140, 234–242.
Wittmann, C., Zeng, A.-P., Deckwer, W.-D., 1995. Growth inhibition by ammonia and
use of a pH-controlled feeding strategy for the effective cultivation of
Mycobacterium chlorophenolicum. Appl. Microbiol. Biotechnol. 44, 519–525.
Zheng, H., Li, D., Stanislaus, M.S., Zhang, N., Zhu, Q., Hu, X., Yang, Y., 2015.
Development of a bio-zeolite fixed-bed bioreactor for mitigating ammonia
inhibition of anaerobic digestion with extremely high ammonium
concentration livestock waste. Chem. Eng. J. 280, 106–114.