Journal of Environmental Management 193 (2017) 32e39

Contents lists available at ScienceDirect

Journal of Environmental Management

journal homepage: www.elsevier.com/locate/jenvman

Research article

Effectiveness of phosphate removal during anaerobic digestion of

waste activated sludge by dosing iron(III)
Xiang Cheng a, *, Jue Wang a, Bing Chen a, Yu Wang b, Jiaqi Liu a, Lubo Liu c
a
Beijing Key Laboratory for Source Control Technology of Water Pollution, Beijing Forestry University, Beijing 100083, China
b
Beijing China Sciences Guoyi Environment Protection Engineering Co., Ltd, No. 8 Caihefang Road, Beijing 100080, China
c
California State University, Fresno, CA 93740, USA

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

Article history: Phosphate-Fe(II) precipitation induced by Fe(III) reduction during the anaerobic digestion of excess
Received 17 November 2016 activated sludge was investigated for the removal of phosphorus and its possible recovery. The experi-
Received in revised form ments were conducted with three Fe(III) sources at 35
C and 55
C. The results show that ferrihydrite-
2 February 2017
Fe(III) was effectively reduced during the anaerobic sludge digestion by 63% and 96% under mesophilic
Accepted 5 February 2017
Available online 9 February 2017
and thermophilic conditions, respectively. Whereas FeCl3-Fe(III) was only mesophilically reducible and
the reduction of hematite-Fe(III) was unnoticeable at either temperature. Efficient precipitation of viv-
ianite was not observed although high saturation index values, e.g., >14 (activity reduction not
Keywords:
Iron(III) reduction
considered), had been reached. This reveals the complexity of vivianite precipitation in anaerobic
Iron(III) sources digestion systems; for example, Fe(II) complexation and organic interference could not be ignored. With
Phosphate removal ferrihydrite amendments at a Fe/TP of 1.5, methane production from sludge digestion was reduced by
Anaerobic sludge digestion 35.1% at 35
C, and was unaffected when the digestion temperature went up to 55
C. But, acidic FeCl3
Vivianite severely inhibited the methane production and consequently the sludge biomass degradation.
© 2017 Elsevier Ltd. All rights reserved.

1. Introduction chemical precipitation and enhanced biological phosphorus

Eutrophication of aquatic environments causes undesirable
growth of phototrophs, depletion of dissolved oxygen, and
appearance of foul tastes. It has been reported that phosphorus,
even at low concentrations of 0.02e0.3 mg/L, can stimulate algae
growth in lakes and rivers (Schindler, 1977). In many freshwater
ecosystems, phosphorus is often the limiting factor of eutrophica-
tion occurrence (Correll, 1999; Teikari et al., 2015). The total
phosphorus in untreated sewage ranges from 10 to 15 mg/L, with
50e70% in the form of orthophosphate (Blackall et al., 2002). In
many countries, strict regulations have been established on the
upper limits of phosphorus discharge. The USEPA recommends the
limits of 0.05 and 0.1 mg/L for total phosphorus in streams that
enter lakes and in flowing waters, respectively (The Cadmus Group,
2009).
To reduce phosphorus loads (mainly phosphate) to the receiving
water bodies, phosphate removal technologies have been adopted
in municipal wastewater treatment plants (WWTPs), including
* Corresponding author.
E-mail address: xcheng@bjfu.edu.cn (X. Cheng).
removal (EBPR) (Huang et al., 2015; Mao et al., 2014; Morse et al.,
1998; Park et al., 2017; Tervahauta et al., 2014). Biological
methods decrease the level of phosphate in the effluent by trans-
ferring phosphate to the sludge phase. However, the resulting
phosphate-rich sludge and the reject water from sludge digestion
and dewatering need additional treatment (often by chemical
methods) to keep the biotreatment system from phosphate over-
load to the aeration tank. Chemical precipitation of phosphate is
based on the formation of insoluble phosphate, e.g., calcium, iron,
and aluminum salts. The process is simple and reliable, but will
incur a considerable increase in the cost of reagents (Gong and
Zhao, 2014). As a shortage of phosphorus being noted, and also
for making benefits from phosphate removal, phosphate crystalli-
zation for its recovery becomes a necessary step (Koppelaar and
Weikard, 2013). Recent research found that crystallization of
phosphate from digested sludge also improved the dewaterability
of the sludge, thereby reducing the sludge disposal costs
(Bergmans, 2011).
To date, the crystallization of phosphate from waste steams is
mostly researched and practiced by forming struvite (magnesium
ammonium phosphate, MAP) and hydroxyapatite (calcium

http://dx.doi.org/10.1016/j.jenvman.2017.02.009
0301-4797/© 2017 Elsevier Ltd. All rights reserved.
 X. Cheng et al. / Journal of Environmental Management 193 (2017) 32e39
phosphate, HAP) (Ye et al., 2017). Some fully operational facilities
have also been developed and operated in Europe, North America
and Japan (Desmidt et al., 2015; Doyle and Parsons, 2002). How-
ever, other precipitation/crystallization pathways are needed for
some specific conditions, lower costs and/or better products.
Invanov and the coworkers reported that phosphorus in reject
water was effectively removed (up to 90%) via ferrous phosphate
(FeHPO4) precipitation induced by the bioreduction of ferric ions
(Ivanov et al., 2005, 2009). Chen et al. (2015) developed a vivianite
([Fe3(PO4)2$8H2O]) precipitation-based phosphate removal process
for septic systems where alternating anaerobic-aerobic conditions
for EBPR are unavailable. In fact, a substantial portion of phosphate
in natural environments under strong reducing conditions, e.g.,
lake sediments or mesic soils, is Fe(II)-bound as vivianite or
vivianite-like phases (Hupfer et al., 1998; Peretyazhko and Sposito,
2005).
Recently, Cheng et al. (2015) investigated the phosphate pre-
cipitation during anaerobic sludge digestion with iron(III) amend-
ments and found that Fe(III) bioreduction, vivianite precipitation
and methane production could occur in a single system. This pro-
vides a new alternative method for phosphate recovery and may
remarkably reduce the reagent costs because the quality require-
ment of iron source for Fe(II) production is much lower than the
reagents in other processes, e.g., Mg2þ for MAP crystallization
(Ivanov et al., 2009). Also, vivianite crystallization is more
competitive for ferrous ions over siderite (FeCO3) formation, which
reduces the demand of iron source to the system or omits a
degassing step for the removal of bicarbonates as an undesired
ferrous sink (Saidou et al., 2009). However, compared with MAP or
HAP crystallization, vivianite-based phosphate crystallization in
wastewater treatment system lacks sufficient understanding. For
example, the bioreduction of Fe(III) for vivianite formation in the
anaerobic sludge digester was almost unclear. Vivianite crystalli-
zation in real waste water/sludge system also needs additional
information.
The aim of this work was to study the performance phosphate
removal in an iron(III) amended anaerobic sludge digester by pro-
moting vivianite precipitation. Sludge digestion was conducted
under mesophilic and thermophilic conditions with three different
iron(III) sources to explore the relationship phosphate removal and
the reduction of Fe(III).
2. Materials and methods
2.1. Sludge
Excess activated sludge was obtained from the return sludge
line in Gaobeidian WWTP (Beijing, China). Anaerobic-anoxic-
aerobic (A2O) systems were adopted in this plant for enhanced
nutrients removal with a sludge retention time of 20 d. The sludge
was thickened at room temperature for 24 h by gravity before use.
The characteristics of the thicken sludge are shown in Table 1.
2.2. Iron(III) sources
Ferrihydrite (a-FeOOH), ferric chloride (FeCl3) and hematite
(Fe2O3) were used as three forms of Fe(III) sources. Ferric chloride
and hematite were purchased from Xilong Chemical Co. Ltd.
(Guangdong, China) and were used without further purification.
Ferrihydrite was prepared by neutralizing 0.1 M of Fe(NO3)3 to
pH7.5 with KOH under vigorous stirring. The resulting slurry was
washed thoroughly and was resuspended in deionized (DI) water.
Concentrations of Fe(III) in the ferrihydrite and hematite suspen-
sions were determined after solubilizing the samples in 3 N HCl
(Fredrickson et al., 1998).
33
2.3. Sludge digestion assays
Anaerobic digestion of the thickened excess sludge with Fe(III)
amendments were conducted in 117-mL serum vials. Suspensions
of Fe(III) sources, as addressed in 2.2, were added into 50 mL of
excess sludge at a Fe/P (TP in the sludge) molar ratio of 1.5 (as in
vivianite, Fe3(PO4)2$8H2O). The vials were then sealed with butyl
rubber stoppers and aluminum caps. The sludge digestion was
performed under both mesophilic (35
C) and thermophilic (55
C)
conditions in thermostat shakers with 120 rpm of agitation. For
Fe2O3eamended sludge, the effectiveness of anthraquino-
nee2,6edisulfonate (AQDS) as an electron mediator for acceler-
ating Fe(III) reduction and the resulting phosphate removal were
also studied. Iron-free vials were used as controls to observe the
phosphate release during the anaerobic sludge digestion. The levels
of pH, total COD, soluble COD, aqueous phosphate, aqueous Fe(II),
aqueous Fe (T) (total Fe ions ¼ ferrous ions þ ferric ions), 0.5 N HCl-
extractable phosphate, 0.5 N HCl-extractable Fe(II), 0.5 N HCl-
extractable Fe(T) and volatile fatty acids (VFAs) were followed for
30 d. Biogas were collected for the measurements of CH4 and CO2.
All the assays and measurements were performed in triplicate and
the results were shown in mean values ± standard deviations.
Additional sludge samples were collected on day 3 and day 30 of
the anaerobic digestion for scanning electron microscopy (SEM)
observation to study the interaction between microbes and the
Fe(III) sources.
2.4. Analytical methods
The levels of pH were measured by a PHS-3C pH meter (Leici Co.
Ltd., Shanghai). Chemical oxygen demand (COD), phosphate, Fe(II)
and Fe(T) were determined according to standard methods (APHA
et al., 1998) on a DR3900 spectrophotometer (Hach, USA). Samples
for Fe(II) analysis were acidified to pH 1 by HCl acid and were
measured immediately to minimize Fe(II) oxidation. Aqueous
concentrations were measured after filtering the water/sludge
samples through 0.22-mm glass-fiber filter papers (Gelman A/E,
Ann Arbor, USA). Extractable phosphate and Fe species were ob-
tained by treating the sludge samples in 0.5 M HCl for >3 h and
were measured as addressed above (Cheng et al., 2015). Biogas
samples were analyzed using an Agilent 7890A gas chromatograph
with a methanizer/flame ionization detector (FID) combination and
a Porapak Q column (Wilmington, USA) for simultaneously deter-
mination of the contents of CO2 and CH4 (Cheng et al., 2015). Vol-
atile fatty acids (VFAs) were determined by a SP3420A gas
chromatograph (Beifen-Ruili, Beijing, China) with a FID and a
capillary column (30 m  0.53 mm).
Electron microscopy imaging was performed by a field-emission
scanning electron microscope (SU8000, Hitachi, Japan). Sludge
samples were fixed by 2.5% glutaraldehyde and were dehydrated
through a graded ethanol series (Chen et al., 2015). The samples
were coated by gold before SEM examination.
3. Results and discussion
3.1. Iron(III) reduction
The rate of Fe(III) reductive dissolution (by ascorbic acid) was
reported to decrease with the crystallinity of Fe(III)-containing
mineral increasing (Larsen and Postma, 2001). In Fig. 1a, the
levels of aqueous Fe(II) in the FeCl3 and ferrihydrite-amended vials
increased in 9e12 d under mesophilic conditions, with final values
accounting for 81% and 54% of the total Fe in the system. This re-
veals that the reduction of these two Fe(III) sources proceeded
effectively in the anaerobic sludge digestion system. In fact, Fe(III)
34 X. Cheng et al. / Journal of Environmental Management 193 (2017) 32e39

Table 1 (Fig. 2). According to Fig. 1a and b, aqueous Fe(II) concentrations

Characteristics of the thickened excess sludge. were close to aqueous Fe(T), indicating the solubilization of the
Parametersa Thickened excess sludge external Fe(III) sources was via reductive reactions. Aqueous Fe(II)
(Mean ± standard deviation) concentrations also approximated 0.5 N HCl-extractable Fe(II),
pH 6.70 ± 0.35 which accounted for 63% of the supplemented total Fe in the sludge
SCOD (mg/L) 3218.21 ± 212.31 (Fig. 1a and c), suggesting that the majority of the reductively
TCOD (mg/L) 17569.10 ± 152.1 produced Fe(II), if not all the Fe(II) since the precipitated Fe(II) could
TSS (mg/L) 16.36 ± 0.62
be resistant to the extraction by 0.5 N HCl, was in the aqueous
VSS (mg/L) 11.65 ± 0.59
Total Phosphorus (mg/L) 752.59 ± 7.87 phase.
Aqueous phosphate (mg/L as P) 674.62 ± 23.37 After 9e12 d, the levels of aqueous Fe(II) in the sludge decreased
Aqueous Fe(II) (mg/L) 133.78 ± 9.74 continuously, in which a large portion was immobilized and a small
Aqueous Fe(T) (mg/L) 161.68 ± 36.77
portion was reoxidized to aqueous Fe(III) (Fig. 1a and b). According
Extractable phosphateb (mg/L as P) 675.96 ± 10.22
Extractable Fe(II)b (mg/L) 183.40 ± 3.13
to Fig. 1d, 59.8%e97.8% of Fe(T) was extractable in the period of
Extractable Fe(T)b (mg/L) 218.85 ± 1.78 sludge digestion. This reveals that on one hand 0.5 N HCl was
a capable of extracting Fe species from the sludge effectively, and on
SCOD and TCOD denote soluble and total chemical oxygen demand, respec-
tively; TSS and VSS denote total and volatile suspended solid, respectively; Fe(II) and the other hand the extraction efficiency was dependent on the form
Fe(T) denote ferrous ion and total Fe (¼ferrous ions þ ferric ions), respectively. of the Fe species. By comparing the levels of aqueous Fe with
b
Extractable concentrations were measured after treating the sample in 0.5 N HCl extractable Fe illustrated in Fig. 1, the immobilized Fe addressed
solution for at least 3 h.
above was also in the form of Fe(III). In other words, all the Fe(II)
was reoxidized in the latter period of sludge digestion. Anaerobic
ions added to the sludge through FeCl3 were rapidly precipitated to oxidization of Fe(II) has been reported in zones of sufficient light
form FeO$OH because of the pH level of the bulk solution. The penetration by the activity of some phototrophic bacteria (Weber
newly formed ferric precipitate was poorly crystallized ferrihydrite, et al., 2006). In dark environments, anaerobic bio-oxidation of
thus favored the reduction of Fe(III). In contrast, hematite-Fe(III) Fe(II) is also demonstrated to be possible in the presence of nitrate,
reduction was negligible for the entire digestion period even with nitrite or (per)chlorate as electron acceptors (Weber et al., 2006).
AQDS as an electron shuttle (Fig. 1a). This indicates that the crys- However, these two pathways are not likely to explain Fe(II)
tallinity of Fe(III) source greatly influenced the fast reduction of oxidation in the anaerobic sludge digestion system where light
Fe(III) in the sludge digestion system as reported in other anaerobic penetration is limited and nitrate is at low levels. Whether there
environments (Bonneville et al., 2004; Lovley, 1991). As illustrated was a new mechanism warrants a detailed study.
in the SEM images, cell aggregates were closely covered by ferri- Under thermophilic conditions, effective Fe(III) reduction was
hydrite gel in FeCl3 or ferrihydrite-amended sludge, whereas he- only observed in ferrihydrite-amended sludge according to the
matite particles were not effectively interacting with the sludge levels of aqueous Fe(II) and 0.5 N HCl-extractable Fe(II), which

Fig. 1. Profiles of aqueous Fe(II) (a), aqueous Fe(T) (b), extractable Fe(II) (c) and extractable Fe(T) (d) during anaerobic digestion of activated sludge amended with different Fe(III)
sources at 35
C.
 X. Cheng et al. / Journal of Environmental Management 193 (2017) 32e39
Fig. 2. SEM images of the sludge amended with different Fe(III) sources (a: control; b: FeCl3; c: ferrihydrite; d: hematite) during anaerobic digestion.
respectively accounted for 90.0% and 96.0% of the total Fe in the
system (Fig. 3a and c). Unlike the results observed at 35
C, the
reduction of FeCl3-Fe(III) at 55
C was incomplete. In fact, many
Fe(III)-reducing extremophiles, including thermophilic, hyper-
thermophilic and acidophilic archaea and bacteria had been re-
ported in pure culture and real samples (Küsel et al., 1999, 2002;
Lovley et al., 2000). And this is also demonstrated by the unaf-
fected reduction of ferrihydrite-Fe(III) under 55
C in this study.
According to Fig. 3b and d, the FeCl3 added to the sludge was
partially transformed into a solid phase, most possibly being
amorphous ferrihydrite which is 0.5 N-HCl extractable when newly
formed. Since both soluble Fe(III) and ferrihydrite- Fe(III) are
microbially reducible for under circumneutral pH at 35
C, the in-
hibition of Fe(III) reduction in the FeCl3-amended sludge under
thermophilic conditions could be a result of the dual pressure from
temperature and pH (~3, also see 3.4) variations.
3.2. Phosphorus removal
The initial concentration of aqueous phosphate was
728.20 mg P/L after 24 h of gravity thickening of the excess sludge
used for anaerobic digestion. Under mesophilic conditions, the
level of aqueous phosphate in the FeCl3-amended sludge dropped
drastically to 104.28 mg/L in 3 d because of the strong adsorption of
phosphate by the newly formed ferrihydrite and/or the precipita-
tion of ferric phosphate (Fig. 4a). As Fe(III) being reduced in 3e12 d,
the immobilized phosphate was gradually released back into the
liquid phase. However, the reoxidation of Fe(II) afterwards and the
resulting Fe(III) precipitation trapped the aqueous phosphate again,
but this interaction between Fe(III) and phosphate was weak and
unstable. Nearly complete desorption of phosphate was observed
when the 30-d assay was terminated, possibly because the crys-
tallinity increase of the Fe(III) compound led to a great loss of its
adsorption capacity for phosphate (KizewskiA, 2010; Wang et al.,
2013). In the ferrihydrite-amended sludge, phosphate was
35
removed in the initial period via adsorption but the removal effi-
ciency was not as good as in the FeCl3-amended sample where
ferrihydrite was newly formed. The level of phosphate increased
gradually as Fe in the sludge being reduced and reoxidized, but was
averagely ~100 mg/L lower than in the control samples. The vari-
ation of phosphate in the two hematite-amended samples basically
did not differ much from that of the controls, which was consistent
with the fact that hematite-Fe(III) did not effectively enter the
sludge. From Fig. 4b, the 0.5 N HCl-extractable phosphate did not
show significant differences among the sludge samples as ex-
pected. An obvious drop of extractable phosphate was seen in the
FeCl3-amended and ferrihydrite-amended sludge, probably
because the protonation of ferrihydrite or sludge by HCl treatment
enhanced the phosphate adsorption on its surface (Rhoton and
Bigham, 2005). The above results indicate that the variation of
aqueous phosphate levels in the FeCl3-amended and ferrihydrite-
amended sludge was Fe biochemistry induced. However, since
the level of Fe(II) had no obvious correlation with the immobili-
zation of phosphate, the expected vivianite precipitation was not
occurring, or was marginal in this study. In fact, considerable
amounts of aqueous Fe(II) and aqueous phosphate were present in
the system with high saturation index (SI) values, e.g., a SI value of
13.9 in the ferrihydrite-amended sludge after 9 d of digestion
(calculated by using MINTEQ; pH effect was considered to deter-
mine the ionization fractions for Fe2þ and PO3 4 , but the effects of
ion strength and interrupting components were excluded from
consideration). This suggests that vivianite precipitation was more
complex in a sludge digestion system than in simple solutions
where the precipitation can be promoted with SI values of 4e9
(unpublished data). The interrupting factors may include Fe(II)
complexants, ion strength and also biomass, which reduced the
activities of Fe(II) and phosphate for vivianite formation (Ohlinger
et al., 1998; Van der Houwen and Valsami-Jones, 2001; Wilfert
et al., 2015).
From Fig. 4c and d, an increase of 100e200 mg/L in the
36 X. Cheng et al. / Journal of Environmental Management 193 (2017) 32e39
Fig. 3. Profiles of aqueous Fe(II) (a), aqueous Fe(T) (b), extractable Fe(II) (c) and extractable Fe(T) (d) during anaerobic digestion of activated sludge amended with different Fe(III)
sources at 55
C.
phosphate level in the control samples was observed during the
anaerobic sludge digestion under thermophilic conditions. This
reveals that a higher temperature enhanced the lysis of sludge cells.
In the FeCl3-amended sludge, the level of aqueous phosphate
closely related to the level of aqueous Fe(III) (¼aqueous FeT e
aqueous Fe(II)), indicating that the immobilization of phosphate
during the sludge digestion was Fe(III)-bound (Figs. 3b and 4c). This
is confirmed by the fact that aqueous Fe(II) and extractable Fe(II)
were at the same level throughout the entire sludge digestion
period; in other words, phosphate in the solid phase, if any, was not
Fe(II) related. In the ferrihydrite-amended sludge where the dosed
Fe(III) was effectively reduced, the resulting phosphate removal
was fairy limited. Aqueous phosphate was more probably adsorbed
by the unreduced ferrihydrite than was precipitated in the form of
vivianite as expected (the SI value reached 14.6 in 12 d). The dosing
of hematite did not affect the levels of phosphate in the sludge
because of the failure of Fe(III) reduction and the low adsorption
capacity as well.
In general, the effectiveness of Fe(III) reduction-induced phos-
phate precipitation was not as satisfactory as reported in previous
literatures. Our lab had observed a phosphate removal efficiency of
53% during the anaerobic digestion of ferrihydrite amended excess
sludge (Cheng et al., 2015). Ivanov et al. reported the precipitation
of phosphate from reject water with ferrous ions from bioreduction
of Fe(III) ore (Ivanov et al., 2009). Certainly, the differences in the
characteristics of the sludge material gave rise to different perfor-
mance of phosphate precipitation. Since effective Fe(III) reduction
was achieved in this study when ferrihydrite was used as an
iron(III) source and the SI value had reached >14 (ion strength
resulting activity reduction not considered), the ineffectiveness of
Fe(II) phosphate precipitation could be attributed to the complex
anaerobic sludge digestion system where Fe(II) complexation and
organic interference as well could not be ignored. Therefore, further
research is warranted to better understand the key factors gov-
erning vivianite precipitation/crystallization in this specific system
and then develop proper measures for enhancing this reaction.
3.3. Influence of iron(III) on sludge digestion system
The level of pH in the Fe(III)-amended sludge was affected by
the Fe(III)/Fe(II) redox reactions through Eq. (1).
2FeO$OH þ NADH þ H2 O42Fe2þ þ NADþ þ 5OH (1)
The reduction of Fe(III) in the FeCl3-amended sludge under
mesophilic conditions increased the pH from 2.2 to 5.2, while under
thermophilic conditions the pH increase was <1 due to the low
efficiency of Fe(III) reduction (Fig. S1). The pH increases resulted
from the Fe(III) reduction were small in the ferrihydrite-amended
sludge, possibly because of the production of VFAs during the
sludge digestion, which to some extent acted as a OH sink (also
see Fig. 5). The dosing of hematite did not impose a noticeable effect
on the sludge pH.
As shown in Fig. 5a, the production of VFAs from the sludge
digestion was not significantly influenced by the Fe(III) dosing
under mesophilic conditions. The VFAs level in the FeCl3-amended
and ferrihydrite-amended sludge was slightly lower than in the
control samples due to the consumption of VFAs for Fe(III)
reduction as an electron donor (Lovley and Phillips, 1986). A
similar small decrease in the VFA level was observed during the
thermophilic digestion of the ferrihydrite-amended sludge
(Fig. 5b). For the FeCl3-amended sludge, VFAs production at 55
C
was severely inhibited due to the low pH level of 2.3e3.3 (Ponsa

et al., 2008).
 X. Cheng et al. / Journal of Environmental Management 193 (2017) 32e39 37

Fig. 4. Profiles of phosphate during anaerobic digestion of activated sludge amended with different Fe(III) sources (35
C: aqueous (a), extractable (b); 55
C:aqueous (c), extractable
(d)).

The performance of VFAs production had a determinative effect
on the biogas production (Fig. S2) and more significantly the
methanogenesis (Fig. 6). With ferrihydrite amendments, CH4 pro-
duction was lowered by 35.1% for the mesophilic sludge digestion
but was not noticeably affected under thermophilic conditions,
compared with the Fe free controls. The removed inhibition on CH4
production at 55
C can be explained by the higher levels of VFAs
(by ~500 mg/L) generated during the thermophilic digestion, which
relieved the competition of electrons between Fe(III) reducer and
methanogens (Achtnich et al., 1995). In the FeCl3-amended sludge,

Fig. 5. Variation of VFAs during anaerobic digestion of activated sludge amended with different Fe(III) sources at 35
C (a) and 55
C (b).
38 X. Cheng et al. / Journal of Environmental Management 193 (2017) 32e39
Fig. 6. Profile of methane production during anaerobic digestion of activated sludge amended with different Fe(III) sources at 35
C (a) and 55
C (b).
methane production was severely inhibited at both temperatures
because of the decreased pH levels, although fermentative VFAs
production was only partly affected under mesophilic conditions
(Fig. 5a). The inhibition of FeCl3 dosing on the acidogenesis and
methanogenesis during the sludge digestion led to a poor perfor-
mance in the COD removal which reflects the degradation of the
sludge biomass. The concentration of residual TCOD in the FeCl3-
amended sludge after 30 d of mesophilic digestion was 1970 mg/L
higher than in the controls, and the gap was even greater (5909 mg/
L) for the thermophilic digestion because of a higher degree of in-
hibition (Fig. S3). The deterioration of SCOD removal only accoun-
ted for 25% and 30% of the decrease in the TCOD degradation under
mesophilic and thermophilic conditions, respectively. This in-
dicates that FeCl3 dosing also interfered the lyses and hydrolysis of
sludge biomass. For other Fe(III) amendments, the degradation of
sludge biomass was not noticeably influenced as revealed by the
COD removal efficiencies. Therefore, soluble and often acidic Fe(III)
materials (e.g., FeCl3 and Fe(NO3)3) may not be proper iron sources
for the Fe chemistry-based phosphate removal due to the negative
effect on anaerobic sludge digestion.
4. Conclusions
The enhancement of vivianite formation in anaerobic sludge
digesters can be a new promising alternative for phosphorus
removal and recovery from sewage, but the process is far from well
understood, including suitable Fe sources and digestion conditions.
The results of this study indicate that during the anaerobic diges-
tion of excess activated sludge with Fe(III) amendments at a Fe/TP
of 1.5, ferrihydrite-Fe(III) can be effectively reduced under both
mesophilic and thermophilic conditions by 63% and 96% respec-
tively, whereas effective FeCl3-Fe(III) reduction could only be ach-
ieved mesophilically and hematite-Fe(III) reduction was
unnoticeable at either temperatures. Precipitation of vivianite was
not observed although high levels of aqueous Fe(II) and phosphate
were present in the same system with high SI values (activity
reduction not considered), e.g., >14, which warranted further re-
searches on the influences of ion strength and organics, etc. With
ferrihydrite amendments, methane production from the sludge
digestion was reduced by 35.1% at 35
C, and was unaffected when
the digestion temperature increased to 55
C. But, acidic FeCl3
severely inhibited the methane production and consequently the
sludge biomass degradation.
Acknowledgement
This work was supported by the Beijing Higher Education Young
Elite Teacher Project (No. YETP0772). The authors would like to
acknowledge also the Fundamental Research Funds for the Central
Universities (No. 2016ZCQ03) and the National Natural Science
Foundation of China (No. 51478040) for the financial support.
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.jenvman.2017.02.009.
References
Achtnich, C., Bak, F., Conrad, R., 1995. Competition for electron donors among ni-
trate reducers, ferric iron reducers, sulfate reducers, and methanogens in anoxic
paddy soil. Biol. Fert. Soils 19, 65e72.
APHA, AWWA, WEF, 1998. Standard Methods for the Examination of Water and
Wastewater. American Public Health Association, Washington, DC.
Bergmans, B., 2011. Struvite Recovery from Digested Sludge at West WWTP. Delft
University of Technology, Delft.
Blackall, L.L., Crocetti, G.R., Saunders, A.M., Bond, P.L., 2002. A review and update of
the microbiology of enhanced biological phosphorus removal in wastewater
treatment plants. Ant. Leeuw. Int. 81, 681e691.
Bonneville, S., Van Cappellen, P., Behrends, T., 2004. Microbial reduction of iron(III)
oxyhydroxides: effects of mineral solubility and availability. Chem. Geol. 212,
255e268.
Chen, X., Cheng, X., Chen, B., Sun, D., Ma, W., Wang, X., 2015. Phosphorus precipi-
tation in septic systems induced by iron reduction: a process for phosphorus
removal under anaerobic conditions. Desalin. Water Treat. 54, 2891e2901.
Cheng, X., Chen, B., Cui, Y., Sun, D., Wang, X., 2015. Iron(III) reduction-induced
phosphate precipitation during anaerobic digestion of waste activated sludge.
Sep. Purif. Technol. 143, 6e11.
Correll, D., 1999. Phosphorus: a rate limiting nutrient in surface waters. Poult. Sci.
78, 674e682.
Desmidt, E., Ghyselbrecht, K., Zhang, Y., Pinoy, L., Van der Bruggen, B.,
Verstraete, W., Rabaey, K., Meesschaert, B., 2015. Global phosphorus scarcity
and full-scale P-recovery techniques: a review. Cri. Rev. Environ. Sci. Technol.
45, 336e384.
Doyle, J.D., Parsons, S.A., 2002. Struvite formation, control and recovery. Water Res.
36, 3925e3940.
Fredrickson, J.K., Zachara, J.M., Kennedy, D.W., Dong, H., Onstott, T.C., Hinman, N.W.,
Li, S.-M., 1998. Biogenic iron mineralization accompanying the dissimilatory
reduction of hydrous ferric oxide by a groundwater bacterium. Geochim. Cos-
mochim. Ac 62, 3239e3257.
Gong, Y., Zhao, D., 2014. Physicalechemical processes for phosphorus removal and
recovery. In: Ahuja, S. (Ed.), Comprehensive Water Quality and Purification.
Elsevier, Waltham, pp. 196e222.
Huang, W., Huang, W., Li, H., Lei, Z., Zhang, Z., Tay, J.H., Lee, D.J., 2015. Species and
distribution of inorganic and organic phosphorus in enhanced phosphorus
removal aerobic granular sludge. Bioresour. Technol. 193, 549e552.
 X. Cheng et al. / Journal of Environmental Management 193 (2017) 32e39
Hupfer, M., Fischer, P., Friese, K., 1998. Phosphorus retention mechanisms in the
sediment of an eutrophic mining lake. Water Air Soil Poll. 108, 341e352.
Ivanov, V., Kuang, S., Stabnikov, V., Guo, C., 2009. The removal of phosphorus from
reject water in a municipal wastewater treatment plant using iron ore. J. Chem.
Technol. Biot. 84, 78e82.
Ivanov, V., Stabnikov, V., Zhuang, W.Q., Tay, J.H., Tay, S.T.L., 2005. Phosphate removal
from the returned liquor of municipal wastewater treatment plant using iron-
reducing bacteria. J. Appl. Microbiol. 98, 1152e1161.
KizewskiA, F.R., 2010. Phosphate sorption to organic matter/ferrihydrite systems as
affected by aging time. In: 19th World Congress of Soil Science, Soil Solutions
for a Changing World, pp. 84e87.
Koppelaar, R.H.E.M., Weikard, H.P., 2013. Assessing phosphate rock depletion and
phosphorus recycling options. Glob. Environ. Chang. 23, 1454e1466.
Küsel, K., Dorsch, T., Acker, G., Stackebrandt, E., 1999. Microbial reduction of Fe (III)
in acidic sediments: isolation of Acidiphilium cryptum JF-5 capable of coupling
the reduction of Fe (III) to the oxidation of glucose. Appl. Environ. Microbiol. 65,
3633e3640.
Küsel, K., Roth, U., Drake, H.L., 2002. Microbial reduction of Fe (III) in the presence of
oxygen under low pH conditions. Environ. Microbiol. 4, 414e421.
Larsen, O., Postma, D., 2001. Kinetics of reductive bulk dissolution of lepidocrocite,
ferrihydrite, and goethite. Geochim. Cosmochim. Ac 65, 1367e1379.
Lovley, D.R., 1991. Dissimilatory Fe(III) and Mn(IV) reduction. Microbiol. Rev. 55,
259e287.
Lovley, D.R., Kashefi, K., Vargas, M., Tor, J.M., Blunt-Harris, E.L., 2000. Reduction of
humic substances and Fe(III) by hyperthermophilic microorganisms. Chem.
Geol. 169, 289e298.
Lovley, D.R., Phillips, E.J.P., 1986. Organic matter mineralization with reduction of
ferric iron in anaerobic sediments. Appl. Environ. Microbiol. 51, 683e689.
Mao, Y., Yu, K., Xia, Y., Chao, Y., Zhang, T., 2014. Genome reconstruction and gene
expression of “Candidatus Accumulibacter phosphatis” Clade IB performing
biological phosphorus removal. Environ. Sci. Technol. 48, 10363e10371.
Morse, G.K., Brett, S.W., Guy, J.A., Lester, J.N., 1998. Review: phosphorus removal and
recovery technologies. Sci. Total Environ. 212, 69e81.
Ohlinger, K.N., Young, T.M., Schroeder, E.D., 1998. Predicting struvite formation in
digestion. Water Res. 32, 3607e3614.
Park, T., Ampunan, V., Maeng, S., Chung, E., 2017. Application of steel slag coated
with sodium hydroxide to enhance precipitation-coagulation for phosphorus
removal. Chemosphere 167, 91e97.
39
Peretyazhko, T., Sposito, G., 2005. Iron (III) reduction and phosphorous solubiliza-
tion in humid tropical forest soils. Geochim. Cosmochim. Ac 69, 3643e3652.
Ponsa
, S., Ferrer, I., Va
zquez, F., Font, X., 2008. Optimization of the hydro-
lyticeacidogenic anaerobic digestion stage (55 C) of sewage sludge: influence of
pH and solid content. Water Res. 42, 3972e3980.
Rhoton, F., Bigham, J., 2005. Phosphate adsorption by ferrihydrite-amended soils.
J. Environ. Qual. 34, 890e896.
Saidou, H., Korchef, A., Ben Moussa, S., Ben Amor, M., 2009. Struvite precipitation by
the dissolved CO2 degasification technique: impact of the airflow rate and pH.
Chemosphere 74, 338e343.
Schindler, D.W., 1977. Evolution of phosphorus limitation in lakes. Natural mecha-
nisms compensate for deficiencies of nitrogen and carbon in eutrophied lakes.
Science 195, 260e262.
€
Teikari, J., Osterholm, J., Kopf, M., Battchikova, N., Wahlsten, M., Aro, E.M.,
Hess, W.R., Sivonen, K., 2015. Transcriptomic and proteomic profiling of Ana-
baena sp. strain 90 under inorganic phosphorus stress. Appl. Environ. Microbiol.
81, 5212e5222.
Tervahauta, T., van der Weijden, R.D., Flemming, R.L., Leal, L.H., Zeeman, G.,
Buisman, C.J., 2014. Calcium phosphate granulation in anaerobic treatment of
black water: a new approach to phosphorus recovery. Water Res. 48, 632e642.
The Cadmus Group, Inc, 2009. Nutrient Control Design Manualestate of the Tech-
nology Review Report. U.S. Environmental Protection Agency, Washington, DC.
EPA/600/R-09/012.
Van der Houwen, J., Valsami-Jones, E., 2001. The application of calcium phosphate
precipitation chemistry to phosphorus recovery: the influence of organic li-
gands. Environ. Technol. 22, 1325e1335.
Wang, X., Li, W., Harrington, R., Liu, F., Parise, J.B., Feng, X., Sparks, D.L., 2013. Effect
of ferrihydrite crystallite size on phosphate adsorption reactivity. Environ. Sci.
Technol. 47, 10322e10331.
Weber, K.A., Achenbach, L.A., Coates, J.D., 2006. Microorganisms pumping iron:
anaerobic microbial iron oxidation and reduction. Nat. Rev. Microbiol. 4,
752e764.
Wilfert, P., Kumar, P.S., Korving, L., Witkamp, G.-J., van Loosdrecht, M.C., 2015. The
relevance of phosphorus and iron chemistry to the recovery of phosphorus
from wastewater: a review. Environ. Sci. Technol. 49, 9400e9414.
Ye, Y., Ngo, H.H., Guo, W., Liu, Y., Li, J., Liu, Y., Zhang, X., Jia, H., 2017. Insight into
chemical phosphate recovery from municipal wastewater. Sci. Total Environ.
576, 159e171.