CHAPTER 2

BACKGROUND

2.1 CYANOBACTERIA 15

2.2 CYANOTOXINS 16

2.3 LEGISLATION AND GUIDELINES 22

2.4 WATER TREATMENT FOR CYANOTOXINS REMOVAL 23

2.5 PAC ADSORPTION 28

2.6 ULTRAFILTRATION 35

2.7 PAC/UF 40

2.8 REFERENCES 45

13
14
 Chapter 2 - Background

2.1 CYANOBACTERIA

Cyanobacteria belong to Bacteria Domain, presenting some characteristics of bacteria and

some of algae. They are prokaryotic organisms, yet they are similar to algae in size and

contain both chlorophyll-a and accessory pigments to perform oxygenic photosynthesis (Mur

et al., 1999; Kaebernick and Neilan, 2001). The accessory pigment phycocyanin, a bluish

phycobilin, is unique to cyanobacteria and, as so, they are also referred to as blue-green algae

(Zurawell et al., 2005).

The basic morphology comprises unicellular, colonial and multicellular filamentous forms

(Mur et al., 1999), with cell sizes varying from less than 2 µm to 40 µm in diameter

(Kaebernick and Neilan, 2001). Cyanobacteria may have special adaptations such as

heterocysts (for nitrogen fixation), gas vacuoles (light triggered, confer buoyancy), and

akinetes (cells with reserve materials, enabling survival under unfavourable conditions),

allowing them an advantage over many competitors (Mur et al., 1999; Kaebernick and Neilan,

2001; Svrecek and Smith, 2004).

Cyanobacteria are important primary producers and generate several metabolites that are

pharmaceutically useful substances. However, under some conditions, cyanobacteria form

undesirable blooms, associated with a number of water-related problems, including tastes and

odours (e.g., geosmin and methylisoborneol), turbidity, and oxygen depletion as the

cyanobacteria decay (Metcalf and Codd, 2004). Furthermore, cyanobacteria blooms cause

poor settling, filter clogging, disinfectant consumption and production of disinfection-by-

products in water treatment plants (WTP) (Paralkar and Edzwald, 1996; Her et al., 2004). Of

particular concern is the ability of several cyanobacteria strains to produce, under certain

conditions of growth, potent toxins as secondary metabolites − cyanotoxins, which function is

15
 Chapter 2 - Background

still unknown. Secondary metabolites are those compounds not used by an organism for its

primary metabolism, i.e., cell division or energy production (Svrecek and Smith, 2004).

The cyanobacteria class includes 150 genera, among which 40 genera are estimated to be

potential toxin producers (Van Apeldoorn, 2007). Surveys of many fresh cyanobacterial

blooms have shown that up to 50-70% of them are toxic (Sivonen and Jones, 1999) and that

the most commonly occurring toxic genera is Microcystis, in particular, the species

M. aeruginosa (WHO, 2003; MHNZ, 2005).

The reasons for cyanobacterial blooms occurrence are not straightforward and continue to be

controversially debated. Blooms are not always associated with pollution and have been

found in reservoirs with near-pristine catchments (Sivonen and Jones, 1999). It has been

suggested that a stable water column, warm water, high nutrient concentrations (P, N),

low N:P and high pH are advantageous conditions for the development of blooms, but no

single factor has been directly related with it (Mur et al., 1999; Zurawell et al., 2005). Timing

and duration of the cyanobacterial bloom season depend largely on the climatic conditions of

the region. In temperate zones, cyanobacterial blooms are most prominent during the late

summer and early autumn and may last for 2-4 months. In warmer climates, like the ones of

Portugal and Spain, blooms may occur for up to 6 months or longer (Sivonen and Jones,

1999).

2.2 CYANOTOXINS

2.2.1 General

Cyanotoxins may occur both within cells (cell bound or intracellular) or dissolved

(extracellular). Intracellular toxin content is typically highest in the late logarithmic growth

16
 Chapter 2 - Background

phase, and the toxin content apparently shows a positive correlation with cyanobacterial

biomass (Carmichael, 2001). Toxin release into water may be natural (natural cell lysis

caused by age or through active release) or induced (toxin release, resulting from cell

destruction in treatment processes caused by mechanical and chemical stresses) (Sivonen and

Jones, 1999; Pietsch et al., 2002; Schmidt et al., 2002). In natural environments, healthy

blooms populations produce little extracellular toxin (with the exception of

Cylindrospermopsis genera which produces similar fraction of intra and extracellular toxins

throughout the cell life) (Sivonen and Jones, 1999). The range of measured concentration for

dissolved cyanotoxins, in all cases except those where a major bloom is obviously breaking

down, is 0.1-10 µg/L (Jones and Orr, 1994; Lahti et al., 1997). Cell-bound concentrations are

several orders of magnitude higher.

Environmental factors affect the toxin content but only within a range of less than an order of

magnitude (Sivonen and Jones, 1999). For instance, for microcystins (toxins produced by

Microcystis), while the toxin variants produced by a particular strain are rather constant, the

ratios of individual microcystins may change with time, temperature and light. However, for

microcystins, it has been shown that toxicity of a strain mainly depends on whether there is

the gene for microcystin production (WHO, 2003).

According to their chemical structure cyanotoxins are classified in cyclic peptides, alkaloids

and lipopolysaccharides (LPS) (Sivonen and Jones, 1999). In terms of their mode of action,

cyanotoxins are hepatotoxins, neurotoxins, cytotoxins and irritant or inflammatory agents

(WHO, 2006; Meriluoto and Spoff, 2007). General important characteristics of cyanotoxins

were compiled and are presented in Table 2.1.

17
 Chapter 2 - Background

Table 2.1 - Properties of cyanotoxins (Carmichael, 1997; Sivonen and Jones, 1999; Hitzfeld et al., 2000; Codd, 2000; Metcalf and Codd, 2004; Van
Apeldoorn et al., 2007; Meriluoto and Spoof, 2007).

LD50 Structural Molecular

Chemical Structure Toxicity Net charge Hydrophobicity Main Producer Genera
(µg/kg)a Variants Weight (Da)

Microcystis, Anabaena,
900-1100 b Relatively
Microcystins Cyclic peptides Hepatotoxic 25- 600 > 80 -2; -1; 0 Planktothrix, Nostoc,
Hydrophobicc Anabaenopsis

Relatively
Nodularin Cyclic peptides Hepatotoxic 50 ca. 6 824 Negative Hydrophobic Nodularia

Anabaena, Planktothrix,
Anatoxin-a Alkaloid Neurotoxic 375 5 166 +1 Hydrophilic Aphanizomenon

Anatoxina-a (S) Alkaloid Neurotoxic 20 1 252 0 Hydrophilic Anabaena

Anabaena,
0 (C-Tox);
Aphanizomenon,
Saxitoxins Alkaloid Neurotoxic 7.6 - 10.5 ca. 20 256 - 491 +1 (GTX); Hydrophilic Cylindrospermopsis,
+2 (STX) Lyngbya

Hepatotoxic Cylindrospermopsis,
Very
Cylindrospermopsin Alkaloid Cytotoxic 200 - 2000 2 415 0 Aphanizomenon,
Hydrophilic
Umezakia

Potential
LPS Lipopolysacharides -- >3 -- Negative -- All
irritant
a
Toxicity measured by mouse intraperitoneal; b Estimated MW are 500-4000 Da, with the most known between 900-1100 Da; c hydrophobicity depends on the variable
amino acids; LPS: Lipopolysacharides; C-Tox: doubly sulphated saxitoxins; GTX: singly sulphated saxitoxins (gonyautoxins); STX: non sulphated saxitoxins.
 Chapter 2 - Background

Based on current knowledge, microcystins (MC) and cylindrospermopsin (CYN) are

considered to present the greatest risk to human health, since they have both acute and chronic

effects, and occur most frequently in fresh water systems whereas the cyanobacterial

neurotoxins appear in high concentrations only occasionally. Nodularin and microcystins

have comparable toxicity, but as Nodularia spumigena (an obligate brackish and saline

tolerant cyanobacteria) is currently the only species known to produce nodularin, this toxin is

unlikely to occur in drinking water (MHNZ, 2005; WHO, 2006). The genera producing CYN

usually form toxic blooms in subtropical, tropical or arid zones water bodies (Sivonen and

Jones, 1999). However, increasing occurrences of C. raciborskii in Europe are reported,

namely in France, Germany, Hungary, Austria, Greece, Slovakia, Spain and Portugal

(Quesada et al., 2006; Van Apeldoorn et al., 2007). Cylindrospermopsin may thus become

relevant in temperate zones in the future (WHO, 2003).

2.2.2 Microcystins

Microcystins are cyclic heptapeptides, containing seven peptide-linked amino acids, with the

two terminal amino acids joined to form a cyclic compound. All microcystins contain a

specific amino acid (Adda) side chain which is largely responsible for their toxicity

(Carmichael, 1994). There are currently more than 80 known variants of microcystins (WHO,

2006), based on five common amino acids and varying with respect to the methyl groups and

the two amino acids in positions 2 and 4 within the ring (Carmichael, 1997). Each variant is

named according to the variable amino acids present. MC-LR, with leucine and arginine

occupying the 2 and 4 positions (Figure 2.1), is among the most frequent and most toxic

microcystins congeners (Svrcek and Smith, 2004; MHNZ, 2005; WHO, 2006). Other

common microcystins analogues are MC-LA, -RR and -YR (Ho et al., 2006 a).

19
 Chapter 2 - Background

Figure 2.1 - Structure of microcystin-LR (Meriluoto and Spoof, 2007).

Microcystins cause both acute and chronic effects. They are hepatotoxic, causing a

breakdown in the liver structure by interfering with enzyme pathways (inhibit the activities of

protein phosphatase 1 and 2A), and death may occur as little as 30 minutes from severe liver

haemorrhage (Codd, 2000). Chronic exposure promotes tumour growth (MHNZ, 2005).

Concerning their chemical structure, microcystins are large compounds (900-1100 Da) in

comparison with usual micropollutants and other cyanotoxins (Sivonen and Jones, 1999).

Using molecular models, the diameter of microcystin-LR has been estimated to be between

1.2 and 2.6 nm (Donati et al., 1994). The variable amino acids are responsible for different

properties, like hydrophobicity and net charge, toxicities and molecular weight (Table 2.2).

All microcystins have two carboxyl groups that are negatively charged at neutral pH (on D-

glutamate and D-erythro-β-methylaspartic acid), but microcystin variants have different net

charges, depending on the variable amino acid groups − e.g., the arginine group has a positive

charge, whereas the alanine and tyrosine groups have zero charge (Cook and Newcombe,

2002; Newcombe et al., 2003; Meriluoto and Spoof, 2007).

20
 Chapter 2 - Background

Table 2.2 - Properties of the most frequent microcystins (Sivonen and Jones, 1999; De Maagd
et al., 1999; Cook and Newcombe, 2002; Newcombe et al., 2003).
Variable amino acids Molecular Net charge
Analogue LD50 a
X Z Weight (g/mol) (at neutral pH)
MC-LR Leucine Arginine 50 994 -1
MC-YR Tyrosine Arginine 70 1044 -1
MC-LY Leucine Tyrosine 90 1001 -2
MC-LW Leucine Tryptophane NR 1024 -2
MC-LF Leucine Phenylalanine NR 985 -2
MC-LA Leucine Alanine 50 909 -2
MC-RR Arginine Arginine 600 1037 0
a
Toxicity measured by mouse intraperitoneal (µg/kg); NR: not reported

Microcystins are relatively polar molecules and are soluble in water (Sivonen and Jones,

1999). Although the Adda residue gives them a partially hydrophobic character, microcystin-

LR and most of its congeners are hydrophilic and generally not able to penetrate vertebrate

cell membranes (Sivonen and Jones, 1999; Höeger, 2003; Van Apeldoorn et al., 2007).

Microcystin-LR becomes more hydrophobic in water with decreasing pH, but the pKa is very

low (De Maagd et al., 1999). Several microcystin variants have been identified as having

greater hydrophobicity than microcystin-LR, such as microcystin-LL, -LF, -LW, -LY, -LA,

-LV, and -LM (Svrecek and Smith, 2004; Cook and Newcombe, 2002). These cyclic peptides

are extremely stable and resistant to chemical hydrolysis or oxidation at near neutral pH. In

natural waters and in the dark, microcystins may persist for months or years. Microcystin-LR

is not degraded by sun-light alone but in the presence of photosensitizers (e.g., pigments,

humic acids), although the degradation will be of significance only in very shallow water

bodies (Sivonen and Jones, 1999).

21
 Chapter 2 - Background

2.3 LEGISLATION AND GUIDELINES

Cyanotoxins may cause waterborne disease mainly when ingested and through direct contact

during recreational exposure. The World Health Organization (WHO) therefore established

three guidance levels to recreational waters and two alert levels to raw drinking water

(Table 2.3). Furthermore, in 1998 WHO established a provisional drinking water guideline

value of 1.0 µg/L for MC-LR (cell-bound and extracellular). The WHO guideline value is

stated as being provisional since it covers only microcystin-LR, given that the toxicology is

limited and new data for toxicity of other cyanotoxins are being generated (WHO, 2006).

Table 2.3 - WHO guidelines for managing recreational waters and drinking waters which may
contain cyanobacteria (Falconer, 1999; WHO, 2003).
Risk level Density of algal cells
(guidance/alert level) Cyanobacterial (cells/mL) Chlorophyll-a (µg/L)
Low (Guidance 1 level) 20 000 10
Bathing waters Moderate (Guidance 2 level) 100 000 50
High (Guidance 3 level) Cyanobacterial scum formationa

Drinking Low (Vigilance) < 2000 <1

waters Medium (Alert 1 level) 2 000 1
(raw waters) High (Alert 2 level) 100 000 50
a
Scum in recreational areas, where whole body contact and/or ingestion/aspiration risk occur.

A number of countries have developed regulations or guidelines for microcystins in drinking

water (Table 2.4). Many are based on or directly incorporated the WHO guideline value,

including Czech Republic, France, Great Britain, China, Italy, Japan, Korea, Norway, Poland

(Burch, 2007), Canada, Australia, New Zealand, Brasil, Spain and, very recently, Portugal

through the DL 306/2007 (dating from 27 August). USA have no guideline values nor

standards, but cyanobacteria and their toxins have been listed on the USEPA contaminant

candidate list (CCL) for further research (Chorus, 2005). Some variability in the expression of

the concentration of microcystins has been observed, which in practice makes the values not

22
 Chapter 2 - Background

comparable. Sometimes the reference value is expressed in µg/L of microcystin-LR (e.g.,

WHO, Canada, Portugal), others in microcystin-LR equivalents (e.g., New Zealand),

microcystin-LR toxicity equivalents (e.g., Australia; it requires the use of toxicity equivalence

factors) and there is also some cases where there is a lack of information, being only referred

µg/L of microcystins (e.g., Brasil and Spain), with no variants specified. Internationally, the

main focus has been upon microcystins for they are widely regarded as the most significant

potential source of human injury from cyanobacteria on a world-wide scale. However, there is

a pronounced demand for a WHO guideline value for cylindrospermopsin (Chorus, 2005).

Table 2.4 - Guidelines/regulations of several countries for microcystins in drinking water

Guidelines/
Country Units Reference
Regulationa
WHO 1.0 µg/L MC-LR WHO, 2006
Australia 1.3 µg/L MC-LR (toxicity equivalent) Chorus, 2005; Burch, 2007
Brasil 1.0 µg/L Microcystins Chorus, 2005
Canada 1.5 µg/L MC-LR FPTCDW, 2002
New Zealand 1.0 µg/L MC-LReq MHNZ, 2005
Spain 1.0 µg/L Microcystins Chorus, 2005; Burch, 2007
Portugal 1.0 µg/L MC-LR DL 306/2007
a
Sum of cell-bound and extracellular toxin.

2.4 WATER TREATMENT FOR CYANOTOXINS REMOVAL

2.4.1 Removal of Cell-Bound Cyanotoxins

The technologies preferentially referred in scientific literature for the removal of

cyanobacteria and cell-bound cyanotoxins are coagulation/flocculation (C/F) with

sedimentation (S) or dissolved air flotation (DAF), and lately, membrane processes, like

ultrafiltration (UF) and nanofiltration (NF). All the referred technologies are reported to be

efficient for cells removal although with different removal efficiencies

23
 Chapter 2 - Background

(C/F/S < C/F/DAF < UF < NF). The main controversy is the effect of these technologies on

cell integrity, with some studies indicating no effect (Kenefick et al., 1993; Velzeboer et al.,

1995; Chow et al., 1998; 1999; Drikas et al., 2001a; Ribau Teixeira and Rosa, 2006 a, b,

2007), while others refer the occurrence of cell lysis (Lam et al., 1995; Hrudey et al., 1999;

Chow et al., 1999; Pietsch et al., 2002; Gijsbertsen-Abrahamse et al., 2006) mostly due to

physical (e.g. shear stresses induced by mixing, pressuring) and/or chemical (e.g., acid pH,

coagulant overdosing) stresses. Several works have demonstrated that oxidation induces

severe algae cell damage, with consequent toxin release (Tsuji et al., 1997; Hoeger et al.,

2002; Pietsch et al., 2002; Svrcek and Smith, 2004; Daly et al., 2007) and should therefore be

avoided in waters containing cyanobacterial cells. Further studies are still necessary to test

cell lysis occurrence, especially for membrane processes like UF, given that Chow et al.

(1997) and Gijsbertsen-Abrahamse et al. (2006) found a slight cell damaging (4-10% of

Microcystis cells damaging or 14-34% of Plantothrix filaments damaging), although the

quality permeate was not degraded.

2.4.2 Removal of Dissolved Cyanotoxins

Regarding dissolved cyanotoxins, the most studied processes are activated carbon adsorption

and oxidation, and more recently biodegradation and membrane filtration (NF and hybrid

adsorption/membrane processes).

PAC has demonstrated an efficient removal of microcystins, provided that the appropriate

carbon and the correct dose is applied (Falconer et al., 1989; Hart et al., 1998; Pendleton et

al., 2001; Cook and Newcombe, 2002; Schmidt et al., 2002; Newcombe and Nicholson,

2004). The exception is microcystin-LA, which is not readily removed by activated carbon,

probably due to electrostatic interactions (Cook and Newcombe 2002; Newcombe et al.,

24
 Chapter 2 - Background

2003; Ho, 2004). However, PAC performance depends on NOM competitive adsorption

(Donati et al., 1994). There is general agreement that to achieve high toxin removal efficiency

(> 85%), high doses of PAC are required (> 20 mg/L) and that the contact time is very

important (Keijola et al., 1988; Donati et al., 1994; Hart et al., 1998; Mouchet and Bonnélye

1998; Hrudey et al., 1999). Although very easy to apply to conventional water treatment

plants, PAC addition is responsible for a significant increase of treatment costs and sludge

production and is therefore only recommended for short periods of time.

GAC filtration has demonstrated efficiency for toxin removal (Falconer et al., 1989; Donati et

al., 1994; Lambert et al., 1996; Hart et al., 1998; Cook and Newcombe, 2002; Wang et al.,

2007), using both adsorption and biodegradation removal mechanisms and is recommended

for a continuous application. However, GAC filtration still has some problems, such as the

limited lifetime of adsorption due to quick GAC exhaustion by NOM, which may result in

toxin breakthrough after a very short period of time (Lambert et al., 1996). The lifetime of

GAC was referred to vary between one month to more than one year depending on the type of

toxin and the water quality (Hart and Stott, 1993; Hart et al., 1998; Newcombe, 2002;

Newcombe and Nicholson, 2004). Another problem is the desorption of toxin after

discontinuing the toxin input, as observed by Mesquita et al. (2006) for MC-LR, which brings

concerns about cyanotoxin release after cyanobacterial blooms. Besides, it is difficult to

understand the lifetime of a GAC filter and its overall removal efficiency as a function of

operation time (Wang et al., 2006).

Biodegradation is an efficient removal mechanism and may substantially increase the GAC

lifetime for it allows a continuous regeneration. However, it may only occur after a lag period

of several days or even months, due to microorganisms acclimatization to toxins (Drikas et

25
 Chapter 2 - Background

al., 2001 b; Newcombe et al., 2002; MHNZ, 2005; Ho et al., 2006 b). NOM was

demonstrated to be important for MC-LR biodegradation, because the process is probably co-

metabolic and depends on the available assimilable organic carbon (AOC), and MC-LR

desorption is intensified by the presence of other soluble organic compounds in water

(Mesquita et al., 2006). The biodegradation shows great potential as a low cost, low

technology process, particularly if the optimum conditions may be identified and, perhaps,

imposed on the filter (Newcombe et al., 2003; Newcombe and Nicholson, 2004). However,

factors affecting the biological removal, such as biofilm mass and composition, acclimation

periods, temperature and water quality, may not be easily controlled (Ho et al., 2006 b; Wang

et al., 2007) and further research is required.

Dissolved cyanotoxins have been found to be efficiently degraded by oxidants, such as

chlorine, permanganate and especially, ozone (Keijola et al., 1988; Nicholson et al., 1994;

Tsuji et al., 1997; Rositano et al., 1998, 2001; Acero et al., 2005; Ho et al., 2006 a; Daly et

al., 2007; Rodríguez et al., 2007 a, b). However, chlorine is relatively ineffective for

anatoxin-a and MC-LA oxidation (Newcombe and Nicholson, 2004; Ho et al., 2006 a;

Rodríguez et al., 2007 a, b). Oxidation depends on several conditions, difficult of

simultaneously control, e.g., residual dose, contact time and water quality, particularly the

NOM content, pH, alkalinity and temperature (Hart and Stott, 1993; Nicholson et al., 1994;

Tsuji et al., 1997; Rositano et al., 1998, 2001; Senogles et al., 2000; Hoeger et al., 2002;

Acero et al., 2005; Ho et al., 2006 a; Daly et al., 2007; Rodríguez et al., 2007 a, b). The

formation of disinfection by-products, such as trihalomethanes (THM) and bromate, may be a

restriction to the application of sufficiently high oxidant doses (Rodríguez et al., 2007 a, b).

The incomplete oxidation of toxins, producing compounds of unknown toxicity is also a

concern, although the characterization of the decomposition products and their potential

26
 Chapter 2 - Background

health implications has yet not been adequately addressed (Hoeger et al., 2002; Newcombe et

al., 2002; Svrcek and Smith, 2004; MHNZ, 2005).

Most of the research on dissolved cyanotoxins removal by membrane processes studied

nanofiltration. NF was proven to be efficient for cyanotoxins removal (Hart and Stott, 1993;

Muntisov and Trimboli, 1996; Smith et al., 2002; Ribau Teixeira and Rosa, 2005 a, 2006 b;

Gijsbertsen-Abrahamse et al., 2006). Ribau Teixeira and Rosa (2005 a, 2006 b, c) and

Gijsbertsen-Abrahamse et al. (2006) achieved a microcystins removal efficiency above 97%

and a anatoxin-a rejection ≥ 96%, regardless of the variations in feed water quality (NOM and

competitive toxin), the water recovery rate (up to 84%) and the pH values (Ribau Teixeira and

Rosa, 2005 a, 2006 b, c). The major NF drawbacks were the permeate fluxes, which may be

significantly impacted by background organics (NOM and microcystins) (Ribau Teixeira and

Rosa, 2005 a, 2006 c, d) and, especially, inorganics (pH and calcium) (Ribau Teixeira and

Rosa, 2005 b, 2006 c), although it was possible to operate at very high water recovery rates.

Membrane hybrid processes, like PAC/UF, have been referred to be promising for

cyanotoxins removal (Mouchet and Bonnélye, 1998; Zhou and Smith, 2002), but published

data on this application are still rather scarce. Lee and Walker (2006) used cellulose acetate

(20 kDa) and polyethersulphone (5 and 20 kDa) flat sheet membranes to study MC-LR

rejection by PAC/UF. The authors obtained 97.7-98.8% MC-LR rejection in the absence of

NOM and 77.7- 89.6% with 5 mg/L of fulvic acid addition (50 µg/L MC-LR; 5 mg/L PAC;

4 h contact time). Further research is recommended addressing conditions (module type,

operating conditions) closer to full-scale applications.

27
 Chapter 2 - Background

2.5 PAC ADSORPTION

2.5.1 General

A typical activated carbon particle has a porous structure consisting of a network of

interconnected macropores (> 50 nm in diameter), mesopores (2-50 nm), secondary

micropores (0.8-2 nm) and primary micropores (< 0.8 nm) that provides a good capacity for

the adsorption of organic molecules due to its high surface area (Aksu and Kabasal, 2005).

Although the great majority of the structures are composed of carbon, all activated carbon

contain some heteroatoms, mainly oxygen and hydrogen. PAC has a charge in aqueous

solution, which is often attributed to oxygen surface groups (Newcombe and Cook, 2004).

Both the raw material and the activation conditions affect activated carbon pore structure and

surface chemistry (Quinlivan, 2001). The primary difference between PAC (powdered) and

GAC (granular) is the smaller particle size of PAC, typically 65-90% of it passing a 44 µm

sieve.

2.5.2 Mechanisms of Adsorption

For adsorption to take place, several transport steps must occur. The molecular size of the

adsorbate and the adsorbent particle size have both important effects on the rate of adsorption.

As the molecular size increases, the diffusion coefficients decrease, and thus longer time is

required to remove the large-molecular-weight substances. The adsorbent particle size

determines the time required for transport within the pore to available adsorption sites

(Snoeyink and Summers, 1999). The transport steps are (Snoeyink and Summers, 1999,

Newcombe and Cook, 2004):

1) Bulk diffusion to the boundary layer surrounding the PAC particle;

28
 Chapter 2 - Background

2) Diffusion through the boundary layer (stationary layer of water) to the external carbon

surface (external mass transfer or film diffusion);

3) Diffusion through the pore structure to the most favourable adsorption site

(intraparticle diffusion);

4) Adsorption.

These steps are influenced by a range of factors. Step 1 is affected by the molecular

dimensions and shape of the adsorbate, but proper mixing may minimise these effects. In step

2, diffusion depends on flow rate (the higher the flow, the shorter the distance), and on

adsorbate dimensions and shape, although it is generally considered to be fast under proper

mixing. Step 3 is affected by pore structure (both external and internal), and molecular

dimensions and shape of the adsorbate. In practical situations, step 3 is most likely to be rate-

limiting (Newcombe and Cook, 2004). Step 4 is very rapid for physical adsorption and, in that

case, one of the preceding diffusion steps controls the adsorption rate. When chemical

adsorption occurs, step 4 may be slower than the diffusion steps and may therefore control the

rate of compound uptake (Snoeyink and Summers, 1999). Physisorption is a readily reversible

reaction and includes both mono and multilayer coverage. Chemisorption may be reversible

(high adsorption energies), involves only monolayer coverage and is a site specific reaction,

occurring at specific functional group locations (Alley, 2006).

The mechanism and extension of adsorption have been shown to depend on: i) adsorbate

structure (size, shape, charge, polarity, functional groups); ii) PAC characteristics (pore

volume distribution, surface charge, and hydrophobicity); iii) solution conditions (pH, ionic

strength, temperature, adsorbate concentration and presence of competing compounds

(Newcombe et al., 1997; Newcombe and Cook, 2004; Aksu and Kabasal, 2005). Both surface

chemistry and pore volume distribution of PAC will play a role in the adsorption, however, its

29
 Chapter 2 - Background

relative importance will vary depending on the adsorbates and carbons. For NOM, at pH 7,

there is strong evidence that electrostatic effects are determinant for adsorption (Newcombe et

al., 1997; Bjelopavlic et al., 1999; Quinlivan, 2001; Newcombe and Cook, 2004). On the

other hand, for microcystins, Pendleton et al. (2001) and Newcombe and Cook (2002)

concluded about the major influence of the pore volume distribution (with a positive

correlation with the volume of secondary micropores and mesopores) whereas Donati et al.

(1994) and Pendleton et al. (2001) showed no significant effect of the carbon surface

chemistry.

The relationship between the size of the adsorbate and the PAC pore size distribution is very

important, since it determines the fraction of the total pore volume that can be accessed.

Compounds are preferentially adsorbed in a pore of approximately its size, where there will

be greater number of contact points and more favourable adsorption energy (Newcombe et al.,

1997, Pelekani and Snoeyink, 1999) and they are size excluded if pores are too small

compared to their size and shape. A correct pore size distribution provides not only the

adsorption sites, but also the appropriate channels to transport, as a high volume of large

transport pores (macro and mesopores), favours rapid diffusion to adsorption sites

(Newcombe et al., 2002).

2.5.3 NOM Competition

PAC performance for microcontaminant removal is greatly affected by NOM competitive

adsorption, with pore blockage and direct site competition being considered the most likely

competing mechanisms. Direct competition is responsible for a reduction in the adsorption

capacity for the target compound (Snoeyink and Summers, 1999) and occurs when the

competing compounds are able to access the same sites (i.e., when the target and the

30
 Chapter 2 - Background

competing compounds have similar size), or when the target compound adsorbs in a larger

pore (with lower adsorption energy) and the larger competing compound (with higher

adsorption energy) is able to displace it (Newcombe et al., 2002). Sometimes, competing

molecules may not adsorb on the same sites as the target compound, because pores are too

small, but are capable of constricting or blocking pores and disturb the target compound

transport to final adsorption sites, reducing its rate of adsorption (Snoeyink and Summers,

1999). Several authors concluded that a broadening of PAC pore size distribution could

reduce, and even avoid, pore blockage by NOM (Donati et al., 1994; Newcombe et al., 1997;

Pelekani and Snoeyink, 1999, 2001; Ebie et al., 2001; Li et al., 2003 a, b; Quinlivan et al.,

2005).

Several studies showed that the low molecular weight NOM compounds exerted higher

competitive effect on micropollutants, which was directly associated with a direct site

competition mechanism. However, recent studies have showed that the smaller NOM

compounds may also participate in pore constriction/blockage (Pelekani and Snoeyink, 1999,

2001; Ebie et al., 2001; Newcombe et al., 2002; Li et al., 2003 a, b). As suggested by Kilduff

et al. (1998), these two mechanisms become indistinguishable as the competing and target

compounds become closer in size.

2.5.4 Adsorption Isotherms

An adsorption isotherm relates the equilibrium surface concentration (qe, i.e., the amount of

adsorbate adsorbed per unit mass of adsorbent, expressed in µg/mg) with the equilibrium

aqueous concentration (Ce, expressed in µg/L). Several models have been developed,

including the well-known models proposed by Freundlich and Langmuir. The Freundlich

isotherm is represented by (2.1):

31
 Chapter 2 - Background

1
q e = KC e n (2.1),

where K is a unit-capacity parameter (amount adsorbed at a value of Ce equal to unity)

(µg/mg)(L/µg)1/n and n is a dimensionless parameter related to the site-energy distribution.

This model applies if the data fit a straight line when plotted on a log-log basis (K is obtained

from the y-axis intercept and 1/n from the slope). For favourably adsorbed chemicals, 1/n is

between 0 and 1 (Alley, 2006), and the smaller the value of 1/n, the stronger the adsorption

energy. Freundlich equation cannot be applied when saturation is achieved (qe is constant and

independent of Ce increase) (Snoeyink and Summers, 1999).

Langmuir equation handles the interaction between the adsorbent and the adsorbate as a

linear, reversible, monolayer chemical reaction. The Langmuir equation is based on four

assumptions: i) the surface of the adsorbent is uniform (all adsorption sites are equal); ii)

adsorbed molecules do not interact; iii) all adsorption occurs through the same mechanism

and iv) at the maximum adsorption, only a monolayer is formed, meaning that molecules of

adsorbate do not deposit on other already adsorbed (Alley, 2006). The Langmuir equation is

expressed by (2.2), and in its linear form by (2.3):

Qmax bC e
qe = (2.2),
1 + bC e

Ce 1 C
= + e (2.3),
q e Qmax b Qmax

where Qmax (µg/mg) and b (L/µg) are Langmuir constants related to maximum adsorption

capacity and energy of adsorption, respectively. The linear plot of Ce/qe vs. Ce is an obligatory

condition for the application of Lagmuir model (Ong et al., 2007).

32
 Chapter 2 - Background

2.5.5 Adsorption Kinetic Models

Kinetic models are used to investigate adsorption mechanisms and to identify the potential

rate-controlling step(s). Some of the most commonly used models are the pseudo-first-order

adsorption model, the pseudo-second order adsorption model and the intraparticle diffusion

model.

The pseudo-first order adsorption model generally takes the form:

ln(q e − q t ) = ln(q e ) − K 1t (2.4),

where qe and qt are the surface concentration (µg/mg) at equilibrium and at time t,

respectively, and K1 is the equilibrium rate constant (min-1). A straight line of

ln (qe-qt) vs. t (min) suggests the applicability of this kinetic model and the adsorption rate

constant, K1, is obtained from the slope of the linear plot. In order to use this model, the

equilibrium sorption capacity must be known (qe), by extrapolating the experimental data to

t = ∝ or by using a trial-and error-method (Aksu and Kabasal, 2005).

The pseudo-second order adsorption model has the linear form:

t 1 1
= + t (2.5),
q t h qe

where h is the initial adsorption rate (µg/(mg.min)) given by K2qe2 and K2 (mg/(µg.min)) is

the rate constant. If the pseudo-second order model is applicable, the plot of t/qt against t

should give a linear relationship from which the constants qe, h and K2 are determined. The

pseudo-second-order equation is in agreement with chemisorption being the rate controlling

step (Ho et al., 2000; Badmus et al., 2007).

33
 Chapter 2 - Background

The intraparticle diffusion equation, introduced by Weber and Morris, is given by (2.6):

qt = Kit0.5+ C (2.6),

where Ki is the intraparticle diffusion rate constant (µg/(mg.min0.5)) and C (µg/mg) is a

constant proportional to the boundary layer thickness (Yalçin et al., 2004). If intraparticle

diffusion occurs, then the experimental qt data against t0.5 will be linear and Ki values may be

obtained from the slopes of the straight-line portions and C from the y-axis intercept. Such

plots may present a number of linear portions implying that two or more steps occur. The

first, sharper portion is the external surface adsorption stage, the second is the intraparticle

diffusion adsorption stage and the third portion is the final equilibrium stage (Yalçin et al.,

2004, Aksu and Kabasal, 2005). If the intra-particle diffusion is the rate limiting step of

adsorption, the line will pass through the origin, otherwise, external mass transfer resistance

may not be neglected (Abdelwahab, 2007). The kinetic data could be further analysed using

the kinetics expressions derived by Boyd, Adamson and Myers (Reichenberg, 1953):

Bt = −2.30258 log10 (1 − F ) − 0.49770 (2.7),

Bt = 6.28318 − 3.2899 F − 6.28318(1 − 1.0470)1 / 2 (2.8),

where Bt is a mathematical function of F (F = qt/qe) and is computed from (2.7) for F values

above 0.85 and from (2.8) for F values between 0 and 0.85 (Reichenberg, 1953). The linearity

of the plot of Bt values against time provides useful information to distinguish between

external-transport (non linear) and intraparticle-transport-controlled rates of adsorption (linear

and passing through origin) (Wang and Li, 2007).

34
 Chapter 2 - Background

2.6 ULTRAFILTRATION (UF)

2.6.1 General

Ultrafiltration is a low-pressure driven membrane process, usually applied for the removal of

particulate and microbial contaminants. UF may be operated under positive (typically 0.2-

2.8 bar) or negative pressure (between -0.2 to -0.8 bar) (USEPA, 2001). Most UF membranes

used for water treatment have a molecular weight cut-off (MWCO) of 100 kDa (i.e., they

reject more than 90% of the compounds above 100 kDa) and nominal pore size of 0.01 µm

(USEPA, 2005). UF membranes are either symmetric or asymmetric and are made from a

wide variety of materials, including polyvinyl difluoride, polysulfone, polyethersulfone and

cellulose acetate. Each material confers specific properties, including resistance to pH and

oxidant, surface charge and hydrophobicity, which affect the exclusion characteristics

(selectivity) of a membrane as well as the operating constraints, such as the potential use of

pre-chlorination to control biological fouling (USEPA, 2001).

The most commonly available UF modules for water treatment are hollow-fibre (HF)

modules, designed to facilitate backwash and yield a high surface area to volume ratio (Zhou

and Smith, 2002; USEPA, 2005). Although specific dimensions vary by manufacturer,

approximate ranges for internal fiber diameter is 0.3-1.0 mm and for fibre length is 1-2 meters

(USEPA, 2005). HF membranes may be operated in either an inside-out or outside-in mode.

Systems may operate in dead-end mode of operation and are periodically backwashed to

remove the accumulated solids (USEPA, 2005) (Figure 2.2). Some systems use cross-flow

(feed flow tangential to the membrane surface) (Figure 2.2), sometimes with the possibility to

work on dead-end by simply closing the discharge valve (Pilutti and Nemeth, 2003). Cross-

flow mode maintains a high scour velocity across the membrane surface, limiting the extent

35
 Chapter 2 - Background

of particle deposition and cake layer formation, but requires additional pumping for

concentrate recirculation and may substantially increase the operating costs (USEPA, 2001,

2005). In this operation mode, the concentration of suspended material on the feed side of the

membrane is higher than in the feed stream, given that the concentrate stream is recirculated

and mixed with the incoming feed stream (USEPA, 2005).

Figure 2.2 - Hydraulic mode of operation: dead-end (left) and cross-flow (right)
(USEPA, 2005).

The operating flux (permeate flow per unit of membrane area; L/(h.m2)) is one of the critical

parameters of UF, affected by water quality and membrane fouling. Membrane systems

running under critical flux result in maximum flux without significant fouling (Choi, 2003).

Another important parameter is water recovery (the ratio of feed water that is converted to

permeate), which is typically 85-97 % for UF systems. Typical UF systems incorporate a

backwash cycle (Figure 2.3) with air, water (sometimes with chlorine) or a combination of

both. Backwashing frequency depends on the feed water quality and usually increases with

flux and recovery rates, due to higher operating pressure and higher solids concentration in

the feed water. Typical backwash frequency for UF is 15-60 minutes, with backwash duration

from 30 s to 3 minutes. Ideally the backwash restores the transmembrane pressure (TMP), but

most systems experience a gradual increase of TMP that must be addressed by chemical

cleaning (acid, caustic, chlorine and/or surfactants). Most systems are chemically cleaned

every one to six months, depending on the system design and operation (USEPA, 2001).

36
 Chapter 2 - Background

Figure 2.3 - Typical UF operation in most applications (Kim et al., 2001).

Size exclusion by the membrane is the primary mechanism of UF separation. However,

electrostatic repulsion and adsorption may significantly affect rejection, especially for

compounds with dimensions similar to the size of UF membrane pores (Wiesner and Buckley,

1996; Scott, 1998). Low molecular-weight solutes, which may pass through UF pores, have

been showed to cause drastic flux reductions due to adsorption (Jönsson et al., 1997). In

addition, these solutes may interact with larger molecules and be retained within the cake

layer, especially when the layer is compressible (Choi, 2003).

2.6.2. Mass Transport and Fouling

Membrane separation involves various mass transport steps and many models have been

developed to characterise it, not only through the membrane, but also in the membrane

boundary layer. Mass transport may lead to the attachment, accumulation, or adsorption of

materials onto membrane surfaces and (or) within membrane pores, causing membrane

fouling.

37
 Chapter 2 - Background

Concentration polarization (CP) is a very important mass transport phenomenon on the

membrane boundary layer. CP is the result of transport and rejection of solutes by membrane

leading to an accumulation of these solutes on membrane wall and creating a solute

concentration gradient for back-diffusion on the boundary layer. This concentration gradient

is responsible for a maximum permeate flux (mass transport-limited flux), and depends on

equilibrium between convective flux and diffusive flux (Wiesner and Aptel, 1996; Scott,

1998; Taylor and Wiesner, 1999). Concentration polarization effects have been shown to be

small in comparison with pore blockage and surface deposition (Taylor and Wiesner, 1999).

However, CP is often a precursor to cake or gel formation and the high concentrations near

the membrane tend to exacerbate precipitative or adsorptive membrane fouling. CP is more

pronounced at higher TMP, lower cross-flow velocity (CFV), and any other conditions which

bring solute to the membrane surface very rapidly (e.g., a very porous membrane). Increased

back transport can be achieved by adoption of turbulence promoters/inserts/baffles to the

membrane system, frequent backflushing of membrane, permeate backpressure, intermittent

jets, pulsating flow, ultrasonic vibration, and establishing an electric field near the membrane

(Choi, 2003).

Membrane fouling is characterised by a reduction in the permeate flux as a result of increased

flow resistance due to concentration polarization, pore blocking (complete, standard or

intermediate) and/or cake formation (compressible or incompressible) (Figure 2.4). Of

particular importance is the size distribution of feed solution components relative to the

membrane pore size (Zhou and Elimelech, 1997; Costa et al., 2006). In the standard blocking

(or pore constriction), the molecules are much smaller than the pores and fouling occurs

because they deposit inside the pores, decreasing their size. In complete blocking (pore

plugging), the molecules are of the same size as pores and block the entire pore entrance,

38
 Chapter 2 - Background

decreasing the total pore area. The intermediate blocking considers particles directly blocking

the pores and also settling on other particles previously deposited, reducing the total pore

area. In cake filtration, particles are much larger than pores and deposit on the membrane

surface, forming a cake and adding a second resistance layer to the membrane surface (Neal,

2006). The cake can be classified as compressible or incompressible depending on the nature

of the particles, such as shape, particle size distribution and rigidity. Fouling inside the

membrane pores (adsorption) is often considered as an irreversible process. Concentration

polarization and surface cake are considered reversible by decreasing the applied pressure or

the permeate flux or by backwash, respectively (Zhou and Smith, 2002).

Cb Sb Ib Ck
Figure 2.4 - Mechanisms of membrane fouling: Cb, complete blocking; Sb, standard blocking
(pore constriction); Ib, intermediate blocking; Ck, cake formation.

Pore plugging is an important fouling mechanism for UF, in addition to particle cake

formation on the membrane surface (Zhou and Elimelech, 1997). Several studies point out a

transition of the fouling mechanism with the filtration time, specifically from pore blocking to

cake formation above the blocked regions (Bowen et al., 1995; Ho and Zydney, 2000; Costa

et al., 2006).

Membrane fouling depends on such parameters as membrane properties, characteristics of the

feed water and hydrodynamic conditions (Anselme and Jacobs, 1996; Choi, 2003). Important

membrane characteristics are pore size, charge, and roughness, with more permeable

membranes showing faster flux decline (Cho and Amy, 1999; Yuan and Zydney, 2000; Costa

39
 Chapter 2 - Background

and Pinho, 2005; Costa et al., 2006), probably associated with higher fluxes and higher

convective transport (Costa et al., 2006). Other important aspect is the feed water

characteristics, i.e., pH, ionic strength, calcium concentration, NOM content and character

(Hong and Elimelech, 1997; Cho and Amy, 1999; Costa et al., 2006). Several recent studies

have shown that the hydrophilic neutral NOM is determinant for UF fouling (Her et al., 2004;

Lee et al., 2004, 2006; Kwon et al., 2005; Kimura et al., 2006; Yamamura et al., 2007).

Hydrodynamic conditions are also determinant for membrane fouling, like CFV and TMP that

governs the permeate flux and the convective transport of foulants towards the membrane

(Hong and Elimelech, 1997). Costa et al. (2006) concluded that at high pressures the structure

of the fouling layer is changed (the compactness and the specific resistance substantially

increase), and significantly affect the extent of flux decline. CFV influences NOM fouling

and its increment improves back-transport into the bulk solution, reducing the rate of cake

deposition and the resultant cake thickness (Zularisam et al., 2006).

2.7 PAC/UF

2.7.1 General

PAC/UF is an emergent hybrid membrane-adsorption system that incorporates the adsorption

capabilities of activated carbon and the microorganism and particle removal ability of the UF

membranes (Snoeyink et al., 2000). In PAC/UF, PAC particles are applied to the influent of a

UF membrane reactor and are retained by it, making possible the removal of the adsorbed

dissolved organic compounds (Clark et al., 1996). There are already several PAC/UF full-

scale installations and the biggest ones are in Lausanne (Switzerland), Vigneux (France),

Kopper (Slovenia) and San Antonio (Texas) (Amy, 2007), where the main applications are

pesticides, taste and odours, and disinfection-by-products control (Clark et al., 1996).

40
 Chapter 2 - Background

PAC/UF integrates the advantages of both UF and PAC adsorption and minimises some of

their disadvantages. Some of the advantages of PAC/UF compared to higher-pressure

membrane processes are (Clark et al., 1996; Pilutti and Nemeth, 2003): a) the use of a low-

pressure membrane, and therefore with lower operating cost, to remove dissolved organic

compounds; b) the process flexibility since the PAC type and doses may be quickly changed

according with the water quality; also, UF may work without PAC addition, at dead-end mode

of operation, minimising the operating costs, and being changed to cross-flow only when

necessary (semi dead-end operation); c) unlike the spiral-wound modules commonly used in

nanofiltration and reverse osmosis, UF hollow fibre modules can be backflushed, minimising

the frequency of chemical cleaning, and some of them have high tolerance to chlorine.

Relatively to PAC conventional applications, the use of an UF membrane to retain PAC

particles allows: a) higher disinfection capacity; b) the use of smaller PAC particles, with

faster adsorption kinetics, but still with very efficient separation; c) the potential regeneration

of PAC, given that it may be isolated from other reagents; d) the recirculation of PAC during

the filtration cycles, which enhances the carbon residence time and adsorption efficiency

while minimising the sludge production.

The effect of PAC on membrane fouling is controversial and requires further research. In

recent studies, some authors referred a positive impact (Konieczny and Klomfas, 2002; Lee et

al., 2007), others a neutral effect (Mozia and Tomaszewaska, 2004; Matsui et al., 2006) or a

negative impact (Zhao et al., 2005; Zularisam et al., 2007). The effect of PAC seems to

depend on the membrane hydrophobicity and the feed water characteristics, but studies on this

subject are rather scarce and inconclusive.

41
 Chapter 2 - Background

To my knowledge, only one study has been published on microcystins removal by PAC/UF

(Lee and Walker, 2006). These authors found high removal efficiencies (78-99%), although

the type of module (flat-sheet) and the operating conditions were not the ones currently used

in PAC/UF full-scale applications.

2.7.2 Operating Conditions

The adsorption performance of PAC/UF depends on the operating conditions, such as

backwashing frequency, reactor size and configuration, filtration mode (dead-end versus cross

flow) and PAC dosing procedure (Campos et al., 2001). One important parameter is the PAC

residence time on the membrane reactor, which is determined by PAC dosing and wasting

procedures. PAC may be dosed at a constant rate (i.e., step input) or at the beginning of the

filtration cycle (i.e., pulse input) and be wasted either continuously or at one time during

backwash. One approach is to maintain constant the concentration of PAC in the membrane

reactor, by continuously adding and wasting PAC, which will also maintain constant the

adsorbate concentration in the permeate (Campos et al., 2000 a). However, in PAC/UF full-

scale applications, PAC is continuously dosed but is wasted only at the end of the filtration

cycle (Campos et al., 2000a), making the permeate adsorbate concentration not constant

throughout the filtration cycle (Figure 2.5). When PAC is added at the beginning of the cycle

and wasted during backwashing, the carbon retention time is the duration of the filtration

cycle, which is typically between 30 and 90 minutes (Baudin et al., 1997). In order to achieve

longer contact times in the latter application, PAC may be dosed in a continuous-flow reactor

(such as a stirred tank reactor (CSTR) or a plug flow reactor (PFR)) placed ahead of the

membrane reactor (Figure 2.6).

42
 Chapter 2 - Background

C/C0

Fig. 2.5 - Adsorbate concentration profiles (C/C0) in the permeate resulting from wasting
PAC at the end of the filtration cycle and using two PAC dosing procedures
(Snoeyink et al., 2000).

Figure 2.6 - Schematics of PAC/UF process configuration (Snoeyink et al., 2000).

43
 Chapter 2 - Background

Another strategy to maximize the carbon surface loading is to recirculate the partially loaded

PAC from the UF loop to an upflow floc blanket reactor (FBR), installed upstream of the

PAC/CSTR-UF process and used for clarification (Baudin et al., 1997; Campos et al.,

2000 a, b; Snoeyink et al., 2000).

2.7.3 Waste Disposal

Membranes generate a concentrated waste that must be disposed according to applicable laws.

UF waste consists of backwash waters in the case of dead-end operation and of backwash

waters and concentrate stream in cross-flow operation. In the hybrid process, the quantity and

composition of PAC/UF waste depends on the rejection and water recovery, but it usually

corresponds to a volume of 5-15% of the feed stream and consists of PAC (with adsorbed

compounds) and other particulate matter and macromolecules retained (Clark et al., 1996).

The waste requires a treatment due to the fairly high solids concentration (> 200 ppm) (Clark

et al., 1996) which sometimes contains high toxic compounds (as cyanotoxins and pesticides).

The best option is the regeneration and recovery of PAC, but its application needs further

research. Anselme and Jacobs (1996) have cited some works where good results where

obtained with PAC regeneration (mass loss of 18-22% and more than 90% of adsorption

capacity recovered). Shende and Mahajani (2002) recovered more than 98% of the adsorption

capacity of PAC and GAC loaded with reactive dyes by wet oxidative regeneration (which

requires no complicated dewatering and drying as all material remains in a slurry).

When PAC/UF is used after C/F/S, the PAC can be returned to the settler and processed with

the normal sludge (Clark et al., 1996). The treatment usually involves dewatering

44
 Chapter 2 - Background

(clarification and/or thickening, followed by conditioning and dewatering by centrifuges or

press-filters) and after the waste goes to a wastewater treatment plant, is incinerated (Anselme

and Jacobs, 1996) or deposited in a landfill.

2.8 REFERENCES

Abdelwahab O. (2007). Kinetic and isotherm studies of copper (II) removal from wastewater using
various adsorbents. Egyptian Journal of Aquatic Research, 33(1), 125-143.
Acero J., Rodriguez E., Meriluoto J. (2005). Kinetics of reactions between chlorine and the
cyanobacterial toxins microcystins. Water Research, 39, 1628-1638.
Aksu Z., Kabasakal E. (2005). Adsorption characteristics of 2,4-dichlorophenoxyacetic acid (2,4-D)
from aqueous solution on powdered activated carbon. Journal of Environmental Science and
Health Part B, 40, 545-570.
Alley R. (2006). Physical Treatment. In Water Quality Control Handbook. Second Edition, McGraw-
Hill, Lisbon.
Amy G. (2007). Hybrid low pressure membranes. In Membrane Technology in Drinking & Industrial
Water Treatment. Intensive short course, 1-5 October, UNESCO-IHE, Delft.
Anselme C., Jacobs E. (1996). Ultrafiltration. In Water Treatment-Membrane Processes, AWWARF,
Lyonnaise des Eaux and WRCSA (Ed.), McGraw-Hill, Lisbon.
Badmus M., Audu T., Anyata B. (2007). Removal of lead ion from industrial wastewaters by activated
carbon prepared from periwinkle shells (Typanotonus fuscatus). Turkish Journal Engineering
Environmental Science, 31, 251-263.
Baudin I., Chevalier M., Anselme C., Cornu S., Lainé J. (1997). L’Apié and Vigneux case studies:
first months of operation. Desalination, 113, 273–275.
Bjelopavlic M., Newcombe G., Hayes R. (1999). Adsorption of NOM onto activated carbon: effect of
surface charge, ionic strength, and pore volume distribution. Journal of Colloid and Interface
Science, 210(2), 271-280.
Bowen W., Calvo J., Hernandez A. (1995). Steps of membrane blocking in flux decline during protein
microfiltration. Journal of Membrane Science, 101, 153–165.
Burch M. (2007). Effective doses, guidelines & regulations. In Proceedings of the Interagency,
International Symposium on Cyanobacterial Harmful Algal Blooms. H. Kenneth Hudnell (Ed.).
North Carolina, Sept. 2005.
Campos C., Mariñas B., Snoeyink V., Baudin I., Laîné J. (2000 a). PAC-Membrane filtration process.
II. Model application. Journal of Environmental Engineering, 126(2), 104-111.

45
 Chapter 2 - Background

Campos C., Snoeyink V., Mariñas B., Baudin I., Laîné J. (2000 b). Atrazine removal by powdered
activated carbon in floc blanket reactors. Water Research, 34(16), 4070-4080.
Campos C., Baudin I., Laîné J. (2001). Adsorption performance of powdered activated carbon-
ultrafiltration systems. Water Science and Technology: Water Supply, 1 (5/6), 13-19.
Carmichael W. (1994). Toxins of cyanobacteria. Scientific American, 270(1), 78-86.
Carmichael W. (1997). The cyanotoxins. Advances in Botanical Research, 27, 211-256.
Carmichael W. (2001). Health effects of toxin-producing cyanobacteria. Human and Ecological Risk
Assessment, 7, 1393-1407.
Cho J., Amy G. (1999). Interactions between natural organic matter and membranes: rejection and
fouling. Water Science and Technology, 40(9) 131-139.
Choi Y. (2003). Critical Flux, Resistance and Removal of Contaminants in Ultrafiltration (UF) of
Natural Organic Materials. PhD thesis. Pennsylvania State University, USA.
Chorus I. (2005). Current Approaches to Cyanotoxin Risk Assessment, Risk Management and
Regulations in Different Countries. Ingrid Chorus (Ed.), Federal Environmental Agency.
Chow C., Panglisch S., House J., Drikas M., Burch M., Gimbel R. (1997). A study of membrane
filtration for the removal of cyanobacterial cells. Journal of Water Supply Research and
Technology – Aqua, 46(6), 324-334.
Chow C., House J., Velzeboer R., Drikas M., Burch M., Steffensen D. (1998). The effect of ferric
chloride flocculation on cyanobacterial cells. Water Research, 32(3), 808-814.
Chow C., Drikas M., House J., Burch M., Velzeboer, R. (1999). The impact of conventional water
treatment processes on cells of the cyanobacterium Microcystis aeruginosa. Water Research,
33(15), 3253-3262.
Clark M., Baudin I., Anselme C. (1996). Membrane-powdered activated carbon reactors. In Water
Treatment-Membrane Processes. AWWARF, Lyonnaise des Eaux and WRCSA (Ed.), McGraw-
Hill, Lisbon.
Codd, G. (2000). Cyanobacterial toxins, the perception of water quality, and the prioritisation of
eutrophication control. Ecological Engineering, 16(1), 51–60.
Cook D., Newcombe G. (2002). Removal of microcystin variants with powdered activated carbon.
Water Science and Technology: Water Supply, 2(5/6), 201-207.
Costa A., Pinho M.N. (2005). Effect of membrane pore size and solution chemistry on the
ultrafiltration of humic substances solutions. Journal of Membrane Science, 255(1–2), 49–56.
Costa A.R., Pinho M. N., Elimelech M. (2006). Mechanisms of colloidal natural organic matter
fouling in ultrafiltration. Journal of Membrane Science, 281, 716-725.
Daly R., Ho L., Brookes J. (2007). Effect of chlorination on Microcystis aeruginosa cell integrity and
subsequent microcystin release and degradation. Environmental Science and Technology, 41,
4447-4453.

46
 Chapter 2 - Background

De Maagd P., Hendriks A., Seinen W., Sijm D. (1999). pH-Dependent hydrophobicity of the
cyanobacteria toxin microcystin-LR. Water Research, 33(3), 677-680.
Donati C., Drikas M., Hayes R., Newcombe G. (1994). Microcystin-LR adsorption by powdered
activated carbon. Water Research, 28(8), 1735-1742.
Drikas M., Chow C., House J., Burch M. (2001a). Using coagulation, flocculation and settling to
remove toxic cyanobacteria. Journal of the American Water Works Association, 93(2), 100-111.
Drikas M., Newcombe G., Nicholson B. (2001b). Water treatment options for cyanobacteria and their
toxins. Proceedings of the Water Quality Technology Conference, American Water Works
Association, Nashville (Denver, Colorado), 11–15 November.
Ebie K., Li F., Azuma Y., Yuasa A., Hagishita T. (2001). Pore distribution effect of activated carbon
in adsorbing organic micropollutants from natural water. Water Research, 35(1), 167-179.
Falconer I., Runnegar M., Buckleley T., Huyin V., Bradshaw P. (1989). Using activated carbon to
remove toxicity from drinking water containing cyanobacterial blooms. Journal of American Water
Works Association, 2, 102-105.
Falconer I. (1999). An overview of problems caused by toxic blue-green algae (cyanobacteria) in
drinking water. Environmental Toxicology, 14, 5-12.
FPTCDW (2002). Cyanobacterial Toxins − Microcystin-LR. In Guidelines for Canadian Drinking
Water Quality-Supporting Documentation. Federal-Provincial-Territorial Committee on Drinking
Water, Canada.
Gijsbertsen-Abrahamse A., Schmidt W., Chorus I, Heijman S. (2006). Removal of cyanotoxins by
ultrafiltration and nanofiltration. Journal of Membrane Science, 276(1-2), 252-259.
Hart J., Stott P. (1993). Microcystin-LR removal from water. Foundation for Water Research, Report
FR 0367, Marlow, UK.
Hart J., Fawell J., Croll B. (1998). The fate of both intra- and extracellular toxins during drinking
water treatment. Water Supply, 16(1/2), 611-616.
Her N., Amy G., Park H., Song M. (2004). Characterizing algogenic organic matter (AOM) and
evaluating associated NF membrane fouling. Water Research, 38, 1427-1438.
Hitzfeld B., Hoeger S., Dietrich D. (2000). Cyanobacterial toxins: removal during drinking water
treatment, and human risk assessment. Environmental Health Perspectives, 108(1), 113–122.
Ho Y., McKay G., Wase D., Foster C. (2000). Study of the sorption of divalent metal ions on to peat.
Adsorption Science and Technology, 18, 639-650.
Ho C., Zydney A. (2000). A combined pore blockage and cake filtration model for protein fouling
during microfiltration. Journal of Colloid and Interface Science, 232(2), 389-399.
Ho L. (2004). The Removal of Cyanobacterial Metabolites from Drinking Water using Ozone and
Granular Activated Carbon. PhD thesis, University of South Australia, Australia.

47
 Chapter 2 - Background

Ho L., Onstad G, Von Gunten U., Rinck-Pfeiffer S., Craig K., Newcombe G. (2006 a). Differences in
the chlorine reactivity of four microcystin analogues. Water Research, 40(6), 1200-1209.
Ho L., Meyn T., Keegan A., Hoefel D., Brookes J., Saint C., Newcombe G. (2006 b). Bacterial
degradation of microcystin toxins within a biologically active sand filter. Water Research, 40(4),
768-774.
Höeger S., Dietrich D., Hitzfeld B. (2002). Effect of ozonation on the removal of cyanobacterial toxins
during drinking water treatment. Environmental Health Perspectives, 110(11), 1127-1132.
Höeger S. (2003). Problems During Drinking Water Treatment of Cyanobacterial-loaded Surface
Waters: Consequences for Human Health. PhD thesis, Universität Konstanz , Germany
Hong S., Elimelech M. (1997). Chemical and physical aspects of NOM fouling of nanofiltration
membranes. Journal of Membrane Science, 132, 159–181.
Hrudey S., Burch M., Drikas M., Gregory R. (1999). Remedial measures. In Toxic Cyanobacteria in
Water. A Guide to Their Public Health Consequences, Monitoring and Management. 1st edition. I.
Chorus and J. Bartram (Eds.), WHO, E& SPON, London.
Jones G., Orr P. (1994). Release and degradation of microcystin following algicide treatment of a
Microcystis aeruginosa bloom in a recreational lake, as determined by HPLC and protein
phosphatase inhibition assay. Water Research, 28(4), 871-876.
Jönsson A., Lindau J., Wimmerstedt R., Brinck J., Jönsson B. (1997). Influence of the concentration of
a low-molecular weight organic solute on the flux reduction of a polyethersulphone ultrafiltration
membrane. Journal of Membrane Science, 135, 117-128.
Kaebernick M., Neilan B. (2001). Ecological and molecular investigations of cyanotoxin production.
FEMS Microbiology Ecology, 35, 1-9.
Kenefick S., Hrudey S., Peterson H., Prepas E. (1993). Toxin release from Microcystis aeruginosa
after chemical treatment. Water Science and Technology, 27(3-4), 433-440.
Keijola A., Himberg K., Esala A., Sivonen K., Hiisvirta L. (1988). Removal of cyanobacterial toxins
in water treatment processes: laboratory and pilot-scale experiments. Toxicity Assessment: An
International Journal, 3, 643-656.
Kilduff J., Karanfil T., Weber Jr. W. (1998). Competitive effects of displaceable organic compounds
on trichloroethylene uptake by activated carbon. II. Model verification and applicability to natural
organic matter. Journal of Colloid and Interface Science, 205, 280-289.
Kim D., Hong S., Cho J. (2001). Effect of cake layer structure on the backwash efficiency of hollow
fiber MF membranes. Journal of Environmental Science and Engineering, 3, 50-58.
Kimura K., Yamamura H., Watanabe Y. (2006). Irreversible fouling in MF/UF membranes caused by
natural organic matters (NOMs) isolated from different origins. Separation Science and
Technology, 41, 1331-1344.

48
 Chapter 2 - Background

Konieczny K., Klomfas G. (2002). Using activated carbon to improve natural water treatment by
porous membranes. Desalination, 147, 109-116.
Kwon B., Lee S., Cho J., Ahn H., Lee D., Shin H. (2005). Biodegrability, DBP formation and
membrane fouling potential of natural organic matter: characterization and controllability.
Environmental Science and Technology, 39, 732-739.
Lahti K., Rapala J., Fardig M., Niemel M., Sivonen K. (1997). Persistence of cyanobacterial
hepatotoxin, microcystin-LR in particulate material and dissolved in lake water. Water Research,
31(5), 1005-1012.
Lam A., Prepas E., Spink D., Hrudey S. (1995). Chemical control of hepatotoxic phytoplankton
blooms: implications for human health. Water Research, 29(8), 1845-1854.
Lambert T., Holmes C., Hrudey S. (1996). Adsorption of microcystin-LR by activated carbon and
removal in full scale water treatment. Water Research, 30(6), 1411-1422.
Lee N., Amy G., Croué J.-P., Buisson H. (2004). Identification and understanding of fouling in low-
pressure membrane (MF/UF) filtration by natural organic matter (NOM). Water Research, 38,
4511-4523.
Lee N., Amy G., Croué J.-P. (2006). Low-pressure membrane (MF/UF) fouling associated with
allochthonous versus autochonus natural organic matter. Water Research, 40, 2357-2368.
Lee J., Walker H. (2006). Effect of processes variables and natural organic matter on removal of
microcystin-LR by PAC-UF. Environmental Science and Technology, 40, 7336-7342.
Lee C., Bae S., Han S., Kang L. (2007). Application of ultrafiltration hybrid membrane processes for
reuse of secondary effluent. Desalination, 202, 239-246.
Li Q., Snoeyink V., Mariñas B., Campos C. (2003 a). Pore blockage effect of NOM on atrazine
adsorption kinetics of PAC: the roles of PAC pore size distribution and NOM molecular weight.
Water Research, 37, 4863-4872.
Li Q., Snoeyink V., Mariñas B., Campos C. (2003 b). Elucidating competitive adsorption mechanism
of atrazine and NOM using model compounds. Water Research, 37(20), 773-784.
Matsui Y., Sanogawa T., Aoki N., Mima S., Matsushita T. (2006). Evaluating submicron-sized
activated carbon adsorption for microfiltration pretreatment. Water Science and Technology, 6(1),
149-155.
Meriluoto J., Spoof L. (2007). Cyanotoxins: sampling, sample processing and toxin uptake. In
Proceedings of the Interagency International Symposium on Cyanobacterial Harmful Algal
Blooms. Chapter 21. H. Kenneth Hudnell (Ed.). RTP North Carolina, Sept. 2005.
Mesquita E., Menaia J., Rosa M.J., Costa V. (2006). Microcystin-LR removal by bench scale
biologically activated carbon filters. In Recent Progress in Slow Sand Alternative Biofiltration
Processes. R. Gimbel, N. Graham and R. Collins (Ed.). IWA, London.

49
 Chapter 2 - Background

Metcalf J., Codd G. (2004). Cyanobacterial toxins in the water environment-a review of current
knowledge. Foundation for Water Research, FR/R0009, United Kingdom.
MHNZ (2005). Guidelines for Drinking Water Quality Management for New Zealand. Vol. III-
Datasheets, second edition, Ministry of Health of New Zealand.
Mouchet P., Bonnélye V. (1998). Solving algae problems: French expertise and world-wide
application. Journal of Water Science Research and Technology - Aqua, 47(3), 125-141.
Mozia S., Tomaszewaska M. (2004). Treatment of surface water using hybrid processes-adsorption on
PAC and ultrafiltration. Desalination, 162, 23-31.
Muntisov M., Trimboli P. (1996). Removal of algal toxins using membrane technology. Water –
Journal of the Australian Water Association, 23(3), 34.
Mur L., Skulberg O., Utkilen H. (1999). Cyanobacteria in the environment. In Toxic Cyanobacteria in
Water: A guide to Their Public Health Consequences, Monitoring and Management. 1st edition.
I. Chorus and J. Bartram (Eds.), WHO, E& SPON, London.
Neal P. (2006). An Examination of the Nature of Critical Flux and Membrane Fouling by Direct
Observation. PhD thesis, University of New South Wales, Australia.
Newcombe G., Drikas M., Hayes R. (1997). Influence of characterised natural organic material on
activated carbon adsorption: II. Effect on pore volume distribution and adsorption of
2-methylisoborneol. Water Research, 31(5), 1065-1073.
Newcombe G. (2002). Removal of algal toxins from drinking water using ozone and GAC. AWWA
Research Foundation Report, American Water Works Association, Denver.
Newcombe G., Morrison J., Hepplewhite C., Knappe D. (2002). Simultaneous adsorption of MIB and
NOM onto activated carbon: II. Competitive effects. Carbon, 40, 2147-2156.
Newcombe G., Cook D., Brooke S., Ho L., Slyman N. (2003). Treatment options for microcystins
toxins: similarities and differences between variants. Environmental Technology, 24(3), 299-308.
Newcombe G., Cook D. (2004). Removal of natural microcontaminants from drinking water using
activated carbon. In Encyclopedia of Surface and Colloid Science. 1st Edition, P. Somasundaran
(Ed.), Taylor & Francis, London.
Newcombe G., Nicholson B. (2004). Water treatment option for dissolved cyanotoxins. Journal of
Water Supply: Research and Technology - AQUA, 53 (4), 227-239.
Nicholson B., Rositano J., Burch M. (1994). Destruction of cyanobacterial peptide hepatotoxins by
chlorine and chloramine. Water Research, 28(6), 1297-1303.
Ong S.-A., Seng C.-E., Lim P.-E. (2007). Kinetics of adsorption of Cu (II) and Cd (II) from aqueous
solution on rice husk and modified rice husk. Electronic Journal of Environmental, Agricultural
and Food Chemistry, 6(2), 1764-1774.
Paralkar A., Edzwald J. (1996). Effect of ozone on EOM and coagulation. Journal of American Water
Works Association, 4, 143-154.

50
 Chapter 2 - Background

Pelekani C., Snoeyink V. (1999). Competitive adsorption in natural water: role of activated carbon
pore size. Water Research, 33(5), 1209-1219.
Pelekani C., Snoeyink V. (2001). A kinetic and equilibrium study of competitive adsorption between
atrazine and Congo red dye on activated carbon: the importance of pore size distribution. Carbon,
39, 25-37.
Pendleton P., Schumann R., Wong S. (2001). Microcystin-LR adsorption by activated carbon. Journal
of Colloid and Interface Science, 240(1), 1-8.
Pietsch J., Bornmann K., Schmidt W. (2002). Relevance of intra and extracellular cyanotoxins for
drinking water treatment. Acta hydrochimica et hydrobiologica, 30(1), 7-15.
Pilutti M., Nemeth J. (2003). Technical and cost review of commercially available MF/UF membrane
products. International Desalination Association, BAH03-029.
Quesada A., Carrasco D., Cirés S. (2006). Cianobacterias y cianotoxinas en aguas de consumo y de
recreo: un problema para todos. In Jornada sobre las Cianobacterias Tóxicas. Problemas
Asociados. Seguimiento y Control. CEDEX, Madrid.
Quinlivan P. (2001). Effects of Activated Carbon Surface Chemistry and Pore Structure on the
Adsorption of Methyl Tertiary-butyl Ether and Trichloroethene from Natural Water. Master of
Science Thesis, North Carolina State University, USA.
Quinlivan P., Li L., Knappe D. (2005). Effects of activated carbon characteristics on the simultaneous
adsorption of aqueous organic micropollutants and natural organic matter. Water Research, 39,
1663-1673.
Reichenberg D. (1953). Properties of ion-exchange resins in relation to their structure.
III. Kinetics of exchange. Journal of American Chemical Society, 75, 589-597.
Ribau Teixeira M., Rosa M.J. (2005 a). Microcystins removal by nanofiltration membranes.
Separation and Purification Technology, 46, 192-201.
Ribau Teixeira M., Rosa M.J. (2005 b). The role of membrane charge on nanofiltration performance.
Journal of Membrane Science, 265, 160-166.
Ribau Teixeira M., Rosa M.J. (2006 a). Comparing dissolved air flotation and conventional
sedimentation to remove cyanobacterial cells of Microcystis aeruginosa. Part I: The key operating
conditions. Separation Purification Technology, 52, 84-94.
Ribau Teixeira M., Rosa M.J. (2006 b). Integration of dissolved gas flotation and nanofiltration for
M. aeruginosa and associated microcystins removal. Water Research, 40, 3612-3620.
Ribau Teixeira M., Rosa M.J. (2006 c). Neurotoxic and hepatotoxic cyanotoxins removal by
nanofiltration. Water Research, 40, 2837-2846.
Ribau Teixeira M., Rosa M.J. (2006 d). The impact of the water background inorganic matrix on the
natural organic matter removal by nanofiltration. Journal of Membrane Science, 279, 513-520.

51
 Chapter 2 - Background

Ribau Teixeira M., Rosa M.J. (2007). Comparing dissolved air flotation and conventional
sedimentation to remove cyanobacterial cells of Microcystis aeruginosa. Part II: The effect of
water background organics. Separation Purification Technology, 53, 126-134.
Rodríguez E., Onstad G., Kull T., Metcalf J., Acero J., Gunten U. (2007 a). Oxidative elimination of
cyanotoxins: comparison of ozone, chlorine, chlorine dioxide and permanganate. Water Research,
41, 3381-3393.
Rodríguez E., Sordo A., Metcalf J., Acero J. (2007 b). Kinetics of oxidation of cylindrospermopsin
and anatoxin-a with chlorine, monochlorine and permanganate. Water Research, 41, 2048-2056.
Rositano J., Nicholson B., Pieronne P. (1998). Destruction of cyanobacterial toxins by ozone. Ozone:
Science & Engineering, 20, 223-238.
Rositano J., Newcombe G., Nicholson B., Sztajnbok P. (2001). Ozonation of NOM and algal toxins in
four treated waters. Water Research, 35(1), 23-32.
Schmidt W., Willmitzer H., Bornmann K., Pietsch J. (2002). Production of drinking water from raw
water containing cyanobacteria − pilot plant studies for assessing the risk of microcystin
breakthrough. Environmental Toxicology, 17(4), 375-385.
Scott K. (1998). Handbook of Industrial Membranes. Second edition, Elsevier Advanced Technology,
Oxford, United Kingdom.
Senogles P., Shaw G., Norris R., Chiswell R., Mueller J., Sadler R., Eaglesham G. (2000).
Degradation of the cyanobacterial toxin cylindrospermopsin from Cylindrospermopsis raciborskii
by chlorination. Toxicon, 38, 1203-1213.
Shende R., Mahajani V. (2002). Wet oxidative regeneration of activated carbon loaded with reactive
dye. Waste Management, 22(1), 73-83.
Sivonen K., Jones G. (1999). Cyanobacterial toxins. In Toxic Cyanobacteria in Water: A Guide to
Their Public Health Consequences, Monitoring and Management. 1st edition. I. Chorus and
J. Bartram (Eds.), WHO, E & SPON, London.
Smith D., Falls V., Levine A., MacLeod B., Simpson M., Champlin T. (2002). Nanofiltration to
augment conventional treatment for removal of algal toxins, taste and odour compounds, and
natural organic matter. In Proceedings of the Water Quality Technology Conference, Seattle,
Washington (USA), November 10-14.
Snoeyink V., Summers S. (1999). Adsorption of organic compounds. In Water Quality & Treatment.
5th Edition, AWWA, McGraw-Hill, Lisbon.
Snoeyink V., Campos C., Mariñas B. (2000). Design and performance of powdered activated
carbon/ultrafiltration systems. Water Science and Technology, 42(12), 1-10.
Svreck C., Smith D. (2004). Cyanobacteria toxins and the current state of knowledge on water
treatment options: a review. Journal of Environmental Engineering Science, 3, 155-185.

52
 Chapter 2 - Background

Taylor J., Wiesner M. (1999). Membranes. In Water Quality & Treatment. 5th Edition, American
Water Works Association (AWWA), McGraw-Hill, Lisbon.
Tsuji K., Watanuki T., Kondo F., Watanabe M., Nakazawa H., Suzuki M., Uchida H., Harada K.-I.
(1997). Stability of microcystins from cyanobacteria − IV. Effect of chlorination on
decomposition. Toxicon, 35(7), 1033-1041.
USEPA (2001). Low-pressure Membrane Filtration for Pathogen Removal: Application,
Implementation and Regulatory Issues. United States Environmental Protection Agency, Office of
Water, EPA 815-C-01-001, April.
USEPA (2005). Membrane Filtration Guidance Manual. United States Environmental Protection
Agency, Office of Water, EPA 815-R-06-009, November.
Van Apeldoorn M., Van Egmond H., Speijers G., Bakker G. (2007). Toxins of
cyanobacteria − Review. Molecular Nutrition& Food Research, 51, 7-60.
Velzeboer R., Drikas M., Donati C., Burch M., Steffensen D. (1995). Release of geosmin by
Anabaena circinalis following treatment with aluminium sulphate. Water Science and Technology,
31(11), 187-194.
Wang H., Newcombe G., Lewis D., Brookes J., Ho L. (2006). Adsorption and biological removal of
microcystins in GAC filter. Proceedings of Young Water Professionals. IWA and Australia Water
Association, New South Wales (Australia), 15-17 February.
Wang H., Ho L., Lewis D., Brookes J., Newcombe G. (2007). Discriminating and assessing adsorption
and biodegradation removal mechanisms during granular activated carbon filtration of microcystin
toxins. Water Research, 41(18), 4262-4270.
Wang S., Li H. (2007). Kinetic modelling and mechanism of dye adsorption on unburbed carbon. Dye
and Pigments, 72, 308-314.
WHO (2003). Guidelines for Safe Recreational Water Environments. Vol. 1 − Costal Water
Environments. World Health Organization, Geneva.
WHO (2006). Guidelines for Drinking Water Quality. First addendum to third edition.
Vol. 1 − Recommendations. World Health Organization, Geneva.
Wiesner M., Aptel P. (1996). Mass transport and permeate flux and fouling in pressure-driven
processes. In Water Treatment Membrane Processes. J. Mallevialle, P. Odendaal, M. Wiesner
(Ed.), McGraw-Hill, Lisbon.
Wiesner M., Buckley C. (1996). Principles of rejection in pressure-driven membrane processes In
Water Treatment Membrane Processes. J. Mallevialle, P. Odendaal, M. Wiesner (Ed.), McGraw-
Hill, Lisbon.
Yalçin M., G]urse A., Dogar C., Sözbilir M. (2004). The adsorption kinetics of
Cethyltrimethylammonium Bromide (CTAB) onto powdered active carbon. Adsorption, 10, 339-
348.

53
 Chapter 2 - Background

Yamamura H., Kimura K., Watanabe Y. (2007). Mechanism involved in the evolution of physically
irreversible fouling in microfiltration and ultrafiltration membranes used for drinking water
treatment. Environmental Science and Technology, 41, 6789-9794.
Yuan W., Zydney A. (2000). Humic acid fouling during ultrafiltration. Environmental Science
Engineering, 34, 5043–5050.
Zhao P., Takizawa S., Katayama H., Ohgaki S. (2005). Factors causing PAC cake fouling in PAC-MF
(powdered activated carbon-microfiltration) water treatment systems. Water Science and
Technology, 51(6-7), 231-240.
Zhou X., Elimelech M. (1997). Colloidal fouling of reverse osmosis membranes: measurements and
fouling. Environmental Science and Technology, 31, 3654-3662.
Zhou H., Smith D. (2002). Advanced technologies in water and wastewater treatment. Journal of
Environmental Engineering and Science, 1(4), 247-264.
Zularisam A., Ismail A., Salim R. (2006). Behaviours of natural organic matter in membrane filtration
for surface water treatment - a review. Desalination, 194, 211-231.
Zularisam A., Ismail A., Salim M., Sakinah M., Ozaki H. (2007). The effects of natural organic matter
(NOM) fractions on fouling characteristics and flux recovery of ultrafiltration membranes.
Desalination, 212, 191-208.
Zurawell R., Chen H., Burke J., Prepas E. (2005). Hepatotoxic cyanobacteria: A review of the
biological importance of microcystins in freshwater environments. Journal of Toxicological and
Environmental Health B, 8, 1–37.

54