Chemosphere 140 (2015) 2–11

Contents lists available at ScienceDirect

Chemosphere
journal homepage: www.elsevier.com/locate/chemosphere

Platforms for energy and nutrient recovery from domestic wastewater:

A review
D.J. Batstone a,b,⇑, T. Hülsen a,b, C.M. Mehta a, J. Keller a,b
a
Advanced Water Management Centre, Gehrmann Building, The University of Queensland, Brisbane, Queensland 4072, Australia
b
CRC for Water Sensitive Cities, PO Box 8000, Clayton, Victoria 3800, Australia

h i g h l i g h t s

 The increasing cost of energy and essential plant nutrients require a shift towards resource recovery.
 Options are low energy mainline (LEM) and partitioning to a solid phase through biological growth.
 LEM generates energy, while partitioning is energy consuming, but can recover nitrogen.
 LEM is more mature, and is likely to see increased application in the short term.

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

Article history: Alternative domestic wastewater treatment processes that recover energy and nutrients while achieving
Received 23 January 2014 acceptable nutrient limits (<5 mg N L 1) are a key challenge. Major drivers are value and availability of
Received in revised form 12 August 2014 phosphorous, nitrogen, and potassium, and increasing energy costs. The two major platforms that can
Accepted 5 October 2014
achieve this are (a) low energy mainline (LEM), with low strength anaerobic treatment, followed by
Available online 6 November 2014
mainline anaerobic nitrogen removal and chemical or adsorptive phosphorous removal and (b) parti-
tion–release–recover (PRR), in which carbon and nutrients are partitioned to solids through either het-
Keywords:
erotrophic or phototrophic microbes, followed by anaerobic digestion of these solids and recovery
Nutrient
Recovery
from the digestate. This paper reviews practical application of these processes, with a focus on energy
Fertilizer costs. Compared to conventional processes which require 0.5 kW h kL 1 electricity (500 mg COD L 1
Domestic influent concentration), PRR requires only 0.05 kW h kL 1 electricity. LEM offers the possibility to recover
Phototrophs 0.1 kW h kL 1 as electricity with net energy generation above 400 mg COD L 1 influent, while PRR
Energy becomes energy generating at >650 mg COD L 1. PRR offers the possibility for recovery of nitrogen and
other nutrients (including potassium) through assimilative recovery. However, the energetic overhead
of this is substantial, requiring 5 kW h kg N 1 as electricity, which compares to ammonia fixation costs.
The lower energy costs, and near to market status of LEM treatment make it likely as a recovery platform
in the shorter term, while ability to recover other elements such as nitrogen and potassium, as well as
enhance favourability on concentrated wastewaters may enhance the desirability of partitioning in the
longer term.
Ó 2014 Elsevier Ltd. All rights reserved.

1. Introduction mg N L 1, P < 1 mg P L 1) (Foley et al., 2010), with potable recycled

Domestic wastewater treatment is now an extremely mature
technology from the point of view of human health, and environ-
mental impact. This is because wastewater can be treated to a san-
itised, low or zero impact waste for <0.6 kW h kL 1 (N < 5
⇑ Corresponding author at: Advanced Water Management Centre, Gehrmann
Building, The University of Queensland, Brisbane, Queensland 4072, Australia. Tel.:
+61 7 3346 9051; fax: +61 7 33654726.
E-mail address: damienb@awmc.uq.edu.au (D.J. Batstone).
water being produced from domestic wastewater for <2 kW h kL 1
(Pearce, 2008). Overall, given the volumes, range of contaminants,
and stringent goals involved in wastewater treatment processes, it
is a remarkable success that society is able to manage its liquid
wastes so effectively. Over the past 20 years though, a number of
major drivers have emerged that emphasize the need to improve
recovery of the resources available in wastewater. These are water
itself, energy, and nutrients. The current activated sludge paradigm
for wastewater treatment requires a substantial amount of energy
(approximately 50% (Foley et al., 2010)) for aeration. This is a limit

http://dx.doi.org/10.1016/j.chemosphere.2014.10.021
0045-6535/Ó 2014 Elsevier Ltd. All rights reserved.
 D.J. Batstone et al. / Chemosphere 140 (2015) 2–11
imposed by the longer sludge ages necessary to retain nitrification
and denitrification capacity (Tchobanoglous et al., 2003). Organic
nitrogen is effectively destroyed (converted to nitrogen gas) by
nitrification and denitrification, and chemical energy is required
by the denitrification process. Phosphorous can be recovered in
the sludge, but this is bulky and contaminated with organics, and
occasionally metals (Yuan et al., 2012), and if aluminium sulfate
is used for phosphate trimming or clarification, it can have
restricted plant availability (Pritchard et al., 2010).
While electricity costs have not changed substantially in the
last 10 years, there have been a number of pricing and manage-
ment changes that have decreased the favourability of constant
loads such as wastewater treatment and there is a high degree of
uncertainty in future pricing (to 2035) due to changes in energy
generation methods that may need load-generator balancing
(DOE/EIA-0484, 2010). There is also likely to be a high degree of
international variation in energy pricing, particularly where there
is a transition in energy sources such as in Japan (DOE/EIA-0484,
2010), or where supply is constrained, but growth is not. As an
example, Australian Energy Market Commission forecasts an
increase in residential electricity prices of around 22% or 8.34 cents
per kW h in the period from 2010–2011 to 2013–2014 (AEMC,
2011) The high degree of uncertainty, as well as various drivers
at the national level mean that there is an urgent need for alterna-
tive low energy wastewater treatment options.
Natural gas pricing in particular is highly variable, has doubled
in the last 10 years, and is expected to double again to 2025 (DOE/
EIA-0484, 2010). This is directly driving an increase in commercial
nitrogen pricing, as ammonia is manufactured using the Haber–
Bosch process using electrons derived from natural gas (Appl,
2000). 60% of the cost of ammonia is natural gas costs, and nitrogen
manufacturing utilises 1–2% of the world’s energy supply (Smil,
2001) (dwarfing the energy costs of wastewater treatment). Natu-
ral gas price increases, as well as an increase in demand have dri-
ven ammonia prices from a low of $150 tonne 1 in 1998–2000 to
its current pricing of approximately $600 per tonne (DiFrancesco
et al., 2010; Fertecon, 2013a).
As a reference, current world fertilizer consumption (2013 pro-
jection) is 111 MT nitrogen as N, 19 MT phosphorous as P, and 26
MT potassium as K (FAO, 2008). Depletion and availability of phos-
phates, as well as market fluctuation have been recently addressed
extensively in the public arena and scientific literature (Cordell
et al., 2009), with peak phosphorous likely to occur within the next
50 years, and as early as 2035 (Cordell et al., 2009). Pricing has also
fluctuated strongly in the past 10 years, rising to $4000 tonne 1 P
in 2009, and currently sitting at $2000 tonne 1 P (calculated from
DAP pricing in Fertecon (2013b)). Pricing increases, as well as a
focus on reducing wastewater treatment costs have driven a sub-
stantial increase in research in, and commercial application of
phosphorous recovery from concentrate streams, mainly through
magnesium ammonium phosphate (struvite) crystallization in
dewatering reject streams (Le Corre et al., 2009; Yuan et al., 2012).
There has been almost no discussion of potassium as a macro-
nutrient target for recovery in the literature. This is perhaps at cur-
rent consumption rates, there are some 330 years of reserves
(Jasinski, 2011), and pricing has historically been <$500 tonne 1
K (Fertecon, 2013c), which given its moderate consumption level,
has not effected a substantial economic impact on farming com-
pared with phosphorous. However, long-term pricing has doubled
over the last 10 years to $1000 tonne 1 K (Fertecon, 2013c), and is
projected to substantially rise in the next 10 years. This does not
account for the accelerated depletion being seen in intensive grain
areas, or its accelerated use to make potassium depleted soils via-
ble (Peverill et al., 1999). There are concerns for developing coun-
tries for long term availability and self-reliance on potash based
conventional fertilizers (Manning, 2010). This is because potash
3
ores have a limited geological distribution, with the bulk of the
world’s potash mined in Canada and Europe (Jasinski, 2012).
The amount of nutrients available in wastewater are substan-
tial. Phosphorous inventories have been best quantified, and glob-
ally, (Cordell et al., 2009) estimated that 20% of the mineral
phosphorous consumed is excreted by humans (and hence recov-
erable). Including domestic animals, the mineral phosphorous
market could technically be fully supplied from excreta streams,
though much of the waste is currently recycled as manure from
grazing animals and is hence not practically or beneficially recov-
erable. Long-term, additional environmental or geological input is
required, though this could be on a much lower level. On a national
level, waste derived phosphorous availability depends heavily on
agricultural fertilizer consumption and food exports. An extreme
case is Australia, where some 5% of the phosphorous can be recov-
ered from domestic effluent, and a total of 20% from humans and
domestic animals (Tucker et al., 2010). This is reversed in food sink
nations with large domestic animal populations, such as the Neth-
erlands, where manure must be exported to maintain national
mineral balances (Henkens and Van Keulen, 2001). Availability of
nitrogen and potassium have not been assessed in the same level
of detail, but nitrogen:phosphorous mass ratio in wastewaters
are on the order of 3:1–5:1 (Tchobanoglous et al., 2003; Tucker
et al., 2010), indicating that a far higher proportion of applied
and naturally assimilated nitrogen is lost during consumption
and treatment. There are generally large amounts of potassium
available in specific wastes such as sugar cane processing, spent
grains, yeast, and manure and processing byproducts from animals
fed with grains and legumes (Tucker et al., 2010). As a global pic-
ture, phosphorous (Cordell et al., 2009, 2011) can be largely ser-
viced from waste streams, with likely moderate environmental,
and minimum geological input, potassium fully serviced from
waste, while approximately 50% of the nitrogen market could be
serviced from waste streams (assuming 4:1 N:P average
concentration).
1.1. Towards resource recovery from wastewater treatment
Addressing global nutrient needs, as well as recovering energy
and water from wastewater streams are powerful drivers for
change in the wastewater industry. This has led to two major posi-
tion papers for novel domestic wastewater processes that are low
energy or energy generating, designed to produce wastewater fit
for reuse (given specific purposes), and designed to allow recovery
of nutrients. In particular, Verstraete et al. (2009) proposed separa-
tion of streams into major and minor (M&M) concentrated and
dilute streams. The default sets of technologies identified were fil-
tration based treatment (gravity–microfiltration–reverse osmosis),
with treatment of solids and concentrate by anaerobic digestion,
and recovery of the nutrients from digestate though Verstraete also
identified alternatives, including biological concentration. The key
limitation of filtration based technology is that if MF/RO is used in
the mainline, it imposes a minimum energy requirement of
approximately 3 kW h kL 1 (Pearce, 2008). Given that the chemical
energy in domestic wastewater is approximately 3 kW h kL 1
(1000 mg COD L 1), and electrical generation efficiency is normally
on the order of 35%, this process will always require electrical
input (though it could be very favourable where potable recycled
water is required anyway).
As an alternative, McCarty et al. (2011) proposed a low energy
mainline (LEM) process, in which domestic wastewater is primary
settled, and treated through low strength anaerobic treatment
(anaerobic membrane bioreactor (AnMBR), or anaerobic filter
membrane bioreactor), which would remove solids and dissolved
organics, and generate methane gas, but not remove substantial
amounts of nitrogen or phosphorous. McCarty et al. (2011) proposed
4 D.J. Batstone et al. / Chemosphere 140 (2015) 2–11
that the effluent from the anaerobic step should be used for irriga-
tion, and hence biological nutrient removal is not required.
Both the Verstraete M&M and McCarty low energy treatment
concepts are a substantial departure from conventional activated
sludge, and offer the possibility to move wastewater treatment
to resource positive platforms. However, they do not focus on sep-
aration and full recovery of resources for a variety of uses. This
paper focuses on further developing these resource recovery plat-
forms towards more general use, and comparison against conven-
tional processes.
2. Generally applicable alternative wastewater treatment
platforms
2.1. Low energy mainline (LEM)
This represents an amendment to McCarty et al., (2011) to
include solids handling and nutrient removal and has also been
discussed with respect to reactor technology in Batstone and
Virdis (2014). It is feasible and likely beneficial to remove the pri-
mary sedimentation unit, and rely on a single-step anaerobic pro-
cess for raw wastewater, since solids will be retained and degraded
in the MBR. An alternative low-energy carbon-removal process to
the AnMBR could be upflow anaerobic sludge blanket (UASB) reac-
tors, but these are ineffective as pathogen barriers, achieving at
best, one log pathogen reduction (Chernicharo, 2006), and hence
requiring post-treatment for sanitation. Issues around low temper-
ature operation have been discussed in Batstone and Virdis (2014)
and dissolved methane recovery can be effectively done at a very
low energy input (<0.002 kW h kLtreated 1) (Cookney et al., 2012).
Both issues favour application of AnMBR over UASB processes,
methane recovery due to the low solids effluent stream, and
cold-temperature because slow-growing biomass can be retained.
Where irrigation is not possible, and nutrients must be
removed, main-line anaerobic ammonium oxidation (anammox)
for nitrogen removal is a possibility (Winkler et al., 2012), at an
energy cost of 1.2 kW h kg N 1 removed (Wett et al., 2013). The
latter is emerging technology, and needs to achieve very low efflu-
ent nitrogen concentrations. If anammox is applied for ammonia
removal, this also favours the use of AnMBR processes over UASB
PARTITION
Domestic sewage
Membrane
photo-bioreactor
Biomass (5%) (ii) Biogas to Cogeneration
Thickened high N, P
Anaerobic Digester
RELEASE
Fig. 1. Partition–release–recover for complete wastewater treatment.
processes, since a UASB generates an effluent with excessively high
residual biodegradable organics. Mainline adsorptive phosphorous
recovery can reduce phosphate levels to <0.1 mg P L 1, with 300
bed volumes, and essentially no sulphate reactivity (Sengupta
and Pandit, 2011). The combination of anaerobic wastewater treat-
ment, mainline anammox, and adsorptive phosphorous recovery
offers an effluent comparable to conventional activated sludge pro-
cesses at a far lower energy input.
2.2. Partition–release–recover (PRR)
As Verstraete et al. (2009) identified, it is possible to partition to
the solid phase through biological uptake. This can be done
through either assimilation (i.e., growth), or accumulation of com-
pounds such as polyhydroxyalkanoates (Johnson et al., 2009) or
poly-phosphates (Yuan et al., 2012). However, relying on microbial
uptake to completely remove and recover nutrients requires an
integrated process as shown in Fig. 1. As an overall process, this
considers one entry point (wastewater), and four key discharge
points
(i) Water, in which the main hydraulic load is dispersed
through reusable water, with a target of N > 10, N < 10, or
N < 5 and P < 1, depending on reuse requirements and tech-
nology options chosen.
(ii) Biogas, which is the main sink stream for excess chemical
energy.
(iii) Biosolids, in which inert organics, non-recoverable nutrients
and excess metals are lost.
(iv) A fertilizer stream, which is the main sink for N, P, and pos-
sibly K.
This section reviews and proposes options for each of the major
stages of partition–release–recover.
2.2.1. Partition
The two key mechanisms by which organisms can partition
organics and nutrients to the solid phase are through growth
(assimilation), and accumulation of polymers as an energy storage
mechanism. Soluble carbon and phosphates are accumulated by
(i) Reusable water (N10)
RECOVER
Nutrient recovery
Dewatering
(iv) Fertilizer Product
NP(K)
(iii) Stable Biosolids
 D.J. Batstone et al. / Chemosphere 140 (2015) 2–11
heterotrophic aerobic organisms as polyhydroxyalkanoates, glyco-
gen, and polyphosphates (Johnson et al., 2009; Yuan et al., 2012).
When taken to an extreme, very short sludge ages are applied to
achieve the A-stage process, which relies completely on absorption,
assimilation, and accumulation for carbon removal (Jimenez et al.,
2005). Recent work (Ge et al., 2013) has indicated that phosphorous
can also be removed at very short sludge ages (2 d), while maintain-
ing anaerobic degradabilities of up to 90% and with a loss of only
25% of the chemical energy to oxidation. This indicates that where
nitrogen removal is not important, activated sludge based accumu-
lation can be used for carbon and phosphorous removal and recov-
ery. An alternative where nitrogen removal is required is to treat
effluent from the A-stage process through anammox, though this
requires effective removal of degradable organics, including solids.
Where nitrogen recovery is required, growth based assimilation
is the only existing option. Given the dissolved COD:N:P ratios in
wastewater are normally on the order of 8:1:0.2 (Tchobanoglous
et al., 2003), and a minimum of 25–50% of the COD is lost to catab-
olism during growth (Tchobanoglous et al., 2003; Ge et al., 2013),
additional electrons are required to correct the COD:N imbalance.
Doing this through addition of soluble organics such as acetate or
methanol is normally ineffective since they are catabolically con-
sumed. However, it is possible to utilise phototrophic organisms
to enhance ammonia recovery, with only sufficient light being
delivered in order to correct the imbalance, or reduce catabolic
consumption of organics.
Algae and plants have been widely proposed for nitrogen
removal from wastewater (Zimmo et al., 2004), particularly in large
area and lagoon process. However, their requirements for high-
energy white light, slower growth rates, and tendency to form mac-
roscopic structures such as streamers makes them less suitable for
engineered systems. Phototrophic bacteria such as cyanobacteria
and purple phototrophic bacteria are widespread in sewage treat-
ment systems (Siefert et al., 1978), and are capable of both PHA
(Khatipov et al., 1998) and poly-P (Hiraishi et al., 1991) accumula-
tion. Particularly interesting are purple phototrophic bacteria,
which utilise long wavelength Bacteriochlorophyll a and b, with
absorption optima above 800 nm (Bryant et al., 2007) and can
assimilate to discharge limits (Hülsen et al., 2014). Energy require-
ments per photon are inversely linear with wavelength and hence
infra-red is low energy. This offers the possibility of light delivery
in an engineered system through low energy infra-red LEDs together
with biomass retention (and very short sludge ages) in a membrane
photo-bioreactor configuration (Chitapornpan et al., 2012), with the
key energy consumption being delivery of light to maximise assim-
ilative uptake, together with membrane fouling control. Emerging
AMBR technologies utilising granular activated carbon for mem-
brane scouring can operate down to 0.058 kW h kL 1 wastewater
treated (Kim et al., 2010) with conventional technology being
Biogas High-VFA supernatant
to partition via P recovery
From partition
Solids to stabilisation
Rotary drum /
First stage thermophilic hydrolysis
gravity thickening
Fig. 2. Options for enhanced thermophilic release.
5
0.25–0.5 kW h kL 1 (Judd, 2011). Providing 100 mg COD L 1 at 10%
photosynthetic efficiency requires 1.0 kW h kL 1, in oxygenic algae
and cyanobacteria, and is not possible in anoxygenic phototrophic
bacteria (since electrons cannot be sourced from water). It is there-
fore important that there be sufficient carbon to drive assimilative
nitrogen uptake since it is either energetically expensive, or not pos-
sible to derive carbon through photosynthesis. Recent work (Hülsen
et al., 2014) has shown assimilation ratios (COD:N:P) of 100:9:2, and
this is sufficient to remove nitrogen to <10 mg N L 1 for moderate-
high COD:N ratios (including the base case here) with just light
200 kW h ML 1 light input, and no additional COD. High-nitrogen
wastewaters will require external carbon, though it this is also the
case for conventional BNR. Where there is insufficient carbon, it
must either be supplied externally (which is very expensive), or
derived internally. One option is enhancing release of organics
(without methanogenesis) by multi-stage anaerobic digestion dur-
ing the release stage (Fig. 2). Nutrients can be removed from this
stream, and high-carbon digestate recycled to the partition stage.
2.2.2. Release
Anaerobic processes are widely used to release organics from
sewage sludges to methane while simultaneously solubilising
nitrogen and phosphorous (Batstone and Jensen, 2011). Having a
short solids retention time in the partition stage is essential to
avoid accumulation of inerts, and at <2 d SRT in partition, degrada-
bility of activated sludge in subsequent stages can be over 80%
(Gossett and Belser, 1982; Jimenez et al., 2005). As the prevalence
of activated sludge streams only has emerged, there has been a
substantial increase in focus on the use of pretreatment both to
intensify the process through enhanced kinetics as well as increase
net methane availability and biodegradability (Carrère et al., 2010).
Increasing temperature in a pretreatment stage is a favoured
method, since while generally less effective per unit energy input
than alternatives such as mechanical treatment (Carrère et al.,
2010), they use lower value heat energy, which is a byproduct of
electricity cogeneration and can be optionally recovered from the
hot digestate. In addition, thermal pretreatment (or thermophilic
main treatment) in semi-batch mode allows digested sludge to
meet modern complete pathogen destruction standards (Paul
et al., 2012). Thermophilic pretreatment (or temperature phased
anaerobic digestion) allows a 5 fold increase in hydrolysis coeffi-
cient (Ge et al., 2011). Sustained thermophilic methanogenesis on
activated sludge at hydraulic retention times down to 3 d may
offer the possibility to reduce the major electrical energy require-
ment of anaerobic digestion, which is mixing at approximately
8 W kL 1 (0.19 kW h kL 1 d 1) (Foley et al., 2010).
Multi-stage thermophilic anaerobic digestion allows the option
to recycle soluble organics from the first stage back to the partition-
ing stage, to assist with assimilation. Operating the first stage at a
Supernatant to recovery
Solids to reuse
Second stage stabilisation Centrifuge
6 D.J. Batstone et al. / Chemosphere 140 (2015) 2–11

lower pH due to a short hydraulic retention time also offers the pos- to the solid phase through digestion, and there will always be a
sibility to avoid anaerobic phosphorous precipitation due to in- need for a (small) residual solids stream, if only to avoid discharg-
digester calcium and magnesium binding, which is a substantial ing metals and persistent organic pollutants in the liquid stream.
technical hurdle to post-phosphorous recovery through struvite
precipitation (Mehta and Batstone, 2013b). Excessive ammonia
3. Analysis: energy and nutrient recovery using conventional
release in the first stage must be avoided, but lower nitrogen com-
and emerging platforms
pounds such as simple carbohydrates and lipids would normally be
released first in any case (Batstone and Jensen, 2011).
3.1. Comparison of options
Selection of dewatering equipment is critical, as subsequent
recovery requires that dilution be avoided (since recovery tech-
There has been a substantial amount of work done on charac-
niques require precipitation or concentration). This means that cen-
terising energy consumption in conventional activated sludge sys-
trifuges or vacuum filters must be used, since belt filter presses
tems from a life-cycle assessment point of view (Foley et al., 2010),
dilute the reject stream by 2 to 5 with wash water (Albertson
with a generalised benchmark of 0.5 kW h kL 1 in moderate sized
et al., 1987). Unfortunately, centrifuges consume approximately
plants that include nutrient removal, and energy recovery through
on order of magnitude higher energy than belt filter presses (0.03
anaerobic digestion. However, this has generally not considered
vs 0.2 kW h kg DS 1) (Albertson et al., 1987). Low wash belt filter
flows of chemical energy within the sludge. Once this is consid-
presses, or those that separate reject and wash water may enable
ered, external electrical inputs are only on the order of 15% total
recovery of high-strength filtrate at low energy consumption.
I/O energy flows, and total electricity consumption some 25% of
total energy flows. A total of 30% of the chemical energy is dis-
2.2.3. Recover
charged as effluent organics and sludge, and the remaining 60%
The stream entering recovery is likely to have approximately
as waste heat, mainly due to aerobic biological processes. Current
500–1000 mg L 1 solids, 200–500 mg P L 1, and 1000–
processes therefore use 0.5 kW h kL 1 of electricity to process or
3000 mg N L 1. For almost any applicable recovery technique, these
convert 3 kW h (11 MJ) chemical energy to heat or waste organics.
solids should be minimised, either through optimal operation of the
As treatment moves to a net energy exporter, it is important to
centrifuge, or post-filtration or sedimentation. Phosphorous recov-
consider use and flows of chemical energy as well as electrical and
ery through struvite (MgNH4PO4
6H2O) precipitation is the most
heat energy. This analysis uses internal energetic analysis as a basis
widely proposed method for phosphorous recovery, and has been
for comparison of conventional BNR with anaerobic primary and
widely reviewed (Le Corre et al., 2009; Yuan et al., 2012). Struvite
activated sludge digestion (Foley et al., 2010) LEM (McCarty
precipitation can be promoted by stripping carbon-dioxide and
et al., 2011), without primary sedimentation, and partition–
hence raising the pH, followed by addition of magnesium chloride,
release–recover (PRR). Design is based on a nominal 50 000 EP
and sodium hydroxide, magnesium hydroxide, or magnesium
(10 ML average dry weather flow) design, as for (Foley et al.,
oxide. The latter options are far lower cost, and do not add salinity,
2010), modified based on median information in Tchobanoglous
but it is more difficult to blend and add the magnesium oxide or
et al. (2003), with influent COD of 500 mg COD L 1, influent total
hydroxide (Le Corre et al., 2009). The technology has been success-
Kjeldahl nitrogen (TKN) of 50 mg N L 1, and influent total phos-
fully implemented at full scale on domestic wastewater (Ueno and
phorous (TP) of 10 mg P L 1. All processes aim at effluent limits
Fujii, 2001; Battistoni et al., 2006; Moerman et al., 2009). Retention
of 5 mg N L 1, 1 mg P L 1 and 20 mg COD L 1.
times can be very short, and concentrations can be reduced to
Assumptions and design parameters are summarised in this
<10 mg P L 1 in <60 min (Mehta and Batstone, 2013a). Struvite is
text, and are provided with full calculations in a Microsoft Excel
an effective fertilizer, and acts as a direct replacement for mono-
spreadsheet in Supporting materials, or available from the corre-
ammonium phosphate (Massey et al., 2009), with particular value
sponding author on request. The models have also been imple-
to applications where a high magnesium fertilizer is required, such
mented in Matlab R2010a in parallel for Monte-Carlo simulations
as dairy grass and grain. Transfer of pathogens via struvite is a con-
and are provided in Supporting materials. Standard energetic costs
cern (Decrey et al., 2011). Struvite is a moderate pathogen barrier
of pumping and aeration, as well as electrical yield on biogas are
due to high pH and low moisture, but the use of thermophilic anaer-
based on the WEF Energy Conservation Manual of Practice No. 32
obic digestion prior to struvite recovery is required to enable effec-
(WEF, 2010). Assumptions around energy consumption and effi-
tive pathogen control.
ciency of new and emerging processes have been referenced
Nitrogen recovery is much more complicated since there is no
below. Where possible, common elements are used (e.g., uniformly
ready precipitate. Struvite only contains 6% nitrogen by weight,
centrifuge based dewatering). Essential energy sinks and sources
and 12% phosphate, and given the concentrations above, only 10%
are given in Table 1. Energy consumption only includes on-site
of the available nitrogen will be removed through struvite precipita-
energy and chemical consumption (e.g., cost of sludge transport
tion. Alternative options include gas stripping (Liao et al., 1995),
are not included), and hence the only penalty for excess sludge
pervaporation through gas permeable membranes (EL-Bourawi
production is increased dewatering costs.
et al., 2007) and membrane assisted absorption (Vanotti and
Szogi, 2010), adsorption on ion exchange media such as zeolite
(Hedström, 2001), and electrodialysis (Mondor et al., 2008). 3.1.1. Conventional BNR
Stripping and membrane assisted processes require high pH, high The BNR base case is Case 6 from (Foley et al., 2010) (case 6Bii
temperature, and normally ammonia concentrations in excess of with chemical phosphorous trimming) which is a MLE Nitrogen
2000 mg N L 1. Adsorption and electrodialysis are both emerging removal reactor with 4.0 internal recycle and 0.7 return activated
technologies, but are less sensitive to feed concentration. It is likely sludge (RAS) recycle. MLSS is 3500 mg L 1 and sludge age is 15 d.
that the eventual process will be pre-concentration through elec- Primary treatment includes primary sedimentation and thickening,
trodialysis or adsorption to 10 000–50 000 mg N L 1 (w/w) range, and secondary sedimentation is by clarification with secondary
followed by stripping (Ippersiel et al., 2012) or pervaporation sludge thickening. Anaerobic sludge digestion is by mesophilic
(including use of cathode) for final recovery as ammonia sulphate, anaerobic digestion with a 22 d HRT. Key energy sinks are aeration
ammonia chloride, or subsequent distillation as pure ammonia. (49% of electricity consumption), internal recycles (29% of electric-
Full nutrient recovery is not required, since low levels can be ity consumption), and dewatering (9% of electricity consumption).
recycled back to the main process. Metals will generally partition Electricity generation from methane provides 21% of plant wide
 D.J. Batstone et al. / Chemosphere 140 (2015) 2–11 7

Table 1 recover energy, and to avoid excessive greenhouse gas emissions.

Key energy sources and sinks. Nitrogen subsequent to the AnMBR is removed in a combined pro-
Energy costs Value Unit Source cess mainline anaerobic nitrogen removal process (i.e., combined
Pumping – low viscosity a
20 kW h ML 1
1 nitritation and anoxic ammonia removal), with an estimated load-
Pumping – high viscosityb 50 kW h ML 1 1 ing rate of 0.3 kg N kL 1 d 1, and a combined energy consumption
Mixing anoxic/low solids 96 kW h ML 1 d 1 1 of 1.2 kW h kg N 1 (Wett, 2006; Wett et al., 2013). It is noted that
Mixing digester/high solids 150 kW h ML 1 d 1 1 down to 0.8 kW h kg N 1 can be achieved in optimal conditions for
Aeration energy 1 kW h kg O2 1 2
metabolised
sidestream nitrogen removal (Wett, 2006), but a 50% additional
Thickening – gravity belt 0.03 kW h kg DS 1 3 consumption is assumed based on loss of efficiency in mainline
Thickening – centrifuge 0.3 kW h kg DS 1 3 applications. The deammonification process is also assumed to
AnMBR Energy costs 250 kW h ML 1 treated 4 remove any residual degradable soluble COD through denitrifica-
Raw wastewater calorific 13 MJ kg COD 1 5
tion. Phosphorous removal post-anaerobic deammonification
value
Methane calorific value 55.5 MJ kg 1 5 occurs by ion adsorption in a packed column (Sengupta and
Methane calorific value 13.875 MJ kg COD 1
5 Pandit, 2011), with a power input of 100 kW h ML 1 for an average
empty bed contact time (EBCT) of 10 min to attain 1000 bed vol-
References. 1. WEF (2010); 2. Foley et al. (2010); 3. Albertson et al. (1987); 4. Judd
(2011); 5. Lide (2001).
umes (BV) prior to breakthrough (reference used 3 min for 300
a
Used for low solids streams, and to calculate energy losses related to low BV in the laboratory) (Sengupta and Pandit, 2011). This includes
pressure units. energy consumption estimates for regeneration but does not
b
Used for high-solids streams. include embedded energy in regenerants. Final sand or micro-fil-
tration has not been included at this point, but could be for an
additional 50 kW h ML 1. This process would result in recovery
electricity requirements. While Foley simulated the system using of 100 kg P d 1, providing additional income of $400 d 1. The
the more complex mechanistic Biowin Model (Barker and Dold, energy flow diagram for this process is shown in Fig. 4, indicating
1997), results in terms of energy consumption are highly compara- that approximately 5% of the chemical energy in the influent is
ble. Sankey energy flow results for this scenario are shown in Fig. 3 captured as electricity export. This is mainly due to the high energy
and demonstrate that the majority of inherent chemical energy is consumption caused by the AnMBR process. Utilising UASB would
lost as heat, with additional energy needed for secondary treat- almost completely remove this energy requirement, at the cost of
ment. Only a fraction (7%) is recovered for internal use, enabled effective downstream nitrogen removal and an effective pathogen
by solids handling. barrier.
Low energy mainline (LEM) consists of an anaerobic membrane
bioreactor (AnMBR) for the main wastewater stream, with a degra- 3.1.2. Partition–release–recover (PRR)
dation assumption of 75%, based on the undegradable particulate Partition–release–recover includes a photo-anaerobic mem-
fraction (fup) of 0.2 g COD g COD 1, as used by Foley (Foley et al., brane bioreactor in the accumulation stage, gravity belt thickening
2010), and solids retention times (SRT) of >20 d (Liao et al., of accumulated material, subsequent thermophilic anaerobic diges-
2006). Sludge is sent directly from the AnMBR to centrifuge at an tion, struvite precipitation for phosphorous recovery, and electrodi-
assumed 1% solids. Nitrogen and phosphorous is removed to alytic nitrogen and possibly potassium recovery. Energy input for
sludge in proportion to the undegradable particulate fraction. the photo-anaerobic membrane bioreactor is as for the anaerobic
Energetic costs of dissolved methane recovery have not been membrane bioreactor, except that additional energy is provided
included, as this is readily done by vacuum or membrane extrac- an infrared light (200 kW h ML 1) purely for phototrophic selection
tion at an energetic cost of <0.002 kW h kL 1 (Cookney et al., rather than to provide COD. The case considered here is anoxygenic
2012), though an essential part of the combined process, both to phototrophic growth by purple phototrophic bacteria, which does

Fig. 3. Sankey (proportional energy flows) for conventional BNR wastewater treatment with an influent of 500 mg COD L 1 achieving effluent TN of 5 mg N L 1. Chemical
energy flows are in yellow, electrical in blue, gas phase is green and heat in red. Values shown are energy fraction (%) of total influent chemical energy (processes scaled to
energy magnitude). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
8 D.J. Batstone et al. / Chemosphere 140 (2015) 2–11
Fig. 4. Sankey (proportional energy flows) for mainline low energy wastewater treatment (anaerobic MBR, mainline anaerobic ammonia removal, adsorptive phosphorous
removal) with an influent of 500 mg COD L 1 achieving effluent TN of 5 mg N L 1. Chemical energy flows are in yellow, electrical in blue, and heat in red. (For interpretation of
the references to colour in this figure legend, the reader is referred to the web version of this article.)
not include photosynthesis from CO2 and H2O. Therefore, light is
assumed to only enable uptake of native carbon with zero effective
catabolic carbon consumption.
Sludge from the phototrophic stage is sent to a thermophilic
anaerobic digester through gravity belt thickening that concen-
trates it to 5% solids. While thermophilic conditions do not result
in additional degradability from highly degradable substrates (Ge
et al., 2013), the very short solids retention times in the primary
stage, and lack of aerobic conditions mean that some form of path-
ogen control is needed. In addition, thermophilic conditions allow
intensification of the anaerobic process for a retention time of
down to 3 d while maintaining thermophilic activity (Ho et al.,
2013). Operation under multistage thermophilic configuration
would enable for additional pathogen destruction. A retention time
of 7 d has been used in this case. A VS destruction of 85% has been
used in accordance with the results of (Ge et al., 2013). Waste
sludge is fed to a high-g centrifuge (18% solids). Phosphorous is
recovered through a conventional stripping (2 h HRT) and crystal-
lisation (2 h HRT) (Le Corre et al., 2009; Mehta and Batstone,
2013a). A magnesium dose of 1.35 kg MgO kg P 1 is required to
achieve an effluent Mg:N:(P 1) ratio of 1000 (mM basis), which
provides rapid crystallisation kinetics and a favourable equilibrium
at a pH of 8.0 (Mehta and Batstone, 2013a).
Electrodialytic ammonia recovery is used to concentrate ammo-
nia from the 3000–5000 mg N L 1 from phosphorous recovery efflu-
ent (at pH 8–9 this is beneficial for electrodialytic and liquid–gas
separation). Direct in-reactor ammonia recovery through liquid–
gas stripping (Maurer et al., 2003) or electrochemically assisted
cathodic recovery (Desloover et al., 2012) are also options. Both of
these methods though are inherently energetically expensive, with
cathodic ammonia recovery achievable at 13 kW h kg N 1,
(Desloover et al., 2012), and gas stripping requiring 8 kW h kg N 1,
(Maurer et al., 2003). Ion exchange to both artificial media and min-
eral zeolites is also an option for nitrogen recovery (Deng et al.,
2014), even at mainline concentrations, but there are challenges in
economic application, including regeneration, and effective recov-
ery without pH adjustment. For electrodialysis, Ippersiel et al.,
(2012) estimated combined electrodialytic-stripping ammonia
recovery costs at 18 kW h kg N 1 at a process voltage of 17.5 V,
but this was with high cell voltage and two concentrate cells. Opti-
mising to a lower process voltage of 10 V, using the same current
efficiency of 70%, and implementing 6 concentration cells,
4.6 kW h kg N 1 is achievable. This would make the process viable
as a reject nitrogen recovery method at a cost competitive with com-
mercial nitrogen production. Residual nitrogen (500 mg N L 1) is
returned to the main reactor. This process recovers 90 kg P d 1 as
struvite, and an additional 300 kg N d 1. However, it uses an addi-
tional 50 kW h ML 1 treated, which is largely related to energy con-
sumption of phototrophic growth and nitrogen recovery
(4.6 kW h kg N 1). This is comparable with nitrogen manufacturing.
The energy flow diagram for this process is shown in Fig. 5 indicating
that approximately energy is still required from the network. This is
mainly driven by the need to recover nitrogen from the process, with
energetics being similar to LEM when nitrogen is not recovered.
3.2. Sensitivity to influent strength
To test sensitivity to influent strength, the models were imple-
mented in Matlab, and influent strength varied according to
Tchobanoglous et al. (2003) considering maximum and minimum
values (95%ile) or COD (250–1000 mg L 1), N (20–90 mg N L 1, P
(4–15 mg P L 1), and mineral solids (20–70 mg L 1). Ranges were
assumed to be normally distributed and a correlation coefficient
of 0.7 applied in general (between COD and other variables) and
0.85 specifically for nitrogen and phosphorous. This is an assump-
tion based on experience across wastewater strengths but also
results from influent generators such as (Gernaey et al., 2011).
Results are shown as a scatter-plot in Fig. 6 against COD as pri-
mary strength variable based on 1000 simulations. Each data point
represents a simulation, with vertical variability due to variation in
other strength measures.
This indicates that LEM has a break-even at approximately
400 mg COD L 1, and PRR at approximately 650 mg COD L 1. All
technologies are roughly equivalent below 300 mg COD L 1. If the
feed is high-energy, or low nitrogen, partition–release–recover
can achieve an additional advantage. This determines the larger
vertical variation of PRR vs other options, due to its greater sensi-
tivity to COD:N ratio.
 D.J. Batstone et al. / Chemosphere 140 (2015) 2–11
Fig. 5. Sankey (proportional energy flows) for partition–release–recover solids accumulation with sidestream digestion, with an influent of 500 mg COD L 1 achieving
effluent TN of 5 mg N L 1. Chemical energy flows are in yellow, electrical in blue, and heat in red. (For interpretation of the references to colour in this figure legend, the reader
is referred to the web version of this article.)
Fig. 6. Sensitivity of energy recovery (as % of feed – top) or yield (per kL) bottom vs
feed concentration. Results from 1000 Monte-Carlo simulations per scenario shown
with variation in feed concentrations.
9
4. Discussion
A key characteristic of the emerging resource recovery field is a
broad range of options. This review has focused mainly on energy
and nutrient recovery, with alternative and potentially valuable
carbon products such as biopolymers, fibres, and alginate (van
Loosdrecht and Brdjanovic, 2014) excluded to limit scope. Utilisa-
tion of carbon in this form is generally compatible with both
LEM and PRR through either main-line accumulation, or the use
of granular aerobic sludge.
Other options may include use of low technology units such as
UASB in place of the AnMBR, which has been discussed earlier in
this paper, or a combination of the configurations discussed here,
utilising a high-rate A-stage aerobic partitioning unit, utilising
anaerobic digestion of the A-stage sludge (A-stage-PRR), and
main-line anammox removal to remove nitrogen (as for LEM). This
results in an energy profile very similar to LEM, but with the pos-
sibility of partial nitrogen recovery, or use of a digestate anammox
process to support to mainline anammox process (Wett et al.,
2013). Disadvantages are mainly the more complex process, and
losses of carbon in the A-stage to catabolism.
Where pathogen free, high nitrogen effluent can be used directly
as irrigation, LEM and A-stage-PRR have an advantage, since nitro-
gen does not need to be concentrated at approx. 5 kW h kg N 1, and
can be used directly without destructive removal.
Dissolved methane has been discussed earlier, and is not a bar-
rier to anaerobic wastewater treatment. Sulphur is also possibly an
issue, and the behaviour will be different in each configuration. For
LEM, it will mainly be released in the gas from the first stage, with
low residual sulphide concentrations (<10 mg S L 1). This does not
cause inhibition to subsequent nitrogen removal (Jin et al., 2013),
and sulphide will be likely oxidised. Purple phototrophic bacteria
will remove sulphur due to assimilation, and oxidise it as an elec-
10 D.J. Batstone et al. / Chemosphere 140 (2015) 2–11
tron source. It will therefore accumulate in the sludge and be
released during anaerobic digestion as sulphides. In all cases, sul-
phur problems are likely to be attenuated compared with conven-
tional BNR, due to an increase in COD:S fraction converted to
methane (hence diluting gas phase sulphides).
Potassium recovery from wastewater has not been substantially
considered in the literature but as outlined in the introduction, is
an emerging issue. Most biological agents, including heterotrophic
phosphorous accumulating organisms (PAOs) (Pastor et al., 2008)
and phototrophs (Naimark et al., 2009) have the ability to accumu-
late potassium, and it is effectively recovered with nitrogen in elec-
trodialysis (Mondor et al., 2008). PRR is a key process in enabling
future potassium recovery, but further research is needed to
enhance uptake of potassium, particularly to avoid competitive
ions such as calcium.
None of the recovery processes analysed here have been imple-
mented in full-scale, or an integrated pilot process to date. However,
individual elements have been implemented in various scale, and
most of the operational parameters and assumptions are based on
physical implementations. Therefore, while this analysis is hypo-
thetical and not experimentally verified, it is practically feasible,
and likely to represent a realistic vision for alternative wastewater
treatment platforms.
5. Conclusions
Of the two alternative options reviewed here, LEM will recover
only phosphorous, and destructively remove nitrogen, though at a
low energetic input. It is also closer to market, with either units in
the field (e.g., UASB, AnMBR), or active pilot work underway. How-
ever, it will not recover nitrogen, and has limited ability to recover
other elements such as potassium. PRR offers the possibility to
recover nitrogen, and potentially potassium. However, energy
recovery is very marginal at domestic wastewater strengths. Nitro-
gen recovery requires rapidly growing assimilative organisms, as
well as potentially high-energy units such as stripping and electro-
dialytic. This means it costs approximately 5 kW h-e kg N 1 to
recover nitrogen rather than through destructive (low energy)
nitrogen removal.
Given its energetic and cost advantages, emerging processes in
the short term are likely to focus on mainline anaerobic low energy
processes, even with reduced possibility for nitrogen and partition-
ing recovery. However, partition–release–recover offers the possi-
bility for recovery of additional nutrients such as potassium,
particularly where strength is higher, where there is a high COD:N
ratio, or if the energetic costs of nutrient recovery can be reduced.
Acknowledgements
This work was funded by the CRC for Water Sensitive Cities under
Project C2.1. Chirag Mehta is a GRDC research fellow, and funding
from the GRDC Project UQ00061 ‘‘Nutrient from Wastes II’’ contrib-
uted to the project. We also acknowledge inputs from a number of
discussions and workshops including collaboration enabled by the
Water Environment Research Foundation project (WERF Nutrient
Recovery in the Global Water Industry NTRY1R12). Input under this
collaboration was provided by Hazen and Sawyer (US), CH2M Hill
(USA and Australia), and Queensland Urban Utilities (Australia).
Appendix A. Supplementary material
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.chemo-
sphere.2014.10. 021.
References
AEMC, 2011. Possible Future Retail Electricity Price Movements: 1 July 2010 to 30 June
2013. Review of Energy market Frameworks in Light of Climate Change Policies:
2nd Interim Report. Australian Energy Market Commission, Sydney, Australia.
<http://www.aemc.gov.au/market-reviews/completed/future-possible-retail-
electricity-price-movements-1-july-2010-to-30-june-2013.html>.
Albertson, O.E., Burris, B.E., Reed, S.C., Semon, J.A., Smith, J.E., Wallace, A.T., 1987.
Design manual: dewatering municipal wastewater sludges. US EPA, Cincinnati,
Ohio. p. 212.
Appl, M., 2000. Ammonia. In: Ullmann’s Encyclopedia of Industrial Chemistry.
Wiley-VCH Verlag GmbH & Co. KGaA.
Barker, P.S., Dold, P.L., 1997. General model for biological nutrient removal
activated-sludge systems: model presentation. Water Environ. Res. 69, 969–
984.
Batstone, D.J., Jensen, P.D., 2011. 4.17 – Anaerobic processes. In: Peter, W. (Ed.),
Treatise on Water Science. Elsevier, Oxford, pp. 615–639.
Batstone, D.J., Virdis, B., 2014. The role of anaerobic digestion in the emerging
energy economy. Curr. Opin. Biotechnol. 27, 142–149.
Battistoni, P., Via Brecce, B., Paci, B., Fatone, F., Pavan, P., 2006. Phosphorus removal
from anaerobic supernatants: start-up and steady-state conditions of a
fluidized bed reactor full-scale plant. Ind. Eng. Chem. Res. 45, 663–669.
Bryant, D.A., Costas, A.M.G., Maresca, J.A., Chew, A.G.M., Klatt, C.G., Bateson, M.M.,
Tallon, L.J., Hostetler, J., Nelson, W.C., Heidelberg, J.F., Ward, D.M., 2007.
Candidatus Chloracidobacterium thermophilum: an aerobic phototrophic
acidobacterium. Science 317, 523–526.
Carrère, H., Dumas, C., Battimelli, A., Batstone, D.J., Delgenès, J.P., Steyer, J.P., Ferrer,
I., 2010. Pretreatment methods to improve sludge anaerobic degradability: a
review. J. Hazard. Mater. 183, 1–15.
Chernicharo, C.A.L., 2006. Post-treatment options for the anaerobic treatment of
domestic wastewater. Rev. Environ. Sci. Biotechnol. 5, 73–92.
Chitapornpan, S., Chiemchaisri, C., Chiemchaisri, W., Honda, R., Yamamoto, K., 2012.
Photosynthetic bacteria production from food processing wastewater in
sequencing batch and membrane photo-bioreactors. Water Sci. Technol. 65,
504–512.
Cookney, J., Cartmell, E., Jefferson, B., McAdam, E.J., 2012. Recovery of methane from
anaerobic process effluent using poly-di-methyl-siloxane membrane contactors.
Water Sci. Technol. 65, 604–610. http://dx.doi.org/610.2166/wst.2012.2897.
Cordell, D., Drangert, J.-O., White, S., 2009. The story of phosphorus: global food
security and food for thought. Global Environ. Change 19, 292–305.
Cordell, D., Rosemarin, A., Schröder, J.J., Smit, A.L., 2011. Towards global phosphorus
security: a systems framework for phosphorus recovery and reuse options.
Chemosphere 84, 747–758.
Decrey, L., Udert, K.M., Tilley, E., Pecson, B.M., Kohn, T., 2011. Fate of the pathogen
indicators phage UX174 and Ascaris suum eggs during the production of struvite
fertilizer from source-separated urine. Water Res. 45, 4960–4972.
Deng, Q., Dhar, B.R., Elbeshbishy, E., Lee, H.S., 2014. Ammonium nitrogen removal
from the permeates of anaerobic membrane bioreactors: economic
regeneration of exhausted zeolite. Environ. Technol. (United Kingdom) 35,
2008–2017.
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.
DiFrancesco, C.A., Kramer, D.A., Apodaca, L.E., 2010. Nitrogen (Fixed) – Ammonia
Statistics. U. S. Geological Survey (USGS) technical report, 2010.
DOE/EIA-0484, 2010. US Energy Information Administration International Energy
Outlook 2010 July 2010. US Energy Information Administration Office of
Integrated Analysis and Forecasting, Washington, DC 20585.
EL-Bourawi, M.S., Khayet, M., Maa, R., Ding, Z., Lia, Z., Zhang, X., 2007. Application of
vacuum membrane distillation for ammonia removal. J. Membr. Sci. 301, 200–
209.
FAO, 2008. Current World Fertilizer Trends and Outlook To 2011/12. Food and
agriculture organization of the United Nations, Rome.
Fertecon, 2013a. Ammonia Report: Weekly Review of the Ammonia Market.
Fertecon, 2013b. Phosphate Report: Weekly Review of the Phosphate Market.
Fertecon, 2013c. Potash Report: Weekly Review of the Potash Market.
Foley, J., de Haas, D., Hartley, K., Lant, P., 2010. Comprehensive life cycle inventories
of alternative wastewater treatment systems. Water Res. 44, 1654–1666.
Ge, H., Jensen, P.D., Batstone, D.J., 2011. Temperature phased anaerobic digestion
increases apparent hydrolysis rate for waste activated sludge. Water Res. 45,
1597–1606.
Ge, H., Batstone, D.J., Keller, J., 2013. Operating aerobic wastewater treatment at
very short sludge ages enables treatment and energy recovery through
anaerobic sludge digestion. Water Res. 47, 6546–6557.
Gernaey, K.V., Flores-Alsina, X., Rosen, C., Benedetti, L., Jeppsson, U., 2011. Dynamic
influent pollutant disturbance scenario generation using a phenomenological
modelling approach. Environ. Model. Soft. 26, 1255–1267.
Gossett, J.M., Belser, R.L., 1982. Anaerobic digestion of waste activated sludge. J.
Environ. Eng. ASCE 108, 1101–1120.
Hedström, A., 2001. Ion exchange of ammonium in zeolites: a literature review. J.
Environ. Eng. 127, 673–681.
Henkens, P.L.C.M., Van Keulen, H., 2001. Mineral policy in the Netherlands and
nitrate policy within the European Community. NJAS – Wageningen J. Life Sci.
49, 117–134.
Hiraishi, A., Yanase, A., Kitamura, H., 1991. Polyphosphate accumulation by
Rhodobacter sphaeroides under different environmental conditions with
 D.J. Batstone et al. / Chemosphere 140 (2015) 2–11
special emphasis on the effect of external phosphate concentrations. Bull. Jpn.
Soc. Microbial Ecol. 6, 25–32.
Ho, D.P., Jensen, P.D., Batstone, D.J., 2013. Methanosarcinaceae and acetate-
oxidizing pathways dominate in high-rate thermophilic anaerobic digestion
of waste-activated sludge. Appl. Environ. Microbiol. 79, 6491–6500.
Hülsen, T., Batstone, D.J., Keller, J., 2014. Purple non-sulfur bacteria for nutrient
recovery from domestic wastewater. Water Res. 50, 18–26.
Ippersiel, D., Mondor, M., Lamarche, F., Tremblay, F., Dubreuil, J., Masse, L., 2012.
Nitrogen potential recovery and concentration of ammonia from swine manure
using electrodialysis coupled with air stripping. J. Environ. Manage. 95, S165–
S169.
Jasinski, S.M., 2011. Potash, USGS Mineral Commodity Summary 2011. USGS.
Jasinski, S.M., 2012. Phosphate rock <http://minerals.usgs.gov/minerals/pubs/
commodity/phosphate_rock/mcs-2012-phosp.pdf. U.S. Geological Survey,
Mineral Commodity Summaries>.
Jimenez, J.A., La Motta, E.J., Parker, D.S., 2005. Kinetics of removal of particulate
chemical oxygen demand in the activated-sludge process. Water Environ. Res.
77, 437–446.
Jin, R.-C., Yang, G.-F., Zhang, Q.-Q., Ma, C., Yu, J.-J., Xing, B.-S., 2013. The effect of
sulfide inhibition on the ANAMMOX process. Water Res. 47, 1459–1469.
Johnson, K., Jiang, Y., Kleerebezem, R., Muyzer, G., van Loosdrecht, M.C.M., 2009.
Enrichment of a mixed bacterial culture with a high polyhydroxyalkanoate
storage capacity. Biomacromolecules 10, 670–676.
Judd, S., 2011. The MBR Book: Principles and Applications of Membrane Bioreactors
for Water and Wastewater Treatment. Butterworth-Heinemann, Burlington.
Khatipov, E., Miyake, M., Miyake, J., Asada, Y., 1998. Accumulation of poly-b-
hydroxybutyrate by Rhodobacter sphaeroides on various carbon and nitrogen
substrates. FEMS Microbiol. Lett. 162, 39–45.
Kim, J., Kim, K., Ye, H., Lee, E., Shin, C., McCarty, P.L., Bae, J., 2010. Anaerobic fluidized
bed membrane bioreactor for wastewater treatment. Environ. Sci. Technol. 45,
576–581.
Le Corre, K.S., Valsami-Jones, E., Hobbs, P., Parsons, S.A., 2009. Phosphorus recovery
from wastewater by struvite crystallization: a review. Critic. Rev. Environ. Sci.
Technol. 39, 433–477.
Liao, P.H., Chen, A., Lo, K.V., 1995. Removal of nitrogen from swine manure
wastewaters by ammonia stripping. Bioresour. Technol. 54, 17–20.
Liao, B.Q., Kraemer, J.T., Bagley, D.M., 2006. Anaerobic membrane bioreactors:
applications and research directions. Critic. Rev. Environ. Sci. Technol. 36, 489–530.
Lide, D., 2001. CRC Handbook of Chemistry and Physics, 82nd ed. CRC Press, Boca
Raton, FL.
Manning, D.A.C., 2010. Mineral sources of potassium for plant nutrition. Agronomy
Sustain. Develop. 2, 188–201.
Massey, M.S., Davis, J.G., Ippolito, J.A., Sheffield, R.E., 2009. Effectiveness of
recovered magnesium phosphates as fertilizers in neutral and slightly
alkaline soils. Agronomy J. 101, 323–329.
Maurer, M., Schwegler, P., Larsen, T., 2003. Nutrients in urine: energetic aspects of
removal and recovery. Water Sci. Technol. 48, 37–46.
McCarty, P.L., Bae, J., Kim, J., 2011. Domestic wastewater treatment as a net energy
producer – can this be achieved? Environ. Sci. Technol. 45, 7100–7106.
Mehta, C.M., Batstone, D.J., 2013a. Nucleation and growth kinetics of struvite
crystallization. Water Res. 47, 2890–2900.
Mehta, C.M., Batstone, D.J., 2013b. Nutrient solubilization and its availability
following anaerobic digestion. Water Sci. Technol. 67, 756–763.
Moerman, W., Carballa, M., Vandekerckhove, A., Derycke, D., Verstraete, W., 2009.
Phosphate removal in agro-industry: pilot- and full-scale operational
considerations of struvite crystallization. Water Res. 43, 1887–1892.
Mondor, M., Masse, L., Ippersiel, D., Lamarche, F., Massé, D.I., 2008. Use of
electrodialysis and reverse osmosis for the recovery and concentration of
ammonia from swine manure. Bioresour. Technol. 99, 7363–7368.
11
Naimark, E.B., Kompantseva, E.I., Komova, A.V., 2009. Interaction between
anoxygenic phototrophic bacteria of the genus Rhodovulum and volcanic ash.
Microbiology 78, 747–756.
Pastor, L., Marti, N., Bouzas, A., Seco, A., 2008. Sewage sludge management for
phosphorus recovery as struvite in EBPR wastewater treatment plants.
Bioresour. Technol. 99, 4817–4824.
Paul, E., Carrère, H., Batstone, D.J., 2012. Thermal Methods to Enhance Biological
Treatment Processes. Biological Sludge Minimization and Biomaterials/
Bioenergy Recovery Technologies. John Wiley & Sons Inc. pp. 373–404.
Pearce, G.K., 2008. UF/MF pre-treatment to RO in seawater and wastewater reuse
applications: a comparison of energy costs. Desalination 222, 66–73.
Peverill, K.I., Reuter, D.J., Sparrow, L.A., 1999. Soil Analysis: An Interpretation
Manual. CSIRO Publishing, Collingwood, Vic.
Pritchard, D.L., Penney, N., McLaughlin, M.J., Rigby, H., Schwarz, K., 2010. Land
application of sewage sludge (biosolids) in Australia: risks to the environment
and food crops. Water Sci. Technol. 62, 48–57.
Sengupta, S., Pandit, A., 2011. Selective removal of phosphorus from wastewater
combined with its recovery as a solid-phase fertilizer. Water Res. 45, 3318–
3330.
Siefert, E., Irgens, R.L., Pfennig, N., 1978. Phototrophic purple and green bacteria in a
sewage treatment plant. Appl. Environ. Microbiol. 35, 38–44.
Smil, V., 2001. Enriching the Earth: Fritz Haber, Carl Bosch, and the Transformation
of World Food Production. MIT Press.
Tchobanoglous, G., Burton, F., Stensel, H., 2003. Metcalf and Eddy Inc., Wastewater
Engineering, Treatment and Reuse. McGraw-Hill, New York, NY (US).
Tucker, R.W., Mehta, C., McGahan, E., Batstone, D.J., 2010. Fertiliser From Waste:
Phase 1 GRDC Project UQ00046: Output 1. Grains Research and Development
Corporation, Canberra.
Ueno, Y., Fujii, M., 2001. Three years experience of operating and selling recovered
struvite from full-scale plant. Environ. Technol. 22, 1373–1381.
van Loosdrecht, M.C.M., Brdjanovic, D., 2014. Anticipating the next century of
wastewater treatment. Science 344, 1452–1453.
Vanotti, M.B., Szogi, A.A., 2010. Removal and Recovery of Ammonia From Liquid
Manure Using Gas-Permeable Membranes. American Society of Agricultural and
Biological Engineers Annual International Meeting, Pittsburgh, Pennsylvania,
USA, pp. 422–427.
Verstraete, W., Van de Caveye, P., Diamantis, V., 2009. Maximum use of resources
present in domestic ‘‘used water’’. Bioresour. Technol. 100, 5537–5545.
WEF, 2010. Energy Conservation in Water and Wastewater Treatment Facilities/
Prepared by The Energy Conservation in Water and Wastewater Treatment
Facilities Task Force of the Water Environment Federation. WEF Press, Water
Environment Federation, New York: McGraw Hill, Alexandria, VA.
Wett, B., 2006. Solved upscaling problems for implementing deammonification of
rejection water. Water Sci. Technol. 53, 121–128.
Wett, B., Omari, A., Podmirseg, S.M., Han, M., Akintayo, O., Gómez Brandón, M.,
Murthy, S., Bott, C., Hell, M., Takács, I., Nyhuis, G., O’Shaughnessy, M., 2013.
Going for mainstream deammonification from bench to full scale for maximized
resource efficiency. Water Sci. Technol. 68, 283–289.
Winkler, M.K.H., Kleerebezem, R., van Loosdrecht, M.C.M., 2012. Integration of
anammox into the aerobic granular sludge process for main stream wastewater
treatment at ambient temperatures. Water Res. 46, 136–144.
Yuan, Z., Pratt, S., Batstone, D.J., 2012. Phosphorus recovery from wastewater
through microbial processes. Curr. Opin. Biotechnol. 23, 878–883.
Zimmo, O.R., van der Steen, N.P., Gijzen, H.J., 2004. Nitrogen mass balance across
pilot-scale algae and duckweed-based wastewater stabilisation ponds. Water
Res. 38, 913–920.