Reviews in Environmental Science and Bio/Technology (2006) 5:73–92
Springer 2006

DOI 10.1007/s11157-005-5683-5

Post-treatment options for the anaerobic treatment of domestic wastewater

C. A. L. Chernicharo
Department of Sanitary and Environmental Engineering, Federal University of Minas Gerais, Av. do Contorno
842/701, 300110-060, Belo Horizonte, Brazil (e-mail: calemos@desa.ufmg.br; phone: +55-31-3238-1020;
fax: +55-31-3238-1879)

Key words: anaerobic digestion, domestic wastewater, emerging technologies, new developments, post-
treatment, UASB reactors, wastewater treatment

Abstract
This paper focuses on the post-treatment options for the anaerobic treatment of domestic wastewater.
Initially, the main limitations of anaerobic systems regarding carbon, nutrients and pathogen removal
are presented. In sequence, the advantages of combined anaerobic/aerobic treatment and the main
post-treatment options currently in use are discussed, including the presentation of flowsheets and a
comparison between various post-treatment systems. Lastly, the paper presents a review of emerging
options and possible improvements of current post-treatment alternatives.

1. Introduction feasibility studies include anaerobic reactors as

A deep discussion on the evolution and applica-
bility of the anaerobic technology for the treat-
ment of domestic sewage is presented elsewhere
(Lettinga et al., 1993; Seghezzo et al., 1998; von
Sperling and Chernicharo, 2005), where the sev-
eral favourable characteristics of the anaerobic
processes are highlighted, such as low cost, oper-
ational simplicity, no energy consumption and
low production of solids. These advantages, asso-
ciated with the favourable environmental condi-
tions in warm-climate regions, where high
temperatures prevail practically throughout the
year, have contributed to establish the anaerobic
systems, particularly the UASB reactors, in an
outstanding position.
Nowadays, it can be said that the high-rate
anaerobic reactors used for treatment of domestic
sewage are a consolidated technology in some
warm-climate countries, especially in Brazil, Colombia
and India, with several treatment systems operating in
full scale (population equivalents from a few thou-
sand up to around one million inhabitants). In
Brazil, practically all the wastewater treatment
one of the main options. Undoubtedly, a great
contribution to the consolidation and dissemina-
tion of the anaerobic technology for the treatment
of domestic sewage in Brazil came from the
National Research Programme on Basic Sanita-
tion – PROSAB, which has been carried out since
1997 (Chernicharo et al. 2001a). Therefore, for the
purpose of this paper, the advances on post treat-
ment of anaerobic effluents will be mainly fo-
cussed on the Brazilian experience, that is believed
to reflect also the reality of other warm climate
countries.
2. Main limitations of anaerobic systems
In spite of their great advantages, anaerobic reac-
tors hardly produce effluents that comply with
usual discharge standards established by environ-
mental agencies. Therefore, the effluents from
anaerobic reactors usually require a post-treat-
ment step as a means to adapt the treated effluent
to the requirements of the environmental legisla-
tion and protect the receiving water bodies.
74
The main role of the post-treatment is to
complete the removal of organic matter, as well
as to remove constituents little affected by the
anaerobic treatment, such as nutrients (N and P)
and pathogenic organisms (viruses, bacteria, pro-
tozoans and helminths).
2.1. Limitations regarding organic matter
Limitations imposed by environmental agencies
for BOD are usually expressed in terms of efflu-
ent discharge standards and minimum removal
efficiencies. These constraints are probably the
cause that has mostly limited the use of anaero-
bic systems (without post-treatment) for sewage
treatment (see typical values in Table 1).
In view of the limitations imposed by the envi-
ronmental legislation for the effluent BOD concen-
tration, or also when the receiving body has
limited capacity for assimilating the effluent from
the treatment plant (which is frequently the case),
it is usually necessary to use aerobic treatment to
supplement the anaerobic stage. However, there
are situations in which the combination of differ-
ent anaerobic processes can meet less restrictive
requirements regarding efficiency and concentra-
tion of the final effluent (e.g. 80% and
60 mgBOD/l, respectively). This is the case of sys-
tems consisting of a septic tank followed by an
anaerobic filter (usually feasible for small popula-
tions, generally fewer than 1000 inhabitants) or of
a UASB reactor followed by an anaerobic filter.
Obviously, the application of these combined
anaerobic systems is conditioned to an appropri-
ate dilution capacity of the receiving body.
Table 1. Usual effluent BOD and removal efficiencies in anaerobic
systems treating domestic sewage
Anaerobic system Effluent BOD removal
BODa (mg/l) efficiencya (%)
Anaerobic pond 70–160 40–70
UASB reactor 60–120 55–75
Septic tank 80–150 35–60
Imhoff tank 80–150 35–60
Septic tank followed 40–60 75–85
by anaerobic filter
a
Ranges of effluent concentration and typical removal efficien-
cies based on Brazilian experience. Lower efficiency limits are
usually associated with poorly operated systems (Chernicharo
et al. 2001a).
In this sense, in situations in which the receiv-
ing body presents a good dilution capacity, the
adoption of less restrictive discharge standards
could enable the construction of simpler and
more economical treatment plants in several
small cities by means of a more intensive use of
anaerobic reactors, particularly UASB reactors.
At a later stage, if it becomes necessary to pro-
duce a better quality effluent, a complementary
treatment unit can be built after some years. The
high costs of sophisticated treatment systems, de-
signed exclusively to meet BOD discharges stan-
dards, make their construction at a single stage
unfeasible for most cities located in developing
countries. On the other hand, the construction in
stages could be decisive and that systems consist-
ing of UASB reactor and a post-treatment unit
become the most feasible ones regarding techni-
cal and economical criteria.
2.2. Limitations regarding nitrogen
and phosphorus
The discharge of nutrients into surface water
bodies may cause increased algal biomass as a re-
sult of the eutrophication process (abnormal al-
gae growth due to the nutrients discharged). It is
known that 1.0 kg of phosphorus can result in
the reconstruction of 111 kg of biomass, which
corresponds to approximately 138 kg of chemical
oxygen demand in the receiving body (Randall
et al. 1992). Similarly, the discharge of 1.0 kg of
nitrogen can result in the reconstruction of
approximately 20 kg of chemical oxygen demand
under the form of dead algae (Randall et al.
1992). The problem can be even worsened due to
the decreased oxygen levels, by means of the
nitrification processes, when at least 4.0 kg of
dissolved oxygen are consumed for each kg of
ammonia discharged into the receiving body
(Grady & Lim 1980).
In cases in which nutrient removal is required
to meet the quality standards of the receiving
water body, the use of anaerobic processes pre-
ceding a complementary aerobic treatment for
biological nutrient removal should be analysed
very carefully, once anaerobic systems present
good biodegradable organic matter removal, but
practically no N and P removal efficiency. This
certainly causes a negative effect on biological
treatment systems aiming at good nutrient

removal, because the effluent from the anaerobic
reactor will have N/COD and P/COD ratios much
higher than the values desired for the good perfor-
mance of biological nutrient removal processes.
When the purpose of the treatment plant is
also a good nitrogen removal, the anaerobic
reactor should be used to treat initially only a
part of the influent raw sewage (possibly no
more than 50–70%), and the remaining part (30–
50%) should be directed to the complementary
biological treatment, aiming at nitrification and
denitrification, so that there is enough organic
matter for the denitrification step. In this case,
the great advantage of the use of the anaerobic
reactor is to receive and stabilise the sludge gen-
erated in the complementary treatment, eliminat-
ing the need for an anaerobic sludge digester.
On the other hand, when the purpose is the
biological phosphorus removal, the use of an
anaerobic reactor is not advisable for two main
reasons: (i) the effluent from the anaerobic reac-
tor presents a P/COD ratio higher than that of
the raw sewage, which harms the performance of
the biological phosphorus removal system; and
(ii) if the phosphorus-rich sludge generated in the
biological phosphorus removal treatment is direc-
ted to the anaerobic reactor for stabilisation, the
phosphorus incorporated to this sludge will be
released under anaerobic conditions and leave
with the effluent from the anaerobic reactor. This
fact makes an efficient phosphorus removal
unfeasible in a treatment plant with an anaerobic
reactor followed by complementary treatment
with biological phosphorus removal.
Phosphorus removal in treatment plants using
anaerobic reactor will only be effective if chemi-
cal products are used for P precipitation (iron or
aluminium salts). In this case, the anaerobic
reactor has the advantage of stabilising the
sludge generated in the complementary biological
aerobic treatment.
2.3. Limitations regarding microbiological
indicators
Regarding the microbiological indicators, low
faecal coliform removal efficiencies have been
reported in anaerobic reactors, usually amount-
ing to around only 1 log-unit (Chernicharo
et al. 2001c; von Sperling et al. 2002, 2004; von
Sperling & Mascarenhas 2004). Regarding other
75
types of microorganisms, such as viruses and
protozoans (mainly Giardia and Cryptosporidi-
um), there are few references covering their
reduction or elimination in anaerobic reactors.
The removal of helminth eggs in anaerobic
reactors, particularly in UASB reactors, has
been reported as amounting to 60–90%
(Chernicharo et al. 2001c; von Sperling et al.
2002, 2004), being therefore insufficient to pro-
duce effluents that may be used in irrigation.
However, it should be mentioned that these
limitations are not exclusive of anaerobic reac-
tors, but are a characteristic of most compact
wastewater treatment systems.
As the risk of human contamination by inges-
tion or contact with water containing pathogenic
organisms is high, many times it may be neces-
sary to disinfect the effluents. This fact becomes
even more serious due to the poor sanitary con-
ditions in developing countries. On the other
hand, the low investments in health and sanita-
tion make the population of these countries bear-
ers of several diseases that can be transmitted by
faeces and, consequently, by the sewage gener-
ated by this population.
Although the domestic sewage is an unques-
tionable source of contamination by pathogenic
organisms, it is worth of mention that the agents
used in the disinfection processes can also cause
harm to human health and the aquatic environ-
ment. It is then concluded that the decision to
either disinfect or not sewage should be taken
from a careful evaluation, based on the specific
characteristics of each situation. In other words,
there are no universal guidelines ruling sewage
disinfection requirements. The decision on the
need to disinfect the sewage of a certain locality
involves (USEPA 1986):
• an investigation on the uses of the water down-
stream the discharge point, and on the public
health risks associated with that water;
• an evaluation of the alternatives available for
pathogens removal from sewage;
• an evaluation of the environmental impacts the
control measures may cause.
Figure 1 presents a flowsheet that can aid the
decision making on the implementation need and
requirements of a sewage disinfection system, tak-
ing into account the public health risks involved
and the possibility to either reduce or eliminate
these risks. Once the risks involved are identified,
76
Figure 1. Flowsheet for local evaluation of the need for and requirements of sewage disinfection (adapted from USEPA 1986).
the environmental aspects start to determine the
applicability of the control alternative.
In cases where disinfection proves to be nec-
essary, a series of processes for the removal of
pathogenic organisms can be used as listed in
Table 2. Only short comments are made for each
process, since the removal of pathogenic organ-
isms, especially by artificial methods, is outside
the scope of this paper.
The processes listed in Table 2 are capable
of reaching a coliform removal of 99.99% or
more. Regarding pathogenic organisms, bacteria
removal efficiency is very high (equal to or
higher than coliform removal), and the other
pathogens (protozoa, virus, helminths) are
usually high, but variable, depending on the
removal mechanism and the resistance of each
species (USEPA 1986).
3. Advantages of the combined (anaerobic/aerobic)
systems
In comparison with a conventional wastewater
treatment plant consisting of primary sedimenta-
tion tank followed by aerobic biological treat-
ment (activated sludge, trickling filter, submerged
aerated biofilter or biodisc), with the primary
and secondary sludge passing through sludge
thickeners and anaerobic digesters prior to dewa-
tering, a treatment consisting of a UASB reactor
followed by aerobic biological treatment (with the
 77

Table 2. Main processes for the removal of pathogenic microorganisms in sewage treatment

Type Process Comment

Natural Maturation ponds • Shallow ponds, where the penetration of solar ultraviolet radiation and unfavourable
environmental conditions causes a high mortality of the pathogens.
• The maturation ponds do not need chemical products or energy, but require large areas.
• They are highly recommended systems (if there is area available), due to their great
simplicity and low costs.
Land treatment • The unfavourable environmental conditions in the soil favour the mortality of the
(infiltration in soil) pathogens.
• In slow-rate systems, there is the possibility of vegetable contamination, depending on the
type of application.
• Chemical products are not needed.
• Requires large areas.
Artificial Chlorination • Chlorine kills pathogenic microorganisms (although protozoan cysts and helminth eggs
are not much affected).
• High dosages are necessary, which may increase operational costs. The larger the previous
organic matter removal, the lower the chlorine dosage required.
• There is a concern regarding the generation of toxic by-products to human beings.
However, the great benefit to public health in the removal of pathogens must be taken into
consideration.
• The toxicity caused by the residual chlorine in the water bodies are also of concern. The
residual chlorine must have very low levels, frequently requiring dechlorination.
• There is large experience with chlorination in the area of water treatment in various
developing countries.
Ozonisation • Ozone is a very effective agent for the removal of pathogens.
• Ozonisation is usually expensive, although the costs are reducing, making this alternative a
competitive option in certain specific circumstances.
• There is less experience with ozonisation in most developing countries.
Ultraviolet radiation • Ultraviolet radiation, generated by special lamps, affects the reproduction of the patho-
genic agents.
• Toxic by-products are not generated.
• Ideally, the effluent must be well clarified for the radiation to penetrate well in the liquid
mass.
• This process has recently shown substantial development, which has made it more com-
petitive or more advantageous than chlorination in various applications.
Membranes • The passage of treated sewage through membranes of minute dimensions (e.g. ultrafil-
tration, nanofiltration) constitutes a physical barrier for the pathogenic microorganisms,
which have larger dimensions than the pores.
• The process is highly interesting and does not introduce chemical products into the liquid.
• The costs are still high, but they have been reducing significantly in recent years.

(Adapted from von Sperling and Chernicharo 2005).

secondary sludge directed to thickening and diges-
tion in the UASB reactor itself and then straight
to dewatering), can present the following advanta-
ges (Alem Sobrinho & Jordão 2001):
• the primary sedimentation tanks, sludge thick-
eners and anaerobic digesters, as well as all
their equipment, can be replaced with UASB
reactors, which do not require the use of
equipment. In this configuration, besides their
main sewage treatment function, the UASB
reactors also accomplish the aerobic sludge
thickening and digestion functions, requiring
no additional volume;
• power consumption for aeration in activated
sludge systems preceded by UASB reactors
will be substantially lower compared to
78
conventional activated sludge systems, and
especially extended aeration systems;
• thanks to the lower sludge production in
anaerobic systems and to their better dewater-
ability, sludge volumes to be disposed of from
anaerobic/aerobic systems will be much lower
than those from aerobic systems alone.
According to studies carried out by Pontes
(2003), a 30% VSS destruction can be reached
when secondary sludge produced in a trickling
filter is returned to a UASB reactor. When the
mass balance is performed, the total sludge
production in a combined UASB/Trickling Fil-
ter system can be 30–50% lower than in a con-
ventional trickling filter system.
• the construction cost of a treatment plant with
UASB reactor followed by aerobic biological
treatment usually amounts 50–80% of the cost
of a conventional treatment plant (20–50%
investment savings). In addition, due to the sim-
plicity, smaller sludge production and lower
power consumption of the combined anaerobic/
aerobic system, the operational costs also repre-
sent an even greater advantage. Savings on
operation and maintenance costs are usually in
the range of 40–50% in relation to a conven-
tional treatment plant (see Table 4 and von
Sperling & Chernicharo 2005).
4. Main post-treatment options currently in use
Taking into consideration the intrinsic limitations
associated with the anaerobic systems and the
need to develop technologies that are more
appropriate to the reality of developing coun-
tries, it is important to include a post-treatment
stage for the effluents generated in anaerobic
reactors. This stage has the purpose of polishing
not only the microbiological quality of the efflu-
ents, in view of the public health risks and limi-
tations imposed on the use of treated effluents in
agriculture, but also the quality in terms of or-
ganic matter and nutrients, in view of the envi-
ronmental damages caused by the discharges of
the remaining loads of these components into the
receiving bodies.
Some of the main possible combinations of
UASB reactors with post-treatment systems are
discussed in the following items. It can be ob-
served that in the UASB+activated sludge and
UASB+biofilm aerobic reactor systems, the aer-
obic biological excess sludge is simply returned
to the UASB reactor, where it undergoes diges-
tion and thickening with the anaerobic sludge,
dispensing separate digestion and thickening
units for the aerobic sludge. Hence, the overall
sludge production of the combined system is
wasted only from the UASB reactor. Since it is
already thickened and stabilised, and can be di-
rectly sent for dewatering and final disposal.
Sludge drying beds have been frequently used in
small-sized plants. Thus a large simplification in
the overall flowsheet is obtained, including the liquid
(sewage) and solid (sludge) phases.
4.1. Polishing pond
Facultative ponds are largely used for post-treat-
ment of effluents from anaerobic ponds. These
systems have the advantage of removing at a
higher efficiency the pathogenic organisms pres-
ent in the sewage, but their main disadvantages
are excessive land requirement and the high con-
centration of algae in the final effluent, which
leads to serious restrictions by some environmen-
tal agencies.
When an efficient anaerobic pre-treatment is
applied prior to the sewage discharge into a
pond, the concentrations of organic matter and
suspended solids are largely reduced, and conse-
quently it will be required only a complementary
removal of these two constituents, needing much
lower hydraulic detention times. In these condi-
tions, the limiting factor that determines the min-
imum detention time (and, therefore, the volume
and the area of a pond system) will usually be
the removal of pathogenic organisms, and not
the stabilisation of the organic matter. For this
reason, the nomenclature polishing pond has been
adopted to name those ponds intended for
the post-treatment of effluents from efficient
anaerobic systems, thus distinguishing them from
the stabilisation pond, which treats raw sewage
(Cavalcanti et al. 2001a, b).
The UASB reactor+polishing pond configu-
ration is a very interesting alternative from
the technical–economical–environmental point of
view, mainly when there are area limitations for
the construction of only stabilisation ponds. In
addition, the problems related to odours from
anaerobic ponds can be avoided in plants utilising

UASB reactor and polishing pond, since the
anaerobic reactor can be installed with odour
control (Cavalcanti et al. 2001a, b). This alterna-
tive is even more attractive when the effluent
from the pond can be used for agricultural pur-
poses, since the polishing ponds aim mainly at the
removal of pathogenic organisms. Because of its
advantages, the post-treatment of effluents from
anaerobic reactors through ponds has been com-
mon in developing countries.
Wastewater treatment plants using UASB
reactors followed by polishing ponds also have a
very simplified flowsheet (Figure 2). Besides the
preliminary treatment units (screen and grit
chamber), the flowsheet comprises the anaerobic
treatment unit, the polishing pond (either a single
baffled pond or ponds in series), and the dewa-
tering unit for the sludge produced in the UASB
reactor which is already thickened and stabilised.
Thus, dewatering units using drying beds are also
usual in smaller plants.
Long term studies conducted by von Sperling
and Mascarenhas (2004) have shown that a
domestic sewage treatment system comprised of a
UASB reactor followed by four very shallow
(0.40 m-depth) polishing ponds in series, oper-
ated with very low detention times (1.4–2.5 days
in each pond), was able to achieve excellent re-
sults in terms of BOD and E. coli removal, and
also good results in terms of ammonia removal.
The average concentrations observed in the final
effluent were 44 mgBOD l)1, 3.8102 MPN
100 ml)1 and 7.3 mgN–NH4 l)1. In relation to
helminth eggs, other studies have shown that
polishing pond systems are capable to produce
effluents with helminth eggs concentrations pre-
dominantly equal to zero, and satisfying the
Figure 2. Typical configuration of a treatment plant with UASB reactor and polishing ponds (von Sperling & Chernicharo 2005).
79
WHO guidelines for unrestricted and restricted
irrigation (von Sperling et al. 2002, 2004).
Since polishing ponds are designed with low
depths (0.40–1.00 m) and relatively short HDT
(usually between 9 and 12 days in a series of 3–4
ponds), care should be taken to the operation of
the UASB reactor in order to avoid excessive
wash out of solids.
4.2. Overland flow system
Sewage treatment by the overland flow method
is the one that presents the lowest relationship
with the type of soil. In this method, the vege-
tation, associated with the top soil layer, acts as
a filter, removing the nutrients and providing
conditions for the retention and transformation
of the organic matter contained in the sewage.
Besides that, it protects the soil against erosion
and creates a support layer on which the micro-
organisms settle. The main mechanisms through
which organic matter and solids are removed
are biological oxidation, sedimentation and fil-
tration (USEPA 1981, 1984; Metcalf & Eddy
1991). The main characteristic that differentiates
this method from the others is the fact that the
effluent flows downward on a slightly inclined
vegetated ramp and the remaining water (efflu-
ent), which is neither absorbed nor evaporated,
is collected downstream and directed for dis-
posal. For more permeable soils, the process is
similar to that of irrigation, but with the gener-
ation of effluent.
Therefore, the method consists in applying
the liquid in the highest part of the ramp. The
effluent then drains all over the slope by gravity,
where part of it is lost by evapotranspiration and
80
the remaining part is collected on the base of the
ramp. Percolation can be insignificant because
this system is initially conceived for low-perme-
ability soils. In spite of that, its use has been also
reported in soils with medium permeability and
impermeable underground (USEPA 1981). Sew-
age application is intermittent and the following
types of feeding can be adopted: (i) high pressure
sprinklers; (ii) low pressure sprinklers; (iii) distri-
bution piping or channels with spaced openings.
The typical configuration of a wastewater
treatment plant consisting of a UASB reactor
and post-treatment by overland flow has a very
simple flowsheet (Figure 3). Besides the pre-
liminary treatment units, the flowsheet comprises
the anaerobic treatment unit, the land treatment
system and the dewatering unit for the sludge
produced in the UASB reactor. The same consid-
erations made for the polishing ponds, regarding
the characteristics of the anaerobic sludge, that is
already thickened and stabilised, are also valid
here. Dewatering units using drying beds can be
used in small-sized plants.
Studies carried out by Coraucci Filho et al.
(2000) and Chernicharo et al. (2001c) with overland
flow systems following anaerobic filters and UASB
reactors, respectively, operated under application
rates varying from 0.10 to 0.50 m3 m)1 h)1, have
conducted to average concentrations in the final
effluent ranging from 98 to 119 mgCOD l)1,
48 to 62 mgBOD l)1 and 17 to 57 mgTSS l)1,
respectively). In relation to the microbiological
quality of the final effluent, an excellent removal
of helminth eggs in the UASB/overland flow sys-
tem was observed, with an average counting of
0.2 egg l)1 in the final effluent. Regarding faecal
coliforms, the removal was only satisfactory,
Figure 3. Typical configuration of a treatment plant with UASB reactor and overland flow system (von Sperling & Chernicharo
2005).
with the whole treatment system removing 2–3
log-units.
4.3. Activated sludge
The essence of the continuous flow activated sludge
process is the integration of the aeration tank
(aerobic biological reactor), secondary sedimenta-
tion tank and sludge recirculation line. These
three components are maintained in the alterna-
tive of activated sludge systems acting as post-
treatment of effluents from anaerobic reactors.
The intermittent flow activated sludge system
(sequencing batch reactors) can also be adopted
as post-treatment, requiring, in this case, only the
tanks that alternate in the functions of reaction
and sedimentation. Recent developments regard-
ing the application of such system for the post-
treatment of anaerobic effluents are discussed in
the last item of this paper.
When the activated sludge system acts as
post-treatment of anaerobic effluents, the anaero-
bic reactor is used instead of the primary sedi-
mentation tank (which is an integral part of the
conventional activated sludge system). The aero-
bic sludge is recirculated in the usual manner,
that is, from the bottom of the secondary sedi-
mentation tank to the entrance of the aerobic
reactor (aeration tank).
The excess aerobic sludge generated in the
activated sludge stage, not yet stabilised, is sent
to the UASB reactor, where it undergoes thick-
ening and digestion, together with the anaerobic
sludge. As the return flow of the excess aerobic
sludge is very low compared with the influent
flow, there are no operational disturbances in the
UASB reactor. The sludge treatment is largely

simplified: there is no need for separate thicken-
ers and digesters, and just the dewatering stage is
necessary. The mixed sludge removed from the
anaerobic reactor is digested, has solids concen-
trations similar to those from sludge thickeners
and presents good dewaterability. Figure 4 pre-
sents the flowsheet of this configuration.
4.4. Submerged aerated biofilter
A submerged aerated biofilter consists of a tank
filled with porous material, through which sew-
age and air flow permanently. In almost all the
existing processes, the porous medium is main-
tained totally submerged by the hydraulic flow.
The biofilters are characterised as three-phase
reactors consisting of:
• solid phase: consisting of the support medium
and colonies of microorganisms present in the
form of a biofilm;
• liquid phase: consisting of the liquid in perma-
nent flow through the porous medium;
Figure 4. Typical configuration of a treatment plant with UASB reactor and activated sludge system (von Sperling & Chernicharo
2005).
Figure 5. Typical configuration of a treatment plant with UASB reactor and submerged aerated biofilters (von Sperling &
Chernicharo 2005).
81
• gas phase: formed by the artificial aeration
and, in a reduced scale, by the gases deriving
from the biological activity.
Sewage treatment plants that use UASB reac-
tors followed by submerged aerated biofilters also
present a simple flowsheet (Figure 5). Besides the
preliminary treatment units, the flowsheet comprises
the sequential anaerobic and aerobic biological treat-
ment units (UASB reactor and submerged aerated
biofilter), as well as the aeration, sludge accumula-
tion, and dewatering units. Also in this configura-
tion, the excess aerobic sludge removed from the
biofilter is returned to the UASB reactor for thicken-
ing and anaerobic digestion. Therefore, with this
flowsheet, primary sedimentation tanks and separate
units for thickening and anaerobic digestion of the
excess aerobic sludge are avoided, differently from
the conventional treatment plants that use sub-
merged aerated biofilters.
Studies conducted by Gonçalves et al. (2000)
have shown that UASB/submerged aerated bio-
filter systems are capable of maintaining stable
82
operational conditions despite influent load vari-
ations and recycle of aerobic sludge discharged
from the BF. The average concentrations of
COD and TSS in final effluent are usually kept
below 90 mgCOD l)1 and 30 mgTSS l)1.
4.5. Trickling filter
A trickling filter consists basically of a tank filled
with a highly-permeable material, onto which
wastewater is loaded under the form of drops or
jets. Wastewater percolates towards the bottom
drains, allowing bacterial growth on the surface
of the packing material, under the form of a
fixed film (biofilm). Wastewater passes over the
biofilm, allowing a contact between the microor-
ganisms and the organic matter.
Although the trickling filters (TF) are waste-
water treatment systems with great potential and
numerous advantages, mainly because of their
simplicity and low operational cost, few units
have been implemented so far with the purpose
of performing the post-treatment of effluents
from anaerobic reactors (von Sperling and Cher-
nicharo 2005).
The main and innovative purpose of the re-
searches developed in the last years was to evaluate
the applicability and behaviour of the trickling
filters, when used for polishing of effluents from
anaerobic reactors, particularly UASB reactors.
This association (UASB reactor+TF) may contrib-
ute significantly to the reduction of the power and
operational costs of the treatment plant.
Wastewater treatment plants that use UASB
reactors followed by trickling filters present a
simple flowsheet (Figure 6). Basically, besides the
preliminary treatment units, the flowsheet
comprises the sequential anaerobic and aerobic
Figure 6. Typical configuration of a treatment plant with UASB reactor and trickling filter (von Sperling & Chernicharo 2005).
biological treatment units (UASB reactor, trick-
ling filter and secondary sedimentation tank), as
well as the dewatering unit. Therefore, with this
flowsheet, primary sedimentation tanks and sepa-
rate units for thickening and anaerobic digestion
of the excess aerobic sludge are avoided, differ-
ently from the conventional treatment plants that
use trickling filters.
Results of the researches developed by Aisse
et al. (2000), Chernicharo and Nascimento (2001),
and Pontes et al. (2003) indicated that that
trickling filters as post-treatment unit of UASB
reactors can be satisfactory operated at OLR
up to 1.5 kgBOD m)3 day)1 and HLR up to
20 m3 m)2 day)1. Under those operating condi-
tions, UASB+TF systems are usually capable of
producing a final effluent with average COD, BOD
and TSS concentrations around 120 mgCOD l)1,
40 mgBOD l)1 and 30 mgTSS l)1. If nitrification is
desired, much lower OLR should be used.
4.6. Anaerobic filter
Until recently, the anaerobic filters were limited
to small populations, usually treating effluents
from septic tanks. Nowadays, anaerobic filters
after UASB reactors are being used even in cities
with population larger than 50,000 inhabitants
(von Sperling & Chernicharo 2005). The comple-
mentary organic matter removal achieved in the
second anaerobic reactor (anaerobic filter) occurs
by:
• the retention of solids in the anaerobic filter,
reflecting on the removal of particulate matter.
In this case, physical removal mechanisms
prevail through the combined effects of coarse
filtration in the packing medium and sedimen-
tation along the column;

• the formation of biofilm on the packing med-
ium and removal of the remaining soluble or-
ganic matter. In this case, the formation of
biofilm and the removal of carbonaceous mat-
ter by biochemical means depend on the
amount of organic matter present in the efflu-
ent from the UASB reactor.
This association of anaerobic processes con-
tributes greatly to the reduction of power and
operational costs of the treatment plant. Waste-
water treatment plants using UASB reactors fol-
lowed by anaerobic filters represent a very simple
flowsheet (Figure 7). Besides the preliminary
treatment units (screen and grit chamber), the
flowsheet comprises basically the two sequential
anaerobic treatment units (UASB reactor and
anaerobic filter) and the dewatering unit. This is
because the sludge produced in the anaerobic
units is already thickened and stabilised.
Chernicharo and Machado (1998) and And-
rade Neto et al. (2000) have been evaluating
the use of pilot and demonstration scale anaer-
obic filters (AF) for the post-treatment of
anaerobic effluents from septic tanks and
UASB reactors. Different packing materials
have been investigated, such as blast furnace
slag (40–60 mm), perforated construction bricks,
granite stones (50–75 mm) and pieces of wasted
electrical tubing (cut in small pieces). The re-
sults of these researches indicate that anaerobic
filters with packing medium height between
0.80 and 3.0 m and operated at HDT ranging
from 5 to 10 h can further reduce particulate
and soluble organic matter of the incoming
anaerobic effluent, being able to maintain the
final concentrations of COD, BOD and TSS
usually below 120 mgCOD l)1, 60 mgBOD l)1
and 30 mgTSS l)1, respectively.
Figure 7. Typical configuration of a treatment plant with UASB reactor and anaerobic filter (von Sperling & Chernicharo 2005).
83
The combination of anaerobic filters (AF) and
anaerobic hybrid (AH) reactors have been re-
searched by Elmitwalli et al. (2002a, b), aiming at
to investigate the treatment of domestic sewage at
low temperature of 13C. The AF/AH system
operated at a HRT of 4+8 h, respectively, pro-
vided high removal efficiencies for all COD frac-
tions, reaching a total COD removal efficiency as
high as 71%, with 60% of the removed COD
being converted to methane.
4.7. Dissolved air flotation
The post-treatment of anaerobic effluents by dis-
solved-air flotation (DAF) was investigated using
batch flotation test equipment (Reali et al. 2001)
and also in a demo-scale (240 m3 day)1) continu-
ous flow system composed by an expanded bed
anaerobic reactor followed by a DAF unit treat-
ing domestic sewage (Penetra et al. 2002). In the
latter experiment, the use of 50 mg l)1 FeCl3 as
coagulant and flocculation under the gradient of
80 s)1 associated with a retention time of 20 min
conducted to the best results: 94.4% COD
removal (53 mgCOD l)1 residual), 87% phos-
phorus removal (0.80 mgP l)1 residual), 96.7%
TSS removal (9 mgTSS l)1 residual).
The use of DAF units for post-treatment of
anaerobic effluents results in a very compact
treatment system (Figure 8) that is capable of
producing very high quality effluents in terms of
COD, TSS and phosphorus. However, the re-
moval of ammonia nitrogen and faecal coliforms
is poor. In relation to the sludge produced in
DAF units, the amount tends to be higher than
the ones observed in biological post-treatment
systems, but it usually presents higher solids
content, favouring its final disposal in landfills.
84
Figure 8. Typical configuration of a treatment plant with UASB reactor and DAF (von Sperling & Chernicharo 2005).
4.8. Constructed wetlands
Constructed wetlands are purposely built waste-
water treatment processes, which consist of
ponds, basins or shallow canals (usually with a
depth of less than 1.0 m) that shelter aquatic
plants, and use biological, chemical and physical
mechanisms to treat the sewage. The constructed
wetlands usually have an impermeable layer of
clay or synthetic membrane, and structures to
control the flow direction, hydraulic detention
time and water level. Depending on the system,
they can contain an inert porous medium such as
stones, gravel or sand.
The subsurface flow wetlands seem to be
more appropriate to receive effluents from septic
tanks and anaerobic reactors because of its lower
potential for the generation of odours and the
appearance of mosquitoes and rats. For effluents
from anaerobic reactors the land requirements
are around 2.5–4.0 m2/hab (von Sperling &
Chernicharo 2005).
The association of anaerobic reactors and con-
structed wetlands contributes greatly to the reduc-
tion of power and operational costs of the
Figure 9. Typical configuration of a treatment plant with UASB reactor and constructed wetland (von Sperling & Chernicharo 2005).
treatment plant. Besides, the wastewater treatment
plant represents a very simple flowsheet (Figure 9).
Results of the 1-year research developed
by Sousa et al. (2001), with a UASB/constructed
wetland system, showed average COD removal
efficiencies in the range of 79–85%, suspended
solids in the range of 48–71% and faecal coli-
forms around 4 log-units. Phosphorus was also
efficiently removed (average of 90% for the low-
est hydraulic load) but nitrogen removal was
only partial (45–70% for ammonia and 47–70%
for TKN).
5. Comparison between various post-treatment
options
Tables 3–7 (adapted from von Sperling & Chernicharo
2005) present a comparative analysis between the
main systems applied to the post-treatment of
effluents from UASB reactors, as follows:
• Quantitative comparison (Table 3): average
effluent concentrations and typical removal effi-
ciencies of the main pollutants of interest in
domestic sewage
Table 3. Average effluent concentrations and typical removal efficiencies of the main pollutants of interest in domestic sewage

System Average quality of the effluent Average removal efficiency

BOD5 COD TSS Ammonia Total N Total P FC Helminth BOD5 COD TSS Ammonia Total N Total P FC
(mg/l) (mg/l) (mg/l) (mg/l) (mg/l) (mg/l) (FC/100 ml) eggs (eggs/l) (%) (%) (%) (%) (%) (%) (log units)

UASB reactor 70–100 180–270 60–100 >15 >20 >4 106–107 >1 60–75 55–70 65–80 <50 <60 <35 1–2
UASB+activated 20–50 60–150 20–40 5–15 >20 >4 106–107 >1 83–93 75–88 87–93 50–85 <60 <35 1–2
sludge
UASB+submerged 20–50 60–150 20–40 5–15 >20 >4 106–107 >1 83–93 75–88 87–93 50–85 <60 <35 1–2
aerated biofilter
UASB+high rate 20–60 70–180 20–40 >15 >20 >4 106–107 >1 80–93 73–88 87–93 <50 <60 <35 1–2
trickling filter
UASB+anaerobic 40–80 100–200 30–60 >15 >20 >4 106–107 >1 75–87 70–80 80–90 <50 <60 <35 1–2
filter
UASB+dissolved-air 20–50 60–100 10–30 > 20 > 30 1–2 106–107 >1 83–93 83–90 90–97 <30 <30 75–88 1–2
flotation
UASB+polishing 40–70 100–180 50–80 10–15 15–20 <4 102–104 <1 77–87 70–83 73–83 50–65 50–65 >50 3–5
ponds
UASB+overland 30–70 90–180 20–60 10–20 >15 >4 104–106 <1 77–90 70–85 80–93 35–65 <65 <35 2–3
flow

Notes:
1. Ranges of effluent concentration and typical removal efficiencies considering systems properly designed and operated.
2. Chemical precipitation of phosphorus with any of the technologies above: P<1 mg/l.
3. Disinfection: e.g. chlorination, ozonisation, UV radiation; Barrier: e.g. membranes (provided the disinfection/barrier process is compatible with the quality of the effluent from
the preceding treatment): CF<103 FC/100 ml; helminth eggs: variable.
(Adapted from von Sperling & Chernicharo 2005).
85
 86

Table 4. Typical characteristics of UASB reactor and various post-treatment systems, expressed as per capita values

System Land requirements Power for aeration Sludge volume Costs

(m2/inhab)
Installed power Consumed power Liquid sludge to be Dewatered sludge Construction Operation and
(W/inhab) (kWh/inhab.year) treated (l/ inhab.year) to be disposed of (US$/inhab) maintenance
(l/ inhab.year) (US$/inhab.year)

UASB reactor 0.03–0.10 0 0 70–220 10–35 12–20 1.0–1.5

UASB+activated sludge 0.08–0.2 1.8–3.5 14–20 180–400 15–60 30–45 2.5–5.0
UASB+submerged aerated 0.05–0.15 1.8–3.5 14–20 180–400 15–55 25–40 2.5–5.0
biofilter
UASB+high rate trickling filter 0.1–0.2 0 0 180–400 15–55 25–35 2.0–3.0
UASB+anaerobic filter 0.05–0.15 0 0 150–300 10–50 20–30 1.5–2.2
UASB+dissolved-air flotation 0.05–0.15 1.0–1.5 8–12 300–470 25–75 25–35 2.5–3.5
UASB+polishing ponds 1.5–2.5 0 0 150–250 10–35 15–30 1.8–3.0
UASB+overland flow 1.5–3.0 0 0 70–220 10–35 20–35 2.0–3.0

Notes:
1. Costs based on Brazilian experience (basis: year 2002–US$ 1.00=R$ 2.50).
2. Per capita costs are applicable inside the typical population ranges within which each treatment system is usually applied (usually, for a certain system, the lower the population,
the greater the per capita costs).
3. Additional disinfection: construction costs – increase US$ 2.0 to 4.0/inhab; operational and maintenance costs: increase US$ 0.2 to 0.6/inhab.year.
4. In compact aerated systems (e.g.: activated sludge, submerged aerated biofilters) or after treatment with a UASB reactor, aeration control allows a certain economy (not all the
installed power is consumed).
(Adapted from von Sperling & Chernicharo 2005).
 87

Table 5. Capacity of UASB reactors and various post-treatment systems in consistently achieving the indicated levels of effluent
quality in terms of BOD, COD and TSS

Table 6. Capacity of UASB reactors and various post-treatment systems in consistently achieving the indicated levels of effluent
quality in terms of Ammonia, total N and total P

Table 7. Capacity of UASB reactors and various post-treatment systems in consistently achieving the indicated levels of effluent
quality in terms of Faecal (thermotolerant) coliforms and Helminth Eggs

• Quantitative comparison (Table 4): typical char-
acteristics of the main sewage treatment sys-
tems, expressed in per-capita values
• Diagrammatic comparison (Tables 5 to 7):
capacity of the various sewage treatment sys-
tems in consistently reaching different quality
levels in terms of BOD, COD, TSS, ammonia,
total nitrogen, total phosphorous, faecal coli-
forms and helminth eggs
6. Emerging options and possible improvements
of current post-treatment alternatives
Aerobic post-treatment of anaerobic effluent
provides further reduction of residual organics
and nitrification of ammonia. A denitrification
step is necessary in order to remove nitrate and
nitrite produced from nitrification. The chal-
lenge of adopting a post-treatment system to
treat an anaerobic effluent is to find a proper,
reliable and efficient process that is simple in
construction, operation and maintenance; have
lower capital costs; have capacity to remain
stable under both hydraulic and organic shock
loads; and is energy efficient (Tai et al. 2004).
Although researches have contributed a lot for
the understanding, improvement and develop-
ment of post-treatment processes in the last
decade, the main contributions were related to
organic matter and pathogen removal. How-
ever, due to the substantial organic material re-
moval reached in the anaerobic step, biological
88
removal of nitrogen (by nitrification/denitrifica-
tion) and phosphorus (by luxury uptake)
becomes problematic due to the lack of biode-
gradable organic carbon. Hence, the main
drawbacks of current technologies are still
related to nutrient removal and that is the
reason why most recent studies have been
focusing on this subject as discussed herein.
6.1. Sequence batch reactors (SBR)
In studies developed by Callado and Foresti
(2001), almost complete removal of COD, nitro-
gen and phosphorus were accomplished in anaer-
obic–aerobic systems composed of SBRs in
series. Nitrification, denitrification and biological
phosphorus removal may occur in the second
SBR treating the effluent of the first anaerobic
SBR supplemented with acetate, operating under
aerobic–anoxic cycles. The results obtained in
bench-scale unit opens the possibility of using
very simple systems to promote the complete
treatment of domestic wastewater. The first unit
can be an anaerobic SBR or any other configura-
tion, since the effluent quality of anaerobic reac-
tors is not expected to change drastically from
one configuration to another. However, the need
of supplementary addition of an external carbon
source for denitrification and biological phos-
phate removal makes this alternative inconve-
nient from the sustainability point of view. On
the other hand, the external carbon source can
be alternatively produced from the controlled
digestion of sludge and domestic solid wastes.
These considerations imply in changes on the
conception of environmental control systems to
integrate solid and liquid wastes treatment.
Van Haandel and Guimarães (2000) evaluated
the sequencing batch reactor as an alternative for
aerobic post-treatment. Even though the required
retention time was very low (an anaerobic HDT
of 5 h and an aerobic HDT of 2.4 h) proved to
be sufficient to produce consistently a very high
effluent quality (BOD and TSS <20 mg l)1). The
produced activated sludge maintained fair to
good settling properties and no bulking was ob-
served. The authors showed that nitrification at
high sludge age did not cause problems during
the settling period: during settling the denitrifica-
tion rate was low (no extra cellular material), so
that not enough nitrogen was formed to cause
sludge flotation.
6.2. Hybrid systems
An alternative to the treatment in activated sludge
system is the utilization of high capacity hybrid
suspended biomass-biofilm systems. These systems
have been successfully employed not only to up-
grade low nitrifying capacity wastewater treatment
plants, but also as a new technology to develop
compact systems for simultaneous nitrogen and
organic matter removal. Further improvement in
these systems include the replacement of the final
settler by membrane filtration units, as in the con-
figuration proposed by Oyanedel et al. (2002) that
includes an anoxic chamber with suspended
biomass, followed by an aerobic circulating bed
reactor (CBR) which contains biofilm and sus-
pended biomass. The aerobic reactor is coupled to
a vessel containing a hollow fibre ultrafiltration
membrane module that allows the separation of
the permeate (effluent) from a retentate (sludge)
that is recycled to the anoxic chamber. With this
system it was possible to reach high COD and
nitrogen removals and no solids in the final effluent.
6.3. Rotating biological contact (RBC)
Tawfik et al. (2002) tested a three stage pilot-scale
RBC for removal of E. coli, COD and ammonia
from anaerobically pre-treated domestic sewage.
The three stage RBC system operated at a HDT
of 10.0 h and an OLR of 5.3 gCOD m)2 day)1
was capable of producing a final effluent containing
43 mgCOD l)1, 3.3 mgNH4-N l)1 and 2.0103
E. coli/100 ml. The authors pointed out that the
removal of E. coli in a RBC system comprises: (i)
sedimentation of coarse particles; (ii) adsorption
onto the biofilm; and (iii) predation by ciliated
protozoa.
Also working with a RBC system, but for
the post-treatment of an ammonium rich anaer-
obic effluent, Pynaert et al. (2002) reported that
the inoculation of the RBC with methanogenic
sludge favoured nitrogen removal via oxygen-
limited oxidation of ammonium with nitrite as
the electron acceptor. The authors state that

the experiment confirms the property of this
system to remove ammonium to nitrogen gas
without the use of heterotrophic carbon source.
6.4. Expanded granular sludge bed reactors
The feasibility of using EGSB reactor for the
post-treatment of very low concentration anaero-
bic effluents was evaluated by Kato et al. (2002). A
pilot-scale 157.5-L EGSB reactor treating the efflu-
ent from a full scale UASB reactor and operating
at 4-h HDT and upflow velocity of 3.75 m h)1
was capable of producing a final effluent with to-
tal and soluble COD concentration below 87 and
55 mg l)1, respectively, and TSS below 32 mg l)1.
The authors also reported stable operation during
all experimental period.
6.5. Fixed bed reactors and alternative support
materials
Daniel and Foresti (2004) carried out studies in a
fixed bed reactor filled of polyurethane foam, fed
with synthetic substrate simulating an anaerobic
effluent with a high concentration of ammonium.
The results obtained with the reactor operated in
a sequential batch mode and each cycle com-
posed by aerobic and anoxic periods indicated a
stable process of nitrogen removal. It was possi-
ble to establish partial nitrification to nitrite and
complete denitrification.
Machdar et al. (1997, 2000), Araki et al.
(1999) and Uemura et al. (2002) present the
development of the downflow hanging sponge –
DHS – reactor, tested for the aerobic post-treat-
ment of effluents from UASB reactors treating
municipal wastewater. In its fourth generation
(Tandukar et al. 2005), the DHS reactor was
constituted of slabs containing long sponge strips
measuring 2.52.550 cm, which were then
stacked one above another but in direction 90
to each other to make 20 rows. This was consid-
ered a module with 300 sponge units and 39%
occupancy of the sponge by volume. Four such
modules were put one above the other with a
certain gap in between for the construction of
the whole reactor, giving a height of 4 m. The
main improvements in the reactor were related to
the enhancement of air dissolution into the
wastewater and to avert the possible clogging of
the reactor. The whole system was operated at a
89
total hydraulic retention time of 8 h (UASB
reactor: 6 h; DHS: 2 h) being capable of remov-
ing 96% of unfiltered BOD and 3.45 log-units of
faecal coliforms. The authors also reported a
high nitrification degree during the start-up peri-
od, with NH4–N removal over 56%. As HDT
was decreased and organic and hydraulic loads
increased after the start-up period, the NH4–N
removal efficiency dropped to less than 30%.
The DHS reactor accommodates both nitrifiers
and denitrifiers giving way to simultaneous nitri-
fication and denitrification.
A novel radial anaerobic/aerobic immobilized
biomass reactor using biogas constituents as elec-
tron donor was investigated by Garbossa et al.
(2005), aiming at simultaneous carbon and nitro-
gen removal from municipal wastewater. The
research meant to confirm that methane and
sulphide present in the biogas could be used as
electron donors for denitrification, as previously
suggested by other authors (Thalasso et al. 1997;
Islas-Lima et al. 2004). The bench-scale reactor
was divided into five concentric chambers with
the second and fourth chambers filled with poly-
urethane foam matrices for biomass immobiliza-
tion. Promising results were obtained, with mean
COD and TKN removal efficiencies of 90% and
92% being observed. Average COD, N-TKN and
N–NO3 were 44, 3.2 and 1.9 mg l)1, respectively.
6.6. Jet loop reactor (JLR)
A novel two-stage anaerobic/aerobic integrated
system consisting of an UASB reactor and jet
loop reactor (JLR) was developed by Tai et al.
(2004) aiming at complementary removal of
organics and simultaneous removal of Total
Kjeldahl Nitrogen and Total Nitrogen. The JLR
is a sort of aeration tank that incorporates a re-
cycle line with a venture and a draft tube, thus
allowing the introduction of air drawn from the
atmosphere and mixing of the reactor contents.
In this system configuration the UASB reactor is
used to achieve denitrification and methanogene-
sis, by means of recycling the nitrified effluent
(taken from the secondary clarifier) jointly with
the influent wastewater to the UASB reactor.
The authors claims that the combined system
was capable of removing more than 85% of total
BOD and COD and more than 95% of soluble
BOD and COD. The JLR presented average
90
removals of 94.0 and 95.4% of TKN at HDTs of 8
and 5 h, respectively. The UASB-JLR achieve an
average of 78.1% TN removal with a 4:1 recycle
ratio at a combined system HDT of 58.8 h.
6.7. Membrane bioreactors (MBR)
Another interesting post-treatment option is the
use of micro and ultrafiltration membranes
associated with anaerobic reactors, aiming at to
increase the quality of the final effluent and to
maintain biomass inside the anaerobic reactor
with greater efficiency. In Membrane Anaerobic
Bioreactors (AnMBR), enhanced biomass reten-
tion can be accomplished by membrane-based
separation techniques (Jeison & van Lier 2005),
favouring the increase of the mean cell residence
time and improving the conditions for the degra-
dation of low degradable compounds. In last
years, important advances have been made in the
development of new types of membranes with
reduced costs, and research is being carried out
in order to find better reactor configurations and
operational procedures that decrease energy
consumption and fouling (Hernandez et al. 2002;
Beal & Monteggia 2004; Fitzke et al. 2004;
Jeison & van Lier 2005).
6.8. Advanced oxidation processes (AOP’s)
The use of advanced oxidation processes can also
become an interesting alternative for post-treat-
ment of anaerobic effluents. The AOP’s involve
the generation of hydroxyl radicals which have a
high oxidation potential and attacks organic
molecules by either abstracting a hydrogen atom
or by adding to double bonds, thus allowing its
mineralization to non-toxic forms such carbon
dioxide or water. Studies carried out by Sigge
et al. (2002) demonstrated the feasibility of this
process in further reducing the COD contents of
anaerobic effluents, when using ozone and ozone/
hydrogen peroxide in combination with a granu-
lar activated carbon contacting column. Colour
and COD reductions ranged from 66 to 90% and
from 27 to 55%, respectively.
6.9. Two stage flotation
Tessele et al. (2004) presented an important
alternative for the improvement of conventional
dissolved air flotation (DAF) process, proposing
a two stage flotation unit. In this configuration,
the first flotation stage is intended to remove sus-
pended solids by the flocculation–flotation pro-
cess. This flotation technique was originally
developed for oil removal and is based on the
formation of aerated flocs, in the presence of
high molecular weight polymer under high shear.
The second stage flotation removes phosphate by
precipitation and coagulation with Fe(FeCl3) and
also acts as a polishing step, separating the resid-
ual fine solids. In trial studies, the application of
5.0–7.5 mg l)1 cationic flocculant was capable to
separate off more than 99% of the suspended
solids, while phosphate ions were completely
recovered using carrier flotation with 5–25 mg l)1
of Fe(FeCl3) at pH varying from 6.3 to 7.0. The
staged flotation leads to high recoveries of water
and allows separating organic matter and phos-
phate bearing sludges. In contrast, the conven-
tional DAF produces significant volumes of
mixed organic and inorganic sludge which may
lead to complex post sludge treatment either to
reuse or to dispose of.
7. Conclusion
The fundamental and practical results obtained
so far have effectively contributed to consolidate
the anaerobic technology as the first stage treat-
ment for domestic and municipal sewage, and
also to offer a series of post-treatment alterna-
tives that take into account the social, economi-
cal and environmental aspects of most
developing countries. Recent developments and
further research on nutrient removal will soon
overcome the few drawbacks that still remain,
which are challenging a wider application of
combined anaerobic/aerobic, anaerobic/anaero-
bic and anaerobic/physico-chemical systems for
domestic sewage treatment.
References
Aisse MM, Nolasco MA, Andreoli FDN, Lobato MB, Savelli
CS, Jurgensen D & Alem Sobrinho P (2000) Pós-tratamento
de efluentes provenientes de reatores anaeróbios tipo UASB.
In: Proceedings VI Latin-American Workshop and Seminar
on Anaerobic Digestion. Recife, Brazil, p. 321–327. (in
Portuguese)

Alem Sobrinho P & Jordão EP (2001) Pós-tratamento de
efluentes de reatores anaeróbios – uma análise crı́tica. Cap. 9.
In: Chernicharo CAL (coordenador). Pós-tratamento de
efluentes de reatores anaeróbios. FINEP/PROSAB, Rio de
Janeiro, Brasil, 544 p. (in Portuguese)
Andrade Neto CO, Pereira MG & Melo HNS (2000) Materiais
alternativos para o enchimento de filtros anaeróbios: condu-
ı́te cortado e tijolo cerâmico vazado. Proceedings VI Latin-
American Workshop and Seminar on Anaerobic Digestion.
Recife, Brazil, p. 28–35. (in Portuguese)
Araki N, Ohashi A, Machdar I & Harada H (1999) Behaviors
of nitrifiers in a novel biofilm reactor employing hanging
sponge-cubes as attachment site. Water Sci. Technol. 39(7):
23–31
Beal L & Monteggia LO (2004) The effect of microfiltration
upon methanogenic activity. In: Proceedings of the 10th
World Congress on Anaerobic Digestion – Anaerobic
Digestion 2004. Montreal, Canada, Vol. 1, pp. 389–393
Callado NH & Foresti E (2001) Removal of organic carbon,
nitrogen and phosphorus in sequential batch reactors inte-
grating the anaerobic/aerobic processes. Water Sci. Technol.
44(4): 263–270
Cavalcanti PFF, van Haandel A & Lettinga G (2001a)
Polishing ponds for post-treatment of digested sewage Part
1: flow-through ponds. Water Sci. Technol. 44(4): 237–245
Cavalcanti PFF, van Haandel A, Kato MT, von Sperling M,
Luduvice ML & Monteggia LO (2001b). Pós-tratamento de
efluentes de reatores anaeróbios por lagoas de polimento.
Cap. 3. In: Chernicharo CAL (coordenador). Pós-tratamento
de efluentes de reatores anaeróbios. FINEP/PROSAB, Rio
de Janeiro, Brasil, 544 p. (in Portuguese)
Chernicharo CAL & Machado RMG (1998) Feasibility of the
UASB/AF system for domestic sewage treatment in devel-
oping countries. Water Sci. Technol. 38(8–9): 325–332
Chernicharo CAL & Nascimento MCP (2001) Feasibility of a
pilot-scale UASB/trickling filter system for domestic sewage
treatment. Water Sci. Technol. 44(4): 221–228
Chernicharo CAL, van Haandel AC, Foresti E & Cybis LF
(2001a). Introdução. Cap. 1. In: Chernicharo CAL (coor-
denador). Pós-tratamento de efluentes de reatores anaerób-
ios. FINEP/PROSAB, Rio de Janeiro, Brasil, 544 p. (in
Portuguese)
Chernicharo CAL, van Haandel AC, Cybis LF & Foresti E
(2001b). Post treatment of anaerobic effluents in Brazil: State
of the art. In: Proceedings of the 9th World Congress on
Anaerobic Digestion – Anaerobic Digestion 2001. Techno-
logisch Instituut, IWA, Netherlands Association for Water
Management. Antuerp, Belgium, 747–752
Chernicharo CAL, Silveira Cota R, Zerbini AM, von Sperling
M & Novy de Castro Brito LH (2001) Pos-treatment of
anaerobic effluents in an overland flow system. Water Sci.
Technol. 44(4): 229–236
Coraucci Filho B, Nour EAA, Figueiredo RF, Stefanutti R,
Klusener Filho LC & Broleze ST (2000) Estudo de um
sistema de pós-tratamento de efluente com aplicação do
método do escoamento superficial no solo: Polimento de
efluentes de filtros anaeróbios. In: Chernicharo CAL (coor-
denador). Pós-tratamento de efluentes de reatores anaerób-
ios. Pós-tratamento de efluentes de reatores anaeróbios –
Coletânea de Artigos Técnicos. FINEP/PROSAB, Rio de
Janeiro, Brasil, Vol. 1. pp. 1–8. (in Portuguese)
Cybis LF & Pickbrenner K (2001) – Uso de RSB em escala
piloto para pós-tratamento de efluente de tratamento
anaeróbio. Pós-tratamento de efluentes de reatores anaerób-
91
ios – Coletânea de Artigos Técnicos – Vol. 1. p. 193–202,
FINEP/PROSAB. (in Portuguese)
Daniel LMC & Foresti E (2004). Nitrogen removal via nitrite in
a sequential immobilized biomass batch reactor. In:
Proceedings of the 10th World Congress on Anaerobic
Digestion – Anaerobic Digestion 2004. Montreal, Canada,
Vol. 1, p. 367–372
Elmitwalli TA, Sklyar V, Zeeman G & Lettinga G (2002a)
Treatment of domestic sewage in a two-step anaerobic filter/
anaerobic hybrid system at low temperature. Water Res. 36:
2225–2232
Elmitwalli TA, Sklyar V, Zeeman G & Lettinga G (2002b) Low
temperature pre-treatment of domestic sewage in an anaer-
obic hybrid or an anaerobic filter reactor. Bioresour. Tech-
nol. 82: 233–239
Fitzke B, Km SM Geiben SU & Vogelpohl A (2004). Anaerobic
treatment of low concentrated wastewater with newly devel-
oped MBR-systems. Proceedings of the 3rd World Water
Congress and Exhibition 2004. Marrakech
Foresti E (2002) Anaerobic treatment of domestic sewage:
established technologies and perspectives. Water Sci. Tech-
nol. 45(10): 181–186
Garbossa LHP, Lapa KR & Foresti E (2005). Fixed-film radial
reactor for nitrogen removal from municipal wastewater
utilizing biogas as electron donor. In: Proceedings VIII
Latin-American Workshop and Seminar on Anaerobic
Digestion. Punta Del Este, Uruguai, p. 195–199
Gonçalves RF, Passamani FRF, Salim FP, Silva ALB, Mart-
ineli G & Bauer DG (2000). Associação de um reator UASB
e biofiltros aerados submersos para tratamento de esgoto
sanitário. Pós-tratamento de efluentes de reatores anaeróbios
– Coletânea de Artigos Técnicos – Vol. 1. pp. 119–134,
FINEP/PROSAB. (in Portuguese)
Grady CPL & Lim HC (1980) Biological Wastewater Treat-
ment: Theory and Application. Marcel Dekker Inc., New
York, 964 p
Hernández AE, Belalcazar LC, Rodrı́gues MS & Giraldo E
(2002) Retention of granular sludge at high hydraulic charges
in an anaerobic membrane bioreactor with immersed filtra-
tion. Water Sci. Technol. 45(10): 169–174
Islas-Lima S, Thalasso F & Gomes-Hernandez J (2004)
Evidence of anoxic methane oxidation coupled to denitrifi-
cation. Water Res. 38: 13–16
Jeison D & van Lier JB (2005). Bio-layer management in
anaerobic membrane bioreactors for wastewater treatment.
In: Proceedings VIII Latin-American Workshop and Semi-
nar on Anaerobic Digestion. Punta Del Este, Uruguai, pp.
117–122
Kato MT, Florencio L & Arantes RFM (2002). Post-treatment
of UASB effluent in an EGSB reactor type using flocculent
sludge. Proceedings VII Latin American Workshop and
Symposium on Anaerobic Digestion, Mérida, Mexico, Vol. 1,
pp. 489–494
Lettinga G, Man A, Last A, Wiegant W, Knippenberg K,
Frijns J & Buren J (1993) Anaerobic treatment of domestic
sewae and wastewater. Water Sci. Technol. 27(9): 67–73
Machdar I, Harada H, Ohashi A, Sekiguchi Y, Okuy H & Ueki
K (1997) A novel and cost-effective sewage treatment system
consisting of UASB pre-treatment and aerobic post-treat-
ment units for developing countries. Water Sci. Technol.
36(12): 189–197
Machdar I, Sekiguchi Y, Sumino H, Ohashi A & Harada H
(2000) Combination of a UASB reactor and a curtain type
DHS (downflow hanging sponge) reactor as a cost-effective
92
sewage treatment system for developing countries. Water Sci.
Technol. 42(3–4): 83–88
Metcalf & Eddy (1991). Wastewater Engineering: Treatment,
Disposal, and Reuse. Metcalf & Eddy, Inc. 3rd Ed., 1334 p
Oyanedel V, Garrido JM, Lema JM & Méndez R (2002). A
membrane assisted hybrid bioreactor for the post treatment of
an anaerobic effluent from a fish canning factory. In: Proceed-
ings VII Latin American Workshop and Symposium on
Anaerobic Digestion, Mérida, Mexico, Vol. 1, pp. 518–5525
Penetra RG, Reali MAP & Campos JR (2002). Influence of
flocculation conditions in the performance of an experimen-
tal domestic sewage treatment plant consisting of an anaer-
obic expanded bed reactor followed by dissolved air flotation.
Proceedings VII Latin American Workshop and Symposium
on Anaerobic Digestion, Mérida, Mexico, Vol. 1, pp. 503–
510
Pontes PP, Chernicharo CAL, Frade EC & Porto MTR (2003)
Performance evaluation of an UASB reactor used for
combined treatment of domestic sewage and excess aerobic
sludge from a trickling filter. Water Sci. Technol. 48(6): 227–
234
Pontes PP (2003). Reatores UASB aplicados ao tratamento
combinado de esgotos sanitários e lodo excedente de filtro
biológico percolador. Ph.D. Thesis. Federal University of
Minas Gerais – School of Engineering, 220 p. (in Portuguese)
Pynaert K, Wyffels S, Sprengers R, Boeckx P, van Cleemput Ol
& Verstraete W (2002) Oxygen-limited nitrogen removal in a
lab-scale rotating biological contactor treating an ammo-
nium-rich wastewater. Water Sci. Technol. 45(10): 357–363
Randall CW, Barnard JL & Stensel HD (1992) Design and
retrofit of wastewater treatment plants for biological nutrient
removal. Technomic Publ., AG, Basel, Switzerland 416 p
Reali MAP, Penetra RG & Carvalho ME (2001) Flotation
technique with coagulant and polymer application applied to
the post-treatment of effluents of anaerobic reactor treating
domestic sewage. Water Sci. Technol. 44(4): 205–212
Seghezzo L, Zeeman G, van Lier JB, Hamelers HVM &
Lettinga G (1998) A review: the anaerobic treatment of
sewage in UASB and EGSB reactors. Biresour. Technol. 65:
190–215
Sigge GO, Britz TJ, Fourie PC, Barnardt CA & Strydom R
(2002) Combining UASB technology and advanced oxida-
tion processes (AOP’s) to treat food processing wastewaters.
Water Sci. Technol. 45(10): 329–334
Sousa JT, van Haandel AC & Guimarães AAV (2001) Post-
treatment of anaerobic effluents in constructed wetland
system. Water Sci. Technol. 44(4): 213–219
Souza JT, van Haandel A, Lima EPC & Guimarães AVA
(2002). Performance of constructed wetland systems treating
anarobic effluents. In: Proceedings VII Latin American
Workshop and Symposium on Anaerobic Digestion, Mérida,
Mexico, Vol. 1, p. 511–517
Tai CS, Singh KS & Grant SR (2004). A novel two-stage
integrated anaerobic–aerobic biotreatment system for low-
strength wastewater. Proceedings of the 10th World Congress
on Anaerobic Digestion – Anaerobic Digestion 2004. Mon-
treal, Canada, Vol. 2, pp. 1051–1057
Tandukar M, Uemura S, Machdar I, Ohashi A & Harada H
(2005) A low-cost municipal sewage treatment system with a
combination of UASB and the ‘‘fourth generation’’ down-
flow hanging sponge reactors. Water Sci. Technol. 52(1–2):
323–329
Tawfik A, Klapwijk B, El-Gohary F & Lettinga G (2002) Post-
treatment of effluent from anaerobic (UASB) reactor treating
domestic sewage by rotating biological contactor. Water Sci.
Technol. 45(10): 371–376
Tessele F, Monteggia LO & Rubio J (2004). Treatment of
municipal wastewater UASB reactor effluent by unconven-
tional flotation and UV disinfection. In: Proceedings of the
10th World Congress on Anaerobic Digestion – Anaerobic
Digestion 2004. Montreal, Canada, Vol. 2, pp. 933–939
Thalasso F, Vallecillo A, Garcia-Encina P & Fernadez-Polanco
F (1997) The use of methane as a sole carbon source for
wastewater denitrification. Water Res. 31(1): 55–60
Uemura S, Takahashi K, Takaishi A, Machdar I, Ohashi A &
Harada H (2002) Removal of indigenous coliphages and fecal
coliforms by a novel sewage treatment system consisting of
UASB and DHS units. Water Sci. Technol. 46(11–12): 303–
309
U.S. Environmental Protection Agency (1981) Process design
manual: Land treatment of municipal wastewater, EPA 625/
1-81-013, Cincinnati, OH
U.S. Environmental Protection Agency (1984). Process design
manual: Land treatment of municipal wastewater: supple-
ment on rapid infiltration and overland flow, EPA 625/1-81-
013a, Cincinnati, OH
U.S. Environmental Protection Agency (1986) Municipal
Wastewater Disinfection – Design Manual, EPA/625/1-86/
021, Cincinnati, 247 p
van Haandel AC & Lettinga G (1994) Anaerobic Sewage
Treatment in Regions With a Hot Climate. John Wiley and
Sons, Chichester, UK
van Haandel AC & Guimarães P (2000) Modelamento do
tratamento anaeróbio-aeróbio usando-se o reator UASB e
lodo ativado em bateladas sequenciais. Pós-tratamento de
efluentes de reatores anaeróbios – Coletânea de Artigos
Técnicos. Vol. 1, p. 143–153, FINEP/PROSAB. (in Portu-
guese)
von Sperling M, Chernicharo CAL, Soares AME & Zerbini
AM (2002) Coliform and helminth eggs removal in a
combined UASB reactor – baffled pond system in Brazil:
performance evaluation and mathematical modelling. Water
Sci. Technol. 45(10): 237–242
von Sperling M & Mascarenhas LCAM (2004). Performance of
very shallow ponds treating effluents from an UASB reactor.
In: Proceedings 6th International Conference on Waste
Stabilization Ponds. IWA. Avignon, France
von Sperling M, Bastos RKX & Kato MT (2004). Removal of
E. coli and helminth eggs in UASB – polishing pond systems.
Proceedings 6th International Conference on Waste Stabil-
ization Ponds. IWA. Avignon, France
von Sperling M & Chernicharo CAL (2005) Biological Waste-
water Treatment in Warm Climate Regions. IWA Publishing,
London 1452