**2.1. Biofouling due to biofilm on membrane**

Biofilm is one of the most challenging issues in membrane technology [16–18]. The adsorption of bacteria cell on the membrane surface depends on membrane properties such as membrane materials, hydrophobicity, and roughness [19]. The adhesive nature of EPS is considered as the most severe problem in membrane biofouling [20, 21]. Biofilm on the membrane surface reduces the permeate flux and salt rejection [22–24]. In membrane technology, the flux and salt rejection are the two primary criteria for characterizing of membrane performance. The more the flux and salt rejection, better the membrane performance is. The volume flux (*J*) of porous membrane is usually calculated by Hagen-Poseuille equation where the pores are assumed to have the same radius,

$$J = \frac{\varepsilon}{8\eta r} \frac{r^2}{\Delta x} \tag{1}$$

Where, Δ*x* is the membrane thickness, Δ*P* is pressure difference across the membrane, *η* is the viscosity, *τ* is tortuosity, *r* is the radius of the pore, and *ε* is the porosity of the membrane. Porosity can be calculated by,

$$
\varepsilon = \frac{n\_p \pi r^2}{A\_n} \tag{2}
$$

*Am* is the membrane surface area and *nP* is the number of pores. Tortuosity is defined by:

$$
\tau = \frac{(2 - \varepsilon)^2}{\varepsilon} \tag{3}
$$

Matin et al. Provided a list of typical bacteria species that can cause biofilm formation on the membrane surface as well as a reduction in flux decline and salt rejection due to the formation of biofilm on the membrane surface [25]. They observed that, without bacterial adhesion, the membrane was able to reject (R) 98.2% salt. The rejection was decreased by 4.6% because of the biofouling on the membrane.

Biofilm is a complex structure due to the viscoelastic nature of EPS that can lead to the formation of memory effect in a material [10, 26]. Rheological measurement of the biofouling layer on the membrane surface is required to understand the EPS nature. Patsios et al. [27] performed some rheological measurements of the biofouling layer on the membrane. They obtained nonlinear behavior of shear stress and strain of the EPS. They claimed that EPS shows more elastic nature than viscous on the membrane surface. The storage modulus Gˊ, the elastic part, was higher than the loss modulus Gˊˊ that is the viscous component. [25].

#### **2.2. Microfluidic approach in biofouling study**

of fouling on a continuous basis. Fouling is the unwanted accumulation of substances on the membrane surface. There are five types of fouling including scaling (by divalent ions), heavy metal fouling, organic fouling, colloidal fouling, and biofouling [1, 2]. Among these fouling types, biofouling is the most severe since it is a dynamic process and is also the most con-

Biofouling due to biofilms (matrix-encapsulated bacterial colonies) and colloidal materials act as the main components of membrane fouling [2]. Moreover, biofilms have a significant impact on the membranes used for different types of water filtration such as brackish and seawater. Once a cell is attached to the membrane surface, it decreases membrane permeability by forming a gel layer [3]. Biological substances always remain in the membrane. Even if 99.9% of these materials are removed by pre-treatment, the remaining 0.1% can grow expo-

Biofouling occurs due to the adsorption of the biological cells on a membrane surface [5]. Biological organisms are usually identified by their length scale. Microorganisms, which lie within very small length scales (1–200 μm), include bacteria, fungi, and algae. Furthermore, length scale >200 μm is referred to macro-organisms such as larvae, barnacles, hydroids, tubeworms, mussels, and bivalves [1]. Bacteria are a common biofouling agent and are found extensively in nature. Bacterial colonization of a surface is an extremely complex process, where several phenomena can take place at multiple length and time scales [6–8]. Colonization on the surface starts with adhesion of bacteria to a solid-liquid interface. The interaction of bacteria with the surface leads to the formation of extracellular polymeric substance (EPS), where bacterial cells are embedded in a matrix. These matrix-encapsulated, surface-associated bacterial communities are referred to as a biofilm [9, 10]. EPS, the binding material of biofilms, is composed of long-chain biomolecules such as polysaccharides, nucleic acids, protein, DNA and lipids [11–14]. Biofilms can play an important role in chronic infections [1]. Moreover, they are prevalent in industrial and shipping environment, causing significant problems related to

Biofilm is one of the most challenging issues in membrane technology [16–18]. The adsorption of bacteria cell on the membrane surface depends on membrane properties such as membrane materials, hydrophobicity, and roughness [19]. The adhesive nature of EPS is considered as the most severe problem in membrane biofouling [20, 21]. Biofilm on the membrane surface reduces the permeate flux and salt rejection [22–24]. In membrane technology, the flux and salt rejection are the two primary criteria for characterizing of membrane performance. The more the flux and salt rejection, better the membrane performance is. The volume flux (*J*) of

fronted one, and can contribute as much as 45% of the total fouling [1].

nentially by using biodegradable substances in the feed (waste) water [4].

**1.2. Biofouling due to bacterial colonization**

environmental impacts and health risks [15].

**2.1. Biofouling due to biofilm on membrane**

**2. Background**

294 Microfluidics and Nanofluidics

In wastewater treatment, microfiltration membranes with the pore sizes lying between 0.1 and 10 μm are used to remove bacteria. Membranes are usually part of an opaque setup, where only the input and the output can be measured. Advancements in micro-/nano-technologies, for example, microfluidic devices can be employed to study membrane processes at the porescale. An example of this is the use of microfluidic-based membrane mimics, which are being used to explore a wide variety of membrane related issues, including biofouling. An essential advantage of microfluidic membrane mimics in studying biofouling is that they make realtime microscopy of biofouling possible. **Figure 1** shows a basic schematic difference between membrane filtration mode and microfluidic approach. The pillars are shown in **Figure 1b** are solid in structures and usually made of polydimethylsiloxane (PDMS). The gap between the pillars is considered as the pore. The coverslip is used to seal the device.

**Figure 1.** Schematics of (a) membrane filtration where feed is wastewater and permeate is the clean water and (b) microfluidic filtration mode.

Design and fabrication are the initial steps to work with microfluidic devices. Different types of fabrication techniques include photolithography, electron lithography, hot embossing and injection molding, etc. Photolithography is a common technique when feature sizes larger than 1 μm are desired. The nanoscale feature can also be fabricated by e-beam lithography where the minimum resolution could go down to 10 nm [28, 29].

flow direction, and the red and yellow ellipses show the formed streamers attached between two pillars. As can be seen, the thickness of the streamers increased with the increase of time of the streamer. Fluorescence microscopy was used to capture the image where only bacteria

Microfluidic Membrane Filtration Systems to Study Biofouling

http://dx.doi.org/10.5772/intechopen.75006

297

**Figure 2.** Streamer formation in a microfluidic channel. Figures are reproduced with permission from Ref [6].

Marty et al. [33, 34] studied the effect of different pore sizes and filtration modes on the lengths of streamers that formed in a microfluidic membrane mimic system. They fabricated a microfluidic device with 25 straight, interconnected and staggered PDMS pillars to observe the nature of biofouling in a membrane mimic. The width and height of pillars were 10 and 50 μm respectively, and the mimic membrane pore size was 10 μm. They found that flow configuration and presence of tortuosity in a microchannel has a significant impact on streamer formation.

Membrane process is an emerging separation technology. The membrane itself is the heart of a membrane process. It can be classified as polymeric and inorganic, porous and dense, isotropic and anisotropic, hydrophilic and hydrophobic, etc. **Figure 3** gives an overview of types and preparation process of the polymeric membranes. Phase inversion (phase separation) and track etching are the most widely used techniques for the preparation of porous membranes [38].

In phase inversion process method, the polymer is transformed in a controlled manner from liquid to solid state by changing the thermodynamic state of the polymer, solvent and the solution [38, 39]. Symmetric porous phase inversion membranes are made by using water vapor as the coagulant. For making asymmetric membranes by phase inversion temperature increase and a liquid nonsolvent is used to precipitate the polymer (**Figure 3**). In track etching method, a high energy particle radiation is applied to the polymeric film, to damage the polymeric matrix and create tracks. By etching the polymeric material along the track uniform cylindrical

cells are visible (green). The fluid media and the EPS appear dark.

**3. Basic overview of fabrication techniques**

**3.1. Membrane fabrication**

#### **2.3. Bacterial streamer due to biofouling**

The impact of hydrodynamic flow on biofilms is the large time-dependent deformations that can result in nonlinear phenomena. An example of such phenomena is the bacterial streamer. Streamers form in flowing water and attach to the surface by the upstream "head" while the downstream "tail" can oscillate [6, 10, 30–32]. Streamers in a microfluidic system are typically tethered at one end to the pillar walls while the rest of the body is suspended in the downstream direction. Their filamentous structure can extend significantly with the flow [6, 33, 34]. Drescher et al. [35] revealed that streamers can cause a sudden and rapid clog in the fluid flow system in comparison with the biofilm attached to the surface. Surface hugging biofilms have a very modest effect on the flow rate whereas; streamers can drastically decrease the flow rate in a very short period [31].

Rusconi et al. [36] reported streamer formation in the microfluidic channel under laminar flow conditions. They observed formation of a single streamer in the middle of the channel connecting the inner corners of the channel. They also claimed that secondary flows in the curved edge of the channel were responsible for the location of the streamer, which was located at the mid-plane. They further investigated the streamer formation behavior by changing the radius of the curvature of a zigzag microchannel and discovered that streamer formation depends on the geometric angle of microchannel [37].

Valiei et al. [6] observed streamers through the height of the channel with 50 × 8 array of micro-pillars and mentioned it as a 'web' of the streamers. They claimed that flow rate has a significant impact on the number of streamer formation. A higher number of streamer formations was reported in the middle of channel height. **Figure 2** shows the formation of bacterial streamers in a microfluidic device with an array of micropillars. The white arrow indicates the

**Figure 2.** Streamer formation in a microfluidic channel. Figures are reproduced with permission from Ref [6].

flow direction, and the red and yellow ellipses show the formed streamers attached between two pillars. As can be seen, the thickness of the streamers increased with the increase of time of the streamer. Fluorescence microscopy was used to capture the image where only bacteria cells are visible (green). The fluid media and the EPS appear dark.

Marty et al. [33, 34] studied the effect of different pore sizes and filtration modes on the lengths of streamers that formed in a microfluidic membrane mimic system. They fabricated a microfluidic device with 25 straight, interconnected and staggered PDMS pillars to observe the nature of biofouling in a membrane mimic. The width and height of pillars were 10 and 50 μm respectively, and the mimic membrane pore size was 10 μm. They found that flow configuration and presence of tortuosity in a microchannel has a significant impact on streamer formation.
