Scope of this study - 以不同次臨界通量操作之生物薄膜反應槽之薄膜積垢的研究

Chapter 1: Introduction

1.3 Scope of this study

Figure 1.1 shows the scope of this study in which the membrane fouling under different sub-critical flux operation were considered. First step of this study was to identify the critical flux of HPI and HPO membrane operated under different operational conditions. After found out the critical flux, sub-critical fluxes were calculated with 60%

and 80% of critical flux. In case of operating the MBR system by 60% of critical flux, HPI membranes were chosen for performing in different activated sludge (AS) concentrations (7,000 – 7,500 mg-MLSS/l and 6,000 – 6,500 mg-MLSS/l). The MBR operated with AS of 7,000 – 7,500 mg-MLSS/l was abbreviated as MBR-1. Another was MBR-2. In case of operating the MBR system by 80% of critical flux, MBR-2 was chosen to carry out the test with HPI and HPO membrane. After finishing the long-term test, EPS concentration, particle size distribution, FTIR, CLSM and EEM methods were applied to determine the fouling behavior in each case.

A comparison between HPI and HPO membrane operated under 80% critical flux in MBR-2; HPI membranes operated in MBR-2 with different sub-critical flux (60% and

80% of critical flux) and HPI membrane operated under the same sub-critical flux in MBR-1 and MBR-2 were also performed to identify the critical factors for fouling during sub-critical

Figure 1.1 Scope of this study

Chapter 2 Literature Review

When the conventional wastewater treatment system no longer meets the stringent standards in an effort to preserve natural water resources, the emerging technology for wastewater treatment has become an urgent requirement for supplying the clean water for daily demands or discharging into the receiving sources such as lake, river, ground water or any other receiving sources. Membrane bioreactors (MBRs) offering several advantages over the conventional processes have gradually been used for satisfying these requirements in each country all over the world. Unfortunately, membrane fouling causing the decline of membrane performance leading to the increasing of operational cost has been preventing the widespread of this application. In this chapter, some concepts relating to MBRs as well as membrane fouling will be introduced.

2.1 Membrane bioreactors (MBRs)

By the late 1960s, a commercialized MBRs process was firstly introduced by Dorr-Oliver with the use of an activated sludge bioreactor coupled a cross-flow membrane (Smith et al., 1969). Due to a poor industrial development in producing a good performance of membrane at that time, the cost of membrane, obviously, was very high.

Beside that, to reduce fouling occurring in MBRs process, the mixed liquor suspension in the activated sludge bioreactor was pumped at high velocity at considerable energy (up to 10 kWh per 1 m³ of water product) (Le-Clech et al., 2006). As a result, the first generation of MBRs couldn’t be applied as an emerging technology. Up to 1989s, the Japanese Government co-operated with many large companies to find out a new direction for implementing a better process for water recycling. Kubota company - a participate member of above cooperation- proposed and developed a flat-sheet submerged MBR.

Yamamoto and his co-worker have pioneered in running a submerged MBR with hollow fiber membrane in this time (Yamamoto et al., 1989). The new era of using MBRs in treating wastewater has been opened ever since then

In general, MBRs are a technology combined between a direct filtration of a selectively permeable membrane and the biological degradation of activated sludge in a

wastewater treatment system. The solvent liquid is separated from the solution through membrane by a certain driving force based on the using purpose of the operational workers. According to the operational mode, MBRs were basically divided into two subgroups: bioreactors with membrane in external circuit and bioreactor with submerged membrane (Visvanathan et al., 2000). Bioreactors with membrane in external circuit appeared in the beginning of using membrane in wastewater treatment. Because of its inconvenience in operation as well as in operational cost, nowadays, no more authors have mentioned to this approach in their researches. In contrary, most MBRs design for wastewater treatment has been focused on the submerged membrane bioreactors. The fouling will easily be removed by shear stress caused air bubbles from aeration.

Nevertheless, the operating cost as well as fouling-cleaning cost will decrease compared to the external circuit membrane bioreactor. A general setup of two types of membrane bioreactors is presented in Figure 2.1.

Figure 2.1 Types of MBRs based on the position of membrane

In addition, with different direction of feed flow, concentrate or permeate, MBRs can also be categorized into end filtration and cross-flow filtration. In case of dead-end filtration, the feed flow perpdead-endicularly moves to the membrane surface. By this way, the fouling easily deposits onto the membrane surface. Then, the operator will often replace the membrane modules. Regarding to cross-flow filtration, the feed flow moves parallel with the surface of membrane. This way is considered as a natural method to mitigate the membrane fouling without stopping the system. Figure 2.2 illustrates the two concepts above.

Figure 2.2 Types of MBRs based on the flow directions

2.1.1 Membrane characteristics

Membrane characteristics are considered as the factors that contribute to the success of MBR performance such as the pore size of membrane, hydrophobic or hydrophilic membrane or membrane morphology. These mentioned factors affect membrane fouling propensity (Madaeni et al., 1999; Le-Clech et al., 2003b; Maximous et al., 2009) and resultantly affect the performance of MBR system. The extent of this part is to introduce some basic information about membrane material, membrane pore size and hydrophobic/hydrophilic characteristic of membrane.

2.1.1.1 Membrane materials

A membrane classification is conveniently accordant to the composition which formed the membranes. To perform with an optimal efficiency, many types of membranes have been experimented either in real MBR systems or in experimental MBR systems.

During 50 years of development and application, two different types of materials, organic/polymeric materials and inorganic/ceramic materials, are used to form membranes served in many purposes of water fields (Judd, 2006). Polymeric membranes are noted as the preferred membranes in membrane system. Ceramic membranes have been successfully applied for several kind of wastewater, such as treatment of high-strength industrial waste and anaerobic applications (Le-Clech et al., 2006). However, as mentioned above, ceramic membrane cannot be used as a popular option due to the high cost compared to polymeric membranes. In conclusion, a suitable membrane materials should have some special feature to stand with the conditions of MBR system such as having a strong resistance to thermal and chemical attack, high temperature, high or low pH in the reactors and low cost. Stephenson et al. (2000) has released some membrane materials by name and simultaneously presents the advantages and disadvantages of it.

With the development of technology, the cost of membrane is day-by-day decreasing. This becomes one of the most advantages of MBR system to invest many MBR plants in most of countries even in the developing countries.

7 2.1.1.2 Membrane pore size

A range of membrane pore size used in lab-scale experiments mainly lies from 0.02 to 0.5 µm (Stephenson et al., 2000). In a submerged flat sheet system, Stephenson also showed the dependence of the decline rate of flux against membrane pore size.

Membrane pore size strongly effects to the performance of membrane by rejecting the higher size of colloid, suspended solid etc…in the sludge solution that tend to form a cake layer on the membrane surface caused the decline of the effluent flux. Gander et al.

(2000) revealed that the initial fouling of the larger membrane pore size dominated over the smaller pore size when conducting a series of membranes with different pore size from 0.4 to 5 µm.

It was reported that the blocking index of 0.4 µm of membrane pore size was always higher than that of 0.2 µm under the same operational conditions (Hwang et al., 2008). Beside that, many researchers have focused on the optimal performance of membrane pore size. Choo and Lee (1996) proposed that membrane with 0.1 µm pore size have a best performance with the least membrane fouling tendency compared to other 0.02, 0.5 and 1 µm of membrane pore size. A study related to the influence of four membrane types (cellulose acetate, polyethersulfone, mixed ester, polycarbonate) with three different pore sizes (0.40 – 0.45, 0.22, 0.10 µm) on cross-flow filtration was conducted by Nadir Dizge and his colleagues. In this study, cellulose acetate membrane with pore size of 0.45 µm presented a worst performance with the most rapid decline of flux among all membranes.

Some authors have found that membrane pore size exhibits a negative effect to the critical flux with the pore size from 0.01 to 0.1µm (Le-Clech et al., 2003a). Madaeni (1999) supposed that the critical flux of membrane was not dependent on the membrane pore size (although the TMP is different with the difference of the pore size). In other words, according to Madaeni, the larger the pore size, the higher the TMP produced although the flux is the same with all cases.

Table 2.1 shows the characteristics of PVDF flat-sheet membranes used in MBRs by many authors in order to making a relationship between critical flux and membrane characteristics.

Table 2.1 Characteristics of PVDF flat-sheet membranes used in MBRs Pore size

Madaeni et al. (1999) Madaeni et al. (1999) B.D.Cho et al. (2002) Wang et al. (2008)

Wu et al. (2008)

2.1.2 Membrane fouling

The development of membrane fouling in a membrane bioreactor is one of the main causes for flux decline during long-term operation, preventing the wide-spread application of MBR in the wastewater treatment. It is caused by the deposition or attachment of small particles such as colloids, solutes, microorganisms, cells-debris and biomass residuals on the membrane surface. Membrane fouling mitigation has been investigated to improve the performance of membrane, and to decrease the operational cost. The carbohydrate fraction from the soluble microbial product has been accepted as the major fouling in membrane bioreactors (Clech et al., 2006; Pan et al, 2008). In which, hydrophilic carbohydrates are the predominant cause for membrane fouling (Pan et al., 2008). Another study has mentioned that the fatty acids from bacterial lipopolysaccharides are mainly related to membrane fouling (Al-Halbouni et al., 2008). Zhang et al (2006) observed the membrane fouling that follows a three stage fouling history. Stage 1 is an initial short-term rise in transmembrane pressure. Stage 2 is a long-term TMP rise or a slow fouling stage. Stage 3 is that the transmembrane pressure suddenly jump due to the fouling exceedingly deposited on the membrane surface. A review of Meng et al. (2008) has summarized the membrane fouling mechanisms as following:

• Adsorption of solutes or colloids

• Deposition of sludge flocs

• Formation of a cake layer

• Detachment of fouling caused by shear forces

• The spatial and temporal changes of the fouling composition

Particle size plays a significant role in categorizing membrane. Three type of membrane fouling are proposed: adsorption, pore blocking and cake layer (Hong et al., 2005). With the complex compositions of membrane fouling such as colloids, solutes .etc, size of some sludge particles that is smaller than the membrane pore size tends to adsorb on pore wall and gradually fulfill the membrane pore. In a case of the sludge particle size is as same as membrane pore size, the foulants will block the membrane pore. It is called as pore blocking. Both adsorption and pore blocking are considered as irreversible fouling (Chang et al., 2002; Drews 2010) or irremovable fouling (Meng et al., 2009).

Cake layer is formed on membrane surface only when the size of sludge particle is much larger than that of membrane. A cake layer is a complex system of the interaction between not only colloids-colloids, solutes-solutes, solutes-colloids but also colloids, solutes-membrane surface. The cake layer which contributes 80% of total resistance of membrane system mainly affects to the membrane fouling formation (Lee et al., 2001).

Fortunately, many researches have pointed out that cake layer is removable from the membrane surface by physical cleaning as well as is removed by shear-stress caused by aeration supported during the running of membrane bioreactors (Chu et al., 2004; Chang et al., 2002; Le-Clech et al., 2006; Meng et al., 2009; Drews 2010).

According to the removable characteristics of membrane fouling on the membrane surface, many scientists have mentioned to membrane fouling as reversible and irreversible fouling. Reversible fouling is a type of membrane fouling that is easy to be removed by physical washing such as back flushing or relaxation. On the other hand, irreversible fouling is generally removed by chemical cleaning (Chang et al., 2002).

However, Meng et al. (2008) supposed that there is existing some types of membrane fouling that neither physical washing nor chemical cleaning can be used to remove.

Therefore, these authors defined three types of fouling that are removable fouling, irremovable fouling and irreversible fouling. The term of irrecoverable fouling has been proposed by Drews (2010) in which this type of membrane fouling can’t be removed by any cleaning means. This definition is familiar with the definition of Meng et al. (2008) about irreversible fouling. Table 2.2 shows various definition of reversible and irreversible fouling occurred during MBR operation.

Table 2.2 Definition of reversible and irreversible fouling

Cleaning methods The term of fouling References Physical cleaning Reversible fouling

Removable fouling

Drews, (2010) Chang et al. (2002) Meng et al. (2009)

Chemical cleaning Irreversible fouling

Irremovable fouling

Drews, 2010 Chang et al. (2002) Meng et al. (2009)

None(*) Irrecoverable fouling Irreversible fouling

Drews, (2010) Meng et al.(2009)

(*) no suitable methods used to remove fouling

Regarding to the constituents of membrane fouling, fouling can be classified into biofouling, organic fouling and inorganic fouling (Meng et al., 2009). For a long-term operation, a microbial biofilm occurs on the membrane surface due to an interaction between microbial community and membrane surface. This microbial biofilm is the accumulation of cells and microorganisms products. In other words, the free-floating microorganisms lived in the sludge solution will colonize the membrane surface by attaching on the surface and develop their biomass. The so-called “biofouling” is also referred to this microbial biofilm and all products excreted from the microbial community activities that effect to the performance of membrane systems. Biofouling is still inevitable due to the complex characteristics of microbial system (Le-Clech et al., 2006).

The term of organic fouling refers to the adsorption of organic matters on the membrane surface and its pore inside due to the small size of particle and suction force pump in the system. Biopolymers such as proteins and polysaccharides are predominant in organic fouling compositions and dependent on the Food/Microorganism (F/M) ratio in which high F/M make the fouling more proteinaceous (Kimura et al., 2005).

Inorganic fouling contributes a small role in membrane resistance due to the low concentration in the activated sludge solution. The inorganic fouling is formed from the chemical precipitation and biological precipitation. Metal ions can easily react with the specific functional group of biopolymer such as CO₃^2-, SO₄^2- (Meng et al., 2009)… and deposited on the cake layer. Otherwise, inorganic fouling is also able to adsorb into the membrane pore and become the irreversible fouling. Chemical cleaning is predominant in removal of the inorganic precipitate compared to the physical cleaning.

2.2 Critical flux and sub-critical flux

The term “flux” refers to how much of clean water that flows through a unit area of membrane per unit time. When running the membrane system, the flux is determined by the following equation seen as an integral form of Darcy law (Bacchin et al., 2006).

∆ ∆

2.1

Where:

J: Permeate flux (m³m^-2h^-1)

∆: A transmembrane pressure (Pa)

∆: An osmotic pressure (Pa) µ: Viscosity (Pa.s)

Rm is the resistance of original membrane; Rads is the resistance caused by surface or pore adsorption; Irreversible fouling R_irrev is caused by cake deposit or gel formation; Reversible fouling R_rev is caused by pore linking or cake deposit.

Critical flux plays an important role in operating a certain membrane bioreactor.

The engineer should controls the system based on critical flux for an effective operation.

Critical flux plays an important role in characterizing the membrane fouling as well as the membrane bioreactor performance. One definition for critical flux is that “The critical flux for MF is that on start-up there exists a flux below which a decline of flux with time does not occur; above it fouling is observed. This flux is the critical flux and its value depends on the hydrodynamics and probably other variables” (Field et al., 1995). The other definition is that the flux below which colloids do not deposit on the membrane surface (Howell, 1995). By these two definitions, membrane fouling will be noticeable at

a flux called “critical flux”. Above the critical flux, all type of activated sludge compositions such as colloids, suspended solid … will quickly attach on the membrane surface caused the rapid decline in flux. The term sub-critical refers to the flux below which the fouling does not. Otherwise, the strong form and weak form of critical fluxes have been developed to compare with the conventional definition (Bacchin et al., 2006).

The strong form of critical flux is defined as the flux point at which the transmembrane pressure curve changes direction from linearity with the assumption of the absence of adsorption and osmotic pressure. In the presence of adsorption, the definition of the weak form of critical flux is also the same with that of the strong form of critical flux but the steady-state is different from that of pure water. Up to now, a standardized methodology has not yet been designed for exactly determining the value of the weak form of critical flux (Guglielmi et al., 2006)

Numerous methods for identifying the critical flux of a certain membrane have been proposed by a number of authors that whose research focused on the membrane area (Le-Clech et al., 2003a; Espinasse et al., 2002). Pierre Le-Clech and his colleagues have established the concept of flux-step method in 2003 (Le-Clech et al., 2003a). Flux-step method is a common method developed for determining the critical flux of membrane used in a membrane bioreactor operating at constant flux. During the flux-step test, smaller step height is required to make the test precisely (Tiranuntakul et al., 2011). Some specific hydraulic parameters obtained from this experiment such as the initial transmembrane pressure increase (∆P₀), the rate of transmembrane pressure increase (dP/dt), permeability (K), and the average transmembrane pressure (Pave) are used to identify critical flux (Le-Clech et al., 2003b). This author also defined critical flux the flux at which permeability decreases to below 90% of initial permeability recorded for the first filtration step. In another paper, Guglielmi et al. (2006) proposed critical flux is the flux at which the initial permeability is higher than 90% of the initial permeability. The concentration of feed-solution, shear rate and membrane characteristics affect the value of critical flux. The critical flux has been found higher at higher cross-flow velocity, lower feed concentration and lower for hydrophobic membrane (Madaeni et al., 1999).

2.3 Parameters affecting membrane fouling in MBR

2.3.1 Effect of membrane hydrophobicity

An interaction between feed solution and membrane existing with two major classes: An interaction between the water and membrane and an interaction between the solutes in the water and membrane. The so-called “hydrophilicity” is referred to the high affinity of membrane surface against the water, while the low affinity is called

“hydrophobicity” (Cardew et al., 1998).

A study of eight types of membrane with the same cut-off but different materials was conducted by Jonsson et al. (1995) with using a low-molecular weight hydrophobic solute (octanoic acid) as foulant to observe the performance of hydrophobic and hydrophilic membrane. As a result, the performance of hydrophilic membrane was better than that of hydrophobic membrane. Chang et al. (1999) have mentioned to the effect of this interaction mentioned above against membrane fouling in which the membrane fouling is easer deposited to the hydrophobic membrane than the hydrophilic membrane.

在文檔中以不同次臨界通量操作之生物薄膜反應槽之薄膜積垢的研究 (頁 13-0)