M. C. M. Van Loosdrecht - 1997 - Biological Phosphate Removal Processes
M. C. M. Van Loosdrecht - 1997 - Biological Phosphate Removal Processes
M. C. M. Van Loosdrecht - 1997 - Biological Phosphate Removal Processes
MINI-REVIEW
Received: 3 April 1997 / Received revision: 9 June 1997 / Accepted: 14 June 1997
Abstract Biological phosphate removal has become a Biological phosphate removal was discovered by ac-
reliable and well-understood process for wastewater cident in full-scale wastewater treatment plants around
treatment. This review describes the historical devel- 1959, and the ®rst full-scale processes were designed and
opment of the process and the most important micro- introduced at the end of the 1970s. Initially most of the
biological and process-engineering aspects. From a research was practically oriented, aimed at achieving
microbiological point of view, the role of poly(hydroxy- systems for BPR and paying limited attention to the
alkanoates) as storage material in a dynamic process and basic mechanisms underlying the phenomenon. In the
the use of polyphosphate as an energy reserve are the 1980s the research ®eld became more interdisciplinary,
most important ®ndings. From a process-engineering with microbiological and process-engineering research
point of view, the study of biological phosphate removal resulting in a better understanding of the basic phe-
has shown that highly complex biological processes can nomena. Biological phosphate removal has been a ®eld
be designed and controlled, provided that the impor- of research where it was virtually impossible to make
tance of the prevailing microbiological ecological pro- progress without an interdisciplinary approach. This is
cesses is recognised. largely due to the complexity of the organisms involved,
which require a good background in microbiological
physiology for continuing research. However, groups
applying a strictly microbial approach have often ig-
Introduction nored important observations, leading to research on
organisms that do not play a signi®cant role in the
Biological phosphate removal (BPR) has become a well- process.
established process and is applied in many full-scale In this review we will brie¯y review the historical
wastewater treatment processes. The process as such developments of the process, indicating the interaction
does not only oer a good opportunity to remove between dierent research groups. Thereafter the es-
phosphate in an ecient way from wastewater, it is also sential microbiological aspects will be discussed brie¯y
interesting to study for microbial ecological research. (for a more detailed discussion the reader is referred to
The organisms involved in BPR have a complex physi- Mino et al. 1997) followed by a longer discussion on the
ology in which the formation and consumption of process-engineering aspects.
storage polymers [polyphosphate, glycogen, poly(hy-
droxyalkanoates) (PHA)] play a dominant role.
Historical development
Levin and Shapiro (1965) conducted the ®rst struc- operate on plug ¯ow and the ®rst part of the reactor
tured investigation of the phosphate-removal phenom- should not be well aerated. Moreover they found that
ena observed in several treatment plants. They there was a maximum capacity of the sludge to accu-
postulated the hypothesis that the removal was biolog- mulate phosphate. Until this stage, the research was
ically mediated because it only occurred under aerobic mainly performed in full-scale systems by civil engineers.
conditions. The phosphate could be stored in volutin This led to great controversy and confusion, basically
granules, as observed in several bacteria. Levin and because of a lack of any proper understanding of mi-
Shapiro studied the phosphate removal process in full- crobial processes in general. One aspect that was highly
scale treatment plants and with batch experiments using important and (when looking back) very obvious, a link
sludge retrieved from these plants. Their main observa- between phosphate release and uptake processes, was
tions were that phosphate was released under non-aer- not really recognised. Full attention was given to the
ated conditions and taken up under aerobic conditions; phosphate-uptake process in the process design. This
moreover, addition of wastewater (substrate) increased was considered to be dependent on aerobic bacteria,
the phosphate uptake. Since phosphate was taken up resulting from stress conditions due to the dynamic
under aerobic conditions, they concluded that the up- feeding of activated sludge or the presence of anaerobic
take occurred via formation of ATP during oxidative conditions in part of the treatment process. The phos-
phosphorylation. Uptake via substrate phosphorylation phate-uptake process was called ``overplus'' or ``luxury''
could have taken place anaerobically via the Embden- phosphate uptake. These phenomena were described by
Meyerhof pathway. They showed that the process was Harold (1966) and observed with pure cultures, sub-
clearly a biological one since aeration and substrate were jected to stress conditions. On the basis of this stress
necessary, and inhibition of oxidative phosphorylation theory, Nicholls and Osborn (1979) arrived at recom-
by 2,4-dichlorophenoxyacetic acid led to inhibition of mendations for process design that led to processes that
phosphate uptake. Observing that, at high pH (9) also, function well although the fundamental assumptions
no phosphate uptake occurred they suggested that made later turned out to be wrong.
chemical precipitation was no causing the phosphate In the second part of the 1970s research in microbi-
removal. In this and other papers from this period it was ology expanded and, by applying more process-engi-
assumed that glucose was the main substrate; fermen- neering principles, Fuhs and Chen (1975) concluded,
tation processes (in the sewer or treatment plant) were from a range of isolation tests, that bacteria of the genus
seemingly not recognised. Acinetobacter were responsible for biological phosphate
Later Shapiro et al. (1967) focused their research removal. These organisms accumulated large amounts
more on the anaerobic stage of the process. They indi- of polyphosphate and could also accumulate poly(hy-
cated that the phosphate removal was not caused by droxybutyrate). Fuhs and Chen (1975) postulated the
cellular decay but could be enhanced by adding poisons hypothesis that an anaerobic phase was needed to pro-
such as KCN. Moreover the release was directly asso- duce volatile fatty acids, which are the substrates for
ciated with the amount of sludge present. This pointed phosphate-removing organisms. Acinetobacter-type or-
again towards a biological basis for the observations. ganisms could use these substrates under aerobic con-
From experimental ®ndings they concluded that the ditions for growth and excessive phosphate uptake, in
redox potential rather than the oxygen concentration line with the prevailing theory among sanitary engineers
was triggering the phosphate release. This conclusion and the work of Harold (1966). It was, therefore, not
in¯uenced many later research projects, even after surprising that Fuhs and Chen (1975) also failed to make
Randall et al. (1970) had clearly shown that it was not a link between anaerobic phosphate release and the oc-
the redox potential but conditions that adversely aect currence of polyphosphate-accumulating bacteria. Nor
cell metabolism (such as lack of oxygen or substrate) was it considered strange, therefore, that the isolated
that caused phosphate release. bacteria showed anaerobic phosphate release but only at
On the basis of the behaviour of systems manifesting very low rates compared to activated sludge. Since there
excess phosphate removal, Levin (1966) ®led a patent for were no good measurements available from activated
the Phostrip process. In this process the observed sludge systems, it was also not recognised that the or-
phosphate release is used to obtain a high concentration ganisms responsible take up phosphate under anaerobic
of phosphate in a separate tank, which can subsequently conditions, whereas Acinetobacter sp. only consumes
be precipitated. This process, developed without a acetate under aerobic conditions.
proper understanding of the mechanism, is still suc- Most subsequent microbiological studies relied on the
cessfully used in treatment plants today. However, the isolation procedure described by Fuhs and Chen (1975)
concept only became accepted widely in the 1980s when (e.g. Deinema et al. 1980; Lotter et al. 1986) so it is not
the basic background of BPR processes became clear. surprising that the same types of organisms were always
In the late 1960s and early 1970s many researchers found. However, these organisms were not involved in
tried to ®nd a good explanation for the observed excess the actual process, as recently clearly demonstrated by
phosphate removal in certain full-scale treatment plants. the use of molecular ecology techniques (Bond et al.
Milbury et al. (1971) de®ned some basic requirements 1995; Wagner et al. 1994; Mino et al. 1997). Neverthe-
for phosphate removal by stating that the reactor should less, microbiological research has greatly helped the
291
growth (Kuba et al. 1994; Brdjanovic et al. 1997). For- Biological phosphate removal processes are depen-
mation of storage materials rather than using substrate dent on the accumulation of bacteria capable of storing
for growth seems to be a basic characteristic of micro- large amounts of polyphosphate inside their cells
organisms in systems with feast/famine regimes such as (Marais et al. 1983). This means that the eciency of the
those that occur in wastewater treatment processes (van process is directly coupled to the formation of poly-
Loosdrecht et al. 1997a). Unfortunately this aspect re- phosphate-accumulating bacteria. For BPR processes
ceives only limited attention from microbial researchers, this means that the incoming wastewater needs to be
who prefer to work with batch or continuous cultures mixed with the sludge in a true anaerobic zone (i.e. no
rather than in dynamic cultures. oxygen or nitrate presence). In this zone VFA present in
It is clear that the BPR process has introduced a the wastewater, or formed by fermentation processes,
range of interesting aspects of applied microbial re- can be accumulated by polyphosphate-accumulating
search, which certainly require further elaboration. bacteria. An adequate design of the anaerobic phase is
First, there is the recognition that ``strictly'' aerobic thus essential for a good BPR process and will depend
organisms can be active under conditions without elec- heavily on the wastewater characteristics. Sewage from
tron acceptors presence. Secondly, the role of storage anaerobic sewers will be partly fermented into VFA,
polymers in microbial competition processes has become therefore small anaerobic reactors can be applied. When
evident and, ®nally, it was found that the growth rate of the wastewater does not contain VFA (as with aerobic
these organisms cannot be directly predicted by sub- sewers, for example), the anaerobic phase has to be
strate availability, for example, as generally assumed. based on the slower fermentation process, resulting in a
The organisms seem to use the available substrate larger anaerobic reactor. The length of the aerobic phase
(PHA) primarily for the formation of polyphosphate in the treatment process will usually not be limited by
and glycogen and for maintenance processes. Growth the phosphate uptake but by the nitri®cation, because of
results from the dierence between PHA consumption the slower growth rate of nitrifying bacteria.
rate and PHA use for the aforementioned processes In practice there are many dierent BPR process
(Murnleitner et al. 1997). con®gurations, all of which can be grouped into two
basic types: full biological processes and combined bio-
logical/chemical processes. A basic scheme for a bio-
logical phosphate removal process is shown in Fig. 2.
Process engineering aspects In the anaerobic phase substrate is taken up by
polyphosphate-accumulating bacteria, this results in
The traditional method for phosphate removal from
wastewater is the addition of precipitating chemicals
(iron or aluminium salts) to wastewater. Besides the fact
that, in waste treatment, addition of chemicals should be
minimised, there are several negative aspects of this
practice. The counterion of the salts (usually chloride)
remains in the water, resulting in increased salinity of
surface waters. The chemical precipitate accumulates in
the sludge, leading to extra costs for the treatment of the
excess sludge. Moreover, since the sludge content in a
treatment system is limited to a maximum amount, large
treatment reactors are required in order to maintain the
same amount of biological sludge. Owing to the low
costs of chemical addition, the ease and stability of such
methods, and the lack of feeling for biological processes
among civil engineers, chemical phosphate removal has
long been the dominant treatment process. In most
countries where phosphate removal is required, how-
ever, biological phosphate removal has become the
preferred process, eventually supplemented by chemical
precipitation. In evaluating full-scale treatment plants it
should be realised that precipitation might always con-
tribute to the overall phosphate-removal process. This
depends strongly on the cation composition of the in-
¯uent and the prevailing pH conditions (Maurer 1997).
The biological process can induce the precipitation be-
Fig. 2 Schematic representation of a biological phosphorus-removal
cause, in the anaerobic phase, phosphate and the cellular process. The graph gives a schematic representation of the change in
counterion magnesium are released into the liquid, concentrations in a plug-type of process. VFA volatile fatty acids,
leading to increased concentrations. PHA poly(hydroxyalkanoate)
293
The negative eect of nitrate on phosphate release is polyphosphate-accumulating organisms. Since this COD
caused by direct substrate competition between hetero- fraction is soluble, presettling as such should not have a
trophic denitri®ers and polyphosphate-accumulating direct eect on the eciency of biological phosphate
bacteria. This competition is won by the denitri®ers. removal.
This observation, and the fact that the bacterial group In a recent study the fate of nitrogen, phosphate and
thought responsible, Acinetobacter, cannot denitrify, COD in a full-scale treatment plant with and without
induced a general belief that phosphate and N removal presettling was followed in detail (van Loosdrecht et al.
are competing processes. 1997b). This study showed that, in the process consid-
Recently several authors (Vlekke et al. 1988; Kuba ered, particulate COD (which was removed during pre-
et al. 1993) have shown that denitrifying bacteria can settling) was not used by polyphosphate-accumulating
also be involved in the removal of biological phosphate. bacteria for cell growth. The addition of extra, particu-
Moreover it was shown that these bacteria can con- late, COD leads to a shorter sludge age (higher growth
tribute signi®cantly to biological phosphate removal in rate). The decreased sludge retention time leads to more
UCT-type processes (van Loosdrecht et al. 1997c; Kuba biomass formation, since less substrate is used for
et al. 1997). In these processes, organisms are cycled maintenance. With presettling, iron had to be used to
between anaerobic and denitrifying conditions, which supplement the biological phosphate removal; without
stimulate their growth. The use of denitrifying poly- presettling, enough polyphosphate-accumulating bio-
phosphate-accumulating bacteria has as its main advan- mass was produced to accumulate all the phosphate.
tage the fact that less COD is needed in the nutrient- The positive eect of adding non-settled sewage (no
removal process, thereby expanding the operational chemicals needed for biological phosphate removal) was
range of the biological processes. The discovery of largely balanced by the extra energy consumption (and
denitrifying polyphosphate-accumulating bacteria has therefore CO2 production) since not enough methane
given the opportunity to design new types of wastewa- was produced in the digestor. Moreover it turned out
ter-treatment processes in which the slow-growing ni- that the amount of nitrate denitri®ed was not increased
tri®ers and fast-growing heterotrophs are separated by addition of the particulate COD; extra nitrogen re-
(Wanner et al. 1992; Kuba et al. 1996; Sorm et al. 1996). moval was due to the extra assimilation of ammonium in
After the anaerobic reactor, the sludge/water mixture is the surplus sludge.
separated. The water containing ammonium is nitri®ed Application of primary settling and sludge digestion
in a separate system and thereafter mixed with the of primary and secondary sludge therefore seem to be
sludge again in a denitri®cation/phosphate-uptake re- logical processes for nutrient removal. This can, how-
actor. ever, lead to a large recycling of nutrients from the
digestor back to the treatment plant. For nitrogen
removal, speci®c treatment technologies have been de-
Eect of presettling and sludge digestion veloped (Hellinga et al. 1997); for phosphate the recy-
cling depends heavily on the prevailing conditions. If
Wastewater-treatment processes have to ful®l a large sucient magnesium or calcium is present in wastewater
number of requirements. These sometimes seem to the majority of the phosphate will be precipitated in the
con¯ict. On one hand it is necessary to have COD sludge digestor (Jardin and Popel 1996). If this is not the
available for nitrogen and phosphate removal, on the case, some iron needs to be added to the digestor. This is
other hand it would be worthwhile to separate as much often done in any case in order to control sulphide levels
COD as possible in a primary settling tank to produce in the methane gas.
methane. The methane could be used in a gas generator
to produce energy for the operation of the wastewater-
treatment plant. Outlook
Siebritz et al. (1983) concluded, from several experi-
mental observations, that only the ``readily biodegrad- Biological phosphate removal has become an established
able fraction'' of the in¯uent COD was used by the biotechnological process operative at full scale. Research
295
has shown that, under dynamic process conditions, Johansson P (1994) SIPHOR a kinetic model for simulation of
bacteria exhibit a complex ecophysiological behaviour, biological phosphate removal. PhD thesis, Department of
Water and Environmental Engineering, Lund University,
which is dicult to observe in standard pure-culture Sweden
work. Although the physiology has been quite well es- Kuba T, Smolders GJF, Loosdrecht MCM van, Heijnen JJ (1993)
tablished, there is still a need to obtain the organisms Biological phosphorus removal from wastewater by anaerobic-
responsible in pure culture. Hereto innovative new en- anoxic sequencing batch reactor. Water Sci Technol 27: 241±252
Kuba T, Wachtmeister A, Loosdrecht MCM van, Heijnen JJ
richment and isolation techniques need to be developed. (1994) Eect of nitrate on phosphorus release in biological
Traditional techniques are usually rather static, which is phosphorus removal systems. Water Sci Technol 30: 263±269
probably the reason why the phosphate-removing or- Kuba T, Loosdrecht MCM van, Heijnen JJ (1996) Phosphorus and
ganisms have not yet been cultured. Obviously only un- nitrogen removal with minimal COD requirement by integra-
der dynamic enrichment conditions (similar to those tion of denitrifying dephosphatation and nitri®cation in a two-
sludge system. Water Res 30: 1702±1710
experienced by the organisms in the treatment plant) will Kuba T, Loosdrecht MCM van, Brandse F, Heijnen JJ (1997)
it be possible to enrich bacteria relying on these condi- Occurrence of denitrifying phosphorus removing bacteria in
tions for their competitive advantage. Biological phos- modi®ed UCT-type wastewater treatment plants. Water Res 31:
phate removal was discovered in wastewater treatment 777±787
Levin GV, Shapiro J (1965) Metabolic uptake of phosphorus by
plants by accident, and only became known because it wastewater organisms. J Water Pollut Control Fed 37: 800±821
was of technological interest. A better study of microbial Levin GV (1966) Sewage treatment process US patent 3236766
competition in the real (continuously changing) envi- Loosdrecht MCM van, Pot MA, Heijnen JJ (1997a) Importance of
ronment, rather than in steady-state pure cultures, might bacterial storage polymers in bioprocesses. Water Sci Technol
35: 41±47
reveal a range of microorganisms with new and inter- Loosdrecht MCM van, Smolders GJF, Kuba T, Heijnen JJ (1997b)
esting capacities. These will probably not only be of Metabolism of microorganisms responsible for biological
academic interest, but also lead to new biotechnological phosphorus removal from wastewater. Antconia van Lee-
processes. uwenhoek 71: 109±116
Loosdrecht MCM van, Kuba T, Van Veldhuizen HM, Brandse
FA, Heijnen JJ (1997c) Environmental impacts of nutrient re-
moval processes: case study. J Environ Eng 123: 33±40
References Loosdrecht MCM van, Brandse FA, De Vries AC (1997d) Up-
grading of wastewater treatment processes for integrated nu-
Arun V, Mino T, Matsuo T (1987) Biological mechanisms of ace- trient removal ± the BCFS process. Water Sci Technol (in press)
tate uptake mediated by carbohydrate consumption in excess Lotter LH, Wentzel MC, Loewenthal RE, Ekama GA, Marais
phosphate removal systems. Water Res 22: 565±570 GvR (1986) A study of selected characteristics of Acinetobacter
Barnard JL (1974) Cut P and N without chemicals. Water Wastes spp. isolated from activated sludge in anaerobic/anoxic/aerobic
Eng 11: 33±36 and aerobic systems. Water SA 12: 203±208
Barnard JL (1975) Biological nutrient removal without the addition Marais GvR, Loewenthal RE, Siebritz IP (1983) Review: obser-
of chemicals. Water Res 9: 485±490 vations supporting phosphate removal by biological excess
Bond P, Hugenholtz P, Keller J, Blackall L (1995) Bacterial com- uptake. Water Sci Technol 15: 15±41
munity structures of phosphate removing and non-phosphate Maurer M, Gujer W, Hany R, Bachmann S (1997) Intracellular
removing activated sludge from sequencing batch reactor. Appl carbon ¯ow in phosphorus accumulating organisms in activated
Environ Microbiol 61: 1910±1916 sludge systems. Water Res 31: 907±917
Brdjanovic D, Slamet A, Loosdrecht MCM van, Hooijmans CM, Milbury WF, McCauley D, Hawthorne CH (1971) Operation of
Alaerts GJ, Heijnen JJ (1997) Impact of excessive aeration on conventional activated sludge for maximum phosphorus re-
biological phosphorus removal from wastewater. Water Res (in moval. J Water Pollut Control Fed 43: 1890±1901
press) Mino T, Loosdrecht MCM van, Heijnen JJ (1997) Microbiology
Comeau Y, Hall KJ, Hancock REW, Oldham WK (1986) Bio- and biochemistry of the enhanced biological phosphate removal
chemical model for enhanced biological phosphorus removal. process. Water Res (in press)
Water Res 20: 1511±1521 Murnleitner E, Kuba T, Loosdrecht MCM van, Heijnen JJ (1997)
Deinema MH, Habets LHA, Scholten J, Turkstra E, Webers Metabolic model for phosphorus removal. Biotechnol Bioeng
HAAM (1980) The accumulation of polyphosphate in Acinet- 54: 434±450
obacter spp. FEMS Microbiol Lett 9: 275±279 Nicholls HA (1975) Full scale experimentation at the new Johan-
Eggers E, Dirkzwager AH, Van den Honing H (1991) Full scale nesburg aeration plants. Water SA 1: 121±132
phosphate crystallisation in a crystalactor. Water Sci Technol Nicholls HA, Osborn DW (1979) Bacterial stress: prerequisite for
24: 333±334 biological removal of phosphorus. J Water Pollut Control Fed
Fuhs GW, Chen M (1975) Microbiological basis of phosphate re- 51: 557±570
moval in the activated sludge process for the treatment of Perreira H, Lemos PC, Reis MA, Crespo JPSG, Carrondo MJT,
wastewater. Microbiol Ecol 2: 119±138 Santos H (1996) Model for carbon metabolism in biological
Harold FM (1966) Inorganic polyphosphates in biology: structure phosphorus removal processes based on in-vivo 13C-NMR la-
mechanisms and function. Bacteriol Rev 30: 772±794 beling experiments. Water Res 30: 2128±2138
Hascoet MC, Florentz M (1985) In¯uence of nitrates on biological Randall CW, Marshall DW, King PH (1970) Phosphate release in
phosphorus removal of wastewater. Water SA 11: 1±8 activated sludge process. J Sanit Eng Div Proc Am Soc Civ Eng
Hellinga C, Schellen AAJC, Mulder JW, Loosdrecht MCM van, 96: 395±408
Heijnen JJ (1997) The SHARON-process; an innovative Rensink JH (1981) Biologische defosfatering en procesbepalende
method for nitrogen removal from ammonium rich wastewater. factoren. In: NVA Symposium Proceedings. Nieuwe on-
In: Balmer P (ed) Proceedings of the Conference ``Upgrading of twikkelingen en praktijkervaringen in Nederland en Zweden (in
wastewater treatment plants''. IAWQ London, UK Dutch)
Jardin N, Popel HJ (1996) Behavior of waste activated sludge from Shapiro J, Levin GV, Zea HG (1967) Anoxically induced release of
enhanced biological phosphorus removal during sludge treat- phosphate in wastewater treatment. J Water Pollut Control Fed
ment. Water Environ Res 68: 965±974 39: 1811±1818
296
Siebritz IP, Ekama GA, Marais GvR (1983) A parametric model Van Groenestijn JW, Bentvelsen MMA, Deinema MH, Zehnder
for biological excess phosphorus removal. Water Sci Technol AJB (1989) Polyphosphate degrading enzymes in Acinetobacter
15: 127±152 spp. and activated sludge. Appl Environ Microbiol 55: 219±223
Smolders GJF, Van der Meij J, Loosdrecht MCM van, Heijnen JJ Vlekke GJFM, Comeau Y, Oldham WK (1988) Biological phos-
(1994a) Model of the anaerobic metabolism of the biological phorus removal from wastewater with oxygen or nitrate in se-
phosphorus removal process; stoichiometry and pH in¯uence. quencing batch reactors. Environ Technol Lett 9: 791±796
Biotechnol Bioeng 43: 461±470 Wagner M, Erhart R, Manz W, Amann R, Lemmer H, Wedi D,
Smolders GJF, Loosdrecht MCM van, Heijnen JJ (1994b) A Schleifer K-H (1994) Development of an rRNA-targetted
metabolic model for the biological phosphorus removal pro- oligonucleotide probe for the genus Acinetobacter and its ap-
cess. Water Sci Technol 31: 79±93 plication for in situ monitoring in activated sludge. Appl En-
Smolders GJF, Loosdrecht MCM van, Heijnen JJ (1996) Steady viron Microbiol 60: 792±800
state analysis to evaluate the phosphate removal capacity and Wanner J, Cech JS, Kos M (1992) New process design for bio-
acetate requirement of biological phosphorus removing main- logical nutrient removal. Water Sci Tech 25: 445±448
stream and sidestream process con®gurations. Water Res 30: Wentzel MC, Lotter LH, Loewenthal RE, Marais GvR (1986)
2748±2760 Metabolic behaviour of Acinetobacter sp. in enhanced biologi-
Sorm R, Bortone G, Saltarelli R, Jenicek P, Wanner J, Tilche A cal phosphorus removal ± a biochemical model. Water SA 12:
(1996) Phosphate uptake under anoxic conditions and ®xed ®lm 209±224
nitri®cation in nutrient removal activated sludge system. Water Wentzel MC, Loewenthal RE, Ekama GA, Marais GvR (1988)
Res 30: 1573±1585 Enhanced polyphosphate organism cultures in activated sludge
Srinath EG, Sastry CA, Pillai SC (1959) Rapid removal of phos- systems ± part 1:enhanced culture development. Water SA 14:
phorus from sewage by activated sludge. Experienta 15: 339±340 81±92