**Meet the editor**

Dr. Yu was trained as a physical chemist and kineticist at the University of Michigan, Ann Arbor, Michigan, USA. She did her postdoctoral research at Brookhaven National Laboratory and Colorado State University, USA. She has been a senior scientist at Pacific Northwest National Laboratory since 2006. She has led the development of a novel mesoscale imaging tool based on microfluidics at PNNL since

2009, which has resulted in two issued patents, a prestigious R&D 100 Awards, and a Federal Laboratory Consortium Technology Transfer Excellence Award. Dr. Yu has developed new concepts in aerosol sampling and led and participated in many field studies for in situ measurements of aerosols. Dr. Yu is the chair of the Department of Energy chemical exposure working group and a member of the TEEL Advisory Group for consequence assessment.

## Contents

**Preface XIII**

#### **Section 1 Biological Applications 1**


#### **X** Contents

### **Section 3 Microfluidics and Electronics 163**


### **Section 5 Microfluidics and Material Synthesis 357**

**Section 3 Microfluidics and Electronics 163**

**Feedback Control 165**

Chapter 8 **Microfluidics for Soft Electronics 189**

**Applications 213**

Chapter 10 **Electroosmotic Flow Pump 237** Meng Gao and Lin Gui

**Applications 293**

Taotao Fu

Chin-Tai Chen

**Medical Applications 257**

John Yan

**VI** Contents

Chapter 7 **Integrated Control of Microfluidics – Application in Fluid Routing, Sensor Synchronization, and Real-Time**

Elishai Ezra, Danny Bavli and Yaakov Nahmias

Chapter 9 **Microfluidic Sensors and Circuits for Internet of Things**

**Section 4 Fabrications and Applications in Energy Research 255**

Daniel Nieto García and Gerard O'Connor

Chapter 12 **Microfluidics in Membraneless Fuel Cells 273**

Chapter 13 **Microfluidics in CO2 Capture, Sequestration, and**

Chapter 14 **Generation and Evaporation of Microsprays 315**

and Anne‐Marie Haghiri‐Gosnet

Chapter 15 **Overview of Materials for Microfluidic Applications 335**

Chapter 11 **Laser-Based Fabrication for Microfluidics Devices on Glass for**

Jesus A. Diaz-Real, Minerva Guerra-Balcázar, Noe Arjona, Francisco Cuevas-Muñiz, Luis Gerardo Arriaga and Janet Ledesma-García

Emmanuel Roy, Antoine Pallandre, Bacem Zribi, Marie‐Charlotte Horny, François Damien Delapierre, Andrea Cattoni, Jean Gamby

Babak Taghavi, Jiantong Li, Mikael Östling and Shi Cheng

#### Chapter 16 **Synthesis of Functional Materials by Non-Newtonian Microfluidic Multiphase System 359** Yong Ren, Kai Seng Koh and Yaping Zhang

#### Chapter 17 **High and Efficient Production of Nanomaterials by Microfluidic Reactor Approaches 385** Victor Sebastian Cabeza

## Preface

Microfluidics has seen a rapid development in the last decade as a result of the seminal work pioneered by Dr. Whitesides group and many others since the 1990s. Integrating sci‐ ence and technology, microfluidics has grown steadily as a new industry and scientific field. This is a field filled with imagination, ingenuity, and enthusiasm from many researchers across disciplines.

The InTech Open Access Publisher kindly asked me to edit a new book presenting recent advances and applications in microfluidics. This book is a timely report from many re‐ searchers who are actively practicing in the field. Each chapter represents the perspective of their views of the past, present, and future of microfluidics and its unique associations with their diverse research. The content of the book is easy to follow. It can be used by a research‐ er or student who is eager to learn more about microfluidics. Moreover, it can serve those who are experts and want to get a quick update from peers.

This book consists of five sections including the following: (1) Biological Applications, (2) Imaging and Spectroscopy, (3) Microfluidics and Electronics; (4) Fabrications and Applica‐ tions in Energy Research, and (5) Microfluidics and Material Synthesis. Each section presents review articles with a particular focus in that area. While some space is given to more general overview of the fabrication techniques and materials used in microfluidics, most chapters are dedicated to new concepts, results, and applications. Specifically, the more widely known biological applications are listed in Section 1. Readers may find intrigu‐ ing applications of microfluidics in imaging and spectroscopy besides conventional optical microscopy and spectroscopy in Section 2. Section 3 showcases examples of the integration of electronics and microfluidics as more researchers are exploring the marriage between MEMS and microfluidics. Section 4 provides some familiar backgrounds of microfluidic fab‐ rication. However, the majority of the content in this part is geared at microfluidic applica‐ tions in energy-related research such as fuel cells and carbon dioxide sequestration. Section 5 contains new results of material synthesis using microfluidics.

This book covers a wide range of new topics. The intent is to provide readers with a fresh view of the future directions of microfluidics based on the current research. It is a pleasure to acknowledge the InTech editors who helped to get everything in a timely manner, espe‐ cially Ms. Iva Simcic for her persistence.

> **Xiao-Ying Yu** Pacific Northwest National Laboratory, Richland, WA, USA

**Biological Applications**

## **Advances in Low Volume Sample Analysis Using Microfluidic Separation Techniques**

Virginie Houbart and Marianne Fillet

Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/64952

#### **Abstract**

During the last decades, a great interest has been shown for miniaturised separation tech‐ niques. The use of microfluidic techniques fulfills the constant needs for increasing sam‐ ple throughput and analysis sensitivity, while reducing costs and sample volume consumption. In this chapter, three microfluidic separation techniques will be addressed: capillary electrophoresis, gas chromatography and liquid chromatography. A special at‐ tention will be paid to miniaturised liquid chromatography, with a deep investigation of its advantages compared with classical liquid chromatography. Sample preparation adapted to low volumes (a few µl) will also be discussed.

**Keywords:** Separation, miniaturisation, microfluidics, sensitivity

## **1. Introduction**

Separation techniques are widely used for the analysis of biomolecules as well as small molecules in various fields, as genomics, proteomics or pharmaceutical sciences. Due to the wide range of separation techniques, numerous studies have been conducted aiming to improve performances in terms of sample preparation, sensitivity, cost or analysis throughput.

Liquid chromatography (LC) is the most employed separation technique, but alternative techniques such as capillary electrophoresis (CE) and gas chromatography (GC) are never‐ theless helpful to provide orthogonal separation capabilities. Ultraviolet, electrochemical and fluorescence detection are used to detect the target compounds, but mass spectrometry (MS) detection offers enhanced sensitivity and additional structural information since co-eluting compounds are differentially detected according to their mass-to-charge ratio.

© 2016 The Author(s). Licensee InTech. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Miniaturisation is a general trend common to many areas in sciences and technology. Down‐ scaling the separation techniques has been initiated in the 1970s, but miniaturisation has mainly experienced an exponential growth since the 1990s. Reducing the size of the separation supports brings valuable advantages as analysis time reduction, increased sensitivity and low sample and reagent consumption. However, the limited loading capacity of microfluidic devices is a drawback. Adequate sample preparation, pre-concentration and appropriate device can circumvent these inherent limitations.

## **2. Microfluidic separation techniques**

#### **2.1. Capillary electrophoresis**

#### *2.1.1. Instrumentation*

Capillary electrophoresis (CE) is a microscale analytical technique based on the separation of compounds according to their charge-to-size ratio. The first CE device was described by Hjertén in 1967 that performed the electrophoretic separations in narrow bore tubes of 300 µm inner diameter (i.d.) for the analysis of various analytes (inorganic ions, nucleotides, proteins) [1]. In 1981, Jorgenson and Lukacs demonstrated for the first time that capillaries with a smaller i.d. (75 µm) could provide high separation efficiency using high voltages (30 kV), due to the small capillary dimensions that allowed good dissipation of Joule heat produced by such a high voltage [2].

Most modern CE instruments are very simple: a high voltage power supply, an autosampler with injection system, a capillary (25–100 µm i.d. and more often 50–75 µm, 20–100 cm length) and a detector coupled to a computer for data acquisition (Figure 1A).

Briefly, a capillary made of fused silica coated with a layer of polyimide is filled with a background electrolyte solution. When an optical detection (commonly UV or fluorescence) is employed, two electrodes are placed in buffer reservoirs to provide the necessary electrical contact between the high voltage supply and the capillary. To perform an analysis, the sample is loaded into the capillary by applying either a pressure difference (hydrodynamic injection) or an electric field (electrokinetic injection) between both extremities of the capillary. Optical detection is performed through a detection window directly on the capillary.

The hyphenation of CE to MS was first presented in 1987 by Olivares et al. that proposed an interface between CE and MS with an electrospray ionisation source (ESI) [3]. Alternative ionisation method has been described for CE-MS [4], as continuous flow-fast atom bombard‐ ment ionisation (CF-FAB) [5], atmospheric pressure chemical ionisation (APCI) [6] or atmos‐ pheric pressure photochemical ionisation (APPI) [7]. The CE-MS coupling provides structural information, enhanced sensitivity and selectivity compared with an optical detection. The CE-MS instrument configuration is modified to allow the direct entrance of the analytes into the mass spectrometer (Figure 1B).

Advances in Low Volume Sample Analysis Using Microfluidic Separation Techniques http://dx.doi.org/10.5772/64952 5

Miniaturisation is a general trend common to many areas in sciences and technology. Down‐ scaling the separation techniques has been initiated in the 1970s, but miniaturisation has mainly experienced an exponential growth since the 1990s. Reducing the size of the separation supports brings valuable advantages as analysis time reduction, increased sensitivity and low sample and reagent consumption. However, the limited loading capacity of microfluidic devices is a drawback. Adequate sample preparation, pre-concentration and appropriate

Capillary electrophoresis (CE) is a microscale analytical technique based on the separation of compounds according to their charge-to-size ratio. The first CE device was described by Hjertén in 1967 that performed the electrophoretic separations in narrow bore tubes of 300 µm inner diameter (i.d.) for the analysis of various analytes (inorganic ions, nucleotides, proteins) [1]. In 1981, Jorgenson and Lukacs demonstrated for the first time that capillaries with a smaller i.d. (75 µm) could provide high separation efficiency using high voltages (30 kV), due to the small capillary dimensions that allowed good dissipation of Joule heat produced by such a

Most modern CE instruments are very simple: a high voltage power supply, an autosampler with injection system, a capillary (25–100 µm i.d. and more often 50–75 µm, 20–100 cm length)

Briefly, a capillary made of fused silica coated with a layer of polyimide is filled with a background electrolyte solution. When an optical detection (commonly UV or fluorescence) is employed, two electrodes are placed in buffer reservoirs to provide the necessary electrical contact between the high voltage supply and the capillary. To perform an analysis, the sample is loaded into the capillary by applying either a pressure difference (hydrodynamic injection) or an electric field (electrokinetic injection) between both extremities of the capillary. Optical

The hyphenation of CE to MS was first presented in 1987 by Olivares et al. that proposed an interface between CE and MS with an electrospray ionisation source (ESI) [3]. Alternative ionisation method has been described for CE-MS [4], as continuous flow-fast atom bombard‐ ment ionisation (CF-FAB) [5], atmospheric pressure chemical ionisation (APCI) [6] or atmos‐ pheric pressure photochemical ionisation (APPI) [7]. The CE-MS coupling provides structural information, enhanced sensitivity and selectivity compared with an optical detection. The CE-MS instrument configuration is modified to allow the direct entrance of the analytes into the

and a detector coupled to a computer for data acquisition (Figure 1A).

detection is performed through a detection window directly on the capillary.

device can circumvent these inherent limitations.

4 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

**2. Microfluidic separation techniques**

**2.1. Capillary electrophoresis**

mass spectrometer (Figure 1B).

*2.1.1. Instrumentation*

high voltage [2].

**Figure 1.** Schematic representation of a classical capillary electrophoresis system with an optical detection (A) and the coupling to a mass spectrometer (B).

The coupling of CE to MS can be achieved using a sheath liquid interface or a sheathless interface. The use of an additional liquid, the sheath liquid, to the electrophoretic effluent allows the formation of an electrical contact at the capillary MS output that is necessary for the electrophoretic separation, and enables the electrospray formation. In addition, the back‐ ground electrolyte composition can be modified by dilution with the sheath liquid to ensure the compatibility with MS detection. However, the dilution process decreases the sensitivity in proportion with the sheath liquid flow rate.

Sheathless interfaces overcome the dilution-related sensitivity limitations encountered when using a sheath liquid interface. In this configuration, the electrical contact cannot be established through a liquid junction; the electrical contact may be established by many different techni‐ ques, e.g. by the insertion of the separation capillary into a conductive sprayer or the coating of the outlet end of the capillary by a conductive material.

#### *2.1.2. Capillary electrophoresis on chip*

In the light of the small dimensions of the separation capillary, classical CE is naturally classified into the category of miniaturised separation techniques. During the past few years, a new trend in CE instrumentation has emerged: the miniaturisation of CE into an integrated chip device for hyphenation to MS [12]. Since the introduction of the first chip-based electro‐ phoresis device by Manz et al. [13], chip design has undergone continuous evolution from a single-channel design to more complex layouts integrating all the analytical steps on a single component. The actual classical chip design is made of two crossed microchannels, solution reservoirs for the sample and the waste, and reservoirs at the cathode and the anode for the buffer (Figure 2A) [14].

Interfacing a CE chip to an ESI-MS detector can be realised by spraying directly from the chip [15] or from a capillary sprayer attached to the chip [16] (Figure 2B).

**Figure 2.** CE chip (A) and CE-MS (B) chip basic configuration.

In 1999 Agilent launched the Bioanalyser 2100, the first commercial microfluidics-based platform for DNA, RNA, protein and cell analysis. Separation is performed by capillary electrophoresis in channels containing a gel matrix. This device is the miniaturised counterpart of gel electrophoresis analysis (e.g. SDS-PAGE for proteins and agarose gel electrophoresis for nucleic acids) [17, 18]. In this device, sample with a volume between 1 and 6 µl moves through the loading channels, and a fraction of this volume is injected into the separation channel filled with a gel matrix. Fluorescence detection is performed on the chip itself. Total analysis time (including sample loading, separation, staining and destaining) is 30–40 min on the Bioana‐ lyser chip, what is much shorter than the few hours (up to 1 day) required for the classical gel electrophoresis process.

Advantages of CE downscaling are reduced analysis times (minutes to seconds), low sample volume requirements (to the picolitre range), low solvent consumption and high throughput capabilities through the possibility of performing simultaneous separations in parallel channels [12].

Many applications using CE on chip have been developed for the analysis of a wide range of matrices and analytes: food analysis (including small molecules, organic acids, heavy metals, toxins, microorganisms or allergens) [19], amino acid analysis [20] or even intact protein characterisation [21].

### **2.2. Gas chromatography**

Sheathless interfaces overcome the dilution-related sensitivity limitations encountered when using a sheath liquid interface. In this configuration, the electrical contact cannot be established through a liquid junction; the electrical contact may be established by many different techni‐ ques, e.g. by the insertion of the separation capillary into a conductive sprayer or the coating

In the light of the small dimensions of the separation capillary, classical CE is naturally classified into the category of miniaturised separation techniques. During the past few years, a new trend in CE instrumentation has emerged: the miniaturisation of CE into an integrated chip device for hyphenation to MS [12]. Since the introduction of the first chip-based electro‐ phoresis device by Manz et al. [13], chip design has undergone continuous evolution from a single-channel design to more complex layouts integrating all the analytical steps on a single component. The actual classical chip design is made of two crossed microchannels, solution reservoirs for the sample and the waste, and reservoirs at the cathode and the anode for the

Interfacing a CE chip to an ESI-MS detector can be realised by spraying directly from the chip

In 1999 Agilent launched the Bioanalyser 2100, the first commercial microfluidics-based platform for DNA, RNA, protein and cell analysis. Separation is performed by capillary electrophoresis in channels containing a gel matrix. This device is the miniaturised counterpart of gel electrophoresis analysis (e.g. SDS-PAGE for proteins and agarose gel electrophoresis for nucleic acids) [17, 18]. In this device, sample with a volume between 1 and 6 µl moves through the loading channels, and a fraction of this volume is injected into the separation channel filled with a gel matrix. Fluorescence detection is performed on the chip itself. Total analysis time (including sample loading, separation, staining and destaining) is 30–40 min on the Bioana‐ lyser chip, what is much shorter than the few hours (up to 1 day) required for the classical gel

Advantages of CE downscaling are reduced analysis times (minutes to seconds), low sample volume requirements (to the picolitre range), low solvent consumption and high throughput capabilities through the possibility of performing simultaneous separations in parallel

[15] or from a capillary sprayer attached to the chip [16] (Figure 2B).

**Figure 2.** CE chip (A) and CE-MS (B) chip basic configuration.

of the outlet end of the capillary by a conductive material.

6 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

*2.1.2. Capillary electrophoresis on chip*

buffer (Figure 2A) [14].

electrophoresis process.

channels [12].

Gas chromatography (GC) was first described by James and Martin in 1952. They presented a separation of volatile fatty acids on diatomaceous earth impregnated with a mixture of silicone oil and stearic acid as stationary phase, and a flow of nitrogen as the mobile phase [22]. GC underwent an explosive progression during the next decade, with applications for the petroleum industry [23], followed by biochemical applications [24, 25].

First GC separations were performed on packed columns of 1–5 mm i.d. Column length limitations due to backpressure drop led to the introduction of capillary GC columns [26, 27]. In such columns, the stationary phase is coated on the inner walls of the capillary to form a thin film (wall-coated open tubular, WCOT), or impregnated into a porous layer (porous layer open tubular, PLOT) [28]. Since capillary GC columns have less than 1 mm i.d. (typically 0.05– 0.53 mm), this technique could already be considered as miniaturised.

**Figure 3.** Angell's GC chip integrating sample injection valve, separation column and detector [29].

However, since the 1970s researchers have been trying to integrate all the components of a gas chromatographer including the detector on a single piece, or chip. In 1979, Angell proposed a silicon wafer chip including a sample injection valve, a 1.5 m column and a detector [29] (Figure 3). In the next decades, other homemade chips were proposed, but to our knowledge no commercial version of a small and portable GC chip has been proposed so far.

A few applications have been developed on microbore GC systems, but research is still more dedicated to reliable miniaturised system development rather than method development [32].

#### **2.3. Liquid chromatography**

Liquid chromatography is the most commonly used separation technique with a wide range of applications. The precursor of liquid chromatography was the Russian scientific Mikhail Semenovich Tswett. He discovered that plant leave extracts poured on a column packed with particles could be separated into distinct coloured bands. In 1956, Van Deemter published his famous work about the fundamental equation of the relationship between mobile phase linear velocity and height equivalent to a theoretical plate (Figure 4) [33]. The modern appellation high pressure (now interchangeable with performance) liquid chromatography was first introduced by Horvath in 1970 to designate liquid chromatography performed on reduced (<10 µm) porous particles. Since the 1970s, LC underwent an explosive popularity to become a standard separation technique with continuous progress in stationary phase variety and performances, hardware features and fields of applications.

**Figure 4.** Van Deemter plot deconvolution: (A) Eddy diffusion term; (B) longitudinal diffusion term; (C) resistance to mass transfer term.

Two major research axes of LC have been developed (Figure 5) to comply with the growing needs in increasing the analysis throughput, enhancing sensitivity and reducing analysis cost and environmental footprint through a decrease in solvent consumption [35].

#### *2.3.1. Stationary phase particle size reduction*

silicon wafer chip including a sample injection valve, a 1.5 m column and a detector [29] (Figure 3). In the next decades, other homemade chips were proposed, but to our knowledge no

A few applications have been developed on microbore GC systems, but research is still more dedicated to reliable miniaturised system development rather than method development [32].

Liquid chromatography is the most commonly used separation technique with a wide range of applications. The precursor of liquid chromatography was the Russian scientific Mikhail Semenovich Tswett. He discovered that plant leave extracts poured on a column packed with particles could be separated into distinct coloured bands. In 1956, Van Deemter published his famous work about the fundamental equation of the relationship between mobile phase linear velocity and height equivalent to a theoretical plate (Figure 4) [33]. The modern appellation high pressure (now interchangeable with performance) liquid chromatography was first introduced by Horvath in 1970 to designate liquid chromatography performed on reduced (<10 µm) porous particles. Since the 1970s, LC underwent an explosive popularity to become a standard separation technique with continuous progress in stationary phase variety and

**Figure 4.** Van Deemter plot deconvolution: (A) Eddy diffusion term; (B) longitudinal diffusion term; (C) resistance to

commercial version of a small and portable GC chip has been proposed so far.

8 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

performances, hardware features and fields of applications.

**2.3. Liquid chromatography**

mass transfer term.

Considerable gains in terms of sensitivity and analysis time (or chromatographic resolu‐ tion) could be obtained by reducing the stationary phase particle size to less than 2 µm, giving rise to ultra-high performance liquid chromatography (UHPLC). The use of smaller parti‐ cles can significantly reduce the height equivalent of a theoretical plate (HETP) generated in a separation.

$$HETP = A + \frac{B}{\mu} + Cu \tag{1}$$

where *u* is the mobile phase velocity, *A* is the Eddy diffusion term, *B* is the longitudinal diffusion term and *C* is the resistance to mass transfer term.

The *C* term mobile phase component *Cm* can be expanded to the following relationship, showing its dependency on the square of particle size:

$$C\_{\rm u} = o \frac{d\_p^2}{D\_{\rm u}} \tag{2}$$

where *k* is the retention factor, *dp* is the particle diameter, *Dm* is the diffusion coefficient of the solute in the mobile phase and *ω* is the pore size distribution, shape and particle size distri‐ bution coefficient.


**Figure 5.** Historical trends in development of HPLC and on-chip LC (adapted from Lavrik et al.'s review [34]).

Particle size reduction has been initiated since the beginning of the spreading of HPLC as a separation technique, but technical limitations related to the pressure drop caused by particle size reduction delayed the commercialisation of sub-2 µm particle columns with classical dimensions [36]. The first pumps able to face ultra-high pressure were presented by Jorgenson [37] shortly followed by Lee et al. in the late 1990s. In 2004, Waters commercialised the first UHPLC system that was design to deliver pressure up to 1000 bar [41].

#### *2.3.2. Column inner diameter reduction*

In parallel with the reduction of particle size, the miniaturisation of LC columns in terms of inner diameter encounters a growing interest since the late 1970s [42].

As previously developed, downscaling the inner diameter of the separation support increases sensitivity with up to 3–4 orders of magnitude in a reduced analysis time. In addition to the advantages of UHPLC, LC miniaturisation reduces drastically the requirements in terms of sample and mobile phase volume.

Miniaturised columns operated on classical LC systems have been described, but void volumes that are very large compared with flow rates and column volumes are responsible for peak dispersion. For that reason, the integration of chromatographic components on a chip (sepa‐ ration channels and electrospray emitters for MS detection, but also additional channels, connections and microvalves) has rapidly been the major strategy to minimise void volumes and efficiency drop [43].

In 1978, Tsuda and Novotny experienced with the performances of packed glass capillaries with 50–200 µm inner diameters.

During the next years, research on chip technology was mainly focused on electroosmosis- or electrophoretic-driven separations due to the technical challenge represented by the connec‐ tion between LC pumps and chips.

#### *2.3.2.1. Open-channel chromatography*

The simplest way to perform miniaturised liquid chromatography on chip is to coat the inner walls of the channels with chemical groups that may interact with the compounds of interest, i.e. to perform open-channel chromatography. In 1990, Manz et al. proposed the first chip prototype for open-tubular liquid chromatography made of silicon and coupled to a minia‐ turised conductometric detector connected to a classical LC pump [46]. Jacobson et al. proposed the first open-channel separation application on a glass chip coated with octadecyl‐ silane chains in 1994, with low theoretical plate heights (4.1–5 µm) [47]. Due to the small specific surface of such systems, researchers conceived coating modifications to increase the phase ratio (ratio between the volume of stationary phase and the volume of mobile phase): porous layer open-tubular (PLOT) columns, functionalised particles embedded in a porous layer [50] or immobilisation of nanoparticles onto the walls [51].

Open-channel chromatography (with an ideal i.d. of 10–20 µm) provides high efficiency since the molecular diffusion is the only contributor to band broadening. However, due to its limited specific surface, column capacity stays low even with stationary phase modifications.

#### *2.3.2.2. Micropillars, collocated monolith support structures and nanotubes*

dimensions [36]. The first pumps able to face ultra-high pressure were presented by Jorgenson [37] shortly followed by Lee et al. in the late 1990s. In 2004, Waters commercialised the first

In parallel with the reduction of particle size, the miniaturisation of LC columns in terms of

As previously developed, downscaling the inner diameter of the separation support increases sensitivity with up to 3–4 orders of magnitude in a reduced analysis time. In addition to the advantages of UHPLC, LC miniaturisation reduces drastically the requirements in terms of

Miniaturised columns operated on classical LC systems have been described, but void volumes that are very large compared with flow rates and column volumes are responsible for peak dispersion. For that reason, the integration of chromatographic components on a chip (sepa‐ ration channels and electrospray emitters for MS detection, but also additional channels, connections and microvalves) has rapidly been the major strategy to minimise void volumes

In 1978, Tsuda and Novotny experienced with the performances of packed glass capillaries

During the next years, research on chip technology was mainly focused on electroosmosis- or electrophoretic-driven separations due to the technical challenge represented by the connec‐

The simplest way to perform miniaturised liquid chromatography on chip is to coat the inner walls of the channels with chemical groups that may interact with the compounds of interest, i.e. to perform open-channel chromatography. In 1990, Manz et al. proposed the first chip prototype for open-tubular liquid chromatography made of silicon and coupled to a minia‐ turised conductometric detector connected to a classical LC pump [46]. Jacobson et al. proposed the first open-channel separation application on a glass chip coated with octadecyl‐ silane chains in 1994, with low theoretical plate heights (4.1–5 µm) [47]. Due to the small specific surface of such systems, researchers conceived coating modifications to increase the phase ratio (ratio between the volume of stationary phase and the volume of mobile phase): porous layer open-tubular (PLOT) columns, functionalised particles embedded in a porous layer [50]

Open-channel chromatography (with an ideal i.d. of 10–20 µm) provides high efficiency since the molecular diffusion is the only contributor to band broadening. However, due to its limited

specific surface, column capacity stays low even with stationary phase modifications.

UHPLC system that was design to deliver pressure up to 1000 bar [41].

10 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

inner diameter encounters a growing interest since the late 1970s [42].

*2.3.2. Column inner diameter reduction*

sample and mobile phase volume.

and efficiency drop [43].

with 50–200 µm inner diameters.

tion between LC pumps and chips.

*2.3.2.1. Open-channel chromatography*

or immobilisation of nanoparticles onto the walls [51].

Micropillars, collocated monolith support structures (COMOSS) or nanotubes may combine small channel dimensions and large specific surfaces. COMOSS were introduced in 1998 by He and Regnier in response to the difficulty to produce chromatographic columns from wafers [52]. They proposed an approach where the stationary phase is not created by polymerisation *in situ*, but by etching the chip material (e.g. quartz, polydimethylsiloxane (PDMS) or cyclic olefin copolymer (COC)) that may be further functionalised. The result is a highly well-ordered structure (Figure 6) obtained as separation support. Eddy diffusion term in the Van Deemter equation is consequently much reduced, leading to high separation efficiency.

**Figure 6.** SEM image of a COMOSS organised structure [52].

PDMS can be considered as a C1 phase, but its hydrophobicity is too low to perform adequate separation; PDMS monolithic pillars could therefore be functionalised by octasilane, octade‐ cylsilane or other groups to improve analyte separation (Figure 7).

**Figure 7.** PDMS functionalisation.

COC stationary phase has been presented for the first time by Gustafsson et al. in 2008 [55]. This material presents interesting features in terms of chemical inertness and stability in hydroorganic solvents. The hydrophobic character of COC allows it to be used as chip substrate and stationary phase. COMOSS chips made of on-porous materials as PDMS and COC have high separation efficiency, but low sample capacity due to the low interaction surface.

In addition to non-porous materials as PDMS or COC, superficially porous pillars have been proposed to circumvent the low sample capacity. Two orders of magnitude could be gained in terms of specific surface, increasing the chip sample capacity [56]. Another approach is the *in situ* growth of nanotubes on a COMOSS structure. Increased sample loading capacity and better retention than with C18-functionalised pillars could be obtained [43].

#### *2.3.2.3. Miniaturised monolithic columns*

Monoliths are continuous stationary phase beds generated by *in situ* polymerisation of monomers in the presence of porogen agents, resulting in a bimodal structure that exhibits macropores (>50–100 nm) that allow the mobile phase to pass through the column, and mesopores (<20 nm) that offers a high interaction surface for analyte retention [57]. Monolith retention properties can be defined before the polymerisation process by adjusting reagent nature and proportion, or by functionalising the polymer bed. Monoliths present undeniable chromatographic features and deserve to be more thoroughly understood in terms of synthesis parameters and their impact on chromatographic properties [43].

### *2.3.2.4. Packed particles*

Besides the above-mentioned novel LC-chip stationary phases, silica particles can also be employed with the advantage of being well-known due to their broad utilisation for decades in classical LC; a wide range of particle functionalisation types and specifications have been commercialised for a long time. However, special attention has to be paid to particle packing homogeneity and immobilisation of the particles inside the microchannel.

Different column packing procedures have been developed to find the best way to obtain homogenous particle beds. Particles could be brought into chromatographic channels and trapped between weirs or frits that prevent further particle movements. Micromachined frits demonstrate better efficiency than sintered frits that generate more band dispersion [61]. Another procedure was developed for the first time in 2002 by Ceriotti et al. [62]. They proposed a fritless configuration where the particulate bed is retained in the chromatographic channel by a tapered profile at the end of the column. Improvements to this concept were proposed by Gomez et al. that presented a packing process with increased particulate bed stability.

#### *2.3.3. The Agilent HPLC-chip*

In 2005, Agilent developed and commercialised a miniaturised HPLC-chip system designed for direct coupling to a mass spectrometer [60]. Polyimide was chosen as chip substrate material due to its chemical and physical inertness, and the low MS background generated. The fabrication process consists in laser ablation of polyimide film to form the microfluidic channels, ports, chambers and columns followed by deposition of electrical contacts for the electrospray. The last step is the packing of the sample enrichment column and LC column with the stationary phase [65]. This latter operation is performed by introducing isopropanol particle slurries into both channels under a pressure of 120 bar. A wide range of particle chemistries, dimensions and porosities are available in classical Agilent LC columns that can be packed into the chip device.

COC stationary phase has been presented for the first time by Gustafsson et al. in 2008 [55]. This material presents interesting features in terms of chemical inertness and stability in hydroorganic solvents. The hydrophobic character of COC allows it to be used as chip substrate and stationary phase. COMOSS chips made of on-porous materials as PDMS and COC have high

In addition to non-porous materials as PDMS or COC, superficially porous pillars have been proposed to circumvent the low sample capacity. Two orders of magnitude could be gained in terms of specific surface, increasing the chip sample capacity [56]. Another approach is the *in situ* growth of nanotubes on a COMOSS structure. Increased sample loading capacity and

Monoliths are continuous stationary phase beds generated by *in situ* polymerisation of monomers in the presence of porogen agents, resulting in a bimodal structure that exhibits macropores (>50–100 nm) that allow the mobile phase to pass through the column, and mesopores (<20 nm) that offers a high interaction surface for analyte retention [57]. Monolith retention properties can be defined before the polymerisation process by adjusting reagent nature and proportion, or by functionalising the polymer bed. Monoliths present undeniable chromatographic features and deserve to be more thoroughly understood in terms of synthesis

Besides the above-mentioned novel LC-chip stationary phases, silica particles can also be employed with the advantage of being well-known due to their broad utilisation for decades in classical LC; a wide range of particle functionalisation types and specifications have been commercialised for a long time. However, special attention has to be paid to particle packing

Different column packing procedures have been developed to find the best way to obtain homogenous particle beds. Particles could be brought into chromatographic channels and trapped between weirs or frits that prevent further particle movements. Micromachined frits demonstrate better efficiency than sintered frits that generate more band dispersion [61]. Another procedure was developed for the first time in 2002 by Ceriotti et al. [62]. They proposed a fritless configuration where the particulate bed is retained in the chromatographic channel by a tapered profile at the end of the column. Improvements to this concept were proposed by Gomez et al. that presented a packing process with increased particulate bed

In 2005, Agilent developed and commercialised a miniaturised HPLC-chip system designed for direct coupling to a mass spectrometer [60]. Polyimide was chosen as chip substrate material due to its chemical and physical inertness, and the low MS background generated.

separation efficiency, but low sample capacity due to the low interaction surface.

12 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

better retention than with C18-functionalised pillars could be obtained [43].

parameters and their impact on chromatographic properties [43].

homogeneity and immobilisation of the particles inside the microchannel.

*2.3.2.3. Miniaturised monolithic columns*

*2.3.2.4. Packed particles*

stability.

*2.3.3. The Agilent HPLC-chip*

Chromatographic separations on the Agilent HPLC chip are performed using pressure-driven mobile phase flow. Interfacing macrodimension pumps to nanodimension channels is made through a Chip-Cube interface in which the chip device is sandwiched between the rotor and stator (Figure 8) of a valve. Transfer capillaries from pumps, injector and to waste are connected to the valve stator, ensuring a tight and zero void-volume connection.

Analysis on HPLC-chip consists of sample loading on an enrichment column pushed by a first pump equipped with a split flow device and working at capillary flow rate. After microvalve switching, a second pump delivering a split nanoflow rate is employed to perform chroma‐ tographic separation, passing through the enrichment column and the separation channel.

HPLC-chip hyphenation to MS is ensured by an electrospray emitter incorporated in the chip device. The electrospray tip is formed of a prolongation of the polyimide laminated films that constitute the chip substrate. The latter is laser ablated to the appropriate shape (45 µm diameter and 2 mm long) and coated with a conductive metal.

The integrated design of this miniaturised device reduces drastically void volumes and leakage possibilities. Moreover, HPLC-chip is easy to use and compatible with classical LC modules (pumps, autosampler/injector), which opened a wide field of applications.

Since its commercialisation in 2005, HPLC-chip has been used in qualitative analysis of tryptic peptides and proteins, and quantitative analysis of small molecules and peptides [72].

### **3. Interests of miniaturised LC**

#### **3.1. Injection volume**

As in classical HPLC, the maximal volume that can be injected without causing a chromato‐ graphic band distortion is expressed by the following equation :

$$V\_{\max} = \frac{\theta \, D.\pi.L.d\_{\text{c}}^2 \, \epsilon\_{\text{c}} \left(1 + k\right)}{\sqrt{N}} \tag{3}$$

where *θ* is the fractional loss of the column plate number caused by the injection, *D* is the constant describing the injection profile, *L* is the column length, *d*c is the column i.d., ∊*c* the is column porosity, *k* the is retention factor and *N* the is column efficiency expressed by the theoretical plate number.

As shown in this equation, *V*max is the proportional to the square of *d*c, and the following relationship can be established:

$$\frac{V\_{\text{max}^{\text{univariate}}}}{V\_{\text{max}^{\text{causal}}}} = \frac{d^2}{d^2}\_{\text{cobalt}}\tag{4}$$

For two columns that have the same length, efficiency and porosity but differ by their inner diameter (4.6 mm for classical dimensions and 75 µm for the miniaturised version), a theoret‐ ical injection volume reduction factor of 3762 should be observed (e.g. 10 µl onto a conventional system to approximately 2.5 nl on a nano-LC column) while keeping the same chromato‐ graphic performances. Such a reduction of the required injection volume represents an undeniable advantage of miniaturised LC systems, since a growing interest is brought, for instance, to the analysis of biological matrices that are often available in limited volumes.

In practice, a great sensitivity gain can be obtained by injecting higher volumes onto the miniaturised chromatographic system, without causing peak distortion due to an overload. In the case of micro-LC, different peak compression techniques have been studied, such as oncolumn concentration or sample plug bracketing. In nano-LC, a trapping column is often connected to the analytical column by a valve, allowing large sample volumes to be loaded onto the system and the sample to be pre-concentrated.

#### **3.2. Peak concentration**

A reduction of the inner diameter of a chromatographic column results in a higher peak concentration at the detector (*C*max), as shown in the following equation:

$$C\_{\max} = \sqrt{\frac{N}{2\pi}} \times \frac{4m}{\pi.L.V\_{\circ}.\epsilon\_{\circ}.(1+k)}\tag{5}$$

Advances in Low Volume Sample Analysis Using Microfluidic Separation Techniques http://dx.doi.org/10.5772/64952 15

**Figure 9.** Schematic representation of band broadening components in a chromatographic system (adapted from Lauer [79]).

where *m* is the total amount of sample loaded on the column and *V*0 is the column volume.

*C*max is the proportional to *m* and to *N*, and inversely proportional to *V*0. Since *V*<sup>0</sup> is directly related to *d*c, *C*max is inversely proportional to the square of column diameter. In other words, Eq. (6) can be used to illustrate the sensitivity gain that can be expected with miniaturised columns.

$$\frac{\mathbf{C}\_{\text{max}^{\text{univator}}}}{\mathbf{C}\_{\text{max}^{\text{conivator}}}} = \frac{V\_{0^{\text{cathest}}}}{V\_{0^{\text{mutativ}}}} = \frac{d\_{c^{\text{cashest}}}^2}{d\_{c^{\text{cashest}}}^2} \tag{6}$$

Downscaling the size from classical dimensions (4.6 mm) to miniaturised dimensions (75 µm) would theoretically result in a gain factor of *C*max of 3762.

#### **3.3. Void volume reduction**

**3. Interests of miniaturised LC**

As in classical HPLC, the maximal volume that can be injected without causing a chromato‐

( ) <sup>2</sup> . . . . . .1 *c c*

(3)

*N*

where *θ* is the fractional loss of the column plate number caused by the injection, *D* is the constant describing the injection profile, *L* is the column length, *d*c is the column i.d., ∊*c* the is column porosity, *k* the is retention factor and *N* the is column efficiency expressed by the

As shown in this equation, *V*max is the proportional to the square of *d*c, and the following

2 2 *miniaturised miniaturised classical classical*

For two columns that have the same length, efficiency and porosity but differ by their inner diameter (4.6 mm for classical dimensions and 75 µm for the miniaturised version), a theoret‐ ical injection volume reduction factor of 3762 should be observed (e.g. 10 µl onto a conventional system to approximately 2.5 nl on a nano-LC column) while keeping the same chromato‐ graphic performances. Such a reduction of the required injection volume represents an undeniable advantage of miniaturised LC systems, since a growing interest is brought, for instance, to the analysis of biological matrices that are often available in limited volumes.

In practice, a great sensitivity gain can be obtained by injecting higher volumes onto the miniaturised chromatographic system, without causing peak distortion due to an overload. In the case of micro-LC, different peak compression techniques have been studied, such as oncolumn concentration or sample plug bracketing. In nano-LC, a trapping column is often connected to the analytical column by a valve, allowing large sample volumes to be loaded

A reduction of the inner diameter of a chromatographic column results in a higher peak

2 . . . .1 *max*

*N m <sup>C</sup>* p p *LV k* = ´ ò

4

*o c*

( )

<sup>+</sup> (5)

concentration at the detector (*C*max), as shown in the following equation:

*<sup>V</sup> <sup>d</sup>* <sup>=</sup> (4)

*D Ld k*

graphic band distortion is expressed by the following equation :

14 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

*max*

q p <sup>+</sup> <sup>=</sup> ò

*max max*

onto the system and the sample to be pre-concentrated.

*V d*

*V*

**3.1. Injection volume**

theoretical plate number.

**3.2. Peak concentration**

relationship can be established:

Void volumes are detrimental to the chromatographic performances in all LC configurations. However, when working with miniaturised systems, the smallest void volume can act as a mixing chamber and result in an important loss in sensitivity and separation efficiency. The total band dispersion occurring in a chromatographic system (Figure 9) can be expressed by the total variance *σtot* 2 that sums the variances due to the column (*σcol* <sup>2</sup> ) and to the rest of the chromatographic system (*σext* <sup>2</sup> ).

$$
\sigma\_{\rm tot}^2 = \sigma\_{\rm ext}^2 + \sigma\_{\rm col}^2 \tag{7}
$$

Band dispersion due to the column, *σcol* <sup>2</sup> , is in particular a function of column volume (and consequently to *dc* 2 ) and efficiency, which are physical properties that cannot be changed for a given column in order to decrease peak broadening:

$$
\sigma\_{col}^2 = \frac{\pi .L.d\_c^2.\epsilon\_c.(1+k)}{4\sqrt{N}}\tag{8}
$$

However, other factors having an influence on band broadening through *σext* 2 can be expressed as:

$$
\sigma\_{ext}^2 = \frac{V\_{in\parallel}^2}{d\_c^2} + \sigma\_0^2 \tag{9}
$$

where *Vinj* 2 is the injection volume, *σ*<sup>0</sup> 2 is the instrument variance and *σext* <sup>2</sup> is the extra-column variance.

As shown in this equation, *Vinj* 2 and *σ*<sup>0</sup> 2 are directly related to *σext* 2 [79]. In other words, the minimisation of extra-column void volumes by using the smallest connection capillaries and fittings possible is clearly beneficial to avoid chromatographic band dispersion.

In the light of these considerations, systems with very low extra-column void volumes have been developed including integrated systems (see Section 2.3.3).

#### **3.4. Low flow rate**

Mobile phase flow rate *F* is a value that is also related to the internal column diameter as seen in Eq. (10):

$$F = \frac{\pi \, d\_c^2 \, . \epsilon\_c \, . \mu}{4} \tag{10}$$

where *u* is the mobile phase velocity.

The following relationship can be written in Eq. (11):

$$\frac{F\_{\text{causal}}}{F\_{\text{stoatored}}} = \frac{d^2\_{\text{causal}}}{d^2\_{\text{voinstuation}}} \tag{11}$$

This drastic flow rate reduction has evident economical and ecological advantages, especially when working with pumping systems that directly deliver the right mobile phase flow rate without involving the use of a split flow system.

#### **3.5. Retention volume**

(8)

2 can be expressed

<sup>2</sup> is the extra-column

2 [79]. In other words, the

( ) <sup>2</sup>

*Ld k N*

<sup>2</sup> . . . .1 4 *c c*

<sup>+</sup> <sup>=</sup> ò

2 2 2 2 0 *inj*

 s

minimisation of extra-column void volumes by using the smallest connection capillaries and

In the light of these considerations, systems with very low extra-column void volumes have

Mobile phase flow rate *F* is a value that is also related to the internal column diameter as seen

<sup>2</sup> . .. 4 *c c d u <sup>F</sup>* p

ò

2 2 *classical classical miniaturised miniaturised*

This drastic flow rate reduction has evident economical and ecological advantages, especially when working with pumping systems that directly deliver the right mobile phase flow rate

=

*F d*

is the instrument variance and *σext*

are directly related to *σext*

= + (9)

(10)

*<sup>F</sup> <sup>d</sup>* <sup>=</sup> (11)

*V d*

*c*

p

However, other factors having an influence on band broadening through *σext*

*ext*

s

2

fittings possible is clearly beneficial to avoid chromatographic band dispersion.

2 and *σ*<sup>0</sup> 2

been developed including integrated systems (see Section 2.3.3).

2 is the injection volume, *σ*<sup>0</sup>

As shown in this equation, *Vinj*

where *u* is the mobile phase velocity.

The following relationship can be written in Eq. (11):

without involving the use of a split flow system.

*col*

s

16 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

as:

where *Vinj*

variance.

**3.4. Low flow rate**

in Eq. (10):

The retention volume *VR* is defined as the mobile phase volume that is required to elute a compound of a given retention time *tR* :

$$\mathbf{V}\_{\mathcal{R}} = \mathbf{t}\_{\mathcal{R}}.\\F = \mathbf{t}\_{\mathcal{R}}. \frac{\boldsymbol{\pi}. \boldsymbol{d}\_c^2}{\mathbf{4}}. \boldsymbol{\epsilon}\_c. \boldsymbol{u} \tag{12}$$

In the light of the reduced column dimensions in miniaturised LC systems compared with classical systems, the mobile phase volume that is needed to elute a compound with a specified *k* value is reduced proportionally to the square of the internal column diameter, as shown in Eq. (13).

$$\frac{V\_{R^{\text{classical}}}}{V\_{R^{\text{univalent}}}} = \frac{d^2}{d^2\_{\text{unnatural}}}\tag{13}$$

#### **3.6. Hyphenation to MS**

When using mass spectrometry, compounds of interest have to carry a net positive or negative charge, depending on the mode that is employed. Analyte electrospray ionisation occurs in three major steps: first, charged droplets are formed from the chromatographic eluent under the action of a strong electric field. The eluent takes the shape of a cone (the Taylor cone) when a critical electric field threshold is reached. A pneumatic assistance is required to provide stable droplet formation in the classical LC [80]. Then, charged droplets undergo Coulomb fission into smaller daughter droplets: eluent solvent progressively evaporates in the heated source until reaching the Rayleigh limit where the electrostatic repulsion forces are exactly equal to the surface tension of the solvent [81]. Beyond the Rayleigh limit, droplets become unstable and divide into smaller droplets. Eq. (14) presents the relationship between droplet charge and Rayleigh radius.

$$\mathbf{Q}^2 = \mathbf{64}\pi^2 \varepsilon\_0 \mathbf{y} \mathbf{R}\_{\mathbf{R}}^3 \tag{14}$$

where *Q* is the droplet charge , *ε*0 is the vacuum permittivity and *RR* is the Rayleigh radius.

The ion transfer from small droplets to the gas phase can happen following two mechanisms. The ion evaporation model described by Iribarne and Thomson is commonly admitted to describe the small ion formation [81]. According to this model, the electric field at the droplet surface becomes strong enough at an intermediate state and before reaching the Rayleigh limit to directly desorb ions from the droplet (Figure 10) [82].

**Figure 10.** Ion evaporation model.

A second model proposed by Dole, or the charged residue model, could be appropriate to describe protein ionisation. This model suggests that successive Coulomb fissions occurring when the Rayleigh limit is reached, finally yielding droplets containing one single charge (Figure 11) [83].

**Figure 11.** Charge residue model.

Nanoelectrospray (nano-ESI) source was first introduced in 1994 as a response to the devel‐ opment of low flow separation devices. Typical flow rates in nano-ESI are 200–1000 nl/min and the i.d. of spray emitter is about 10–20 µm. The interest of such a miniaturised ionisation source is the improvement of the overall ionisation efficiency (the number of ions recorded at the detector divided by the number of analyte molecule sprayed) [86]. Since signal intensity with ESI sources is concentration sensitive rather than mass sensitive, low analyte amount are advantageously detected at lower flow rates with higher peak concentrations thanks to the miniaturised technique, as previously explained. Lower flow rates as well as narrower emitter tip orifice produce smaller droplets (2–3 orders of magnitude reduction), and desolvation efficiency is increased: smaller initial droplet size requires less Coulomb fission and solvent evaporation to release charged compounds into gas phase, making a larger portion of ions available to detection.

#### **4. Sample preparation**

In the light of the previously described features of miniaturised separation techniques, having low volume samples with the highest concentration possible is a clear objective. On the other hand, analysis of complex media (e.g. environmental, forensic, food, pharmaceutical or biological samples) requires preliminary purification to isolate analyte from contaminants and interferences, and to avoid column or capillary blockage, reduced separation phase lifetime and MS ion suppression. In addition, sample preparation may allow analyte concentration and analyte matrix simplification to make the sample fully compatible with separation technique and detection.

The combination of miniaturised sample preparation and separation techniques offers the main advantages of high throughput, high sensitivity and low costs. The most employed miniaturised sample preparation techniques are briefly described below.

**Figure 10.** Ion evaporation model.

**Figure 11.** Charge residue model.

available to detection.

**4. Sample preparation**

(Figure 11) [83].

A second model proposed by Dole, or the charged residue model, could be appropriate to describe protein ionisation. This model suggests that successive Coulomb fissions occurring when the Rayleigh limit is reached, finally yielding droplets containing one single charge

18 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

Nanoelectrospray (nano-ESI) source was first introduced in 1994 as a response to the devel‐ opment of low flow separation devices. Typical flow rates in nano-ESI are 200–1000 nl/min and the i.d. of spray emitter is about 10–20 µm. The interest of such a miniaturised ionisation source is the improvement of the overall ionisation efficiency (the number of ions recorded at the detector divided by the number of analyte molecule sprayed) [86]. Since signal intensity with ESI sources is concentration sensitive rather than mass sensitive, low analyte amount are advantageously detected at lower flow rates with higher peak concentrations thanks to the miniaturised technique, as previously explained. Lower flow rates as well as narrower emitter tip orifice produce smaller droplets (2–3 orders of magnitude reduction), and desolvation efficiency is increased: smaller initial droplet size requires less Coulomb fission and solvent evaporation to release charged compounds into gas phase, making a larger portion of ions

In the light of the previously described features of miniaturised separation techniques, having low volume samples with the highest concentration possible is a clear objective. On the other

Liquid-liquid extraction (LLE) is a sample preparation technique that relies on the partition of analytes between two immiscible liquid phases. The best results are obtained for compounds showing a clear preference for one liquid over the other one. Factors that influence compound partition include liquid phase polarity, pH, analyte p*K*a and polarity, and mixing and contact duration. Micro liquid-liquid extraction (MLLE) is a simple downscaling of classical LLE procedure. The use of lower sample volumes has economical and ecological advantages since the apolar liquid phase is often constituted of alkanes (e.g. pentane, hexane and cyclohexane) or chlorinated solvents; moreover, reduced solvent volumes may lead to the increased analyte concentration.

Solid-phase microextraction (SPME) is a miniaturised sample preparation process involving a fused-silica rod coated with a polymeric layer employed as extraction medium. This technique is applied for the extraction of trace compounds from liquid or gas samples [94] (Figure 12). Analyte desorption is performed by heating the SPME fibre in a classical GC injector for volatile and thermally stable compounds, or by a special desorption device for nonvolatile or thermally unstable compounds for subsequent LC [95, 96] or CE [97] analysis.

Dried spots are an expanding way of microsampling and purifying biological samples as blood (dried blood spots, DBS), serum (dried serum spots, DSS) or plasma (dried plasma spots, DPS). A few microliters of a biological fluid are collected on a filter paper and allowed to dry. The dried spot is then punched out and desorbed in an appropriate mixture of solvent chosen to enable maximal analyte extraction while minimising interference desorption (Figure 13). In addition to analytical advantages as small sample volume requirements and low cost, dried spots are very convenient from a sampling point of view: the collection technique is not invasive and can be performed without pain, e.g. for pharmacokinetic studies on laboratory animals or for systematic disease screening on newborns.

Finally, solid-phase extraction (SPE) follows the miniaturisation trend by reducing cartridge and solid phase bed volume (Figure 14A and B). In this technique, sample is loaded in a tube containing a few mg to a few tens mg particles maintained in the bottom of the cartridge by two frits. Sample loading solvent has to be carefully chosen to ensure analyte retention on the particles. Washing steps are then performed to remove a maximal amount of contaminants and interferences that are co-retained on the solid phase, while maintaining analyte-particle interactions. Elution is the final step of SPE to collect a sample containing the analyte for further analysis. Downscaling SPE support allows preparing sample volumes as low as 10 µl, and

**Figure 12.** Extraction from aqueous sample solution by conventional SPME device. (A) Liquid phase sampling and (B) headspace sampling [99].

**Figure 13.** DBS sampling and extraction procedure.

analyte elution by similar volumes. Moreover, SPE or micro-SPE supports are increasingly available in 96-well format (Figure 14C) to provide high extraction throughput by the use of multichannel pipettes or extraction automation.

Advances in Low Volume Sample Analysis Using Microfluidic Separation Techniques http://dx.doi.org/10.5772/64952 21

**Figure 14.** (A) Classical SPE cartridge. (B) Miniaturised SPE cartridge. (C) 96-well SPE plate.

## **5. Conclusions and perspectives**

To summarise, the advantages of microfluidic devices include their small size, improved sensitivity, low sample volume requirements, rapid analysis, potential disposability, and importantly their ease of use that eliminates the need for skilled personnel to perform the assays. In the same time, ethical, analytical and sample availability considerations are a challenge faced by many (bio)analytical laboratories and have resulted in a drive to limit sample volume.

Integration of various nanotechniques through microfabrication processes and advances in detection devices and informatics drive new types of analysis facilitating on-site multicom‐ ponent analysis resulting in rapid diagnostic tools and rapid screening methods in various application fields (clinical, pharmaceutical and biopharmaceutical, environmental, food analysis, etc.).

## **Acknowledgements**

The authors thank the Fund for Scientific Research (F.R.S.-FNRS, Belgium), the Walloon Region (WR), the Leon Fredericq Fund and University of Liege for financial support.

## **Author details**

analyte elution by similar volumes. Moreover, SPE or micro-SPE supports are increasingly available in 96-well format (Figure 14C) to provide high extraction throughput by the use of

**Figure 12.** Extraction from aqueous sample solution by conventional SPME device. (A) Liquid phase sampling and (B)

20 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

multichannel pipettes or extraction automation.

**Figure 13.** DBS sampling and extraction procedure.

headspace sampling [99].

Virginie Houbart and Marianne Fillet\*

\*Address all correspondence to: marianne.fillet@ulg.ac.be

Laboratory for the Analysis of Medicines (LAM), Department of Pharmacy, CIRM, University of Liege, Liege, Belgium

## **References**


[12] Schappler J, Veuthey J-L, Rudaz S. 18 Coupling CE and microchip-based devices with mass spectrometry. In: Satinder A, Jimidar MI, editors. Separation Science and Technology. Volume 9. Academic Press; 2008. p. 477–521.

**References**

10.1016/0009-5907(67)80003-6.

10.1021/ac00135a034.

10.1002/rcm.1290030310

6. doi:10.1021/ac00104a027.

30. doi:10.1021/ac990535e.

10.1016/0021-9673(94)85026-7.

emitters. Analytical Chemistry. 1999;71:4115–8.

1082.

[1] Hjertén S. Free zone electrophoresis. Chromatographic Reviews. 1967;9:122–219. doi:

[2] Jorgenson JW, Lukacs KD. Zone electrophoresis in open-tubular glass capillaries. An‐

[3] Olivares JA, Nguyen NT, Yonker CR, Smith RD. On-line mass spectrometric detec‐ tion for capillary zone electrophoresis. Analytical Chemistry. 1987;59:1230–2. doi:

[4] Hommerson P, Khan AM, de Jong GJ, Somsen GW. Ionization techniques in capillary electrophoresis-mass spectrometry: principles, design, and application. Mass Spec‐

[5] Moseley MA, Deterding LJ, Tomer KB, Jorgenson JW. Capillary-zone electrophore‐ sis/fast-atom bombardment mass spectrometry: design of an on-line coaxial continu‐ ous-flow interface. Rapid Communications in Mass Spectrometry. 1989;3:87–93. doi:

[6] Takada Y, Sakairi M, Koizumi H. Atmospheric pressure chemical ionization interface for capillary electrophoresis/mass spectrometry. Analytical Chemistry. 1995;67:1474–

[7] Nilsson SL, Andersson C, Sjöberg PJR, Bylund D, Petersson P, Jörntén-Karlsson M, et al. Phosphate buffers in capillary electrophoresis/mass spectrometry using atmos‐ pheric pressure photoionization and electrospray ionization. Rapid Communications

[8] Chang YZ, Her GR. Sheathless capillary electrophoresis/electrospray mass spectrom‐ etry using a carbon-coated fused-silica capillary. Analytical Chemistry. 1999;72:626–

[9] Nilsson S, Klett O, Svedberg M, Amirkhani A, Nyholm L. Gold-coated fused-silica sheathless electrospray emitters based on vapor-deposited titanium adhesion layers. Rapid Communications in Mass Spectrometry. 2003;17:1535–40. doi:10.1002/rcm.

[10] Wahl JH, Gale DC, Smith RD. Sheathless capillary electrophoresis-electrospray ioni‐ zation mass spectrometry using 10 µm I.D. capillaries: analyses of tryptic digests of cytochrome *c*. Journal of Chromatography A. 1994;659:217–22. doi:

[11] Barnidge DR, Nilsson S, Markides KE. A design for low-flow sheathless electrospray

alytical Chemistry. 1981;53:1298–302. doi:10.1021/ac00231a037.

22 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

trometry Reviews. 2011;30:1096–120. doi:10.1002/mas.20313.

in Mass Spectrometry. 2003;17:2267–72. doi:10.1002/rcm.1182.


[40] Wu N, Collins DC, Lippert JA, Xiang Y, Lee ML. Ultrahigh pressure liquid chroma‐ tography/time-of-flight mass spectrometry for fast separations. Journal of Microcol‐ umn Separations. 2000;12:462–9. doi:10.1002/1520-667X(2000)12:8<462::AID-MCS5>3.0.CO;2-F.

[26] Martin A, Desty D. Vapour Phase Chromatography. Butterworths, London; 1957. p.

[28] Bartle KD, Myers P. History of gas chromatography. TrAC Trends in Analytical

[29] Terry SC, Jerman JH, Angell JB. A gas chromatographic air analyzer fabricated on a

[30] Gross GM, Grate JW, Synovec RE. Monolayer-protected gold nanoparticles as an effi‐ cient stationary phase for open tubular gas chromatography using a square capillary. Journal of Chromatography A. 2004;1029:185–92. doi:10.1016/j.chroma.2003.12.058.

[31] Narayanan S, Alfeeli B, Agah M. A micro gas chromatography chip with an embed‐ ded non-cascaded thermal conductivity detector. Procedia Engineering. 2010;5:29–32.

[32] Tranchida PQ, Mondello L. Current-day employment of the micro-bore open-tubular capillary column in the gas chromatography field. Journal of Chromatography A.

[33] van Deemter JJ, Zuiderweg FJ, Klinkenberg A. Longitudinal diffusion and resistance to mass transfer as causes of nonideality in chromatography. Chemical Engineering

[34] Lavrik NV, Taylor LT, Sepaniak MJ. Nanotechnology and chip level systems for pressure driven liquid chromatography and emerging analytical separation techni‐ ques: a review. Analytica Chimica Acta. 2011;694:6–20. doi:10.1016/j.aca.2011.03.059.

[35] Lin S-L, Bai H-Y, Lin T-Y, Fuh M-R. Microfluidic chip-based liquid chromatography coupled to mass spectrometry for determination of small molecules in bioanalytical

[36] Snyder LR. Peer reviewed: HPLC: past and present. Analytical Chemistry.

[37] MacNair JE, Lewis KC, Jorgenson JW. Ultrahigh-pressure reversed-phase liquid chromatography in packed capillary columns. Analytical Chemistry. 1997;69:983–9.

[38] Lippert JA, Xin B, Wu N, Lee ML. Fast ultrahigh-pressure liquid chromatography: on-column UV and time-of-flight mass spectrometric detection. Journal of Microcol‐ umn Separations. 1999;11:631–43. doi:10.1002/(SICI)1520-667X(199911)11:9<631::AID-

[39] Wu N, Clausen AM. Fundamental and practical aspects of ultrahigh pressure liquid chromatography for fast separations. Journal of Separation Science. 2007;30:1167–82.

applications. Electrophoresis. 2012;33:635–43. doi:10.1002/elps.201100380.

[27] Golay M. Gas chromatography. Academic Press. New York; 1958.

24 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

Chemistry. 2002;21:547–57. doi:10.1016/S0165-9936(02)00806-3.

doi:10.1016/j.proeng.2010.09.040.

2012;1261:23–36. doi:10.1016/j.chroma.2012.05.074.

2000;72:412 A–20 A. doi:10.1021/ac002846r.

doi:10.1021/ac961094r.

MCS1>3.0.CO;2-I.

doi:10.1002/jssc.200700026.

Science. 1995;50:3869–82. doi:10.1016/0009-2509(96)81813-6.

silicon wafer. IEEE Transactions on Electron Devices. 1979;26:1880–6.

1.


[64] Gaspar A, Hernandez L, Stevens S, Gomez FA. Electrochromatography in microchips packed with conventional reversed-phase silica particles. Electrophoresis. 2008;29:1638–42. doi:10.1002/elps.200700489.

[52] He B, Regnier F. Microfabricated liquid chromatography columns based on collocat‐ ed monolith support structures. Journal of Pharmaceutical and Biomedical Analysis.

[53] Slentz BE, Penner NA, Lugowska E, Regnier F. Nanoliter capillary electrochromatog‐ raphy columns based on collocated monolithic support structures molded in poly(di‐ methyl siloxane). Electrophoresis. 2001;22:3736–43. doi:

[54] Pumera M. Microchip-based electrochromatography: designs and applications. Ta‐

[55] Gustafsson O, Mogensen KB, Kutter JP. Underivatized cyclic olefin copolymer as substrate material and stationary phase for capillary and microchip electrochroma‐

[56] De Malsche W, Clicq D, Verdoold V, Gzil P, Desmet G, Gardeniers H. Integration of porous layers in ordered pillar arrays for liquid chromatography. Lab on a Chip.

[57] Cabrera K. Applications of silica-based monolithic HPLC columns. Journal of Separa‐

[58] Smith NW, Evans MB. The analysis of pharmaceutical compounds using electrochro‐

[59] Oleschuk RD, Shultz-Lockyear LL, Ning Y, Harrison DJ. Trapping of bead-based re‐ agents within microfluidic systems: on-chip solid-phase extraction and electrochro‐

[60] Yin H, Killeen K, Brennen R, Sobek D, Werlich M, van de Goor T. Microfluidic chip for peptide analysis with an integrated HPLC column, sample enrichment column, and nanoelectrospray tip. Analytical Chemistry. 2005;77:527–33. doi:10.1021/

[61] Ehlert S, Kraiczek K, Mora J-A, Dittmann M, Rozing GP, Tallarek U. Separation effi‐ ciency of particle-packed HPLC microchips. Analytical Chemistry. 2008;80:5945–50.

[62] Ceriotti L, de Rooij NF, Verpoorte E. An integrated fritless column for on-chip capil‐ lary electrochromatography with conventional stationary phases. Analytical Chemis‐

[63] Gaspar A, Piyasena ME, Gomez FA. Fabrication of fritless chromatographic micro‐ chips packed with conventional reversed-phase silica particles. Analytical Chemistry.

matography. Chromatographia. 1994;38:649–57. doi:10.1007/BF02277170.

matography. Analytical Chemistry. 1999;72:585–90. doi:10.1021/ac990751n.

tography. Electrophoresis. 2008;29:3145–52. doi:10.1002/elps.200800131.

10.1002/1522-2683(200109)22:17<3736::AID-ELPS3736>3.0.CO;2-Y.

lanta. 2005;66:1048–62. doi:10.1016/j.talanta.2005.01.006.

tion Science. 2004;27:843–52. doi:10.1002/jssc.200401827.

2007;7:1705–11. doi:10.1039/B710507J.

ac049068d.

doi:10.1021/ac800576v.

try. 2002;74:639–47. doi:10.1021/ac0109467.

2007;79:7906–9. doi:10.1021/ac071106g.

1998;17:925–32. doi:10.1016/S0731-7085(98)00060-0.

26 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences


spectrometry: practical considerations. Journal of Mass Spectrometry. 1999;34:244–54. doi:10.1002/(SICI)1096-9888(199904)34:4<244::AID-JMS775>3.0.CO;2-0.

[89] Juraschek R, Dülcks T, Karas M. Nanoelectrospray—more than just a minimizedflow electrospray ionization source. Journal of the American Society for Mass Spec‐ trometry. 1999;10:300–8. doi:10.1016/S1044-0305(98)00157-3.

[76] Leon-Gonzalez ME, Rosales-Conrado N, Perez-Arribas LV, Polo-Diez LM. Large in‐ jection volumes in capillary liquid chromatography: Study of the effect of focusing on chromatographic performance. Journal of Chromatography A. 2010;1217:7507–13.

[77] Tao D, Zhang L, Shan Y, Liang Z, Zhang Y. Recent advances in micro-scale and nano-scale high-performance liquid-phase chromatography for proteome research. Analytical and Bioanalytical Chemistry. 2011;399:229–41. doi:10.1007/

[78] Lauer HH, Rozing GP. The selection of optimal conditions in HPLC II. The influence of column dimensions and sample size on solute detection. Chromatographia.

[79] Lauer HH, Rozing GP. The selection of optimum conditions in HPLC I. The determi‐ nation of external band spreading in LC instruments. Chromatographia. 1981;14:641–

[80] Ikonomou MG, Blades AT, Kebarle P. Electrospray-ion spray: a comparison of mech‐ anisms and performance. Analytical Chemistry. 1991;63:1989–98. doi:10.1021/

[81] Kebarle P. A brief overview of the present status of the mechanisms involved in elec‐ trospray mass spectrometry. Journal of Mass Spectrometry. 2000;35:804–17. doi:

[82] Iribarne JV, Thomson BA. On the evaporation of small ions from charged droplets.

[83] Dole M, Mack LL, Hines RL, Mobley RC, Ferguson LD, Alice MB. Molecular beams

[84] Wilm MS, Mann M. Electrospray and Taylor-Cone theory, Dole's beam of macromo‐ lecules at last? International Journal of Mass Spectrometry and Ion Processes.

[85] Emmett MR, Caprioli RM. Micro-electrospray mass spectrometry: ultra-high-sensi‐ tivity analysis of peptides and proteins. Journal of the American Society for Mass

[86] Wilm M, Mann M. Analytical properties of the nanoelectrospray ion source. Analyti‐

[87] El-Faramawy A, Siu KW, Thomson BA. Efficiency of nano-electrospray ionization. Journal of the American Society for Mass Spectrometry. 2005;16:1702–7. doi:10.1016/

[88] Abian J, Oosterkamp AJ, Gelpí E. Comparison of conventional, narrow-bore and ca‐ pillary liquid chromatography/mass spectrometry for electrospray ionization mass

10.1002/1096-9888(200007)35:7<804::AID-JMS22>3.0.CO;2-Q.

of macroions. The Journal of Chemical Physics. 1968;49:2240–9.

Spectrometry. 1994;5:605–13. doi:10.1016/1044-0305(94)85001-1.

The Journal of Chemical Physics. 1976;64:2287–94.

doi:10.1016/j.chroma.2010.09.076.

28 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

1982;15:409–13. doi:10.1007/BF02261599.

s00216-010-3946-7.

7. doi:10.1007/BF02291104.

ac00018a017.

1994;136:167–80.

cal Chemistry. 1996;68:1–8.

j.jasms.2005.06.011.


**Chapter 2**

## **Molecular Microfluidic Bioanalysis: Recent Progress in Preconcentration, Separation, and Detection**

[101] Abu-Rabie P, Spooner N. Dried matrix spot direct analysis: evaluating the robustness of a direct elution technique for use in quantitative bioanalysis. Bioanalysis.

[102] McDade TWS, Snodgrass J. What a drop can do: dried blood spots as a minimally invasive method for integrating biomarkers into population-based research. Demog‐

[103] Li W-T. Dried blood spot sampling in combination with LC-MS/MS for quantitative

[104] Britz-McKibbin P. Expanded newborn screening of inborn errors of metabolism by capillary electrophoresis-electrospray ionization-mass spectrometry (CE-ESI-MS).

[105] Gilar M, Bouvier ES, Compton BJ. Advances in sample preparation in electromigra‐ tion, chromatographic and mass spectrometric separation methods. Journal of Chro‐

[106] Ekström S, Wallman L, Hök D, Marko-Varga G, Laurell T. Miniaturized solid-phase extraction and sample preparation for MALDI MS using a microfabricated integrated selective enrichment target. Journal of Proteome Research. 2006;5:1071-81. doi:

analysis of small molecules. Biomed Chromatogr. 2010;24:49-65.

matography A. 2001;909:111–35. doi:10.1016/S0021-9673(00)01108-0.

2011;3:2769–81. doi:10.4155/bio.11.270.

30 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

Methods Mol Biol. 2013;919:43-56.

raphy. 2007;44:899-925.

10.1021/pr050434z.

Emmanuel Roy, Antoine Pallandre, Bacem Zribi, Marie-Charlotte Horny, François-Damien Delapierre, Andrea Cattoni, Jean Gamby and Anne-Marie Haghiri-Gosnet

Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/65772

#### **Abstract**

This chapter reviews the state-of-art of microfluidic devices for molecular bioanalysis with a focus on the key functionalities that have to be successfully integrated, such as preconcen‐ tration, separation, signal amplification, and detection. The first part focuses on both pas‐ sive and electrophoretic separation/sorting methods, whereas the second part is devoted to miniaturized biosensors that are integrated in the last stage of the fluidic device.

**Keywords:** microfluidic bioanalysis, separation, concentration, on-a-chip optical detec‐ tion, electrochemical sensors

## **1. Introduction**

Advances in biochemistry and technology for enhancing sensitivity and selectivity of bioa‐ nalysis play a central role in clinical chemistry and medical diagnostics. The latter are per‐ formed much earlier to prevent disease or in a repetitive manner to define more specific and personal therapies. However, such analytical protocols are often implemented at the macro‐ scale level where large volumes of samples are needed. The development of microfluidic bioanalysis thus becomes important, since these platforms can offer short analysis time to result in volume smaller than 1 µl, low cost, multiplexed analysis of several analytes, and portability. Therefore, the development of extremely sensitive, highly selective, simple, robust and yet inexpensive miniaturized platforms has become essential for a wide range of appli‐ cations, including clinical diagnostics, environmental monitoring, and food safety testing. This chapter reviews the state-of-art of microfluidic devices for molecular bioanalysis with a focus

© 2016 The Author(s). Licensee InTech. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

on the key functionalities that have been successfully integrated in the chip, such as precon‐ centration, separation, and detection.

Impressive innovations have been demonstrated allowing selectively sorting, concentrating, and amplifying analytes of interest [1–3]. Therefore, the first section of this chapter mainly targets passive and electrical-based strategies for biomolecular and cellular purification, and concentration toward genomic, proteomic, and metabolite applications. The second part focuses on sensitive and selective detection techniques. Two routes are described: (i) on-chip detection based on nanostructured biophotonic sensors and (ii) electrochemical detection. These sensors are integrated in small microchannels or microchambers where specific enhanced sensing properties are generated. The correlation between the biosensor interface and its microfluidic environment for optimal detection of analytes is reviewed. Fundamental kinetics and mass-transport versus the microfluidic properties in terms of fluidic characteris‐ tics is introduced for highly operative and sensitive microfluidic applications.

#### **2. Microfluidic filtration, concentration, and sorting**

Molecular extraction and purification from a biofluid for diagnostic and therapeutic purposes using microfluidics are sensitive issues. Such objectives are challenging regarding the manip‐ ulation of the complex biofluid system such as blood or sputum. Moreover, the extraction and purification steps are also important for the quality and the pertinence of the analyte identifi‐ cation and quantification. Each progress and advance will definitively help the clinicians for proper medication. The advantages of such a microfluidic device for genomic pathogenic recognition are also of tremendous importance, because such a platform can shorten the sample analysis time compared to classical methods. Conventional processing methods for preconcentration and separation of analytes entail numerous manual and time-consuming steps. Consequently, it requires most often highly skilled operators, who do not always guarantee the absence of mishandling and contaminants. As a unique example of the needs for rapid molecular extraction from blood, sepsis and blood stream infections that are a major cause of death [4] impose daily patient follow-up to doctors in virology and bacteriology. Antimicrobial therapy should thus be curtailed as early as practicable, ideally just after the identification of the causative pathogen. Indeed, delay in effective treatment reduces the survival rate on average of 8% for the each following 6-hour period [5–7]. However, extraction of pathogen agents in the range of 1–5 units per ml with clinical relevance is a critical issue. Meanwhile, the gold standard blood culture (BC) approach for pathogenic and antibiotic susceptibility identification remains yet a major limitation with turnaround time of 2–5 days [8, 9]. Moreover, less than 10% of blood samples processed from hospitalized patients with blood infections are positive [10, 11]. Decreasing the sample analysis time compared to classical methods with microfluidics thus appears very challenging for diagnosis. In this section, main sorting, extraction, and purification methods are introduced.

#### **2.1. Passive approaches**

on the key functionalities that have been successfully integrated in the chip, such as precon‐

Impressive innovations have been demonstrated allowing selectively sorting, concentrating, and amplifying analytes of interest [1–3]. Therefore, the first section of this chapter mainly targets passive and electrical-based strategies for biomolecular and cellular purification, and concentration toward genomic, proteomic, and metabolite applications. The second part focuses on sensitive and selective detection techniques. Two routes are described: (i) on-chip detection based on nanostructured biophotonic sensors and (ii) electrochemical detection. These sensors are integrated in small microchannels or microchambers where specific enhanced sensing properties are generated. The correlation between the biosensor interface and its microfluidic environment for optimal detection of analytes is reviewed. Fundamental kinetics and mass-transport versus the microfluidic properties in terms of fluidic characteris‐

Molecular extraction and purification from a biofluid for diagnostic and therapeutic purposes using microfluidics are sensitive issues. Such objectives are challenging regarding the manip‐ ulation of the complex biofluid system such as blood or sputum. Moreover, the extraction and purification steps are also important for the quality and the pertinence of the analyte identifi‐ cation and quantification. Each progress and advance will definitively help the clinicians for proper medication. The advantages of such a microfluidic device for genomic pathogenic recognition are also of tremendous importance, because such a platform can shorten the sample analysis time compared to classical methods. Conventional processing methods for preconcentration and separation of analytes entail numerous manual and time-consuming steps. Consequently, it requires most often highly skilled operators, who do not always guarantee the absence of mishandling and contaminants. As a unique example of the needs for rapid molecular extraction from blood, sepsis and blood stream infections that are a major cause of death [4] impose daily patient follow-up to doctors in virology and bacteriology. Antimicrobial therapy should thus be curtailed as early as practicable, ideally just after the identification of the causative pathogen. Indeed, delay in effective treatment reduces the survival rate on average of 8% for the each following 6-hour period [5–7]. However, extraction of pathogen agents in the range of 1–5 units per ml with clinical relevance is a critical issue. Meanwhile, the gold standard blood culture (BC) approach for pathogenic and antibiotic susceptibility identification remains yet a major limitation with turnaround time of 2–5 days [8, 9]. Moreover, less than 10% of blood samples processed from hospitalized patients with blood infections are positive [10, 11]. Decreasing the sample analysis time compared to classical methods with microfluidics thus appears very challenging for diagnosis. In this section, main

tics is introduced for highly operative and sensitive microfluidic applications.

**2. Microfluidic filtration, concentration, and sorting**

32 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

sorting, extraction, and purification methods are introduced.

centration, separation, and detection.

#### *2.1.1. Surface binding technique*

Solid-phase extraction technique (SPE) is the most widely used preconcentration and purifi‐ cation technique. Capture is performed through a hydrophobic interaction between the analyte and a confined monolith element. Subsequently, a washing step to eliminate interference components is achieved. Finally, the elution step releases the trapped components. From cardiac biomarkers, nucleic acids to proteins, monolith materials with or without functional‐ ization are used as the preferred on-chip approach for purification and concentration. Indeed, monoliths are intensely used, because of ease of preparation, wide ranges of formulation, and adjustable surface area and porosity. Monoliths are commonly prepared via photo-polymer‐ ization or sol-gel approaches. Recently, in a PCO microfluidic chip based on an octyl-metha‐ crylate-based polymer monolith, Yang et al. [12] have shown that the ratio of monomer to pyrogen can be adjusted to change the porosity of the column. In addition, the system features a fluorescence labeling capability where model HSP90 proteins were labeled on-column, prior to their elution. Extending the range of suitable materials for SPE integration in an all polymer approach, Lounsbury et al. [13] developed an original PMMA-based column extraction method within a fully integrated device for the sample-to-PCR products collection. From buccal swabs to whole blood samples, they showed a ~5-fold reduction in processing time for complete amplicon purification and extraction. Beyond continuous and static monolith approaches, bead materials alternatively might be used for capture and extraction. Indeed, using silica beads, and others chitosan-coated beads among others methods, many systems have been implemented. However, some limitations may occur due to packing difficulties in such complicated designs. Using a dual-weir filtration strategy for bead immobilization, Zhong et al. [14] were able to extract and purify λDNA in a glass chip system, finally eluted in a small volume of ∼8 µL. In order to circumvent the relative difficulties of bead packing, static silicon micropillars have been fabricated directly inside the microchannel. Repeatedly separated micropillars promote recombined multiple flow streams favoring DNA contact with the silica surface areas. Since extreme high flow rates can be operated (10 ml min–1), this strategy appears of great interest for medical applications. Finally, the system featured a binding capacity of 57 ± 5 ng cm–2 [15]. For medical applications, the direct molecular concentration from a real biofluid remains a difficult issue; therefore, it might be preferred to extract a cellular group instead of a molecular species. Indeed, super-macroporous cryogels, with interlinking pores ranging from 10 to 100 µm, with added targeted ligands inside the gel trapped cells of interests [16]. In addition to blood cells, bacteria such as *Escherichia coli* have been shown to absorb over the gel.

The definitive easiness and reported performance of SPE strategy position this approach at the leading position for on-chip purification and concentration. Additionally, since two decades, numerous similar off-chip SPE and bead technologies have been developed and even been commercialized, their integration inside lab-on-a-chip (LOC) system should, therefore, benefit from those achievements. Agilent (Santa Clara, CA, USA) and Bio-Rad (Hercules, CA, USA) companies developed their 2100 Bioanalyzer and Experion™ automated electrophoresis systems, respectively, for nucleic acids and protein analysis. These systems perform all of the steps of gel-based electrophoresis synthesis, molecular staining and final imaging.

#### *2.1.2. Solvent extraction*

As solid phase extraction, solvent phase extraction is an off-chip mature technique for separa‐ tionandpreconcentrationofanalytes.Liquid-liquidextraction(LLE)intheLOCsystembenefits from short molecular diffusion, low consumption and high extraction efficiency due to large specific interfacial areas. Chen et al. [17] reported a concentration factor of over 1000 for butyl-Rhodamine B using only few hundreds picoliters of organic solvents. Kitamor's group developed several glass chip extraction systems with channels ranging from 20 to 200 µm for successive extractionofNickel andCobalt complexes [18, 19],drugsofabuse species concentrat‐ ed from urine (amphetamine, mephentermine) [20] and finally carbamate pesticides for agronomicspurposes [21].Especiallyformolecularmicrobiology,LLEisaconventionalmethod for nucleic acid purification and extraction from lysate of various clinical isolates. Indeed, the cellularmembrane components andproteinsmove towardthe organic/aqueous interface while the DNA stays in the aqueous phase, which is then subsequently removed. Based on phenolaqueous extraction system, Reddy and Zahn [22] performed either genomic or plasmid DNA extraction and concentration in a 30 µm deep and 80 µm wide microchannel glass chip. Several optimization strategies based on flow velocities and their profile properties for enhanced interface stability and surface modification have been proposed. Concerning microstructures and surfactants that are the most concerned parameters for extraction performance improve‐ ments, the readers could further read to the following complete review [23].

To sum up this section, regarding materials, one could clearly observe that all reported systems have been realized using either glass, silicon, or quartz interface*.* Straightforwardly under‐ stood, it should be noticed that with the involved organic solvent, polymer approach lack of intrinsic chemical resistance, avoiding their use in the field. But on the other side, glass bonding, typically achieved in between 350 and 650°C for several tens of minutes, induced major drawbacks (i.e., bioreagents integration and biochemical surface functionalization), heavy cost issues and limited throughputs. Dedicated efforts to implement highly chemically resistance polymers such as a Teflon-like (e.g., PFA, PEFE) or polyimide interface should be a way for significant advances of liquid-liquid extraction methodology for cost-effective LOC platforms, where additional downstream amplification and detection means with on-board reagents would be, therefore, possible to be integrated.

#### *2.1.3. Microscale filters*

Compared to SPE or liquid-liquid phase extraction methods, microscale filtration is performed in a single buffer system. Typical filtration approaches involve two large mechanisms. First, the filtering effects those permit sorting through their pores by steric exclusion. Second, electric fields that will induce transport of the analyte inside or at the vicinity of the pores sites. Only representative works based on the first steric aspect, considered as "passive approach," are addressed in this section. Exclusion/enrichment effects based on electrical fields are presented later in (Section 3.2).

Mainly four types of filtering approaches are commonly used: membrane, weir, pillars, and cross-flow. First, integration of optically clear polycarbonate track-etched membrane is one of the preferred options for membrane based filtration approach. At low cost, an impressive range of membrane features are available with thicknesses ranging from 5 to 50 µm, pores sizes from 10 nm to 10 µm, and pores density from 10 + 5 to 10 + 9 pores cm–2. Additionally, such membranes can be easily integrated in a multilayered microfluidic system. In order to reduce their clogging and fouling issues specific microfluidic protocols have been investigated. Using the periodic reverse flow strategy, Redkar and Davis [24] have shown that the final average flux obtained with backflushing is still 20–30 times higher than filtration without the reserve flow strategy. Recently, in order to separate diverse multicomponents from a sample, Lo and Zahn [25] have integrated three membranes with pore size of 0.4, 2, and 5 µm in a microfluidic chip. Therefore, from one side to the other, they demonstrated the separation of sheep blood into its components. Conventional membrane filtration has been also optimized through the cross-flow approach. Recently, Aran et al. [26, 27] reported a cross-flow filtration using the track etched membrane integrated in a microchannel for the continuous bacteria extraction from whole blood for sepsis interest and also plasma protein extraction during cardiac surgery. Compared to isoporous planar membrane integration, the weir and pillar-based filtration approaches benefit of added advantages such as the flexibility of microfluidic design and microfabrication since pore size, filter orientation, and geometry can be modified on demand in order to improve microscale interaction in between filters structures and filtrate. Recently, using a weir-type filter, Chung et al. [28], developed a microfluidic cell sorter for circulating tumor cell (CTC) concentration and analysis. Moreover, the continuously separation is achieved at a high flow rate of up to 20 ml/h (see Figure 1a). Micropillar arrays for CTC isolation, fetal red blood cell enrichment and removal of pathogens have been also intensively used in microfluidic environments [29, 30, 31]. For further interest, we refer the readers to the following reviews [32, 33].

#### *2.1.4. Hydrodynamic sorting and concentration techniques*

systems, respectively, for nucleic acids and protein analysis. These systems perform all of the

As solid phase extraction, solvent phase extraction is an off-chip mature technique for separa‐ tionandpreconcentrationofanalytes.Liquid-liquidextraction(LLE)intheLOCsystembenefits from short molecular diffusion, low consumption and high extraction efficiency due to large specific interfacial areas. Chen et al. [17] reported a concentration factor of over 1000 for butyl-Rhodamine B using only few hundreds picoliters of organic solvents. Kitamor's group developed several glass chip extraction systems with channels ranging from 20 to 200 µm for successive extractionofNickel andCobalt complexes [18, 19],drugsofabuse species concentrat‐ ed from urine (amphetamine, mephentermine) [20] and finally carbamate pesticides for agronomicspurposes [21].Especiallyformolecularmicrobiology,LLEisaconventionalmethod for nucleic acid purification and extraction from lysate of various clinical isolates. Indeed, the cellularmembrane components andproteinsmove towardthe organic/aqueous interface while the DNA stays in the aqueous phase, which is then subsequently removed. Based on phenolaqueous extraction system, Reddy and Zahn [22] performed either genomic or plasmid DNA extraction and concentration in a 30 µm deep and 80 µm wide microchannel glass chip. Several optimization strategies based on flow velocities and their profile properties for enhanced interface stability and surface modification have been proposed. Concerning microstructures and surfactants that are the most concerned parameters for extraction performance improve‐

steps of gel-based electrophoresis synthesis, molecular staining and final imaging.

34 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

ments, the readers could further read to the following complete review [23].

reagents would be, therefore, possible to be integrated.

*2.1.3. Microscale filters*

later in (Section 3.2).

To sum up this section, regarding materials, one could clearly observe that all reported systems have been realized using either glass, silicon, or quartz interface*.* Straightforwardly under‐ stood, it should be noticed that with the involved organic solvent, polymer approach lack of intrinsic chemical resistance, avoiding their use in the field. But on the other side, glass bonding, typically achieved in between 350 and 650°C for several tens of minutes, induced major drawbacks (i.e., bioreagents integration and biochemical surface functionalization), heavy cost issues and limited throughputs. Dedicated efforts to implement highly chemically resistance polymers such as a Teflon-like (e.g., PFA, PEFE) or polyimide interface should be a way for significant advances of liquid-liquid extraction methodology for cost-effective LOC platforms, where additional downstream amplification and detection means with on-board

Compared to SPE or liquid-liquid phase extraction methods, microscale filtration is performed in a single buffer system. Typical filtration approaches involve two large mechanisms. First, the filtering effects those permit sorting through their pores by steric exclusion. Second, electric fields that will induce transport of the analyte inside or at the vicinity of the pores sites. Only representative works based on the first steric aspect, considered as "passive approach," are addressed in this section. Exclusion/enrichment effects based on electrical fields are presented

*2.1.2. Solvent extraction*

Hydrodynamic methods that are passive approaches represent the most widely used strategies for sorting and concentrating micron-sized objects. Hydrodynamic methods rely on any external field other than the forces developed by the fluid on the analyte. These techniques not only ease the overall concerns of integration, but also feature promising potential for highthroughput and enhanced parallel purposes. Additionally, hydrodynamic techniques are label-free. Relying mainly on microscale microfluidic effects based on deterministic lateral displacement (DLD) [34], hydrodynamic focusing [35], and inertial migration mechanism [36, 37], various demonstrations have been already reported. In this section, we only report for few significant examples for each approach, for in-depth review and exhaustive hydrodynamic understanding, we refer the readers to aforementioned publications [38].

Last refinement for DLD systems focused on the optimization of the geometrical boundary of micropillar components which drive the separation performance of the system. In that sense, Inglis [39] has proposed fluidic aberrant corrections for enhanced separation characteristics. In order to reduce clogging and fluidic resistance for high DLD performance, Loutherback et

**Figure 1.** (a) Schematic illustration of a cross-flow microfluidic cell sorter used to separate circulating tumor cells from normal blood cells. Large cancer cells move along the weir-type barrier and are collected at one outlet while smaller hematological cells pass through a gap underneath the barrier, and are directed to a separate outlet. (b) An integrated blood barcode chip for a protein assay. Plasma is separated from blood collected from a fingerprick by harnessing the Zweifach-Fung effect. Proteins in plasma are skimmed and analyzed *in situ* within the antibody barcode arrays. (c) Schematic of particle separation using a spiral microchannel with a trapezoidal cross-section. At the outlet of the spiral microchannel (the A-A cross-section), CTCs are focused near the inner wall due to the combination of inertial lift force and Dean drag force at the outlet; white blood cells (WBCs) and platelets are trapped inside the core of the Dean vor‐ tex, which is formed closer to the outer wall. (d) (i) Magnetic opsonin and biospleen device. Design scheme of native mannose-binding lectin (MBL) to produce the generic opsonin FcMBL and its coating on magnetic nanobeads to en‐ sure stable and to facilitate protein purification. (ii) Pseudocolored scanning electron micrographs showing magnetic beads (128 nm) bound to the bacteria *S. aureus* (orange/brown; left) and *E. coli* (blue; right). Scale bars, 1 µm; arrows indicate pathogen with bound beads. (iii) Schematics of a venous sinus in the red pulp of the spleen (left) and a longi‐ tudinal view of the biospleen (right), with a photograph of an engineered device (right).

al. [40] have integrated several pillar structures (circular vs. triangular) and their gaps. Finally, combining those two corrections, 85% of CTC recovery was achieved operating at an impres‐ sive flow rate of 10 ml/min [41]. *E. coli* and trypanosomes pathogenic cell extraction by DLD have been reported using circular posts of 6 and 20 µm, respectively [42, 43].

Hydrodynamic focusing refers to the use of flow streams to achieve cell or bead concentration and isolation. It is a size-based approach controlled by the flow rates, channel geometry and downstream channel configurations. For continuous plasma extraction, Vermesh et al. [44], developed an integrated blood bar code system displaying two units. First acted for plasma microfluidic skimming, when the size of the cells are comparable to a main central channel width the cells at a bifurcation point migrate toward this high flow rate channel while plasma penetrates adjacent perpendicular channels. The harvested proteins can then react on probes deposited inside the adjacent channels as illustrated in Figure 1(b).

Finally, due to its improved performance when operating at high flows, inertial focusing in recent years has gained tremendous interest for diagnostics, therapeutics and cell applications. Related to the combined effects of the two size-dependent forces, the inertial lift force (*FL* ∝ *a*4), and when operating in a curved channel, the additional Dean drag force (*FD* ∝ *a*), diverse lateral positions across the channel section will be favored for specific particle geometry for sorting and concentration purposes [45]. In a recent study, using optimized spiral devices with trapezoidal cross-section, a recovery of 80% of CTC from 7.5 ml of whole blood have been reported in 8 min (Figure 1c) [46]. Using a combination of hydrodynamics and inertial focusing methods, Clime et al. [47] recently, devised a system for the concentration, and the cleaning of a wide range of pathogenic agents from a ground beef sample. The enrichment strategies and numerous alternatives related to the use of both microfilter and hydrodynamic methods close the loop for the concentration and sorting of micron-sized objects (cells, coupling microparti‐ cles to nucleic acids, proteins, etc.). Therefore, it is subsequently possible to reach molecular extraction and eventual higher concentration with the help of aforementioned SPE and LLE means.

In a natural manner too, those considerations might be extended to the use of magnetic labeled approaches where nanoparticles and microbeads can be implemented for enhanced sorting and concentration in combination with the aforementioned tools. However, such topics are not covered in this section. But as a unique example, we wish to report a microfluidic extrac‐ orporeal blood-cleansing device recently published in the Nature Medicine Journal [48], where therapy has been used for large spectrum of pathogen and toxin removal from blood. The capture was performed with magnetic nanobeads coated with human mannose-binding lectin under the application of an external magnet through a micron-sized nanoporous membrane, which pull the opsonin-bound pathogens and toxins out from the blood. The system was operated at a tremendous throughput up to 1.25 l/h in an *in vitro* approach and the extraction performance cleared more than 90% of bacteria out stream (Figure 1d).

#### **2.2. Electrophoretic separation and preconcentration on-a-microchip**

al. [40] have integrated several pillar structures (circular vs. triangular) and their gaps. Finally, combining those two corrections, 85% of CTC recovery was achieved operating at an impres‐ sive flow rate of 10 ml/min [41]. *E. coli* and trypanosomes pathogenic cell extraction by DLD

**Figure 1.** (a) Schematic illustration of a cross-flow microfluidic cell sorter used to separate circulating tumor cells from normal blood cells. Large cancer cells move along the weir-type barrier and are collected at one outlet while smaller hematological cells pass through a gap underneath the barrier, and are directed to a separate outlet. (b) An integrated blood barcode chip for a protein assay. Plasma is separated from blood collected from a fingerprick by harnessing the Zweifach-Fung effect. Proteins in plasma are skimmed and analyzed *in situ* within the antibody barcode arrays. (c) Schematic of particle separation using a spiral microchannel with a trapezoidal cross-section. At the outlet of the spiral microchannel (the A-A cross-section), CTCs are focused near the inner wall due to the combination of inertial lift force and Dean drag force at the outlet; white blood cells (WBCs) and platelets are trapped inside the core of the Dean vor‐ tex, which is formed closer to the outer wall. (d) (i) Magnetic opsonin and biospleen device. Design scheme of native mannose-binding lectin (MBL) to produce the generic opsonin FcMBL and its coating on magnetic nanobeads to en‐ sure stable and to facilitate protein purification. (ii) Pseudocolored scanning electron micrographs showing magnetic beads (128 nm) bound to the bacteria *S. aureus* (orange/brown; left) and *E. coli* (blue; right). Scale bars, 1 µm; arrows indicate pathogen with bound beads. (iii) Schematics of a venous sinus in the red pulp of the spleen (left) and a longi‐

have been reported using circular posts of 6 and 20 µm, respectively [42, 43].

tudinal view of the biospleen (right), with a photograph of an engineered device (right).

36 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

The first experimental examples of µTAS have been published in the 1990s. Whereas the conventional separation systems are based on chromatographic methods, the separation ona-chip most often relies on electromigration methods and particularly with free zone electro‐ phoresis [49–51]. The conventional chromatographic systems are generally considered to be more robust and reliable for quality control or medical purposes. However, the miniaturization of analytical systems has shown that electrophoresis on a chip is a versatile analytical separa‐ tion technique that may provide high separation efficiency. Manz et al. [52] have pioneered the electrophoretic separation of sample mixtures in glass microchips. Their work that is one of the most cited papers has demonstrated the ability of a simple chip with optical detection to separate several amino acids in 15 s. The electrophoretic and electroosmotic mobility are the two contributions of transport in electrophoresis:

$$
\Delta \nu = \left(\mu\_e + \mu\_{\alpha \circ}\right) \tag{1}
$$

where *ν* is the migration velocity of the analyte, *µ*eof is the electroosmotic mobility and *E* is the electric field.

Compared to the parabolic profile with Poiseuille flow obtained for chromatographic techni‐ ques, electrophoresis in a microchannel produces a plug-like flow with more homogenous distribution of the velocity vectors. Thus, the electrophoretic profile presents sharp peak and it helps to improve the analytical resolution. The surface to volume ratio is rather high in microfluidics and the electroosmotic mobility expression renders electrophoresis very sensitive to the surface state of the inner wall of the microfluidic channels. From this assess‐ ment, we could think that the modulation of surface charge is one way to improve the robustness of analytical performances of electrophoresis-based separation on-a-chip. The resolution in this kind of separation is also directly linked to the electroosmotic mobility by the following equation [53]:

$$R = \frac{1}{4\left(2\right)^{\frac{1}{2}}} \left(\mu\_{\epsilon1} - \mu\_{\epsilon2}\right) \left[\frac{E}{D\left(\overline{\mu\_{\epsilon}} + \mu\_{\epsilon\eta'}\right)}\right]^{1/2} \tag{2}$$

where *µ*e1 and *µ*e2 are, respectively, the electrophoretic mobilities of two different analytes and upper lined *µ*e is the mean electrophoretic mobility, and *D* is the diffusion coefficient.

By considering Eq. (2), if the mean electrophoretic mobility and the electroosmotic mobility haveoppositesignandsimilarabsolutevaluestheresolutionparameter*R*ismaximized.Various strategies have been published in the literature to adjust the surface charge and thus the electroosmoticmobility.Thesimplestwaytotunethisparameteris toperformsurfacetreatment with polymers, self-assembled monolayers by adding organic solvents in the buffer [54].

Another strategy consists of integrating a fluidic transistor in the separation channel. The gate voltage of the fluidic transistor will modify the surface state in a more versatile manner. Van den Berg et al. [55] pioneered the introduction of insulated electrodes in the separation microchannel. The electrode and the upper insulating layer interact with the ionic species of the liquid/solid interface to adjust the surface charge. This first microfluidic transistor has shown its ability to drastically change the zeta potential. This microfluidic device is capable of adjusting the electroosmotic mobility and even choosing the direction of its vector. This dynamic control of electroosmotic flow (EOF) requires a calibration of the gate voltage versus the electroosmotic mobility in the chosen buffer.

tion technique that may provide high separation efficiency. Manz et al. [52] have pioneered the electrophoretic separation of sample mixtures in glass microchips. Their work that is one of the most cited papers has demonstrated the ability of a simple chip with optical detection to separate several amino acids in 15 s. The electrophoretic and electroosmotic mobility are the

where *ν* is the migration velocity of the analyte, *µ*eof is the electroosmotic mobility and *E* is the

Compared to the parabolic profile with Poiseuille flow obtained for chromatographic techni‐ ques, electrophoresis in a microchannel produces a plug-like flow with more homogenous distribution of the velocity vectors. Thus, the electrophoretic profile presents sharp peak and it helps to improve the analytical resolution. The surface to volume ratio is rather high in microfluidics and the electroosmotic mobility expression renders electrophoresis very sensitive to the surface state of the inner wall of the microfluidic channels. From this assess‐ ment, we could think that the modulation of surface charge is one way to improve the robustness of analytical performances of electrophoresis-based separation on-a-chip. The resolution in this kind of separation is also directly linked to the electroosmotic mobility by

= + (*µ µ e eof* ) (1)

n

38 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

( )

4 2

1

2

( )

*e e*

ê ú = - ê ú <sup>+</sup> ê ú ë û

upper lined *µ*e is the mean electrophoretic mobility, and *D* is the diffusion coefficient.

1 1 2

*<sup>E</sup> R µµ*

( )

é ù

*Dµ µ*

where *µ*e1 and *µ*e2 are, respectively, the electrophoretic mobilities of two different analytes and

By considering Eq. (2), if the mean electrophoretic mobility and the electroosmotic mobility haveoppositesignandsimilarabsolutevaluestheresolutionparameter*R*ismaximized.Various strategies have been published in the literature to adjust the surface charge and thus the electroosmoticmobility.Thesimplestwaytotunethisparameteris toperformsurfacetreatment with polymers, self-assembled monolayers by adding organic solvents in the buffer [54].

Another strategy consists of integrating a fluidic transistor in the separation channel. The gate voltage of the fluidic transistor will modify the surface state in a more versatile manner. Van den Berg et al. [55] pioneered the introduction of insulated electrodes in the separation microchannel. The electrode and the upper insulating layer interact with the ionic species of the liquid/solid interface to adjust the surface charge. This first microfluidic transistor has shown its ability to drastically change the zeta potential. This microfluidic device is capable of adjusting the electroosmotic mobility and even choosing the direction of its vector. This

*e eof*

1/2

(2)

two contributions of transport in electrophoresis:

electric field.

the following equation [53]:

**Figure 2.** (a) Photography of the Wheatstone fluidic bridge bearing a fluidic transistor in the middle. The blue dashed lines are showing the fluidic network of the device. A zoom of the red box of the (i) photography is given in (ii). This optical microscopy image shows the three pairs of electrodes in the central microchannel. The brown scheme of the central channel bearing a fluidic field effect transistor. Again a zoom of the red window of the B image is proposed in (ii). In this latest image we distinguish the reference electrode of the right part of the fluidic transistor. The gate elec‐ trode (green) is placed between the ITO adhesion layer and the SiC polarizable layer. (b) Velocities of the fluorescent microbeads measured from the PIV image analysis as function of the gate voltage of the fluidic transistor. Electropho‐ resis in the central channel is still 3.33 V/cm.

In Haghiri's group, a new generation of fluidic field effect transistor (FFET) with a direct contact between a polarizable electrode and the buffer has been proposed and studied since 2007 [56]. This fluidic component requires less voltage to adjust the surface charge and avoids the problem of current leakage from the insulating layer that is observed during ageing with the Van den Berg's transistor. On the other side, the polarizable interface will be more sensitive to unspecific adsorption or any surface contaminations. Our first attempt to adjust the surface charge with such polarizable interface was already published and we partially succeeded to control the zeta potential with three different materials. With the first generation of FFETs, some electrochemical reactions have been observed, leading to the partial destruction of the transistor. Moreover, the linear variation of the electroosmotic mobility versus the gate voltage could be surprising since the equivalent electrical model includes a capacitance. For the second generation of FFETs to avoid these parasitic electrochemical reactions, the gate voltage of the polarizable interface should be adjusted from a reference electrode close to the transistor. In addition, voltage followers have been integrated in the electric system of the transistor. The electric potential of two reference electrodes at both side of the transistor could thus be connected to these voltage followers to precisely adjust the gate voltage. Moreover, the SiC polarizable interface does not directly adhere onto the glass surface since a layer of ITO (indium titanium oxide) between the substrate and the polarizable layer allows to electrically isolate the SiC polarizable interface. This configuration of flow field effect transistor was integrated in the Wheatstone fluidic bridge to measure the EOF mobility as a function of time. Compared to the first generation FFET transistor without reference electrodes a drastic decrease of the electrochemical reaction was observed. A microfabrication process including WL-5150 photosensitive resist and metal deposition onto glass substrates was optimized. The inlets and outlets of the device are opened by gently sandblasting the upper glass substrate. The total thickness of the double ITO/SiC layer is 200 nm. Pictures of the Wheatstone fluidic bridge with a transistor are given in Figure 2(a). The fluidic flow in the central channel is controlled by the modulation of the EOF flow and thus can be adjusted by the value of the gate voltage. A 5 V transverse electric field was applied with the extreme electrodes for electrophoresis. The gate voltage values were less than 1 V in the polarizable window of the SiC. Particle image velocimetry (PIV) was used to characterize the fluidic flow as a function of the gate voltage. Indeed, 1 µm diameter fluorescent bead was injected in an aqueous 10–3 mol/l KCl solution. Finally, the modulation of the fluidic flow as function of the gate voltage is shown in Figure 2(b). It should be noticeable that the amplitude of the velocity is decreased by more than a factor three between a gate voltage of 0.9 and 0 V. The gate voltages used in these experiments are very low compared to values used in the MIE (metal-insulator-electrolyte) configuration (few hundreds Volts).

The development of electrophoresis on-a-chip has led to high-throughput microfabrication methods to produce cost-effective miniaturized fluidic devices. Recently, Liedert et al. [57] reported a foil-based PMMA chips fabricated by high-throughput roll-to-roll (R2R) process for the identification of the antibiotic resistance gene *mecA* in *Staphylococcus epidermidis*.

Pu et al. [58] pioneered molecular enrichment using nanofluidic devices. Electropreconcen‐ tration with nanoslit is used to rapidly and locally increase the concentration of low abundant species. The group of Santiago [59], leader in the field, performed most efficient and innovative ways to preconcentrate several analytes and address the most significant theoretical concerns. However, the influence of several parameters such as ionic strength, chemical nature of the buffer or the surface charge of the inner walls of the device is not yet fully understood and there is still a need for pertinent preconcentration diagrams. The micro-nano-microstructure and real pictures of the electropreconcentration chip are presented in Figure 3(a) [60]. More‐ over, we add in this figure the velocity profiles by considering the electrophoretic and the electroosmotic contributions and the influence of an additional hydrodynamic pressure. This additional pressure permits modification of the position of the preconcentration plug and even stabilizes some propagating regimes. Moreover, such hydrodynamic pressure stabilizes the concentration polarization effect and controls the selective preconcentration of analytes in terms of preconcentration rate and localization as compared to pure electrical preconcentration (see Figure 3b).

Molecular Microfluidic Bioanalysis: Recent Progress in Preconcentration, Separation, and Detection http://dx.doi.org/10.5772/65772 41

addition, voltage followers have been integrated in the electric system of the transistor. The electric potential of two reference electrodes at both side of the transistor could thus be connected to these voltage followers to precisely adjust the gate voltage. Moreover, the SiC polarizable interface does not directly adhere onto the glass surface since a layer of ITO (indium titanium oxide) between the substrate and the polarizable layer allows to electrically isolate the SiC polarizable interface. This configuration of flow field effect transistor was integrated in the Wheatstone fluidic bridge to measure the EOF mobility as a function of time. Compared to the first generation FFET transistor without reference electrodes a drastic decrease of the electrochemical reaction was observed. A microfabrication process including WL-5150 photosensitive resist and metal deposition onto glass substrates was optimized. The inlets and outlets of the device are opened by gently sandblasting the upper glass substrate. The total thickness of the double ITO/SiC layer is 200 nm. Pictures of the Wheatstone fluidic bridge with a transistor are given in Figure 2(a). The fluidic flow in the central channel is controlled by the modulation of the EOF flow and thus can be adjusted by the value of the gate voltage. A 5 V transverse electric field was applied with the extreme electrodes for electrophoresis. The gate voltage values were less than 1 V in the polarizable window of the SiC. Particle image velocimetry (PIV) was used to characterize the fluidic flow as a function of the gate voltage. Indeed, 1 µm diameter fluorescent bead was injected in an aqueous 10–3 mol/l KCl solution. Finally, the modulation of the fluidic flow as function of the gate voltage is shown in Figure 2(b). It should be noticeable that the amplitude of the velocity is decreased by more than a factor three between a gate voltage of 0.9 and 0 V. The gate voltages used in these experiments are very low compared to values used in the MIE (metal-insulator-electrolyte) configuration

40 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

The development of electrophoresis on-a-chip has led to high-throughput microfabrication methods to produce cost-effective miniaturized fluidic devices. Recently, Liedert et al. [57] reported a foil-based PMMA chips fabricated by high-throughput roll-to-roll (R2R) process for the identification of the antibiotic resistance gene *mecA* in *Staphylococcus epidermidis*.

Pu et al. [58] pioneered molecular enrichment using nanofluidic devices. Electropreconcen‐ tration with nanoslit is used to rapidly and locally increase the concentration of low abundant species. The group of Santiago [59], leader in the field, performed most efficient and innovative ways to preconcentrate several analytes and address the most significant theoretical concerns. However, the influence of several parameters such as ionic strength, chemical nature of the buffer or the surface charge of the inner walls of the device is not yet fully understood and there is still a need for pertinent preconcentration diagrams. The micro-nano-microstructure and real pictures of the electropreconcentration chip are presented in Figure 3(a) [60]. More‐ over, we add in this figure the velocity profiles by considering the electrophoretic and the electroosmotic contributions and the influence of an additional hydrodynamic pressure. This additional pressure permits modification of the position of the preconcentration plug and even stabilizes some propagating regimes. Moreover, such hydrodynamic pressure stabilizes the concentration polarization effect and controls the selective preconcentration of analytes in terms of preconcentration rate and localization as compared to pure electrical preconcentration

(few hundreds Volts).

(see Figure 3b).

**Figure 3.** (a) (i) Scheme of the central channel of the preconcentration device. The nanoslit is connected by two micro‐ channels to obtain the microchannel/nanoslit/microchannel structure (MNM). (ii) Microscopy picture of the MNM structure during an electropreconcentration experiment. The fluorescent analytes are injected at both sides of the de‐ vice. On the right we show a photography of the whole preconcentration device. (iii) The mechanism of pressure-as‐ sisted preconcentration and separation. The local transport rate profile is represented in the MNM structure (velocity of the molecule vs. distance in the structure). The black curve gives the classical preconcentration global velocity of BSA when an electric field is applied as a function of the distance. Four other cases are depicted: two cases with the application of a positive pressure (from the anode to the cathode), which has the effect of moving up the curve in the diagram and obtaining the two anodic regimes depending on the value of the pressure (anodic stacking in purple and anodic focusing in red), two cases with the application of a negative pressure (from the cathode to the anode), which has the effect of moving down the curve and obtaining the two cathodic regimes (cathodic stacking in green and catho‐ dic focusing in blue). (b) Experiments with fluorescein show the role of the ionic strength and the addition of a hydro‐ dynamic pressure over the evolution of the preconcentration profiles (left) a conventional electropreconcentration compared to (right) a cathodic-pressure-assisted electropreconcentration [60].

## **3. Detection components and microfluidic strategy**

#### **3.1. Introduction to MEMS detection**

This section focuses on miniaturized sensors designed for getting, which are integrated in the last stage of the fluidic lab-on-a-chip (LOC) device after preconcentration and separation previously discussed. Integrated microfluidic biosensors can be classified into two main categories: (1) bulk detection**,** more often named labeled detection allowing identification of analytes flowing inside the fluidic channel with prelabeling of the target with a marker (fluorescent or electroactive marker) and (2) "label-free" detection, where physical effects during biochemical recognition are directly measured after binding of the target analyte on the chemical probe. More often this detection mode occurs on a surface, which has been functionalized with the bioreceptor (probe) [61]. Figure 4(a) illustrates these two routes for a classical "Primary antibody/Antigen" couple. The biochemical recognition is mainly governed by the choice of the appropriate biochemical receptor that will specifically bind to the target of interest. Receptors are thus integrated in architectures specifically designed to be well adapted for both analyte and transduction method. Figure 4(b) presents the different existing architectures [2]. In brief, transducers have been paired with antibodies (or antibody frag‐ ments, i.e., proteins that are produced by the immune system) [62, 63] see Figure 4(b) (i), with aptamers [64] (see Figure 4b(ii)) and other receptors as recognition elements (Figures 4b(iii)– (v)). For surface detection, if the affinity of the biomolecular recognition is high (*K*a = *k*on/*k*off ranging from 109 up to 1013 for the case of biotin and streptavidin), the identification of the target can be highly selective during its capture by the bioreceptor. If the specificity of the biochemical recognition is fully determined by the nature of bioreceptor, the sensitivity, also called "Limit-of-Detection" (LoD) and the dynamic range of the sensor are strongly related to the intrinsic properties of the transducer.

It is thus of interest to review different intrinsic properties of the transduction, which can be mechanical, magnetic, optical, or electrical (Figure 5) [65–71]. They are very few comparative studies on intrinsic sensitivity in the literature. However, the reader could refer to the excellent review paper of Arlett et al. [73] published in Nature in 2011, which summarizes and compares the performance of mechanical, optical, and electrical transduction methods. Mechanical sensors are based on cantilever assays where the specific binding of analytes induces lateral stress, resulting in bending of the free-end of the cantilever. One major limitation of surface cantilever detection (Figure 5b) concerns nonspecific binding occurring at the bottom cantile‐ ver surface that can negate the bending and thus alter the detection. Recently, Ndieyira et al. [66] have shown that it is possible to overcome competing stresses from opposing cantilever surfaces allowing direct capture of HIV molecules at 500 fM within 15 min. Magnetic sensors have been mainly used to manipulate magnetic beads in fluidic channels that are used as magnetic label for cell sorting or bead detection [67]. We do not detail such sensors in this chapter since extensive reviews exist. The reader can refer to the excellent reviews from Pamme and Gijs that give a general overview of magnetic integrated sensors [74, 67]. The last part of this chapter is devoted to intrinsic performances of both optical and electrochemical biosensors Molecular Microfluidic Bioanalysis: Recent Progress in Preconcentration, Separation, and Detection http://dx.doi.org/10.5772/65772 43

**Figure 4.** Comparison between (a) labeled and (b) label-free detection methods, reprinted from [61], examples of the different architectures for surface functionalization, reprinted from [2].

and, finally, we conclude with recommendations for optimizing fluidic parameters in order to enhance the capture of biomolecules.

#### **3.2. Optical and photonic detection**

**3. Detection components and microfluidic strategy**

42 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

This section focuses on miniaturized sensors designed for getting, which are integrated in the last stage of the fluidic lab-on-a-chip (LOC) device after preconcentration and separation previously discussed. Integrated microfluidic biosensors can be classified into two main categories: (1) bulk detection**,** more often named labeled detection allowing identification of analytes flowing inside the fluidic channel with prelabeling of the target with a marker (fluorescent or electroactive marker) and (2) "label-free" detection, where physical effects during biochemical recognition are directly measured after binding of the target analyte on the chemical probe. More often this detection mode occurs on a surface, which has been functionalized with the bioreceptor (probe) [61]. Figure 4(a) illustrates these two routes for a classical "Primary antibody/Antigen" couple. The biochemical recognition is mainly governed by the choice of the appropriate biochemical receptor that will specifically bind to the target of interest. Receptors are thus integrated in architectures specifically designed to be well adapted for both analyte and transduction method. Figure 4(b) presents the different existing architectures [2]. In brief, transducers have been paired with antibodies (or antibody frag‐ ments, i.e., proteins that are produced by the immune system) [62, 63] see Figure 4(b) (i), with aptamers [64] (see Figure 4b(ii)) and other receptors as recognition elements (Figures 4b(iii)– (v)). For surface detection, if the affinity of the biomolecular recognition is high (*K*a = *k*on/*k*off

up to 1013 for the case of biotin and streptavidin), the identification of the

target can be highly selective during its capture by the bioreceptor. If the specificity of the biochemical recognition is fully determined by the nature of bioreceptor, the sensitivity, also called "Limit-of-Detection" (LoD) and the dynamic range of the sensor are strongly related to

It is thus of interest to review different intrinsic properties of the transduction, which can be mechanical, magnetic, optical, or electrical (Figure 5) [65–71]. They are very few comparative studies on intrinsic sensitivity in the literature. However, the reader could refer to the excellent review paper of Arlett et al. [73] published in Nature in 2011, which summarizes and compares the performance of mechanical, optical, and electrical transduction methods. Mechanical sensors are based on cantilever assays where the specific binding of analytes induces lateral stress, resulting in bending of the free-end of the cantilever. One major limitation of surface cantilever detection (Figure 5b) concerns nonspecific binding occurring at the bottom cantile‐ ver surface that can negate the bending and thus alter the detection. Recently, Ndieyira et al. [66] have shown that it is possible to overcome competing stresses from opposing cantilever surfaces allowing direct capture of HIV molecules at 500 fM within 15 min. Magnetic sensors have been mainly used to manipulate magnetic beads in fluidic channels that are used as magnetic label for cell sorting or bead detection [67]. We do not detail such sensors in this chapter since extensive reviews exist. The reader can refer to the excellent reviews from Pamme and Gijs that give a general overview of magnetic integrated sensors [74, 67]. The last part of this chapter is devoted to intrinsic performances of both optical and electrochemical biosensors

**3.1. Introduction to MEMS detection**

ranging from 109

the intrinsic properties of the transducer.

In general optical biosensors are divided in fluorescence-based and label-free detection. In fluorescent-based detection (bulk), the evanescent field is used to enhance the excitation (or the emission) of the fluorescent dye used to tag the analyte of interest. In contrast, in label-free detection, the target molecules are not labeled and are detected in their natural forms. Optical sensing remains an important route for which record sensitivities were demonstrated through the considerable progress of photonic nanostructures. We focus in this section on integrated photonics nanosensors that are based on direct coupling between light and fluid since the strength of this interaction determines the intrinsic sensitivity [75]. Before, discussing the specific properties of the different photonic sensors, it is of interest to illustrate how such photonic nanostructures could be integrated inside the microfluidic chip.

The optofluidics device thus integrating source and detector exhibits new optical properties related to the nature of liquids. This concept of optofluidics multilevel platform has been introduced by Psaltis at the University of Caltech in the Unite States [76]. In such platform three levels are stacked: (1) the base of the device (bottom layer) that contains the optical

**Figure 5.** Examples of integrated biosensors presented as function of the nature of the transduction for both bulk and surface routes, reprinted from (a) [65] and (b) [66], (c) [67] and (d) [68]; (e) [69] and (f) [70]; (g) [71], and (h) [72].

elements, namely, the springs, the waveguides and the optical sensors (photonic-crystal or plasmonic nanostructures), (2) the intermediate layer that includes the fluidic microchannels for the circulation of biological fluid, and (3) the top layer that contains the actuators for liquid handling, i.e., valves and pumps. Figure 6 presents the hypothetical architecture of such ideal optofluidic platform based on photonic nanostructure as sensor [75, 77]. If the optical sensors are structured in the bottom layer to dimensions smaller than the wavelength of the order of 100 nm, the fluidic channels have, in turn, typical widths of hundreds of micrometers and lengths of several millimeters. There are still very few complete demonstration platforms, which integrates all the optical components (source, waveguides, and detectors) and fluid control tools. All current researches converge toward this ultimate goal of integration, hoping to increase the portability of the chip and the sensitivity of optical detectors.

Molecular Microfluidic Bioanalysis: Recent Progress in Preconcentration, Separation, and Detection http://dx.doi.org/10.5772/65772 45

**Figure 6.** Hypothetical architectures for integration of photonic nanostructures with a cross-section of the fluidic cham‐ ber containing the sensor (left) and a global view of both chip and reader (right), from references [75, 77].

elements, namely, the springs, the waveguides and the optical sensors (photonic-crystal or plasmonic nanostructures), (2) the intermediate layer that includes the fluidic microchannels for the circulation of biological fluid, and (3) the top layer that contains the actuators for liquid handling, i.e., valves and pumps. Figure 6 presents the hypothetical architecture of such ideal optofluidic platform based on photonic nanostructure as sensor [75, 77]. If the optical sensors are structured in the bottom layer to dimensions smaller than the wavelength of the order of 100 nm, the fluidic channels have, in turn, typical widths of hundreds of micrometers and lengths of several millimeters. There are still very few complete demonstration platforms, which integrates all the optical components (source, waveguides, and detectors) and fluid control tools. All current researches converge toward this ultimate goal of integration, hoping

**Figure 5.** Examples of integrated biosensors presented as function of the nature of the transduction for both bulk and surface routes, reprinted from (a) [65] and (b) [66], (c) [67] and (d) [68]; (e) [69] and (f) [70]; (g) [71], and (h) [72].

to increase the portability of the chip and the sensitivity of optical detectors.

44 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

**Figure 7.** Scanning electron micrograph of various microcavities that can be used for biosensing with (a) suspended polymer microring from [79], (b) silica suspended microdisk (inset) from [80] and silica microtoroid from [81], (c) sili‐ con microring from [82], and (d) planar photonic crystal in an InGaAsP membrane from [83].

In an optical biosensor the strength of light-matter interaction is enhanced by the presence of a guided or a localized optical mode with an evanescent field with subwavelength spatial extension [84, 85]. Guided mode can be excited in dielectric or metallic waveguide. A typical example is the case of the surface plasmon resonances (SPR) [80], an electromagnetic guided mode excited at the interface between a noble metal and a dielectric. Localized modes can be excited in dielectric structures like photonic crystals [86, 87] (Figure 7d) or ring resonators [79, 82] (Figure 7a and c) and in metallic nanoantenna resulting in the so-called localized surface plasmon resonance (LSPR) (Figure 7d) [83].

The light confinement can be used to increase the scattering (or the absorption) cross-section of the analyte molecules or to measure the refractive index (RI) change induced by molecular interactions. The most studied example of the first category is surface-enhanced Raman spectroscopy (SERS). SERS is a spectroscopic technique in which the inelastic scattering of monochromatic light provides information about vibrational, rotational, and other low frequency transitions in molecules of the analyte. Here, we limit the scope of this section to the so-called RI-based label-free detection. The presence of the analyte immobilized on the sensor surface through a specific biochemical reaction results in a RI change in the near field region of the optical mode. The RI change induces a modification of the dispersion relation of the guided mode or a shift in the position of the localized resonance that is monitored in real time. The performance of an RI-based optical sensor is most commonly characterized through the bulk sensitivity *S* = Δ*λ*/Δ*n* [nm/RIU (Refractive Index Unit)] in which *λ* is the resonance wavelength of the optical mode and *n* is the refractive index of the medium probed by the near-field of the optical mode. Since it is easier to detect a given resonance shift for narrow lines, the figure of merit FOM = S/FWHM (where FWHM is the full width at half-maximum of the resonance) is a more meaningful measure of the performance of the sensor. An important parameter that is difficult to quantify is the extension of the evanescent field of the optical mode. In an optical biosensor, the analyte (with RI ~ 1.5) is specifically immobilized at the sensor surface, where the intensity of the evanescent field is higher. The rest of the evanescent field probes the RI of the buffer solution (RI ~ 1.33). The effective RI change probed by the whole near field of the optical mode depends on the overlap between the evanescent electro‐ magnetic field and the analyte. Highly confined modes are therefore more sensitive to small analytes. This explains why, despite their relative high sensitivity and FOM (typically 3300 nm/RIU and 50), SPR-based biosensors fail to detect small molecules like biotin because of the relatively large extension of their evanescent field that exponentially decays over 200–300 nm away from the surface. On the other side, LSPR-based biosensors have relatively smaller sensitivity and FOM (typically 400 nm/RIU and 2), but their evanescent field exponentially decays over a distance 10 times smaller than that of a SPR mode [88, 89].

To our knowledge the highest FOM and field confinement achieved to date for a localized optical mode was reported by Cattoni et al. [90] in arrays of plasmonic nanocavities fabricated by soft UV nanoimprint lithography (Figure 8). The plasmonic nanocavity is designed and ensures total absorption of light at the plasmonic resonance. Sensitivity, FOM, and optical field confinement are parameters that strictly depend on the physics behind the nanophotonic element used as a sensor. These parameters alone cannot be related to the biosensor perform‐

Molecular Microfluidic Bioanalysis: Recent Progress in Preconcentration, Separation, and Detection http://dx.doi.org/10.5772/65772 47

In an optical biosensor the strength of light-matter interaction is enhanced by the presence of a guided or a localized optical mode with an evanescent field with subwavelength spatial extension [84, 85]. Guided mode can be excited in dielectric or metallic waveguide. A typical example is the case of the surface plasmon resonances (SPR) [80], an electromagnetic guided mode excited at the interface between a noble metal and a dielectric. Localized modes can be excited in dielectric structures like photonic crystals [86, 87] (Figure 7d) or ring resonators [79, 82] (Figure 7a and c) and in metallic nanoantenna resulting in the so-called localized surface

46 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

The light confinement can be used to increase the scattering (or the absorption) cross-section of the analyte molecules or to measure the refractive index (RI) change induced by molecular interactions. The most studied example of the first category is surface-enhanced Raman spectroscopy (SERS). SERS is a spectroscopic technique in which the inelastic scattering of monochromatic light provides information about vibrational, rotational, and other low frequency transitions in molecules of the analyte. Here, we limit the scope of this section to the so-called RI-based label-free detection. The presence of the analyte immobilized on the sensor surface through a specific biochemical reaction results in a RI change in the near field region of the optical mode. The RI change induces a modification of the dispersion relation of the guided mode or a shift in the position of the localized resonance that is monitored in real time. The performance of an RI-based optical sensor is most commonly characterized through the bulk sensitivity *S* = Δ*λ*/Δ*n* [nm/RIU (Refractive Index Unit)] in which *λ* is the resonance wavelength of the optical mode and *n* is the refractive index of the medium probed by the near-field of the optical mode. Since it is easier to detect a given resonance shift for narrow lines, the figure of merit FOM = S/FWHM (where FWHM is the full width at half-maximum of the resonance) is a more meaningful measure of the performance of the sensor. An important parameter that is difficult to quantify is the extension of the evanescent field of the optical mode. In an optical biosensor, the analyte (with RI ~ 1.5) is specifically immobilized at the sensor surface, where the intensity of the evanescent field is higher. The rest of the evanescent field probes the RI of the buffer solution (RI ~ 1.33). The effective RI change probed by the whole near field of the optical mode depends on the overlap between the evanescent electro‐ magnetic field and the analyte. Highly confined modes are therefore more sensitive to small analytes. This explains why, despite their relative high sensitivity and FOM (typically 3300 nm/RIU and 50), SPR-based biosensors fail to detect small molecules like biotin because of the relatively large extension of their evanescent field that exponentially decays over 200–300 nm away from the surface. On the other side, LSPR-based biosensors have relatively smaller sensitivity and FOM (typically 400 nm/RIU and 2), but their evanescent field exponentially

decays over a distance 10 times smaller than that of a SPR mode [88, 89].

To our knowledge the highest FOM and field confinement achieved to date for a localized optical mode was reported by Cattoni et al. [90] in arrays of plasmonic nanocavities fabricated by soft UV nanoimprint lithography (Figure 8). The plasmonic nanocavity is designed and ensures total absorption of light at the plasmonic resonance. Sensitivity, FOM, and optical field confinement are parameters that strictly depend on the physics behind the nanophotonic element used as a sensor. These parameters alone cannot be related to the biosensor perform‐

plasmon resonance (LSPR) (Figure 7d) [83].

**Figure 8.** Left: SEM image of the 2D metal-insulator-metal nanocavities fabricated by Soft UV NIL using a hard-PDMS/ PDMS stamp. Center: picture of the biosensor integrated in a microfluidic chamber and silicon master mold used to fabricate the hard-PDMS/PDMS stamp. Right: Spectral shift of the second-order mode for different RI solutions: water (black line), ethanol-water solution (blue line), and pure ethanol (red line). The green line corresponds to the FOM cal‐ culated as function of the wavelength using the equation in the inset, from [90].

ance, since it is only an indication of signal strength. The parameter typically used to charac‐ terize the performance of a biosensor is the LOD. The LOD can be deduced by taking into account the noise in the transduction signal, *σ*, i.e., the minimum resolvable signal: LOD = *σ*/*S*, where *S* is the sensitivity. The LOD can therefore be improved by increasing the sensitivity (and the light confinement) and by reducing the noise level. For an optical RI-based label-free sensor, there are typically three ways to specify LOD: in units of refractive index units (RIU), in surface mass density of the analyte (pg/mm) or in analyte concentration (ng/mL or molarity). The LOD specified in terms of RIU is easy to measure and useful to roughly compare the sensing performances of different optical sensor.

As previously mentioned, it does not take into account the extension of the field confinement because the measure is made varying the RI over the whole extension of the probing near field. Reported values for SPR-based sensor have LOD up to 10–8 RIU, dielectric waveguide and ring resonator LOD up to 10–7 RIU and photonic crystals LOD up to 10–5 RIU [87, 89]. The extension of the field confinement is taken into account when the detection limit is specified in terms of surface mass density, which is what a biosensor actually measures. LOD in terms of surface mass density is difficult to determine accurately, but it can be used to compare more precisely the performance of different optical sensors. Finally, LOD defined in terms of sample concen‐ tration is easy to determine and it can be used to compare more precisely the sensor perform‐ ance. Of course it depends on the specific analyte and its affinity to the biorecognition molecule grafted to the biosensor surface. For this reason, LOD, chemical affinity between a specific analyte and the relative bioreceptors and microfluidic parameters must be considered all together in the optimization of the overall performances of the sensor. Acímovic et al. [91] demonstrated state-of-the-art paralleled LSPR-based lab-on-a-chip composed with 32 sensing sites distributed across eight independent microfluidic channels with very high reproducibil‐ ity/repeatability (Figure 9). In particular they demonstrated the fast detection of relevant cancer biomarkers (human alpha-feto-protein and prostate specific antigen) down to concentrations of 500 pg/ml in a complex matrix consisting of 50% human serum.

**Figure 9.** Description of the sensing platform: schematic of the flow and control layers (a) and final connected chip. (b) The inset shows a standard SEM image of the plasmonic gold sensors. Scale bar = 200 nm. (c) Overview of the optical setup. From [91].

### **3.3. Electrochemical detection**

#### *3.3.1. Introduction to microfluidic detection*

Initially sensing research focused mainly on the performances of the transducer and the biomolecular recognition, without discussing the role of microfluidic parameters. To obtain a well performing analytical device these components have to be considered all together in the optimization of the overall performances of the sensor. Achieving fast time less than one minute, specific and sensitive detection at concentration on the femtomolar level or even less appear thus an actual challenge for the microfluidic community. If the main advantage of microfluidic devices is to reduce the volume and to efficiently deliver target molecules to the sensor surface, working with such extremely reduced size and concentration raises funda‐ mental questions about physical and chemical limits. Independent of the intrinsic sensibility of the sensor (optical or electrochemical), all the fluidic conditions have to be chosen in order to favor binding mechanisms. For most of applications, if the fluidic system works under "extreme fast" flow at a high Péclet number (*PeH*), the number of captured molecules per unit of time is largely enhanced (see Box 1 that resumes the pertinent fluidic parameters). The geometry of the fluidic channel (the height *H* and the width *Wc*), the size of the sensor (the width *Ws* and the length *L*) and the volumetric flow rate *Q* have thus to be precisely defined to fix if the system will work in a reaction-limited or diffusion-limited regime.

sites distributed across eight independent microfluidic channels with very high reproducibil‐ ity/repeatability (Figure 9). In particular they demonstrated the fast detection of relevant cancer biomarkers (human alpha-feto-protein and prostate specific antigen) down to concentrations

**Figure 9.** Description of the sensing platform: schematic of the flow and control layers (a) and final connected chip. (b) The inset shows a standard SEM image of the plasmonic gold sensors. Scale bar = 200 nm. (c) Overview of the optical

Initially sensing research focused mainly on the performances of the transducer and the biomolecular recognition, without discussing the role of microfluidic parameters. To obtain a well performing analytical device these components have to be considered all together in the optimization of the overall performances of the sensor. Achieving fast time less than one minute, specific and sensitive detection at concentration on the femtomolar level or even less appear thus an actual challenge for the microfluidic community. If the main advantage of microfluidic devices is to reduce the volume and to efficiently deliver target molecules to the sensor surface, working with such extremely reduced size and concentration raises funda‐

setup. From [91].

**3.3. Electrochemical detection**

*3.3.1. Introduction to microfluidic detection*

of 500 pg/ml in a complex matrix consisting of 50% human serum.

48 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

In 2008, Squires, Messinger and Manalis proposed a theoretical approach to estimate the interplay between diffusion, convection, and reaction [92]. Their complete analysis gives some guidelines to estimate some fundamental quantities, such as fluxes, collection rates, and equilibration times. Based on dimensionless parameters that are straightforward to compute, their simple approach is very useful in characterizing and designing systems. Two Péclet numbers, *PeH* and *PeS*, characterize the nature of the mass-transport depletion zone around the sensor. At low values of *PeH* (*PeH*≪ 1 at extremely slow flow), the system is in a diffusive regime, with a depletion zone that extends far upstream in the whole thickness of the channel. Even if all molecules are collected, the time of capture is so high (hours or days) that such diffusive regime appears very limited for real applications. At high *PeH* (*PeH* ≫ 1 at extremely fast flow), a depletion zone thinner than the channel (*δS*) exists and a second pertinent Péclet number *PeS* can be used to calculate whether this depletion is thicker or thinner than the sensor itself. Box 1 gives the more important equations useful for optimizing integrated microfluidic devices that involve very small volumes of sample. One should ensure that the sensor exhibits enough sites to bind target molecules maximally. It appears that the sensor should be preferably microstructured with dimensions ranging from 100 µm to several hundreds of microns instead of tens of nanometers as for nanowires.

#### *3.3.2. Current status and state-of-the-art of electrochemical detection*

The need for a miniaturized sensor, portable, with fewer components has allowed the development of electric sensors (chemical or biochemical) [93]. Successful integration of microelectrodes opens the way for the development of electrical and electrochemical detection in microsystems. Indeed, the development of microfabrication techniques to produce micro‐ electrodes was decisive for bioelectroanalysis. The advantage of making electrical transducer sensors is to have portable electrical and electrochemical sensors with easy operation, a great performance in the detection and a low power consumption (applied and measured signals require only a voltage generator). In addition, it contains few components on the very small dimension reducing drastically the size of the final microsystem [94]. In general, electrical sensors operate an electrical signal (current or potential) by amperometry, voltammetry [95, 96], or electrochemical impedance spectroscopy [97]. Electrochemical strategy has been used over a wide range of biochemical identification and analysis purposes from traditional genomics and proteomics areas but it find through cellomics and gases detection increasing interests. For traditional, micro-array DNA electrochemical detection, many works have been reported, and even additional on-board components such as cellular lysis, and genomic preamplification been also incorporated. Indeed, Ferguson et al. [98] have demonstrated the integration of loop-mediated isothermal amplification (LAMP) means coupled with a se‐ quence-specific electrochemical detection in a disposable, monolithic chip. Using this platform, the authors have demonstrated detection of genomic DNA from *Salmonella enterica serovar Typhimurium* LT2 with a limit of detection of 10 aM. On the proteomics side, a recent paper from O. Kelley et al. [99], demonstrated clinical relevance for an electrochemical enzyme-linked immunosorbent assay for HIV antibodies identification. The current method derived from the oxidation increased linearly over a wide antibody concentration range (0.001–1 µg. ml–1), with a detection limit of 1 ng. ml–1 (6.7 pM). For cellomics purposes, Zór et al. [100] recently, demonstrated a powerful electrochemical based integrated platform for real-time monitoring of cellular dynamics. Their system performed, the complete cell based assays comprising online electrode cleaning, sterilization, surface functionalization, cell seeding, cultivation, and electrochemical real-time monitoring of cellular dynamics. To demonstrate the versatility and the multifunctionality of the platform, additionally the authors reported for the amperometric monitoring of intracellular redox activity in yeast (*Saccharomyces cerevisiae*), and detection of exocytotically released dopamine from rat pheochromocytoma cells. As mentioned previous‐ ly, the electrical sensors seem to be more appropriate for easy integration on µTAS develop‐ ment. We shall consider in the following sections the most widely used techniques such as potentiometry, amperometry or voltammetry, and impedance spectroscopy. Concerning the latter, the advantages of contactless microelectrodes integration on-a-chip for impedance measurement and recent applications are discussed.

### *3.3.3. Potentiometry*

sites to bind target molecules maximally. It appears that the sensor should be preferably microstructured with dimensions ranging from 100 µm to several hundreds of microns instead

The need for a miniaturized sensor, portable, with fewer components has allowed the development of electric sensors (chemical or biochemical) [93]. Successful integration of microelectrodes opens the way for the development of electrical and electrochemical detection in microsystems. Indeed, the development of microfabrication techniques to produce micro‐ electrodes was decisive for bioelectroanalysis. The advantage of making electrical transducer sensors is to have portable electrical and electrochemical sensors with easy operation, a great performance in the detection and a low power consumption (applied and measured signals require only a voltage generator). In addition, it contains few components on the very small dimension reducing drastically the size of the final microsystem [94]. In general, electrical sensors operate an electrical signal (current or potential) by amperometry, voltammetry [95, 96], or electrochemical impedance spectroscopy [97]. Electrochemical strategy has been used over a wide range of biochemical identification and analysis purposes from traditional genomics and proteomics areas but it find through cellomics and gases detection increasing interests. For traditional, micro-array DNA electrochemical detection, many works have been reported, and even additional on-board components such as cellular lysis, and genomic preamplification been also incorporated. Indeed, Ferguson et al. [98] have demonstrated the integration of loop-mediated isothermal amplification (LAMP) means coupled with a se‐ quence-specific electrochemical detection in a disposable, monolithic chip. Using this platform, the authors have demonstrated detection of genomic DNA from *Salmonella enterica serovar Typhimurium* LT2 with a limit of detection of 10 aM. On the proteomics side, a recent paper from O. Kelley et al. [99], demonstrated clinical relevance for an electrochemical enzyme-linked immunosorbent assay for HIV antibodies identification. The current method derived from the oxidation increased linearly over a wide antibody concentration range (0.001–1 µg. ml–1), with a detection limit of 1 ng. ml–1 (6.7 pM). For cellomics purposes, Zór et al. [100] recently, demonstrated a powerful electrochemical based integrated platform for real-time monitoring of cellular dynamics. Their system performed, the complete cell based assays comprising online electrode cleaning, sterilization, surface functionalization, cell seeding, cultivation, and electrochemical real-time monitoring of cellular dynamics. To demonstrate the versatility and the multifunctionality of the platform, additionally the authors reported for the amperometric monitoring of intracellular redox activity in yeast (*Saccharomyces cerevisiae*), and detection of exocytotically released dopamine from rat pheochromocytoma cells. As mentioned previous‐ ly, the electrical sensors seem to be more appropriate for easy integration on µTAS develop‐ ment. We shall consider in the following sections the most widely used techniques such as potentiometry, amperometry or voltammetry, and impedance spectroscopy. Concerning the latter, the advantages of contactless microelectrodes integration on-a-chip for impedance

of tens of nanometers as for nanowires.

*3.3.2. Current status and state-of-the-art of electrochemical detection*

50 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

measurement and recent applications are discussed.

We only discuss some potentiometric microsystems, i.e., the systems that have been manu‐ factured at least partly using photolithography or other micromachining techniques and incorporate electrical or electrochemical measurement. This definition excludes a large group of potentiometric electrochemical sensors using microelectrodes. The reason for this exclusion is that the microelectrodes are a smaller version of their macroscopic analogue and their function is similar to larger electrode [101]. From the practical point of view, a potentiometric transducer within microsystem should include a conductive contact, a reference electrode and a microchannel for the fluid flow over the sensing surfaces [102].

Microfabrication of potentiometric sensors has several advantages over conventional electro‐ des, in particular by the dimensions of the measuring system, which is less expensive [103]. Indeed, emergence of lab-on-a-chip applications have benefited from the development of potentiometric sensors on small surfaces in contact with fluidic. Microfluidic systems with potentiometric detection have been developed and characterized. Among them, it can be mentioned the microanalyzer prototypes based on potentiometric measurements for various applications in analytical chemistry and biochemistry, such as ions detection [104], proteins [105], and Ph [106, 107]. The achievement of receptors based on low temperature co-fired ceramic or LTCC for potentiometric microsystems (called µPOT) offers good electrical and mechanical properties, as well as reliability and stability measures. Schöning et al. [108] developed potentiometric microsensor porous silicon, where microporous layer is formed by electrochemical etching. Thus, the microstructured surface allows the enlargement of the active area of the microsensor, which increases the measured capacitance. For instance Lakard et al. [107] developed potentiometric pH microsensors based on films of electrosynthesized polypyrrole, whose electrodes are fabricated by photolithography. The study showed that thin polypyrrole films present the best potentiometric linear responses in pH in the range of 2–11.

#### *3.3.4. Amperometry and voltammetry*

Amperometry and voltammetry techniques use the measurement of the electric current response of a working electrode against the applied potential in the electrochemical cell. The amperometric transductions within microdevices have wide applications in biosensing [109, 110]. Research in biosensors domain using amperometric sensors were initiated by Clark in 1956 by studying the oxygen electrodes [111]. In amperometric microsystems, the current measured with high sensitivity is linearly dependent on the concentration to be detected. This sensitivity is highlighted in the work of Pijanowska et al. [95] about the detection of glucose. These biosensors also have the advantage of being faster, cheaper, and more available than conductimetric and potentiometric biosensors [109]. Other studies have focused on the measurement of enzyme activity measured within miniaturized amperometric and voltam‐ metric cells, using grafted proteins on the surface of the microelectrodes. These microdevices have the characteristic to give fast response [110]. The results showed a high linearity between the measured responses of the biosensor and the concentrations of the sample. Furthermore, microsystems in amperometric and voltammetric transduction have been successfully adapted in capillary electrophoresis, conducted in a chip for toxicity detection of phenolic compounds [112] and synthetic food colorants [113].

The integration of a reference microelectrode in a microfluidic chip is often a hard step. Recently, Faure et al. [72] have proposed an alternative strategy requiring less microfabrication steps according to a configuration with two microelectrodes for electrochemical detection in glass/PDMS microfluidic chips. As displayed in Figure 10(b), it consists of using two micro‐ band electrodes of the same material with a surface ratio of 22 for the counter-electrode (*S* = 0.31 mm2 ) with respect to the working microelectrode area (S = 0.014 mm2 ). Therefore, the counter-electrode can be considered as a pseudo-reference since the current density flowing through it is much smaller than that flowing in the working electrode, thus, limiting possible variations on the rest of potentials. To this end, the redox couple [Fe(III) (CN)6]3–/[Fe (II) (CN)6]4– is used to impose a 0 V as reference potential (see Figure 10). The chip performance using this geometry has been characterized using cyclic voltammetry according to the hydrodynamic conditions in chip, while differential pulse voltammetry (DPV) as it appears in Figure 10, with regard to its analytical performance was preferred for the LOD of transthyretin (TTR). The quantification of transthyretin peptide is a major interest for the diagnosis of familial amyloid polyneuropathy at transthyretin (ATTR) [114]. The obtained LOD for the TTR was determined at 25 nM, a value of 100-fold lower than that reported in conventional capillary electrophoresis coupled to the laser-induced fluorescence (LIF) under the same experimental conditions [115].

#### *3.3.5. Electrical impedance spectroscopy*

The electrochemical impedance spectroscopy is a well-established technique. Thus, the need to access to more sensitive and more precise measurements within microdevices opened novel strategies for sensor development that enable integration of electrical or electrochemical impedance spectroscopy. Among the areas where the impedance measurements are the most used, its noninvasive characteristic is crucial for many applications such as for counting, identification, and detection of particles or cells in biology and biochemistry [116]. In the last years, impedance measurements in microsystems have been mainly pursued to measure the electrical conductivity of liquid [117], the dielectric properties of the particles in suspension or in flow [116, 118], the properties of the cell membranes [119], the kinetics of enzymatic reactions [120], and the adsorption of red bloods on the sensor surfaces [121]. For instance, Park et al. [122] have improved a microfluidic device developed earlier by Ferrier et al. [123]. The microdevice was used for the detection of polystyrene microbeads and cancerous cells. In brief, the device comprised a microfluidic channel and two parallel planar electrodes for particles (microbeads and cells) actuation using dielectrophoresis (DEP), and a trap reservoir containing the electrodes for the impedance measurements. The study showed that measured impedance variations were in relation to the trapping and release of the microbeads and the cell in the reservoir. Ayliffe et al. [124] have also demonstrated the ability to detect the presence of a single particle using flow cytometry combined with the impedance measurement in a microfluidic device. The principle consists in measuring the electric impedance at the passage of a particle or a cell through a microchannel with electrodes disposed on its surface in order to detect each

Molecular Microfluidic Bioanalysis: Recent Progress in Preconcentration, Separation, and Detection http://dx.doi.org/10.5772/65772 53

in capillary electrophoresis, conducted in a chip for toxicity detection of phenolic compounds

The integration of a reference microelectrode in a microfluidic chip is often a hard step. Recently, Faure et al. [72] have proposed an alternative strategy requiring less microfabrication steps according to a configuration with two microelectrodes for electrochemical detection in glass/PDMS microfluidic chips. As displayed in Figure 10(b), it consists of using two micro‐ band electrodes of the same material with a surface ratio of 22 for the counter-electrode (*S* =

counter-electrode can be considered as a pseudo-reference since the current density flowing through it is much smaller than that flowing in the working electrode, thus, limiting possible variations on the rest of potentials. To this end, the redox couple [Fe(III) (CN)6]3–/[Fe (II) (CN)6]4– is used to impose a 0 V as reference potential (see Figure 10). The chip performance using this geometry has been characterized using cyclic voltammetry according to the hydrodynamic conditions in chip, while differential pulse voltammetry (DPV) as it appears in Figure 10, with regard to its analytical performance was preferred for the LOD of transthyretin (TTR). The quantification of transthyretin peptide is a major interest for the diagnosis of familial amyloid polyneuropathy at transthyretin (ATTR) [114]. The obtained LOD for the TTR was determined at 25 nM, a value of 100-fold lower than that reported in conventional capillary electrophoresis coupled to the laser-induced fluorescence (LIF) under the same experimental

The electrochemical impedance spectroscopy is a well-established technique. Thus, the need to access to more sensitive and more precise measurements within microdevices opened novel strategies for sensor development that enable integration of electrical or electrochemical impedance spectroscopy. Among the areas where the impedance measurements are the most used, its noninvasive characteristic is crucial for many applications such as for counting, identification, and detection of particles or cells in biology and biochemistry [116]. In the last years, impedance measurements in microsystems have been mainly pursued to measure the electrical conductivity of liquid [117], the dielectric properties of the particles in suspension or in flow [116, 118], the properties of the cell membranes [119], the kinetics of enzymatic reactions [120], and the adsorption of red bloods on the sensor surfaces [121]. For instance, Park et al. [122] have improved a microfluidic device developed earlier by Ferrier et al. [123]. The microdevice was used for the detection of polystyrene microbeads and cancerous cells. In brief, the device comprised a microfluidic channel and two parallel planar electrodes for particles (microbeads and cells) actuation using dielectrophoresis (DEP), and a trap reservoir containing the electrodes for the impedance measurements. The study showed that measured impedance variations were in relation to the trapping and release of the microbeads and the cell in the reservoir. Ayliffe et al. [124] have also demonstrated the ability to detect the presence of a single particle using flow cytometry combined with the impedance measurement in a microfluidic device. The principle consists in measuring the electric impedance at the passage of a particle or a cell through a microchannel with electrodes disposed on its surface in order to detect each

). Therefore, the

) with respect to the working microelectrode area (S = 0.014 mm2

[112] and synthetic food colorants [113].

52 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

0.31 mm2

conditions [115].

*3.3.5. Electrical impedance spectroscopy*

**Figure 10.** (a) Picture of several pairs of two microelectrodes networks located on the microchannel junction for multi‐ detection possibility. The fluidic microchannels allowing the sample injection with several configurations (inlet 1, 2, 3). (b) Schematic view of the detection area with a pair of microelectrodes (CE, counter electrode, WE, working electrode). The fluidic channel, WE, and CE microband electrodes characteristic dimensions: *h*, *d*, *w*, *x*e, and *l* are indicated and the B scheme is not to the scale. (c) Differential pulse voltammetry of 25 µm Transthyretin peptide (CBI-PN) in the pres‐ ence of 1 mm [Fe(III)(CN)6]3–/Fe(II)(CN)6]4– redox couple as potential reference in buffer 100 mm borate pH 9/MeOH (50/50, v/v). Scan rate 25 mV s–1 modulation time 100 ms modulation amplitude 7.5 mV under a flow rate of 0.05 µL s–1 from [72].

particle flowing [125]. Hediger et al. [126] have investigated on microfluidic devices for detection using impedance spectroscopy in the medical domain. The microdevice was achieved by plasma etching on a silicon substrate with platinum electrodes for detection. The preliminary test in case of device filled with sodium chloride at various concentrations underlined that the measured impedance module plateaus recorded between 10 kHz and 100 Hz decreased with an increase of NaCl concentration. As they expected, the resistance of the microchannel was found as being conversely proportional to solution conductivity.

In the last decades, other kinds of microdevices were developed with electrodes in contactless configuration with the electrolyte [128, 129]. The latter have emerged to overcome some limitations due to microelectrodes contamination, corrosion, or degradation [130]. In the next part, we focus on this original electrode configuration for which the start-of-art and the advantages are discussed for sensing and biosensing development.

Microfluidic chips with a detection module in noncontact mode have the particularity of having galvanically isolated microelectrodes embedded in a dielectric layer. The configura‐ tion in noncontact mode has several advantages in comparison with the traditional mode in contact. The encountered inconvenience for solution with microelectrode in contact is

**Figure 11.** (a) Scanning electron microscopy (SEM) images of the PET photoablated microchannel with a cross- section of 45 × 100 µm and a length of 1.4 cm. (b) Optical image of the detection zone global view including the planar micro‐ electrodes (inverted optical microscope), vertically, the two band electrodes beneath the horizontal flow channel. The detection zone represents the area where capacitive coupling takes place. (c) The SEM image of the trapezoidal section of the microchannel. (b) Side view of the contactless zone, which is about 5 µm and represents the thickness separation between the flow microchannel and the planar microelectrodes [127].

avoided, for instance, the microelectrode passivation or corrosion, the bubble generation due to uncontrolled faradic reactions on electrodes [132]. The first configuration of photoablat‐ ed polyethylene terephthalate PET as flexible microchips with microelectrodes galvanically isolated into the PET was patented in 2004 [133], and then published by Gamby et al. [127]. Indeed, this system was called "SuperCapacitive Admittance Tomoscopy" (SCAT), based on the observation of a thin solvent layer on a dielectric thin film through two embedded microband electrodes (Figure 12). Indeed, the PET dielectric layer coating the two parallel microelectrodes contains a microchannel with chemically modified surface for adsorption of biomolecules [134].

The study is carried out by applying an alternating voltage of 0.1–3 V in the high frequencies (1 kHz–10 MHz) between the two microelectrodes. Indeed, a capacitive coupling effect appears

**Figure 12.** (a) Schematic representation of the BIACORE optical biosensor. The hydrogel is 100 nm high approximately. The ligand is immobilized in the hydrogel the analyte flows in the channel and diffuses in the gel where it binds to the ligand. (b) Cross-section of the dielectric interface microelectrode/ PET/ microchannel and the corresponding electrical equivalent circuit: CPET,2 for the 120 µm-PET layer impedance (distance separation), CPET,1 for the 5 µm-PET layer impe‐ dance (noncontact layer thickness), the element CPE, Qint, for the interfacial impedance (photoablated surface) [134].

through the dielectric layer, on the microelectrodes hand with electronic charges and on the other hand with microchannel filled with ionic charges. This enables the admittance meas‐ urement, which is itself related to the surface state of the PET chemically modified. The SCAT is suitable for the real-time study of electrostatic interactions analyte/substrate on any dielectric having a modified surface. Gamby et al. [127] have investigated the microchannel internal surface modification in the noncontact microdevice and they have showed that the surface charge can be turned and provides an opportunity to enhance the polymer-protein interac‐ tions. For instance, adsorption of β-Lactoglobulin in PET microchannel modified (poly(Llysine) (PLL) adsorbed on gold nanoparticles bearing thiol-carboxylate functions) has been studied in the range of high-frequencies from 1 MHz to 1 kHz with an amplitude of 0.5 V, and finally, an LOD of 4.5 × 10−16 M was reported. The proof of concept of affinity biosensor development by using dielectric impedance detection on insulating PET was investigated and compared to an optical transduction such as surface plasmon resonance (SPR) illustrated on Figure 12 [131]. Proteins are not labeled, as in optical biosensor, even if they need to be attached to the polymer surface coupled with the microelectrodes when a biomolecular interaction occurs. As displayed in Figure 12, modeling the microchip interface using an appropriate equivalent circuit permits to extract the value of the interfacial capacitance for ultra-low protein concentration. The promising results obtained with this methodology make it a competing method in comparison with other transductions for bioanalytical developments. The equili‐ brium association constant was calculated for the affinity between the probe and the target and was estimated equal to 5 × 107 M−1 in agreement with the one determined with SPR technique [78]. The promising results obtained with this strategy make it a competitive biosensor in comparison with SPR.

avoided, for instance, the microelectrode passivation or corrosion, the bubble generation due to uncontrolled faradic reactions on electrodes [132]. The first configuration of photoablat‐ ed polyethylene terephthalate PET as flexible microchips with microelectrodes galvanically isolated into the PET was patented in 2004 [133], and then published by Gamby et al. [127]. Indeed, this system was called "SuperCapacitive Admittance Tomoscopy" (SCAT), based on the observation of a thin solvent layer on a dielectric thin film through two embedded microband electrodes (Figure 12). Indeed, the PET dielectric layer coating the two parallel microelectrodes contains a microchannel with chemically modified surface for adsorption of

between the flow microchannel and the planar microelectrodes [127].

54 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

**Figure 11.** (a) Scanning electron microscopy (SEM) images of the PET photoablated microchannel with a cross- section of 45 × 100 µm and a length of 1.4 cm. (b) Optical image of the detection zone global view including the planar micro‐ electrodes (inverted optical microscope), vertically, the two band electrodes beneath the horizontal flow channel. The detection zone represents the area where capacitive coupling takes place. (c) The SEM image of the trapezoidal section of the microchannel. (b) Side view of the contactless zone, which is about 5 µm and represents the thickness separation

The study is carried out by applying an alternating voltage of 0.1–3 V in the high frequencies (1 kHz–10 MHz) between the two microelectrodes. Indeed, a capacitive coupling effect appears

biomolecules [134].

## **4. Conclusion**

The development of a microfluidic system involves a myriad of issues and the proper selection of pertinent strategies regarding the overall feasibility, fabrication processes and targeted detection goals. At a starting point, due to its sensitivity and unfortunately, due to the fact that many materials platform and their related fabrication and conditioning means are yet be clearly identified and developed as a protocol, the coupled materials and fabrication relation‐ ship should be one of the top priorities. Undoubtedly, the most significant concern should be oriented toward the adaptation of the bioassay protocol in the specific microfluidic format, at this stage the robustness and the stability of the microfluidic operability should be taken into account in order to efficiently contribute to novel successful collaborations between different biologists, clinicians and the microfluidic communities.

## **Author details**

Emmanuel Roy, Antoine Pallandre, Bacem Zribi, Marie-Charlotte Horny, François-Damien Delapierre, Andrea Cattoni, Jean Gamby and Anne-Marie Haghiri-Gosnet\*

\*Address all correspondence to: anne-marie.haghiri-gosnet@lpn.cnrs.fr

Centre for Nanoscience and Nanotechnology, CNRS, University Paris Sud, University Paris Saclay, Marcoussis, France

## **References**


[6] R. P. Dellinger, M. N. Levy, A. Rhodes, D. Annane, H. Gerlach and S. M. Opal, Sur‐ viving sepsis campaign: International guidelines for management of severe sepsis and septic shock, *Intensive Care Med.*, 39(2), 165–228, 2013.

**4. Conclusion**

**Author details**

Saclay, Marcoussis, France

19(6), 501–9, 2013.

**References**

The development of a microfluidic system involves a myriad of issues and the proper selection of pertinent strategies regarding the overall feasibility, fabrication processes and targeted detection goals. At a starting point, due to its sensitivity and unfortunately, due to the fact that many materials platform and their related fabrication and conditioning means are yet be clearly identified and developed as a protocol, the coupled materials and fabrication relation‐ ship should be one of the top priorities. Undoubtedly, the most significant concern should be oriented toward the adaptation of the bioassay protocol in the specific microfluidic format, at this stage the robustness and the stability of the microfluidic operability should be taken into account in order to efficiently contribute to novel successful collaborations between different

biologists, clinicians and the microfluidic communities.

56 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

Emmanuel Roy, Antoine Pallandre, Bacem Zribi, Marie-Charlotte Horny,

\*Address all correspondence to: anne-marie.haghiri-gosnet@lpn.cnrs.fr

munodiagnostics, *Adv*. *Mater.*, 23, H151–76, 2011.

François-Damien Delapierre, Andrea Cattoni, Jean Gamby and Anne-Marie Haghiri-Gosnet\*

Centre for Nanoscience and Nanotechnology, CNRS, University Paris Sud, University Paris

[1] A. Arora, G. Simone, G. B. Salieb-Beugelaar, J. Tae Kim and A. Manz, Latest develop‐

[2] L. Gervais, N. de Rooij and E. Delamarche, Microfluidic chips for point-of-care im‐

[3] N. T. Tran, I. Ayed, A. Pallandre and M. Taverna, Recent innovations in protein sepa‐ ration by electrophoretic methods: An update*, Electrophoresis*, 31, 147–73, 2010.

[4] M. Goto and M. N. Al-Hasan, Overall burden of bloodstream infection and nosoco‐ mial bloodstream infection in North America and Europe, *Clin. Microbiol. Infect*.,

[5] A. Kumar, D. Roberts, K. E. Wood, B. Light, J. E. Parrillo and S. S. Sharma, Duration of hypotension before initiation of effective antimicrobial therapy is the critical deter‐ minant of survival in human septic shock, *Crit. Care Med.*, 34(6), 1589–96, 2006.

ments in micro total analysis systems, *Anal*. *Chem*., 82, 4830–47, 2010.


[18] A. Aota, M. Nonaka, A. Hibara and T. Kitamori, Countercurrent laminar microflow for highly efficient solvent extraction, *Angew. Chem. Int. Ed*., 46, 878–80, 2007.

[19] T. Minagawa, M. Tokeshi and T. Kitamori, Integration of a wet analysis system on a glass chip: Determination of Co(II) as 2-nitroso-1-naphthol chelates by solvent extrac‐

[20] H. Miyaguchi, M. Tokeshi, Y. Kikutani, A. Hibara, H. Inoue and T. Kitamori, Micro‐ chip-based liquid-liquid extraction for gas-chromatography analysis of ampheta‐

[21] A. Smirnova, K. Shimura, A. Hibara, M. A. Proskurnin and T. Kitamori, Pesticide analysis by MEKC on a microchip with hydrodynamic injection from organic extract, *J. Sep. Sci*., 31, 904–8, 2008 (Special Issue: *Micellar Electrokinetic Chromatography*).

[22] V. Reddy and J. D. Zahn, Interfacial stabilization of organic-aqueous two-phase mi‐ croflows for a miniaturized DNA extraction module, *J. Colloid Interface Sci.*, 286, 158–

[23] N. Assmann, A. Ladosz and P. R. von Rohr, Continous micro liquid-liquid extraction

[24] S. G. Redkar and R. H. Davis, Cross*-*flow microfiltration *with* high*-*frequency reverse

[25] M.-C. Lo and J. D. Zahn, Development of a multi-compartment microfiltration device for particle fractionation, 16th international conference on miniaturized systems for

[26] K. Aran, A. Fok, L. A. Sasso, N. Kamdar, Y. Guan, Q. Sun, A. Ündar and J. D. Zahn, Microfiltration platform for continuous blood plasma protein extraction from whole

[27] K. Aran, M. Morales, L. A. Sasso, J. Lo, J. Zheng, I. Johnson and J. D. Zahn, Microfli‐ tration device for continuous label-free bacteria separation from whole blood for sep‐ sis, the 15th international conference on miniaturized systems for chemistry and life

[28] J. Chung, H. Shao, T. Reiner, D. Issadore, R. Weissleder and H. Lee, Microfluidic cell sorter (µFCS) for on-chip capture and analysis of single cells, *Adv. Healthc. Mater*.,

[29] H. Mohamed, M. Murray, J. N. Turner and M. Caggana, Isolation of tumor cells us‐

[30] D. Lee, P. Sukuma, A. Mahyuddin, M. Choolani and G. Xu, Separation of model mix‐ tures of ε-globin positive fetal nucleated red blood cells and a nucleate erythrocytes

ing size and deformation, *J. Chromatogr. A,* 1216, 8289–95, 2009.

using a microfluidic device, *J. Chromatogr. A,* 1217(11), 1862– 66, 2010.

tion and thermal lens microscopy, *Lab Chip*, 1, 72–5, 2001.

58 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

(Review), *Chem. Eng. Technol.*, 36(6), 921–36, 2013.

chemistry and life sciences, Okinawa, Japan, 2012.

blood during cardiac surgery, *Lab Chip*, 11, 2858–68, 2011.

filtration, *AlChE J*., 41, 501–8, 1995.

sciences, MicroTAS, 2011.

1(4), 432–6, 2012.

65, 2005.

mine-type stimulants in urine, *J*. *Chromatogr. A*, 1129, 105–10, 2006.


[59] T. A. Zangle, A. Mani and J. G. Santiago, Theory and experiments of concentration polarization and ion focusing at microchannel and nanochannel interfaces, *Chem. Soc. Rev.*, 39, 1014–35, 2010.

[46] M. E. Warkiani, G. Guan, K. B. Luan and A. A. S. Bhagat, Slanted spiral microfluidics for the ultra-fast, label-free isolation of circulating tumor cells, *Lab Chip*, 14, 128–37,

[47] L. Clime, X. D. Hoa, N. Corneau, K. J. Morton, C. Luebbert, M. Mounier, D. Brassard, M. Geissler, S. Bidawid, Jeff Farber and T. Veres, Microfluidic filtration and extrac‐ tion of pathogens from food samples by hydrodynamic focusing and inertial lateral

[48] J. H Kang, M. Super, C. Wing Yung, R. M. Cooper, K. Domansky, A. R. Graveline, T. Mammoto, J. B. Berthet, H. Tobin, M. J. Cartwright, A. L. Watters, M. Rottman, A. Waterhouse, A. Mammoto, N. Gamini, M. J. Rodas, A. Kole, A. Jiang, T. M. Valentin, A. Diaz, K. Takahashi and Donald E Ingber, An extracorporeal blood-cleansing de‐

[50] "Capillary Electrophoresis Methods and Protocols" Editors: Schmitt-Kopplin, Phil‐ ippe (Ed.) 2008 Springer ISBN 978-1-59745-376-9, Humana Press Inc, Totowa, NJ.

[51] "Capillary Electrophoresis and Microchip Capillary Electrophoresis: Principles, Ap‐ plications, and Limitations", Carlos D. Garcia, Karin Y. Chumbimuni-Torres, Ema‐

[52] A. Manz, N. Graber and H. M. Widmer, Miniaturized total chemical analysis sys‐ tems: A novel concept for chemical sensing, *Sens. Actuators B Chem*., 1(1), 244–48,

[53] J. W. Jorgenson and K. D. Lukacs, Capillary zone electrophoresis*, Science*, 222, 266–

[54] A. Pallandre, B. de Lambert, R. Attia, A. M. Jonas, J.-L. Viovy, Surface treatment and characterization: Perspectives to electrophoresis and Lab-on-Chips, *Electrophoresis*,

[55] R. B. M. Schasfoort, S. Schlautmann, L. Hendrikse and A. van den Berg, Field-effect flow control for microfabricated fluidic networks, *Science*, 286, 942–45, 1999.

[56] A. Plecis, J. Tazid, A. Pallandre, P. Martinhon, C. Deslouis, Y. Chen and A. M. Ha‐ ghiri-Gosnet, Flow field effect transistors with polarisable interface for EOF tunable

[57] R. Liedert, L. K. Amundsen, A. Hokkanen, M. Mäki, A. Aittakorpi, M. Pakanen, J. R. Scherer, R. A. Mathies, M. Kurkinen, S. Uusitalo, L. Hakalahti, T. K. Nevanen, H. Sii‐ tari and H. Söderlund, Disposable roll-to-roll hot embossed electrophoresis chip for detection of antibiotic resistance gene *mecA* in bacteria, *Lab Chip*, 12, 333–39, 2012.

[58] Q. Pu, J. Yun, H. Temkin and S. Liu, Ion-enrichment and ion-depletion effect of nano‐

microfluidic separation devices, *Lab Chip*, 10, 1245–53, 2010.

channel structures, *Nano Lett.*, 4, 1099–1103, 2004.

nuel Carrilho 2013 Wiley, San Antonio, TX, USA, ISBN: 978-0-470-57217-7

migration, *Biomed. Microdevices*, 17, 17, 2015.

60 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

vice for sepsis therapy, *Nat. Med.*, 20(10), 1211–21, 2014.

[49] H. J. Hjertén, Free zone electrophoresis, *Chromatogr. Rev*., 9, 122–219, 1967.

2014.

1990.

72, 1983.

27, 584–610, 2006.


tic laminates for electrochemical and chemiluminescent biodetection of DNA, *Biomi‐ crofluidics*, 5,044115,2011.


[88] D. Stuart, A. Haes, C. Yonzon, E. Hicks and R. P. Van Duyne, Biological applications of localised surface plasmonic phenomenae, *IEEE Proc*. *Nanobiotechnol*., 152, 13–32, 2005.

tic laminates for electrochemical and chemiluminescent biodetection of DNA, *Biomi‐*

[72] M. Faure, A. Pallandre, S. Chebil, I. Le Potier, M. Taverna, B. Tribollet, C. Deslouis, A.-M. Haghiri-Gosnet and J. Gamby, Improved electrochemical detection of a trans‐ thyretin synthetic peptide in the nanomolar range with a two-electrode system inte‐

[73] J. L. Arlett, E. B. Myers and M. L. Roukes, Comparative advantages of mechanical bi‐

[75] C. Monat, P. Domachuk and B. J. Eggleton, Integrated optofluidics: A new river of

[76] D. Psaltis, S. R. Quake and C. Yang, Developing optofluidic technology through the

[78] A. Kausaite, M. van Dijk, J. Castrop, A. Ramanaviciene, J. P. Baltrus, J. Acaite and A. Ramanavicius, Surface plasmon resonance label-free monitoring of antibody antigen

[79] C. Y. Chao, L. J. Guo, Biochemical sensors based on polymer microrings with sharp

[80] D. K. Armani, T. J. Kippenberg, S. M. Spillane and K. J. Vahala, Ultra-high-Q toroid

[81] T. J. Kippenberg, S. M. Spillane, D. K. Armani and K. J. Vahala, Fabrication and cou‐ pling to planar high-Q silica disk microcavities, *Appl*. *Phys*. *Lett*., 83, 797–99, 2003.

[82] J. Niehusmann, A. Vörckel, P. H. Bolivar, T. Wahlbrink, W. Henschel and H. Kurz, Ultrahigh-quality-factor silicon-on-insulator microring resonator, *Opt. Lett.*, 29, 2861–

[83] M. Loncar, A. Scherer and Y. Qiu, Photonic crystal laser sources for chemical detec‐

[84] X. Fan, I. M. White, S. I. Shopova, H. Zhu, J. D. Suter and Y. Sun, Sensitive optical biosensors for unlabeled targets: A review, *Anal. Chim. Acta*, 620, 8–26, 2008.

[85] S. M. Borisov and O. S. Wolfbeis, Optical Biosensors, *Chem*. *Rev*., 108, 423−61, 2008.

[86] J. D. Joannopoulos and S. G. Johnson, Photonic crystals: Molding the flow of light,

[87] C. Fenzl, T. Hirsch and O. S. Wolfbeis, Photonic crystals for chemical sensing and bi‐

grated in a glass/PDMS microchip, *Lab Chip*, 14, 2800–05, 2014.

[74] N. Pamme, Magnetism and microfluidics, *Lab Chip*, 6, 24–38, 2006.

fusion of microfluidics and optics, *Nature*, 442, 381–86, 2006.

asymmetrical resonance, *Appl. Phys. Lett*., 83, 1527–29, 2003.

microcavity on a chip, *Nature*, 421, 925–28, 2003.

tion, *Appl*. *Phys*. *Lett*., 82, 4648–4650, 2003.

Princeton University Press, 2008, Princeton, New Jersey.

osensing, *Angew*. *Chem*. *Int*. *Ed*., 53, 3318–35, 2014.

63, 2004.

[77] A. G. Brolo, Plasmonics for future biosensors, *Nat. Photon.*, 6, 709–13, 2012.

interactions in real time, Biochem. Mol. Boil. Educ., 35, 57–63, 2007.

osensors, *Nat. Nanotechnol*. 6, 203–15, 2011.

62 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

light, *Nat. Photon.*, 1, 106– 114, 2007.

*crofluidics*, 5,044115,2011.


exploring cellular dynamics in real-time using electrochemical detection, *RSC Adv*., 4, 63761–71, 2014.


[114] M. Faure, S. Korchane, I. Le Potier, A. Pallandre, C. Deslouis, A.-M. Haghiri-Gosnet and J. Gamby, Investigating of labeling and detection of transthyretin synthetic pep‐ tide derivatized with naphthalene-2,3-dicarboxaldehyde, *Talanta*, 116, 8–13, 2013.

exploring cellular dynamics in real-time using electrochemical detection, *RSC Adv*., 4,

[102] S. A. Almeida, E. Arasa, M. Puyol, C. S. Martinez-Cisneros, J. Alonso-Chamarro, M. C. Montenegro and M. G. Sales, Novel LTCC-potentiometric microfluidic device for biparametric analysis of organic compounds carrying plastic antibodies as iono‐ phores: Application to sulfamethoxazole and trimethoprim, *Biosens. Bioelectron.*, 30,

[103] S. S. Hassan, H. E. Sayour and S. S. Al-Mehrezi, A novel planar miniaturized poten‐ tiometric sensor for flow injection analysis of nitrates in wastewaters, fertilizers and

[104] N. Ibáñez-García, M. Baeza, M. Puyol, R. Gómez, M. Batlle and J. Alonso-Chamarro, Biparametric potentiometric analytical microsystem based on the green tape technol‐

[105] T. Ahuja, I. A. Mir and D. Kumar, Potentiometric urea biosensor based on BSA em‐ bedded surface modified polypyrrole film, *Sens. Actuators B: Chem.*, 134, 140–45, 2008.

[106] M. J. Natan, D. Belanger, M. K. Carpenter and M. S. Wrighton, pH-sensitive nickel(II) hydroxide-based microelectrochemical transistors, *J. Phys. Chem.*, 91, 1834–42, 1987.

[107] B. Lakard, O. Segut, S. Lakard, G. Herljem and T. Gharbi, Potentiometric miniatur‐ ized pH sensors based on polypyrrole films, *Sens. Actuators B: Chem.*, 122, 101–8, 2007

[108] M. J. Schöning, F. Ronkel, M. Crott, M. Thust, J. W. Schultze, P. Kordos and H. Lüth, Miniaturization of potentiometric sensors using porous silicon microtechnology*,*

[109] S. V. Dzyadevych, V. N. Arkhypova, A. P. Soldatkin, A. V. El'skaya, C. Martelet and N. Jaffrezic-Renault, Amperometric enzyme biosensors: Past, present and future,

[110] Y. Chao, H. Yue, B. L. Hassler, R. M. Worden and A. J. Mason, Amperometric electro‐ chemical microsystem for a miniaturized protein biosensor array, biomedical circuits

[111] L. C. Clark, Monitor and control of blood and tissue oxygen tensions, *Trans. Am. Soc.*

[112] J. Wang, M. P. Chatrathi and B. Tian, Capillary electrophoresis microchips with thick-film amperometric detectors: separation and detection of phenolic compounds,

[113] N. Dossi, R. Toniolo, A. Pizzariello, S. Susmel, F. Perennes and G. Bontempelli, A ca‐ pillary electrophoresis microsystem for the rapid in-channel amperometric detection

of synthetic dyes in food, *J. Electroanal. Chem*., 601, 1–7, 2007.

[101] J. Janata, Potentiometric microsensors, *Chem. Rev.*, 90, 691–3, 1990.

64 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

pharmaceuticals, *Anal. Chim. Acta*, 581, 13–18, 2007.

ogy, *Electroanalysis*, 22, 2376–82, 2010.

*Electrochim. Acta*, 42, 3185–93, 1997.

*Artif. Intern. Organs*, 2, 41, 1956.

*Anal. Chim. Acta*, 416, 9–14, 2000.

and systems, *IEEE Transactions on*, 3, 160–68, 2009.

*IRBM*, 29, 171–80, 2008.

63761–71, 2014.

197–203, 2011.


## **Application of Microfluidics in Stem Cell Culture**

Shinji Sugiura, Kohji Nakazawa, Toshiyuki Kanamori and Kiyoshi Ohnuma

Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/64714

#### **Abstract**

[127] J. Gamby, J. P. Abid, B. Tribollet and H. H. Girault, Nanomosaic network for the de‐ tection of proteins without direct electrical contact, *Small*, 4, 802–9, 2008.

[128] A. J. Zemann, E. Schnell, D. Volgger and G. K. Bonn, Contactless conductivity detec‐

[129] M. Pumera, Contactless conductivity detection for microfluidics: Designs and appli‐

[130] B. Gaš, M. Demjaněnko and J. Vacík, High-frequency contactless conductivity detec‐

[131] J. Gamby, J.-P. Abid and H. H. Girault, Supercapacitive admittance tomoscopy, *J.*

[132] J. Gamby, J. P. Abid, M. Abid, J. P. Ansermet and H. H. Girault, Nanowires network for biomolecular detection using contactless impedance tomoscopy technique, *Anal.*

[133] J.-P. Abid, J. O. Gamby and H. H. Girault, Adsorption monitoring device for contact‐ less testing of sensors using capacitive admittance, *in, Ecole Polytechnique Federale de*

[134] M. Kechadi, B. Sotta, L. Chaal, B. Tribollet and J. Gamby, A real time affinity biosen‐ sor on an insulated polymer using electric impedance spectroscopy in dielectric mi‐

tion for capillary electrophoresis, *Anal. Chem.*, 70, 563–67, 1998.

66 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

tion in isotachophoresis, *J. Chromatogr. A*, 192, 253–57, 1980.

cations, *Talanta*, 74, 358–64, 2007.

*Am. Chem. Soc.*, 127, 13300–04, 2005.

*Lausanne, Switzerland* , p. 33, 2006.

crochips, *Analyst*, 139, 3115–21, 2014.

*Chem*., 78, 5289–95, 2006.

In this chapter, we review the recent developments, including our studies on the micro‐ fabricated devices applicable to stem cell culture. We will focus on the application of plu‐ ripotent stem cells including embryonic stem cells and induced pluripotent stem cells. In the first section, we provide a background on microfluidic devices, including their fabri‐ cation technology, characteristics, and the advantages of their application in stem cell cul‐ ture. The second section outlines the use of micropatterning technology in stem cell culture. The use of microwell array technology in stem cell culture is explored in the third section. In the fourth section, we discuss the use of the microfluidic perfusion culture sys‐ tem for stem cell culture, and the last section is a summary of the current state of the art and perspectives of microfluidic technologies in stem cell culture.

**Keywords:** Embryonic stem (ES) cells, induced pluripotent stem (iPS) cells, microfluidic perfusion culture, micropatterning, microwell array

## **1. Introduction**

This section provides a general background on microfluidic devices and explains the general microfabrication technologies applicable to stem cell culture including embryonic stem (ES) cells and induced pluripotent stem (iPS) cells. We will discuss the importance of smallscale patterning, three-dimensional structure, and medium flow in terms of microenviron‐ ment control and the importance of small volumes, in terms of research cost, for industrial application.

#### **1.1. Microfabrication technology available for stem cell culture**

Microfabrication technology progressed rapidly with the development of semiconductor industry in the 20th century. By the end of the 20th century, the application of microfabrication

technology started to grow in different research areas including biotechnology. In biotechnol‐ ogy, microfabrication technology was initially used for molecular analyses of DNA and proteins and gradually its application diversified to cell culture. This technology enabled precise fabrication of structures with sizes as small as submicrometer, replication of the fabricated structure, liquid manipulation in very small volumes, portability of the devices, and usage of small amounts of expensive reagents. Owing to these advantages, microfabrication technology is expected to create new applications in the cell culture including stem cells.

Many types of materials, including inorganic materials, metals, polymers, and plastics, are applicable to microfabrication. Silicon and glass have been used to fabricate microstructures and semiconductor devices [1, 2]. Polydimethylsiloxane (PDMS), a silicone elastomer, is the most popular material used for the fabrication of microfluidic cell culture devices due to the ease of fabrication, optical transparency, gas permeability, low chemical reactivity, and inexpensiveness. In addition, microstructure of PDMS is generally fabricated by soft lithog‐ raphy in a few days [3]. This easy and quick process broadens the use of microfluidic devices in cell culture applications. In soft lithography process, replica of microstructure in PDMS can be repeatedly fabricated from a microstructure of photoresist that is originally made using photolithography [4, 5]. A multilayered microstructure of PDMS can also be fabricated by using multilayered photoresist pattern [6, 7]. Details of the fabrication method used for soft lithography have been described in previous studies [3, 8]. Many biologists are currently using this convenient microfabrication technology.

Soft lithography is a convenient method for fabricating microstructures on a laboratory scale. Scientists can fabricate dozens of microfluidic devices by themselves for their research. However, for industrial applications, hundreds or thousands of microfluidic devices are required. In addition, it is known that PDMS absorb small hydrophobic molecules [9]. Therefore, other materials applicable to mass production and capable of avoiding molecular adsorption are desired for the industrial application of microfluidic devices. Low-cost fabrication technologies such as injection molding [10, 11] and rapid prototyping [12, 13] are promising fabrication technologies that can address the above-mentioned issues.

In addition, cell culture often requires extracellular matrixes (ECMs). Therefore, microfabri‐ cation of biomaterials, such as hydrogel, is of interest to biologists and engineers. Photofabri‐ cated hydrogels have been studied extensively to create microstructure in the hydrogels [14]. These microfabricated hydrogels have been used for tissue engineering.

#### **1.2. Characteristics of microfabricated cell culture device**

Significant features that affect a microfluidic device are flow viscosity, interfacial tension, laminar flow, fast diffusion, etc. [15]. The characteristic flow profile enabled the formation of special microenvironment including chemical [16] and temperature gradients [17]. Also, a microfluidic perfusion culture continuously supplies nutrient and removes waste, and therefore keeps the culture condition more stable and constant compared with a static cell culture [18]. Furthermore, a microfluidic perfusion culture potentially provides new oppor‐ tunities for cell culture applications because of the precise control of the microscale environ‐ ment [19–22]. For example, some cell types, such as endothelial cells, are sensitive to shear stress caused by the flow of the medium [23]. Another example, in 3D culture condition, such as spheroid culture, molecular transport in the microchannels can be controlled by convection flow and the controlled molecular transport affects the state of inner cells in the spheroids [24]. Therefore, microfluidic perfusion culture can be used for both fundamental research and drug development.

Another feature of cell culture in microfluidic device is its small volume. The miniaturized assays are expected to increase experimental throughput and reliability for drug discovery applications [25–27]. This is an important aspect for the application of microfluidic technology to stem cell culture because culturing stem cells, especially human iPSCs, is cost-prohibitive [28]. Microfluidic systems are cost-effective because these systems need small quantity of culture medium and reagents.

In addition, microfluidic device can generate many cell culture conditions using microfluidic network. For example, we have developed a microfluidic network to generate step-wise serial dilutions [29]. We also reported the method to fabricate combinatorial microenvironment array on a microfluidic device [30]. Titmarsh et al. reported a microfluidic network to generate combinatorial array of culture conditions composed of multiple soluble factors at different concentrations [31], and applied this device for the analysis of human embryonic stem cell culture conditions. We think these examples are just the beginning of the application of microfabrication technology in stem cell culture. Possibly, there are additional scopes for applications because many unknown phenomena regarding stem cell culture are yet to be understood.

#### **1.3. Control of cell culture microenvironment**

technology started to grow in different research areas including biotechnology. In biotechnol‐ ogy, microfabrication technology was initially used for molecular analyses of DNA and proteins and gradually its application diversified to cell culture. This technology enabled precise fabrication of structures with sizes as small as submicrometer, replication of the fabricated structure, liquid manipulation in very small volumes, portability of the devices, and usage of small amounts of expensive reagents. Owing to these advantages, microfabrication technology is expected to create new applications in the cell culture including stem cells.

68 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

Many types of materials, including inorganic materials, metals, polymers, and plastics, are applicable to microfabrication. Silicon and glass have been used to fabricate microstructures and semiconductor devices [1, 2]. Polydimethylsiloxane (PDMS), a silicone elastomer, is the most popular material used for the fabrication of microfluidic cell culture devices due to the ease of fabrication, optical transparency, gas permeability, low chemical reactivity, and inexpensiveness. In addition, microstructure of PDMS is generally fabricated by soft lithog‐ raphy in a few days [3]. This easy and quick process broadens the use of microfluidic devices in cell culture applications. In soft lithography process, replica of microstructure in PDMS can be repeatedly fabricated from a microstructure of photoresist that is originally made using photolithography [4, 5]. A multilayered microstructure of PDMS can also be fabricated by using multilayered photoresist pattern [6, 7]. Details of the fabrication method used for soft lithography have been described in previous studies [3, 8]. Many biologists are currently using

Soft lithography is a convenient method for fabricating microstructures on a laboratory scale. Scientists can fabricate dozens of microfluidic devices by themselves for their research. However, for industrial applications, hundreds or thousands of microfluidic devices are required. In addition, it is known that PDMS absorb small hydrophobic molecules [9]. Therefore, other materials applicable to mass production and capable of avoiding molecular adsorption are desired for the industrial application of microfluidic devices. Low-cost fabrication technologies such as injection molding [10, 11] and rapid prototyping [12, 13] are

In addition, cell culture often requires extracellular matrixes (ECMs). Therefore, microfabri‐ cation of biomaterials, such as hydrogel, is of interest to biologists and engineers. Photofabri‐ cated hydrogels have been studied extensively to create microstructure in the hydrogels [14].

Significant features that affect a microfluidic device are flow viscosity, interfacial tension, laminar flow, fast diffusion, etc. [15]. The characteristic flow profile enabled the formation of special microenvironment including chemical [16] and temperature gradients [17]. Also, a microfluidic perfusion culture continuously supplies nutrient and removes waste, and therefore keeps the culture condition more stable and constant compared with a static cell culture [18]. Furthermore, a microfluidic perfusion culture potentially provides new oppor‐ tunities for cell culture applications because of the precise control of the microscale environ‐ ment [19–22]. For example, some cell types, such as endothelial cells, are sensitive to shear

promising fabrication technologies that can address the above-mentioned issues.

These microfabricated hydrogels have been used for tissue engineering.

**1.2. Characteristics of microfabricated cell culture device**

this convenient microfabrication technology.

Conventionally, cell culture has been carried out in Petri dishes as static culture. In Petri dishes, the actively growing cells form monolayer sheet and culture media is placed on the cells. In this static monolayer culture, cells grow at randomly arranged positions and medium is exchanged regularly in batches. The stem cells are cultured in a similar manner. In contrast, cells in our bodies form highly ordered 3D microstructures and respond to their surrounding microenvironments including soluble factors, ECMs, contact-dependent intercellular signals, and mechanical signals. Therefore, the 3D nature of native, complex microenvironments is not accurately recapitulated in traditional cell culture on Petri dishes[32]. Microfabrication technology has the potential to control the parameters to simulate these complex 3D micro‐ environments.

## **2. Micropatterning technology in stem cell culture**

Monolayer cultures of stem cells garnered considerable attention after human ES/iPS cells were established, because these cells are cultured as a monolayer colony and cannot survive without adhering to the surface of the culture dish. In addition, the differentiation of these cells is sensitive to cell density because cell-cell interactions affect stem cell differentiation. Thus, regulation of cell adhesion and control of shape and size of the stem cell monolayer colony are very important for maintaining stem cell potential and for inducing these stem cells to differentiate into specific cells types. In this section, we reviewed the micropatterning tech‐ nology and its application in human iPS cell culture.

#### **2.1. Micropatterning for human ES/iPS cells: difference between human and mouse ES/iPS cells**

Human ES/iPS cells can differentiate all kinds of human body cells [33, 34]. Human ES cells generated by somatic cell nuclear transfer and human iPS cells contain the donor's genetic information. Therefore, ES/iPS cells can be a good source of cells for rejection-free transplan‐ tation of tissues and disease-specific drug screening [34, 35].

Although ES and iPS cells share most of the properties, there are clear differences between mouse and human ES/iPS cells (Figure 1). Mouse ES/iPS cells can survive after single-cell dissociation, but human ES/iPS cells undergo apoptosis1 [36] following their single-cell dissociation. Thus, the culture conditions and adhesion of human ES/iPS cells need to be carefully controlled to ensure their survival and growth.

**Figure 1.** Difference between mouse and human ES/iPS cells.

Moreover, there are morphological differences between mouse and human ES/iPS cell colonies. Mouse ES/iPS cells form spherical aggregates [37–39]and human ES/iPS cells form flat monolayer colonies [33, 34]. Because the colony size is important for maintaining pluri‐ potency and controlling cell differentiation in ES/iPS cells (see section 3), two-dimensional cell

<sup>1</sup> Apoptosis caused by single cell dissociation is partially rescued by adding Y-27632, a selective inhibitor of p160- Rhoassociated coiled-coil kinase (ROCK).

patterning is especially important for human ES/iPS cells. Warmflash et al. [40] showed that the differentiation pattern of human ES/iPS cells depended on the size of monolayer colonies.

In addition, although both mouse and human ES cells are derived from inner cell mass of the blastocysts, the properties of mouse ES/iPS cells are closer to that of the inner cell mass, while the properties of human ES/iPS cells are closer to that of epiblast, which is a monolayer of cells arising from the inner cell mass [41]. The epiblast is functionally and molecularly distinct from inner cell mass, and is also pluripotent. Thus, micropatterning technology is important, especially for human ES/iPS cells.

### **2.2. Micropatterning technology**

very important for maintaining stem cell potential and for inducing these stem cells to differentiate into specific cells types. In this section, we reviewed the micropatterning tech‐

**2.1. Micropatterning for human ES/iPS cells: difference between human and mouse ES/iPS**

Human ES/iPS cells can differentiate all kinds of human body cells [33, 34]. Human ES cells generated by somatic cell nuclear transfer and human iPS cells contain the donor's genetic information. Therefore, ES/iPS cells can be a good source of cells for rejection-free transplan‐

Although ES and iPS cells share most of the properties, there are clear differences between mouse and human ES/iPS cells (Figure 1). Mouse ES/iPS cells can survive after single-cell dissociation, but human ES/iPS cells undergo apoptosis1 [36] following their single-cell dissociation. Thus, the culture conditions and adhesion of human ES/iPS cells need to be

Moreover, there are morphological differences between mouse and human ES/iPS cell colonies. Mouse ES/iPS cells form spherical aggregates [37–39]and human ES/iPS cells form flat monolayer colonies [33, 34]. Because the colony size is important for maintaining pluri‐ potency and controlling cell differentiation in ES/iPS cells (see section 3), two-dimensional cell

1 Apoptosis caused by single cell dissociation is partially rescued by adding Y-27632, a selective inhibitor of p160-

nology and its application in human iPS cell culture.

70 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

tation of tissues and disease-specific drug screening [34, 35].

carefully controlled to ensure their survival and growth.

**Figure 1.** Difference between mouse and human ES/iPS cells.

Rhoassociated coiled-coil kinase (ROCK).

**cells**

Although there are many cell micropatterning techniques available [42–47], two important features need to be considered while applying this technique to human ES/iPS cells. The human ES/iPS cell differentiation protocols take a few days to few months, and the cell pattern needs to be long lasting. Although there have been reports of successful patterning of ES/iPS cells, the cells have been found to escape from the pattern within a few days [43, 46]. The other feature is ease of pattern preparation. Many micropatterning methods require some special equipment and techniques that are not easy to perform routinely in cell biology laboratory.

### **2.3. Micropatterning technology in human ES/iPS cells**

We succeeded in forming human iPS cells pattern on the PDMS surface by a simple technique using plasma2 oxidation with perforated mask and defined culture conditions [48, 49]. As described above, PDMS is one of the most popular biocompatible materials for research and development of cell culture microdevices. Plasma treatment on PDMS oxidized ≡Si–CH3 groups to generate ≡Si–O–Si≡ groups suggests that hydrophilic and siliceous layers were formed on the surface [5, 49].

We first studied the effects of vitronectin and γ-globulin on hiPSC adhesion to plasma-treated and untreated PDMS surfaces under defined culture conditions [49]. We chose vitronectin and γ-globulin because they have contrasting properties. Vitronectin as well as fibronectin and laminin mediate hiPSC attachment, because vitronectin and fibronectin are ligands of integrin α5β1 and αvβ1, and laminin is a ligand of integrin α6β1 and αvβ5, all of which are known to be expressed on ES/iPS cells surface [50–53]. Moreover, vitronectin is especially suitable for coating on glass (≡Si–O–Si≡) [54]. On the other hand, γ-globulin is adsorbed by hydrophobic surfaces and does not mediate cell adhesion [55]. Immunostaining showed that vitronectin and γ-globulin were adsorbed on both plasma-treated and plasma-untreated PDMS surfaces when these proteins were applied separately. However, vitronectin was preferentially adsorbed on plasma-treated surfaces whereas γ-globulin was adsorbed on untreated surfaces when the mixture of vitronectin and γ-globulin was applied. Human iPSCs adhered to the vitronectin-rich plasma-treated surfaces but not to the γ-globulin-rich untreated surfaces.

<sup>2</sup> In this chapter, plasma refers to low-pressure plasma, not blood plasma, unless otherwise stated.

Based on the results, we succeeded in making a monolayer pattern of human iPS cells by using perforated masks to prepare plasma-patterned PDMS substrates [49]. The patterned human iPS cells expressed undifferentiated-cell markers and did not escape from the patterned area for at least 7 days. The patterned PDMS could be stored for up to 6 days before hiPSCs were plated. Furthermore, we demonstrate that not only γ-globulin but also bovine serum albumin (BSA) could be used to block human iPS cell adhesion on plasma-untreated PDMS surfaces (Figure 2) [48]. The hiPSCs proliferated without escaping from the patterned area and finally detached spontaneously from the discs to form spheroids.

**Figure 2.** Micropatterned human iPS cells created by plasma patterning of PDMS surfaces and single-step coating of vitronectin and BSA. (A, B) Perforated PMMA masks. (C) Schematic of the micropatterning procedure. (D–F) Immu‐ nostaining of patterned cells with undifferentiated cells marker anti-SSEA4 (green) and anti-OCT3/4 (red). Nuclei were stained with DAPI (blue). Lower panels show high magnification images. Reproduced from Yamada et al. [48]with permission from Begell House.

Our micropatterning method presents four advantages over previously reported methods [42, 56]. (1) The plasma treatment through perforated masks enables equal patterning on a wide area, therefore a large number of homogeneously patterned cells can be created reproducibly. (2) Single-step coating of a mixture of proteins is quite simple and easy. Similar methods of producing cell patterning required additional steps, including multistep protein coatings of BSA followed by ECM [56]. (3) The cost-effectiveness and availability of γ-globulin, especially BSA is high. (4) Although there are many types of micro-fabrication tools to make equally sized spheroids, most of them are expensive and difficult to use in cell culture labs [42]. Our method is an easy and cost-effective way to fabricate hiPSC discs and spheroids.

#### **2.4. Harvesting micropatterned human ES/iPS cells by controlling divalent cation**

Cell sheets such as retinal pigment epithelium and cardiomyocytes derived from human iPS cells have been developed for applications in regenerative medicine [57, 58]. We tried to har‐ vest micropatterned human ES/iPS cells without their splitting off. Conventionally, cell sheets are harvested using special equipments, such as a temperature-responsive surface and magnet [59, 60]. In contrast, we focused on integrin and cadherin, which are adhesion molecules on the cell surface. Cadherins mediate cell-cell adhesion at physiological concen‐ tration of Ca2+[61]. On the other hand, integrins mediate cell-ECM adhesion and depend largely on Mg2+ [62]. Thus, a solution containing physiological concentration of Ca2+, but no Mg2+, could be used to harvest cells as large cell clumps under serum-free culture condition. As expected, simple incubation in PBS with Ca2+ without Mg2+ followed by gentle pipetting enabled us to harvest the cells as sheets without cells splitting off (Figure 3) [63]. Similar re‐ sults were obtained for early-differentiated cells and for hepatic progenitors derived from human iPS cells. These results suggest that the cells can be routinely and simply harvested as a large sheet by using a solution with Ca2+and without Mg2+.

Based on the results, we succeeded in making a monolayer pattern of human iPS cells by using perforated masks to prepare plasma-patterned PDMS substrates [49]. The patterned human iPS cells expressed undifferentiated-cell markers and did not escape from the patterned area for at least 7 days. The patterned PDMS could be stored for up to 6 days before hiPSCs were plated. Furthermore, we demonstrate that not only γ-globulin but also bovine serum albumin (BSA) could be used to block human iPS cell adhesion on plasma-untreated PDMS surfaces (Figure 2) [48]. The hiPSCs proliferated without escaping from the patterned area and finally

**Figure 2.** Micropatterned human iPS cells created by plasma patterning of PDMS surfaces and single-step coating of vitronectin and BSA. (A, B) Perforated PMMA masks. (C) Schematic of the micropatterning procedure. (D–F) Immu‐ nostaining of patterned cells with undifferentiated cells marker anti-SSEA4 (green) and anti-OCT3/4 (red). Nuclei were stained with DAPI (blue). Lower panels show high magnification images. Reproduced from Yamada et al. [48]with

Our micropatterning method presents four advantages over previously reported methods [42, 56]. (1) The plasma treatment through perforated masks enables equal patterning on a wide area, therefore a large number of homogeneously patterned cells can be created reproducibly. (2) Single-step coating of a mixture of proteins is quite simple and easy. Similar methods of producing cell patterning required additional steps, including multistep protein coatings of BSA followed by ECM [56]. (3) The cost-effectiveness and availability of γ-globulin, especially BSA is high. (4) Although there are many types of micro-fabrication tools to make equally sized spheroids, most of them are expensive and difficult to use in cell culture labs [42]. Our method

is an easy and cost-effective way to fabricate hiPSC discs and spheroids.

**2.4. Harvesting micropatterned human ES/iPS cells by controlling divalent cation**

Cell sheets such as retinal pigment epithelium and cardiomyocytes derived from human iPS cells have been developed for applications in regenerative medicine [57, 58]. We tried to har‐ vest micropatterned human ES/iPS cells without their splitting off. Conventionally, cell sheets are harvested using special equipments, such as a temperature-responsive surface

detached spontaneously from the discs to form spheroids.

72 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

permission from Begell House.

Our methods introduced here can be also used practically to mimic epiblast (Figure 1) in early human embryonic development. We believe that our cell patterning method will be useful for the development of new bioengineering tools to search for effective cell differentiation methods and to test drug safety for early human embryonic development.

 **Figure 3.** Human iPS cells sheet harvesting. (A) Schematics of spot sheet formation and harvest. (B) ALP staining of the hiPSCs plated on 2-mm-diameter fibronectin spots. Phase-contrast micrographs before (C) and after (D, E) 15 min in PBS with Ca2+ followed by pipetting. The white arrows indicate the same cellspot sheet (C–E). The red spots in (C, D) are position makers. Scale bars are 1 cm (B), 1 mm (C–E). (F) Schematics of the effects of Mg2+ and Ca2+ on hPSCs cul‐ ture. Cell-cell and cell-ECM adhesion depend largely on Ca2+ (abscissa) and Mg2+ (ordinate), respectively. Large cell clumps and sheets can then be harvested by dissociating in lowMg2+ and high Ca2+ solution (lower right).Reproduced from Ohnuma et al. [63] under a Creative Commons Attribution 3.0 Unported License.

## **3. Microwell array technology in stem cell culture**

The formation of three-dimensional cell aggregates called embryoid bodies (EBs) that resemble the embryo structure is the principle behind in vitro differentiation of stem cells. Microwell array is a promising platform for generating EBs, in which microwells of several hundred micrometers size are regularly fabricated on a culture substratum. It controls EB size and produces large number of homogenous EBs. An added advantage of microwell array culture is that it can influence the fate of differentiating cells in EBs. In this section, we review the architectures of microwell arrays, microwell array culture of ES/iPS cells, and the relationships between microwell conditions and EB properties.

### **3.1. Embryoid body (EB) culture**

For in vitro differentiation of stem cells, one of the superior strategies is to imitate in vivo development processes. Three-dimensional aggregate of stem cells called embryoid body resembles early embryo. The multicellular interactions generated with the EB formation trigger cell differentiation. Thus, the EB culture has been used as a principal method for in vitro early differentiation of stem cells.

The EB is formed by the rearrangement and compaction of stem cell aggregates. Therefore, the differences in EB sizes affect the diffusion of soluble molecules, the extent of cell-ECM and cell-cell interactions, and the generation of mechanical forces (Figure 4A).Consequently, it affects the differentiation and fate of stem cells [64–66]. Thus, the culture technique capable of modulating EB size is important to regulate stem cell differentiation.

**Figure 4.** (A) EB characteristics and (B) methods for EB formation.

#### **3.2. Methods for EB formation**

The EB formation occurs if cell-cell adhesion is stronger than cell-surface material adhesion. Therefore, we can lead the EB formation by designing a culture environment, which promotes cell-cell adhesion. Typical methods for EB formation are hanging drop culture, roundbottomed 96-well culture, and agitation or rotational culture (Figure 4B) [67, 68]. The agitation culture can achieve mass production of EBs; however, it is difficult to control the EB size. In contrast, the hanging drop and round-bottomed 96-well cultures can control the EB size, but their scale-up is difficult. Furthermore, these methods pose difficulties in handling of the formed EBs.

Recently, microwell array has been advocated as a promising technique over the current methods. It is a culture platform in which microwells of several hundred micrometers size are regularly fabricated on a culture plate (Figure 4B). Recently, various microwell arrays have been developed by researchers [69]. Generally, the number of microwells is from tens to thousands per culture plate, and these microwells are laid as triangular or square arrangements on the plate. The microwells having various shapes such as column, square, and pyramid have been fabricated by photolithography, soft lithography mold, micromachining, etc. [70–72].

**Figure 5.** (A) Microwell array, (B) manufacturing processes, (C) EB formation, and (D) culture in the microwell array.

Furthermore, most microwell arrays are designed with cell nonadhesion surfaces to promote the EB formation. Such microwell arrays can generate a large number of homogenous EBs, with controlled size, and allow easy EB handling.

#### **3.3. Our microwell array**

**3.1. Embryoid body (EB) culture**

differentiation of stem cells.

For in vitro differentiation of stem cells, one of the superior strategies is to imitate in vivo development processes. Three-dimensional aggregate of stem cells called embryoid body resembles early embryo. The multicellular interactions generated with the EB formation trigger cell differentiation. Thus, the EB culture has been used as a principal method for in vitro early

The EB is formed by the rearrangement and compaction of stem cell aggregates. Therefore, the differences in EB sizes affect the diffusion of soluble molecules, the extent of cell-ECM and cell-cell interactions, and the generation of mechanical forces (Figure 4A).Consequently, it affects the differentiation and fate of stem cells [64–66]. Thus, the culture technique capable of

The EB formation occurs if cell-cell adhesion is stronger than cell-surface material adhesion. Therefore, we can lead the EB formation by designing a culture environment, which promotes cell-cell adhesion. Typical methods for EB formation are hanging drop culture, roundbottomed 96-well culture, and agitation or rotational culture (Figure 4B) [67, 68]. The agitation culture can achieve mass production of EBs; however, it is difficult to control the EB size. In contrast, the hanging drop and round-bottomed 96-well cultures can control the EB size, but their scale-up is difficult. Furthermore, these methods pose difficulties in handling of the

Recently, microwell array has been advocated as a promising technique over the current methods. It is a culture platform in which microwells of several hundred micrometers size are regularly fabricated on a culture plate (Figure 4B). Recently, various microwell arrays have been developed by researchers [69]. Generally, the number of microwells is from tens to thousands per culture plate, and these microwells are laid as triangular or square arrangements on the plate. The microwells having various shapes such as column, square, and pyramid have been fabricated by photolithography, soft lithography mold, micromachining, etc. [70–72].

modulating EB size is important to regulate stem cell differentiation.

74 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

**Figure 4.** (A) EB characteristics and (B) methods for EB formation.

**3.2. Methods for EB formation**

formed EBs.

We developed a microwell array comprising columnar microwells in triangular arrangement on a poly-methylmethacrylate (PMMA) plate with the surface modified by polyethylene glycol (PEG) to render cell nonadhesion (Figure 5A) [73, 74]. This was fabricated as follows (Figure 2B). The microwell structure of array was fabricated using a programmable micromilling system. Subsequently, the array surface was coated with thin layer of platinum using an ion sputter unit, and immersed in a PEG-SH solution. The PEG molecule formed covalent attachment via its thiol group to the platinum layer, thus modifying the surface. The microwell array manufactured via this process was used for the stem cell culture.

Figure 5C shows the changes in cell morphologies of mouse ES cells within the microwell. The cells that were seeded on the array began to aggregate within several hours of inoculation, and they spontaneously formed a single EB in each microwell of array within 1 day of culturing. Although none of the EBs adhered on the microwell surface, all EBs were held within the microwells throughout the culture period. Consequently, the mass production of homogenous EBs was achieved in the single array (Figure 5D).

#### **3.4. EB properties in the microwell array culture**

The microwell array could arbitrarily vary the microwell conditions such as well number, diameter and depth of well, distance of wells, and cell density. To clarify the characteristics of microwell array culture, we evaluated the effects of microwell diameter and cell density on the EB properties of mouse ES cells [75, 76].

Four similar arrays comprising 195 microwells were fabricated with microwell diameters of 400, 600, 800, and 1000 µm to evaluate the relationship between the microwell diameter and EB property. The hanging drop (HD) culture was used as a control method. In this experiment, the inoculated cell density was at 1000 cells/well or 1000 cells/drop, and the cells were cultured in a medium without LIF and inducers.

Figure 6A shows the changes in EB sizes. The EB sizes at the initial stage were almost same (approximately 150 µm) under all culture conditions. Although the EB in the HD culture grew drastically with increasing culture time, the growth in the microwell culture was repressed compared with that in HD culture. Furthermore, the changes in EB sizes depended on the scale of microwell, and the EB growth in larger microwells was higher than that in smaller micro‐ wells. Figure 6B shows the gene expression levels of hepatic (AFP), cardiac (αMHC), and vascular (Flk1) differentiation markers. The differentiation fates of EBs were nearly the same under all conditions, but the gene expression levels varied with culture conditions. The expressions of differentiation markers were highest in HD culture, and gradually decreased in 1000, 800, 600, 400 µm arrays in that order. These results indicate that the EB growth and differentiation rate can be controlled by diameters of microwell, and that they are promoted in larger microwell conditions. These differences may be attributed to special culture envi‐ ronments provided by the microwell culture. The existence of micro spaces (microwell) might have facilitated accumulation of various soluble factors which were secreted from the cells. Additionally, interference effects caused by the neighboring EBs could occur because of extensive EBs on an array. Consequently, the accumulation of paracrine/autocrine factors and/ or the concentrations of oxygen and nutrients in the culture medium vary by the array conditions, and the balances of these factors may regulate the EB properties.

To evaluate the relationship between the inoculated cell density and EB property, the cells at densities of 100, 1000, or 10000 cells/well were inoculated to the array that comprised 195 microwells with 600 µm diameter. Figures 7A and B show the changes in EB sizes and the gene expression levels of differentiation markers, respectively. The change in the EB size of array at 100 cells/well was higher than that of array at 1000 cells/well. In contrast, the cells of array at 10000 cells/well hardly proliferated, and the EB size that formed at the initial stage was maintained throughout the culture period. The expression levels of differentiation markers were the smallest in the array at 10000 cells/well. The expression of vascular (Flk1) differen‐ tiation was highest in the array at 100 cells/well, but the array at 1000 cells/well promoted hepatic (AFP) and cardiac (αMHC) differentiations rather than vascular differentiation, indicating that the inoculated cell density affects the differentiation fate of EBs. These differ‐

microwells throughout the culture period. Consequently, the mass production of homogenous

The microwell array could arbitrarily vary the microwell conditions such as well number, diameter and depth of well, distance of wells, and cell density. To clarify the characteristics of microwell array culture, we evaluated the effects of microwell diameter and cell density on

Four similar arrays comprising 195 microwells were fabricated with microwell diameters of 400, 600, 800, and 1000 µm to evaluate the relationship between the microwell diameter and EB property. The hanging drop (HD) culture was used as a control method. In this experiment, the inoculated cell density was at 1000 cells/well or 1000 cells/drop, and the cells were cultured

Figure 6A shows the changes in EB sizes. The EB sizes at the initial stage were almost same (approximately 150 µm) under all culture conditions. Although the EB in the HD culture grew drastically with increasing culture time, the growth in the microwell culture was repressed compared with that in HD culture. Furthermore, the changes in EB sizes depended on the scale of microwell, and the EB growth in larger microwells was higher than that in smaller micro‐ wells. Figure 6B shows the gene expression levels of hepatic (AFP), cardiac (αMHC), and vascular (Flk1) differentiation markers. The differentiation fates of EBs were nearly the same under all conditions, but the gene expression levels varied with culture conditions. The expressions of differentiation markers were highest in HD culture, and gradually decreased in 1000, 800, 600, 400 µm arrays in that order. These results indicate that the EB growth and differentiation rate can be controlled by diameters of microwell, and that they are promoted in larger microwell conditions. These differences may be attributed to special culture envi‐ ronments provided by the microwell culture. The existence of micro spaces (microwell) might have facilitated accumulation of various soluble factors which were secreted from the cells. Additionally, interference effects caused by the neighboring EBs could occur because of extensive EBs on an array. Consequently, the accumulation of paracrine/autocrine factors and/ or the concentrations of oxygen and nutrients in the culture medium vary by the array

conditions, and the balances of these factors may regulate the EB properties.

To evaluate the relationship between the inoculated cell density and EB property, the cells at densities of 100, 1000, or 10000 cells/well were inoculated to the array that comprised 195 microwells with 600 µm diameter. Figures 7A and B show the changes in EB sizes and the gene expression levels of differentiation markers, respectively. The change in the EB size of array at 100 cells/well was higher than that of array at 1000 cells/well. In contrast, the cells of array at 10000 cells/well hardly proliferated, and the EB size that formed at the initial stage was maintained throughout the culture period. The expression levels of differentiation markers were the smallest in the array at 10000 cells/well. The expression of vascular (Flk1) differen‐ tiation was highest in the array at 100 cells/well, but the array at 1000 cells/well promoted hepatic (AFP) and cardiac (αMHC) differentiations rather than vascular differentiation, indicating that the inoculated cell density affects the differentiation fate of EBs. These differ‐

EBs was achieved in the single array (Figure 5D).

76 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

**3.4. EB properties in the microwell array culture**

the EB properties of mouse ES cells [75, 76].

in a medium without LIF and inducers.

**Figure 6.** (A) Changes in the EB sizes and (B) the gene expression levels of differentiation markers after 7 days of cul‐ ture.

ences may be caused mainly by the differences in size of EB formed at the initial stage, because the EB size affects the generation of intercellular interactions which trigger the cell differen‐ tiation. This explanation is also supported by the facts that similar results were also observed in the HD culture.

**Figure 7.** (A) Changes in the EB size and (B) the gene expression levels of differentiation markers after 7 days of cul‐ ture.

In conclusion, the microwell array culture could control EB size and allow mass production of homogenous EBs. Furthermore, we demonstrated that differentiation fates of stem cells in the microwell array culture were similar to previous HD culture, and that the architectures of microwell array could control the EB growth and differentiation rates. These characteristics offer advantages over previous methods. Thus, the microwell array may be applicable as a cellular platform that can control the in vitro properties of EB.

## **4. Microfluidic perfusion system in stem cell culture**

The control over microenvironment is important for controlling the stem cell fate, which is affected by various soluble factors supplied by the culture medium and by autocrine and paracrine mechanisms. Microfluidic perfusion enables the control of spatial and temporal profiles of the concentration of soluble factors. In this section, we review the use of the microfluidic perfusion culture system, and recent applications of the microfluidic perfusion culture system in the culture of human ES/iPS cells.

#### **4.1. Microfluidic perfusion culture**

Generally, microfluidic perfusion culture in a microfluidic device is carried out using syringe pumps [77, 78]. However, liquid handling using syringe pumps is cumbersome because it requires connection of many tubes from syringe pumps to the microfluidic device. To address this issue, we developed a pressure-driven perfusion culture system, in which multiple liquids could be handled by simply applying pressure in the liquid reservoir. This is a convenient system to create different culture conditions in a single microfluidic device [79]. We also developed a microfluidic network to generate serial dilution concentration profiles using this pressure-driven perfusion culture system [80].We applied this pressure-driven perfusion culture system to determine IC50 by using the serial dilution microfluidic network [29]. In these studies, we developed the perfusion culture microchamber array chip equipped with 12 perfusion culture microchambers. The culture microchambers were connected to a serial dilution microfluidic network that could generate 12 different stepwise concentration profiles (Figure 8A). We have successfully applied this pressure-driven microfluidic perfusion culture system to a dose-response assay of model anticancer drug, paclitaxel. The obtained IC50 of paclitaxel was similar to that obtained by using traditional microplate assay (Figure 8B).

We have also developed an integrated microfluidic device that contains 384 microchambers in a single device. (Figure 9) [81]. In this device, 12 different drugs were diluted stepwise in the serial dilution microfluidic network into eight different concentrations. Each microchannel for each concentration is connected to four cell culture microchambers (Figure 9B). In total, 384 assays (12 drugs × 8 concentrations ×4 microchambers) could be carried out simultaneously. The culture media with 12 different drugs, the culture media without drug, and cell suspension were loaded in the liquid reservoir using a micropipette (Figure 9A). By applying appropriate pressure, serial dilution concentration profiles spanning 3 orders of magnitude were generated in six dilution steps. Theoretically, IC50 of 12 drugs could be determined in the single-experi‐ ment setup.

In conclusion, the microwell array culture could control EB size and allow mass production of homogenous EBs. Furthermore, we demonstrated that differentiation fates of stem cells in the microwell array culture were similar to previous HD culture, and that the architectures of microwell array could control the EB growth and differentiation rates. These characteristics offer advantages over previous methods. Thus, the microwell array may be applicable as a

The control over microenvironment is important for controlling the stem cell fate, which is affected by various soluble factors supplied by the culture medium and by autocrine and paracrine mechanisms. Microfluidic perfusion enables the control of spatial and temporal profiles of the concentration of soluble factors. In this section, we review the use of the microfluidic perfusion culture system, and recent applications of the microfluidic perfusion

Generally, microfluidic perfusion culture in a microfluidic device is carried out using syringe pumps [77, 78]. However, liquid handling using syringe pumps is cumbersome because it requires connection of many tubes from syringe pumps to the microfluidic device. To address this issue, we developed a pressure-driven perfusion culture system, in which multiple liquids could be handled by simply applying pressure in the liquid reservoir. This is a convenient system to create different culture conditions in a single microfluidic device [79]. We also developed a microfluidic network to generate serial dilution concentration profiles using this pressure-driven perfusion culture system [80].We applied this pressure-driven perfusion culture system to determine IC50 by using the serial dilution microfluidic network [29]. In these studies, we developed the perfusion culture microchamber array chip equipped with 12 perfusion culture microchambers. The culture microchambers were connected to a serial dilution microfluidic network that could generate 12 different stepwise concentration profiles (Figure 8A). We have successfully applied this pressure-driven microfluidic perfusion culture system to a dose-response assay of model anticancer drug, paclitaxel. The obtained IC50 of paclitaxel was similar to that obtained by using traditional microplate assay (Figure 8B).

We have also developed an integrated microfluidic device that contains 384 microchambers in a single device. (Figure 9) [81]. In this device, 12 different drugs were diluted stepwise in the serial dilution microfluidic network into eight different concentrations. Each microchannel for each concentration is connected to four cell culture microchambers (Figure 9B). In total, 384 assays (12 drugs × 8 concentrations ×4 microchambers) could be carried out simultaneously. The culture media with 12 different drugs, the culture media without drug, and cell suspension were loaded in the liquid reservoir using a micropipette (Figure 9A). By applying appropriate pressure, serial dilution concentration profiles spanning 3 orders of magnitude were generated in six dilution steps. Theoretically, IC50 of 12 drugs could be determined in the single-experi‐

cellular platform that can control the in vitro properties of EB.

78 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

**4. Microfluidic perfusion system in stem cell culture**

culture system in the culture of human ES/iPS cells.

**4.1. Microfluidic perfusion culture**

ment setup.

**Figure 8.** Photographs of the perfusion culture microchamber array chip equipped with a serial dilution microfluidic network. (A) Serial dilution microfluidic network and the cell culture microchambers. (B) Fluorometric cell growth measurement for drug cytotoxicity assay. Reprinted with permission from Sugiura et al. [29]. Copyright (2010) Ameri‐ can Chemical Society.

**Figure 9.** Microplate-sized integrated perfusion culture microchamber array chip. (A) Photographs of the integrated perfusion culture microchamber array chip. (B) Enlarged view of a microchamber array unit. Reproduced from Hattori et al. [81] with permission from CBMS.

#### **4.2. Microfluidic perfusion system for human iPS cells under defined culture conditions**

We tried to control the state of human ES/iPS cell cultures by using microfluidics perfusion system and defined the culture conditions [82]. Microfluidics perfusion system enables us to control spatial and temporal application of the soluble factors to the cells. On the other hand, defined culture conditions enable us to control the kinds of factors that are applied to the cells. Thus, the combination of microfluidic perfusion system and defined culture conditions enable us to control microenvironment and replicate stem cells niche in vitro.

Conventional culture methods for human ES/iPS cells use many undefined supplements including liquid additives such as knockout serum replacement (KSR; Life technologies, Grand island, NY, USA) and coating matrix such as Matrigel (BD Biosciences, Mississauga, Canada) [83]. These undefined supplements contain unknown quantity of growth factors, hormones, and integrin receptors, all of which affect human ES/iPS cells states. Thus, they are not suitable for being used to control state of human ES/iPS cells. Therefore, culture systems, which consist of defined supplements including hormones and cytokines and a defined coating matrix such as fibronectin or laminin, have been developed [84–86]. These defined culture systems enabled us to assess the direct responses of added factors without masking by undefined factors and to control differentiation of human ES/iPS cells.

We designed a perfusion culture microchamber array chip that was suitable for ECM coating, cell loading, and human iPS cells perfusion culture based on our previous reports [30, 79]. In this system[82], ECM-coating solution and cell suspensions were loaded in all microchambers through a cell-inlet port (Figure 10A, right) via cell-inlet main channels (Figures 10Band C). The culture media are supplied from a medium-inlet port (Figure 10A, left) via medium-inlet main channels (Figures 10B and C). By applying air pressure to these channels, four different culture conditions can be generated (Figure 10B).

To coat microchambers with purified ECM-coating matrix, we applied two modifications to our system. First modification was the reduction of medium flow resistance at the mediuminlet channel3 and the second was intermittent pressure application for medium perfusion. ECM-coating solution was compulsorily loaded in dried microchambers prior to cell suspen‐ sion loading. Because the flow resistance of the liquid was much higher than that of the air, this extracoating step decreased the cell loading flow-rate at the same pressure compared with our previous system, in which the cells were loaded into dried microchambers without an ECM-coating solution.4 Thus, the medium-inlet channels were modified to be thick to increase the flow rate of cell suspension, resulting in an increase in the medium perfusion flow rate at the same pressure. As a result, intermittent pressure was applied during perfusion culture to keep the average flow rate almost the same as in previous systems. Based on the coating experiment, we chose fibronectin as ECM for coating microchambers.

<sup>3</sup> The medium-inlet branch channel is much more shallow than other microchannels; therefore, the flow rate in each microchamber is determined by the maximum fluidic resistance of the medium-inlet branch channel.

<sup>4</sup> The microchambers were coated unintentionallybecause serum contains ECM such as fibronectin, laminin, and vitronectin.

We found that the growth rate of human iPS cells under pressure-driven perfusion culture conditions was higher than under static culture conditions in the microchamber array. We also applied our new system to self-renewal and differentiation cultures of human iPS cells. Immunocytochemical analysis showed that the state of the human iPS cells was successfully controlled (Figure 9). Moreover, the effects of three antitumor drugs on human iPS cells were comparable between microchamber array and 96-well plates.

We believe that our system will be a platform technology in future for large-scale screening of fully defined conditions for differentiation cultures on integrated microfluidic devices.

**Figure 10.** Structure of the perfusion culture microchamber array chip. (A) Overview of the perfusion culture micro‐ chamber array chip. (B) Enlarged view of the array with four dye solutions. (C) Enlarged view of the microchamber. (D) Undifferentiated and differentiated human iPS cells in the microchamber array chip. The four lanes of the array were perfused with four types of defined culture medium: hESF-9a (including a growth factor to keep the cells in un‐ differentiated state), hESF-6 medium (without growth factor), and hESF-6 medium supplemented with 10 ng/mL BMP4 (+10 BMP) or 50 ng/mL BMP4 (+50 BMP). BMP4 induces early differentiation. Microphotographs were taken at day 3. PhC: the top panels show phase-contrast micrographs. The lower panels show immunocytochemistry of the self-renewal marker, Oct 3/4 (red), the early differentiation marker, SSEA1 (green), nuclear staining with DAPI (blue), and the merged image. The white dotted lines represent the edges of cell culture microchamber. Reproduced from Ya‐ mada et al. [82]with permission from John Wiley and Sons.

## **5. Summary and perspectives**

**4.2. Microfluidic perfusion system for human iPS cells under defined culture conditions**

We tried to control the state of human ES/iPS cell cultures by using microfluidics perfusion system and defined the culture conditions [82]. Microfluidics perfusion system enables us to control spatial and temporal application of the soluble factors to the cells. On the other hand, defined culture conditions enable us to control the kinds of factors that are applied to the cells. Thus, the combination of microfluidic perfusion system and defined culture conditions enable

Conventional culture methods for human ES/iPS cells use many undefined supplements including liquid additives such as knockout serum replacement (KSR; Life technologies, Grand island, NY, USA) and coating matrix such as Matrigel (BD Biosciences, Mississauga, Canada) [83]. These undefined supplements contain unknown quantity of growth factors, hormones, and integrin receptors, all of which affect human ES/iPS cells states. Thus, they are not suitable for being used to control state of human ES/iPS cells. Therefore, culture systems, which consist of defined supplements including hormones and cytokines and a defined coating matrix such as fibronectin or laminin, have been developed [84–86]. These defined culture systems enabled us to assess the direct responses of added factors without masking by undefined factors and

We designed a perfusion culture microchamber array chip that was suitable for ECM coating, cell loading, and human iPS cells perfusion culture based on our previous reports [30, 79]. In this system[82], ECM-coating solution and cell suspensions were loaded in all microchambers through a cell-inlet port (Figure 10A, right) via cell-inlet main channels (Figures 10Band C). The culture media are supplied from a medium-inlet port (Figure 10A, left) via medium-inlet main channels (Figures 10B and C). By applying air pressure to these channels, four different

To coat microchambers with purified ECM-coating matrix, we applied two modifications to our system. First modification was the reduction of medium flow resistance at the medium-

ECM-coating solution was compulsorily loaded in dried microchambers prior to cell suspen‐ sion loading. Because the flow resistance of the liquid was much higher than that of the air, this extracoating step decreased the cell loading flow-rate at the same pressure compared with our previous system, in which the cells were loaded into dried microchambers without an

the flow rate of cell suspension, resulting in an increase in the medium perfusion flow rate at the same pressure. As a result, intermittent pressure was applied during perfusion culture to keep the average flow rate almost the same as in previous systems. Based on the coating

3 The medium-inlet branch channel is much more shallow than other microchannels; therefore, the flow rate in each

4 The microchambers were coated unintentionallybecause serum contains ECM such as fibronectin, laminin, and

experiment, we chose fibronectin as ECM for coating microchambers.

microchamber is determined by the maximum fluidic resistance of the medium-inlet branch channel.

and the second was intermittent pressure application for medium perfusion.

Thus, the medium-inlet channels were modified to be thick to increase

us to control microenvironment and replicate stem cells niche in vitro.

80 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

to control differentiation of human ES/iPS cells.

culture conditions can be generated (Figure 10B).

inlet channel3

vitronectin.

ECM-coating solution.4

In this chapter, we introduced the current state of the art of microfabrication technologies used in stem cell culture. As described above, this technology has already been applied to micro‐ patterning, EB formation, and microfluidic perfusion culture. The microenvironment control‐ led by these microfabrication technologies provided sophisticated culture conditions compared to that of conventional static monolayer culture. As a result, stem cell cultures could be carried out in a controlled manner. So far, some obvious advantages of the use of micro‐ fabrication technology in stem cell culture include control of stem cell fate, easy cultivation method, and reduction in culture volume.

Although methods discussed achieved certain success to control the culture microenviron‐ ment, the use of microfluidics in human ES/iPS cell studies is still limited [77, 87–91]. For example, the induction of fully functional organs and the fabrication of complicated 3D structure mimicking an intact organ is still difficult. Moreover, culture conditions available in current microfluidic technology are basically limited to conventional 2D cultures. Even cultivation of cellular aggregate in microfluidic device is still challenging. We believe that these challenges will be addressed gradually in the future by interdisciplinary approaches including mechanical engineering, material science, device design, and cell biology.

## **Acknowledgements**

This work was supported in part by funding to KO from the Japan Agency for Medical Research and Development. The funding bodies had no role in the study design, data collection or analysis, decision to publish, or preparation of the manuscript.

## **Author details**

Shinji Sugiura1 , Kohji Nakazawa2 , Toshiyuki Kanamori1 and Kiyoshi Ohnuma3\*

\*Address all correspondence to: kohnuma@vos.nagaokaut.ac.jp

1 Biotechnology Research Institute for Drug Discovery, National Institute of Advanced Industrial Science and Technology (AIST), Tsukuba, Ibaraki, Japan

2 Faculty of Environmental Engineering, The University of Kitakyushu, Kitakyushu, Fukuoka, Japan

3 Department of Bioengineering, NagaokaUniversity of Technology, Nagaoka, Niigata, Japan

## **References**

[1] Manz A, Harrison DJ, Verpoorte EMJ, Fettinger JC, Paulus A, Ludi H, et al. Planar chips technology for miniaturization and integration of separation techniques into monitoring systems—Capillary electrophoresis on a chip. Journal of Chromatogra‐ phy. 1992 Feb 28;593(1–2):253–8. PubMed PMID: ISI:A1992HJ35500037.

[2] Harrison DJ, Fluri K, Seiler K, Fan ZH, Effenhauser CS, Manz A. Micromachining a miniaturized capillary electrophoresis-based chemical-analysis system on a chip. Sci‐ ence. 1993 Aug 13;261(5123):895–7. PubMed PMID: ISI:A1993LR89700032.

fabrication technology in stem cell culture include control of stem cell fate, easy cultivation

Although methods discussed achieved certain success to control the culture microenviron‐ ment, the use of microfluidics in human ES/iPS cell studies is still limited [77, 87–91]. For example, the induction of fully functional organs and the fabrication of complicated 3D structure mimicking an intact organ is still difficult. Moreover, culture conditions available in current microfluidic technology are basically limited to conventional 2D cultures. Even cultivation of cellular aggregate in microfluidic device is still challenging. We believe that these challenges will be addressed gradually in the future by interdisciplinary approaches including

This work was supported in part by funding to KO from the Japan Agency for Medical Research and Development. The funding bodies had no role in the study design, data collection

, Toshiyuki Kanamori1

1 Biotechnology Research Institute for Drug Discovery, National Institute of Advanced

2 Faculty of Environmental Engineering, The University of Kitakyushu, Kitakyushu,

3 Department of Bioengineering, NagaokaUniversity of Technology, Nagaoka, Niigata,

phy. 1992 Feb 28;593(1–2):253–8. PubMed PMID: ISI:A1992HJ35500037.

[1] Manz A, Harrison DJ, Verpoorte EMJ, Fettinger JC, Paulus A, Ludi H, et al. Planar chips technology for miniaturization and integration of separation techniques into monitoring systems—Capillary electrophoresis on a chip. Journal of Chromatogra‐

and Kiyoshi Ohnuma3\*

mechanical engineering, material science, device design, and cell biology.

82 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

or analysis, decision to publish, or preparation of the manuscript.

\*Address all correspondence to: kohnuma@vos.nagaokaut.ac.jp

Industrial Science and Technology (AIST), Tsukuba, Ibaraki, Japan

, Kohji Nakazawa2

method, and reduction in culture volume.

**Acknowledgements**

**Author details**

Shinji Sugiura1

Fukuoka, Japan

**References**

Japan


[27] Kang LF, Chung BG, Langer R, Khademhosseini A. Microfluidics for drug discovery and development: from target selection to product lifecycle management. Drug Dis‐ covery Today. 2008 Jan;13(1–2):1–13. PubMed PMID: WOS:000252883000001.

[14] Zorlutuna P, Annabi N, Camci-Unal G, Nikkhah M, Cha JM, Nichol JW, et al. Microfab‐ ricated biomaterials for engineering 3D tissues. Advanced Materials. 2012;24(14):1782–

[15] JensenKF.Microchemical systems: status, challenges, andopportunities.Aiche Journal.

[16] Jeon NL, Baskaran H, Dertinger SKW, Whitesides GM, Van de Water L, Toner M. Neutrophil chemotaxis in linear and complex gradients of interleukin-8 formed in a microfabricated device. Nature Biotechnology. 2002 Aug;20(8):826–30. PubMed PMID:

[17] Lucchetta EM, Lee JH, Fu LA, Patel NH, Ismagilov RF. Dynamics of Drosophila embryonicpatterningnetworkperturbedinspace andtimeusingmicrofluidics.Nature.

[18] Wu MH, Huang SB, Lee GB. Microfluidic cell culture systems for drug research.Lab on

[19] Inamdar NK, Borenstein JT. Microfluidic cell culture models for tissue engineering. Current Opinion in Biotechnology. 2011 Oct;22(5):681–9. PubMed PMID: WOS:

[20] Neuzil P, Giselbrecht S, Lange K, Huang TJ, Manz A. Revisiting lab-on-a-chip tech‐ nology for drug discovery. Nature Reviews Drug Discovery. 2012 Aug;11(8):620–32.

[21] van Midwoud PM, Verpoorte E, Groothuis GMM. Microfluidic devices for in vitro studies on liver drug metabolism and toxicity. Integrative Biology. 2012;3(5):509–21.

[22] van der Meer AD, van den Berg A. Organs-on-chips: breaking the in vitro impasse. Integrative Biology. 2012;4(5):461–70. PubMed PMID: WOS:000303321100001. Eng‐

[23] Hattori K, Munehira Y, Kobayashi H, Satoh T, Sugiura S, Kanamori T. Microfluidic perfusion culture chip providing different strengths of shear stress for analysis of vascular endothelial function. Journal of Bioscience and Bioengineering.2014 ;118(3):

[24] Sakai Y, Hattori K, Yanagawa F, Sugiura S, Kanamori T, Nakazawa K. Detachably as‐ sembled microfluidic device for perfusion culture and post-culture analysis of a

[25] El-Ali J, Sorger PK, Jensen KF. Cells on chips. Nature. 2006 07/27/print;442(7101):403–

[26] Hong J, Edel JB, deMello AJ. Micro- and nanofluidic systems for high-throughput bi‐ ological screening. Drug Discovery Today. 2009 Feb;14(3–4):134–46. PubMed PMID:

2005 Apr 28;434(7037):1134–8. PubMed PMID: ISI:000228693300044. English.

1999 Oct;45(10):2051–4. PubMed PMID: ISI:000083057000002.

84 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

a Chip. 2010;10(8):939–56. PubMed PMID: ISI:000276218900001.

PubMed PMID: WOS:000307101100020. English.

PubMed PMID: WOS:000290170600001. English.

spheroid array. Biotechnology Journal. 2014;9(7):971–9.

804.

lish.

327–32.

11.

ISI:000263638400004.

ISI:000177182500035.

000296114600012. English.


man embryonic stem cells. Biochemical and Biophysical Research Communications. 2008 Oct 10;375(1):27–32. PubMed PMID: 18675790. Epub 2008/08/05.eng.

[52] Rowland TJ, Miller LM, Blaschke AJ, Doss EL, Bonham AJ, Hikita ST, et al. Roles of integrins in human induced pluripotent stem cell growth on matrigel and vitronectin. StemCellsandDevelopment.2009Oct7.PubMedPMID:19811096.Epub2009/10/09.eng.

[39] Takahashi K, Yamanaka S. Induction of pluripotent stem cells from mouse embryon‐ ic and adult fibroblast cultures by defined factors. Cell. 2006 Aug 25;126(4):663–76.

[40] Warmflash A, Sorre B, Etoc F, Siggia ED, Brivanlou AH.A method to recapitulate ear‐ ly embryonic spatial patterning in human embryonic stem cells.Nature Methods.

[41] Nichols J, Smith A. Naive and primed pluripotent states. Cell stem cell. 2009;4(6):

[42] Mrksich M, Whitesides GM. Patterning self-assembled monolayers using microcon‐ tact printing: a new technology for biosensors? Trends Biotech. 1995;13(6):228–35. [43] Park J, Cho CH, Parashurama N, Li Y, Berthiaume F, Toner M, et al. Microfabrica‐ tion-based modulation of embryonic stem cell differentiation. Lab on a Chip.

[44] Gupta K, Kim DH, Ellison D, Smith C, Kundu A, Tuan J, et al. Lab-on-a-chip devices as an emerging platform for stem cell biology. Lab on a Chip. 2010;10(16):2019–31.

[45] Saha K, Mei Y, Reisterer CM, Pyzocha NK, Yang J, Muffat J, et al. Surface-engineered substrates for improved human pluripotent stem cell culture under fully defined conditions. Proceedings of the National Academy of Sciences. 2011;108(46):18714–9.

[46] Paik I, Scurr DJ, Morris B, Hall G, Denning C, Alexander MR, et al. Rapid micropat‐ terning of cell lines and human pluripotent stem cells on elastomeric membranes. Bi‐

[47] Hattori K, Yoshimitsu R, Sugiura S, Maruyama A, Ohnuma K, Kanamori T. Masked plasma oxidation: simple micropatterning of extracellular matrix in a closed micro‐ chamber array. Rsc Advances. 2013;3(39):17749–54. PubMed PMID: WOS:

[48] Yamada R, Hattori K, Tagaya M, Sasaki T, Miyamoto D, Nakazawa K, et al. Plasmapatterned polydimethylsiloxane surface with single-step coating of a mixture of vi‐ tronectin and albumin enables the formation of small discs and spheroids of human

[49] Yamada R, Hattori K, Tachikawa S, Tagaya M, Sasaki T, Sugiura S, et al. Control of adhesion of human induced pluripotent stem cells to plasma-patterned polydime‐ thylsiloxane coated with vitronectin and γ-globulin. Journal of Bioscience and Bioen‐

[50] Barczyk M, Carracedo S, Gullberg D. Integrins. Cell and tissue research. 2010;339(1):

[51] Miyazaki T, Futaki S, Hasegawa K, Kawasaki M, Sanzen N, Hayashi M, et al. Re‐ combinant human laminin isoforms can support the undifferentiated growth of hu‐

PubMed PMID: 16904174. Epub 2006/08/15.eng.

86 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

2014; 11(8): 847-54.

2007;7(8):1018–28.

000325275300022.

269–80.

PubMed PMID: WOS:000280394800001.

iPS. Plasma Medicine. 2014;4(1–4):165–76.

gineering. 2014;118(3):315–22.

otechnology and Bioengineering. 2012;109(10):2630–41.

487–92.


[76] Sakai Y, Yoshiura Y, Nakazawa K. Embryoid body culture of mouse embryonic stem cells using microwell and micropatterned chips. Journal of Bioscience and Bioengin‐ eering. 2011;111(1):85–91.

[64] Bratt‐Leal AM, Carpenedo RL, McDevitt TC. Engineering the embryoid body micro‐ environment to direct embryonic stem cell differentiation.Biotechnology Progress.

[65] Ng ES, Davis RP, Azzola L, Stanley EG, Elefanty AG. Forced aggregation of defined numbers of human embryonic stem cells into embryoid bodies fosters robust, repro‐

[66] Burridge PW, Anderson D, Priddle H, Barbadillo Munoz MD, Chamberlain S, Alle‐ grucci C, et al. Improved human embryonic stem cell embryoid body homogeneity and cardiomyocyte differentiation from a novel V‐96 plate aggregation system high‐

[67] Kinney MA, Sargent CY, McDevitt TC. The multiparametric effects of hydrodynamic environments on stem cell culture. Tissue Engineering Part B: Reviews. 2011;17(4):

[68] Kurosawa H. Methods for inducing embryoid body formation: in vitro differentia‐ tion system of embryonic stem cells. Journal of Bioscience and Bioengineering.

[69] Hsiao C, Palecek SP. Microwell regulation of pluripotent stem cell self-renewal and

[70] Kinney MA, Saeed R, McDevitt TC. Systematic analysis of embryonic stem cell differ‐ entiation in hydrodynamic environments with controlled embryoid body size. Inte‐

[71] Azarin SM, Lian X, Larson EA, Popelka HM, de Pablo JJ, Palecek SP. Modulation of Wnt/β-catenin signaling in human embryonic stem cells using a 3-D microwell array.

[72] Schukur L, Zorlutuna P, Cha JM, Bae H, Khademhosseini A. Directed differentiation of sizecontrolledembryoid bodies towards endothelial and cardiac lineages in RGD‐ modified poly (ethylene glycol) hydrogels. Advanced Healthcare Materials.

[73] Sakai Y, Yoshida S, Yoshiura Y, Mori R, Tamura T, Yahiro K, et al. Effect of microwell chip structure on cell microsphere production of various animal cells. Journal of Bio‐

[74] Sakai Y, Nakazawa K. Technique for the control of spheroid diameter using micro‐

[75] Nakazawa K, Yoshiura Y, Koga H, Sakai Y. Characterization of mouse embryoid bodies cultured on microwell chips with different well sizes. Journal of Bioscience

ducible hematopoietic differentiation. Blood. 2005;106(5):1601–3.

lights interline variability. Stem Cells. 2007;25(4):929–38.

88 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

differentiation. Bionanoscience. 2012;2(4):266–76.

science and Bioengineering. 2010;110(2):223–9.

and Bioengineering. 2013;116(5):628–33.

fabricated chips. ActaBiomaterialia. 2007;3(6):1033–40.

grative Biology. 2012;4(6):641–50.

Biomaterials. 2012;33(7):2041–9.

2009;25(1):43–51.

249–62.

2007;103(5):389–98.

2013;2(1):195–205.


## **Advanced Microfluidic Assays for** *Caenorhabditis elegans*

Natalia A. Bakhtina, Neil MacKinnon and Jan G. Korvink

Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/64283

#### **Abstract**

[88] Titmarsh DM, Hudson JE, Hidalgo A, Elefanty AG, Stanley EG, Wolvetang EJ, et al. Microbioreactor arrays for full factorial screening of exogenous and paracrine factors

[89] Khoury M, Bransky A, Korin N, Konak LC, Enikolopov G, Tzchori I, et al. A micro‐ fluidic traps system supporting prolonged culture of human embryonic stem cells

[90] Villa-Diaz LG, Torisawa YS, Uchida T, Ding J, Nogueira-de-Souza NC, O'Shea KS, et al. Microfluidic culture of single human embryonic stem cell colonies. Lab on a Chip.

[91] Kamei K, Guo S, Yu ZT, Takahashi H, Gschweng E, Suh C, et al. An integrated mi‐ crofluidic culture device for quantitative analysis of human embryonic stem cells.Lab on a Chip. 2009 Feb 21;9(4):555–63. PubMed PMID: 19190791. Epub 2009/02/05.eng.

2009 Jun 21;9(12):1749–55. PubMed PMID: 19495459. Epub 2009/06/06.eng.

in human embryonic stem cell differentiation. PLoS ONE. 2012;7(12):e52405.

aggregates. Biomedical Microdevices. 2010;12(6):1001–8.

90 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

The *in vivo* analysis of a model organism, such as the nematode *Caenorhabditis elegans*, en‐ ables fundamental biomedical studies, including development, genetics, and neurobiolo‐ gy. In recent years, microfluidics technology has emerged as an attractive and enabling tool for the study of the multicellular organism. Advances in the application of microflui‐ dics to *C. elegans* assays facilitate the manipulation of nematodes in high-throughput for‐ mat and allow for the precise spatial and temporal control of their environment. In this chapter, we aim to illustrate the current microfluidic approaches for the investigation of behavior and neurobiology in *C. elegans* and discuss the trends of future development.

**Keywords:** *C. elegans*, chip-based, manipulation, microfluidics, model organism

## **1. Introduction**

The invertebrate *Caenorhabditis elegans*, *Drosophila melanogaster*, and the vertebrate zebrafish (*Danio rerio*) are the most widely studied multicellular organisms. The *in vivo* analysis of these model organisms allows the understanding of many complex physiological processes, addressing many of the questions relevant to human biology. The choice of model organism depends on the biological question under investigation. For example, *C. elegans* is simple enough to be experimentally tractable. It has a short life cycle (3 days at 25 °C) and lifespan (15–17 days at 25 °C), passes through four larva (L1–L4) stages and an adult stage [1]. Its small size (1–1.2 mm long and 80 µm wide), transparent body at all life stages, and preferred food source (*Escherichia coli*) simplify its maintenance on agar plates or liquid cultures allowing visualization of individual cells and organs in intact animals. *C. elegans* possesses one of the simplest central nervous systems (the adult hermaphrodite has 302 neurons). Because it is so well studied, rapid identification of signaling pathways, for instance, in studies of aging, has

© 2016 The Author(s). Licensee InTech. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

become possible. About 40 % of human disease genes have an orthologue in the genome of *C. elegans*, including those genes associated with Alzheimer disease, Parkinson's disease, Huntington's disease (HD), and many other neurodegenerative disorders [1–3]. This aston‐ ishing degree of correspondence permits the modeling of human ailments in a simple inver‐ tebrate without involving actual human subjects and provides a meaningful insight into the pathogenesis of a complex disease phenotype.

Traditionally, behavioral genetics is employed as a prime method for neurobiological studies in *C. elegans*. It is based on manual worm manipulation on a Petri dish or a multiwell plate, and monitoring the effects on various biological processes, such as growth and fertility, by visual inspection. Refreshment of old buffer solutions by a fresh solution is invasive and causes stress both to the larvae and adults. For drug screening, the concentration of the active compounds in the exposing solution might not be precisely controlled because of evaporation and non-selective adsorption effects on the wall of the wells. Permanent immobilization of the worm for further neuronal analysis is performed by means of glue or anesthetics. These methods are time consuming, expensive, tedious, prone to human flaws, and frequently result in failure. To address these problems, novel technologies for the manipulation of multicellular organisms are needed.

Microfluidics has recently been adopted as an instrument both to expand and accelerate progress related to the treatment of human diseases and injuries. Due to precise and automated manipulation of fluids and samples (e.g., single cell, multicellular organism, etc.) in a system of channels (10 –150 µm), a microfluidic-based approach is able to open up aspects that would remain hidden from traditional laboratory techniques. The technology provides a junction between engineering and pure sciences with an immense potential for offering simple and practical solutions. The unique properties of this technology are highlighted by several aspects. First, the dimensions of microfluidic channels perfectly match to the size of samples, allowing precise manipulation. With moving parts, flowing fluids, or other passive mechanisms, microsystems can be used to align samples with a particular orientation with ease as compared to hand-manipulations. Second, the ability to manipulate small amounts of liquid makes it suitable for the precise delivery of small amounts of reagent. Due to the laminar nature of the flow at the micro scale, efficient mass and energy transfer can be controlled in a completely predictable manner (e.g., diffusion of dissolved gases across tens of microns through fluids or polymer membrane materials). Third, based on relatively inexpensive polymer-based fabri‐ cation techniques, such as polydimethylsiloxane (PDMS) replica molding, it has become feasible to realize disposable, economic, and biocompatible systems [4]. Complex structures, adapted to different applications, can be easily fabricated in a short time. Finally, the capability to realize large-scale integration makes it possible to handle a large population of samples in parallel or in series for high -throughput assays. For example, COPAS BIOSORT highthroughput analysis system from Union Biometrica, Inc. (USA), enables the performing of high-speed imaging and offers the possibility of studying a large quantity of individual worms, thereby providing detailed statistical information on the biological variance within the same population.

Substantial advances in microfluidic techniques and particular research interest in *C. elegans* have driven the development of numerous microchip-based systems. They have been re‐ viewed a number of times focusing on various aspects of miniaturized systems, their advan‐ tages, application challenges, and scientific potential [5–19]. A summary of microfluidic-based systems with respect to the organism, organ, or tissue studies was presented by Sivagnanam et al. [5]. All available on-chip approaches for *C. elegans* investigations were systemized by the authors [6]. A classification diagram for structuring of approximately 100 references that simplifies their search according to five evaluated aspects (measured output data, and method for sorting, immobilization, stimulation, and detection of *C. elegans*) is included. In addition, we listed the relevant sorting, immobilization, and imaging methods that have been reported in recent literature, and indicated the main qualitative and quantitative characteristics for each.

become possible. About 40 % of human disease genes have an orthologue in the genome of *C. elegans*, including those genes associated with Alzheimer disease, Parkinson's disease, Huntington's disease (HD), and many other neurodegenerative disorders [1–3]. This aston‐ ishing degree of correspondence permits the modeling of human ailments in a simple inver‐ tebrate without involving actual human subjects and provides a meaningful insight into the

Traditionally, behavioral genetics is employed as a prime method for neurobiological studies in *C. elegans*. It is based on manual worm manipulation on a Petri dish or a multiwell plate, and monitoring the effects on various biological processes, such as growth and fertility, by visual inspection. Refreshment of old buffer solutions by a fresh solution is invasive and causes stress both to the larvae and adults. For drug screening, the concentration of the active compounds in the exposing solution might not be precisely controlled because of evaporation and non-selective adsorption effects on the wall of the wells. Permanent immobilization of the worm for further neuronal analysis is performed by means of glue or anesthetics. These methods are time consuming, expensive, tedious, prone to human flaws, and frequently result in failure. To address these problems, novel technologies for the manipulation of multicellular

Microfluidics has recently been adopted as an instrument both to expand and accelerate progress related to the treatment of human diseases and injuries. Due to precise and automated manipulation of fluids and samples (e.g., single cell, multicellular organism, etc.) in a system of channels (10 –150 µm), a microfluidic-based approach is able to open up aspects that would remain hidden from traditional laboratory techniques. The technology provides a junction between engineering and pure sciences with an immense potential for offering simple and practical solutions. The unique properties of this technology are highlighted by several aspects. First, the dimensions of microfluidic channels perfectly match to the size of samples, allowing precise manipulation. With moving parts, flowing fluids, or other passive mechanisms, microsystems can be used to align samples with a particular orientation with ease as compared to hand-manipulations. Second, the ability to manipulate small amounts of liquid makes it suitable for the precise delivery of small amounts of reagent. Due to the laminar nature of the flow at the micro scale, efficient mass and energy transfer can be controlled in a completely predictable manner (e.g., diffusion of dissolved gases across tens of microns through fluids or polymer membrane materials). Third, based on relatively inexpensive polymer-based fabri‐ cation techniques, such as polydimethylsiloxane (PDMS) replica molding, it has become feasible to realize disposable, economic, and biocompatible systems [4]. Complex structures, adapted to different applications, can be easily fabricated in a short time. Finally, the capability to realize large-scale integration makes it possible to handle a large population of samples in parallel or in series for high -throughput assays. For example, COPAS BIOSORT highthroughput analysis system from Union Biometrica, Inc. (USA), enables the performing of high-speed imaging and offers the possibility of studying a large quantity of individual worms, thereby providing detailed statistical information on the biological variance within the same

pathogenesis of a complex disease phenotype.

92 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

organisms are needed.

population.

This chapter provides a comprehensive overview of recent microfluidic-based approaches for investigations of worm behavior and neurobiology (Figure 1). This includes a discussion on tools and approaches needed to ensure high-throughput manipulation (culturing, sorting, and immobilization) and assaying for behavioral and neuronal studies. In addition, a perspective of novel methods for studies of metabolic activity facilitated by microfluidics is presented.

**Figure 1.** Schematic illustration of a microfluidic platform for *C. elegans* assaying. The structure of this chapter is as follows: in Section 2 we summarize the most frequently utilized imaging (A) and detection (B) methods; in Section 3 we concentrate our attention on different techniques for worm-on-a-chip manipulation, such as culturing, sorting, and immobilization (C, D, and E); Section 4 provides an overview of behavioral and neuronal phenotypes of *C. elegans* facilitated by different analysis techniques, including electrochemical impedance spectroscopy (F), microsurgery (G), and microinjection (H); additionally, Section 5 presents the methods utilized for metabolic activity studies (I).

## **2. Microfluidic approaches for** *C. elegans* **detection**

Conventional optical imaging is an established detection technique for the observation of biological samples (e.g., cells, microorganisms, etc.). Microscopy-based (fluorescent, nonfluor‐ escent, or their various combinations) systems can be used to extract valuable and unique data (e.g., image the activity of specific neurons) from biological samples. Combined with micro‐ fluidics, these systems offer several important advantages required for high-throughput screening [7–10]. Fully automated components, software control, and image processing tools make commercial confocal microscopes extremely versatile for real-time and high -resolution diagnosis. However, conventional optical imaging systems are quite expensive, bulky, and limit the miniaturization of chip-based systems. An overview of different optical imaging approaches in microfluidics (e.g., conventional optical imaging, lensless imaging, etc.) and their applications was recently presented by several research groups [20–22].

To overcome limitations mentioned above, researchers utilize on-chip or lensless imaging technologies. On-chip imaging systems for *C. elegans*, including contact optofluidic imaging [23, 24], direct shadow imaging [25], holographic imaging [26–28], in combination with automated data processing have enabled the observation and characterization of key behav‐ ioral parameters *in vivo* at micrometer and nanometer resolution (Figure 1A and B). Lensless imaging has the advantage of cointegration of microfluidics, microelectronics, and optical components into one platform. This has guaranteed an increase of the image quality, and has provided an ultimate spatial resolution of approximately 0.9 µm and a throughput approach‐ ing 40 worms per minute [23]. The combination with fluorescence imaging holds a great potential for screening of cellular processes [28].

Another promising approach is a lensless and sensor-less monitoring of the nematodes' movement in various microenvironments [29]. In a micro-electro-fluidic (MEF) grid, a moving nematode is detected by change in the electrical impedance at the intersection regions of the microelectrode grid, formed by two identical orthogonally arranged arrays of metal lines (Figure 1B). The approach ensured the real-time readout of the crawling nematode with a spatial resolution of 30 µm (the distance between grid lines) of the reconstructed images at the frequency of 174 Hz per readout.

Usually, the use of fluorescence-based techniques, such as calcium imaging or green fluores‐ cent protein (GFP) expression, and microfluidics to image the activity of specific neurons requires chemically or genetically labeled animals to be immobilized for imaging at a cellular level [30–32]. An "immobilization-free" approach detection is achieved via two pairs of integrated optical fibers. Through the measurements of optical density and fluorescence, the fibers can detect and differentiate wild-type and green fluorescent protein (GFP)-type *C. elegans* even when they flow at high speeds (switching time of 1 s per worm) [33]. This has proven to be a well-controlled method for automated handling of worms in a high-throughput manner with a sorting accuracy of more than 96 %.

## **3. Microfluidic techniques for** *C. elegans* **manipulation**

**2. Microfluidic approaches for** *C. elegans* **detection**

94 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

Conventional optical imaging is an established detection technique for the observation of biological samples (e.g., cells, microorganisms, etc.). Microscopy-based (fluorescent, nonfluor‐ escent, or their various combinations) systems can be used to extract valuable and unique data (e.g., image the activity of specific neurons) from biological samples. Combined with micro‐ fluidics, these systems offer several important advantages required for high-throughput screening [7–10]. Fully automated components, software control, and image processing tools make commercial confocal microscopes extremely versatile for real-time and high -resolution diagnosis. However, conventional optical imaging systems are quite expensive, bulky, and limit the miniaturization of chip-based systems. An overview of different optical imaging approaches in microfluidics (e.g., conventional optical imaging, lensless imaging, etc.) and

To overcome limitations mentioned above, researchers utilize on-chip or lensless imaging technologies. On-chip imaging systems for *C. elegans*, including contact optofluidic imaging [23, 24], direct shadow imaging [25], holographic imaging [26–28], in combination with automated data processing have enabled the observation and characterization of key behav‐ ioral parameters *in vivo* at micrometer and nanometer resolution (Figure 1A and B). Lensless imaging has the advantage of cointegration of microfluidics, microelectronics, and optical components into one platform. This has guaranteed an increase of the image quality, and has provided an ultimate spatial resolution of approximately 0.9 µm and a throughput approach‐ ing 40 worms per minute [23]. The combination with fluorescence imaging holds a great

Another promising approach is a lensless and sensor-less monitoring of the nematodes' movement in various microenvironments [29]. In a micro-electro-fluidic (MEF) grid, a moving nematode is detected by change in the electrical impedance at the intersection regions of the microelectrode grid, formed by two identical orthogonally arranged arrays of metal lines (Figure 1B). The approach ensured the real-time readout of the crawling nematode with a spatial resolution of 30 µm (the distance between grid lines) of the reconstructed images at the

Usually, the use of fluorescence-based techniques, such as calcium imaging or green fluores‐ cent protein (GFP) expression, and microfluidics to image the activity of specific neurons requires chemically or genetically labeled animals to be immobilized for imaging at a cellular level [30–32]. An "immobilization-free" approach detection is achieved via two pairs of integrated optical fibers. Through the measurements of optical density and fluorescence, the fibers can detect and differentiate wild-type and green fluorescent protein (GFP)-type *C. elegans* even when they flow at high speeds (switching time of 1 s per worm) [33]. This has proven to be a well-controlled method for automated handling of worms in a high-throughput

their applications was recently presented by several research groups [20–22].

potential for screening of cellular processes [28].

manner with a sorting accuracy of more than 96 %.

frequency of 174 Hz per readout.

Environmental control and manipulation of whole animal poses significant challenges (e.g., animal's body orientation, precise delivery of chemicals, etc.). Transferring traditional neurobiology and behavioral investigation techniques to the microfluidic platform has the potential to overcome these challenges. This is driven by substantial progress in integration of functional components (e.g., valves, detectors, etc.) that allow the monitoring of various steps, such as administration, distribution, metabolism, and toxicity during drug screening. The advanced microfluidic approach offers both qualitative and quantitative data from a single organism by automatic high-throughput manipulation. For example, the worms can be oriented at regular positions on a substrate due to hydrodynamic forces in a microfluidic chip for the determination of gene function in a high-throughput manner [34]. In this section, we discuss general manipulation techniques, such as culturing, sorting, and immobilization.

**Culturing.** To interpret the underlying metabolic changes and specific developmental processes during nematode ageing, longitudinal experiments over the entire lifespan are necessary. Imaging and monitoring of the embryogenesis require specific techniques, which include single embryo isolation and mounting. Worm culturing can be dramatically improved using an automated microfluidic platform for culturing, phenotyping, and long-term live imaging of *C. elegans* embryo and larvae using microfluidic chambers (Figure 2A) and droplet encapsulation (Figure 2B) [35–47]. In combination with image recognition algorithms these "worm-chips" have successfully demonstrated their high potential at enhancing worm handling (e.g., automatic nutrient and waste exchange), accurate imaging, and automated analysis of embryonic morphogenesis during embryonic development [48]. Requiring the loading of only a few adult worms into the chip, the ensuing *C. elegans* embryo population could be processed at the same time.

**Figure 2.** Schematic illustration of the microfluidic platforms for *C. elegans* culturing (A) [35] and sorting based on droplet encapsulation (B) [42–44] or electrotaxis (C) [49].

**Sorting.***C. elegans* exhibits age-dependent specific neuron and behavioral responses. For instance, usage of both the early-stage and adult worms may increases the physiological relevance of drug candidates during the identification process and reveals potential toxic effects. Therefore, sorting (age or size synchronization) of worm populations or individuals is often required for further diagnostics. For example, a passive sorting method is based on selfregulated worm distribution and loading into an array of narrowing channels [36, 50, 51] or microchambers [36, 37, 40, 52–55] with an average loading effectiveness rate of approximately 65 %. Once the worm enters the microchannel, the hydrodynamic resistance increases dramatically, thereby locking a single worm inside the chamber.

When considering high-throughput manipulation, automatic classification of worms (e.g., wild-type from mutants) becomes of high relevance. Typically, sorting involves individual *C. elegans* loading and separation, for instance, according to genetic phenotype for downstream analysis [31, 32, 34, 36, 56–60]. Together with real-time rapid image extraction and data processing, media flow in the microfluidic channel is driven by a syringe and is controlled by on-chip functional components, such as PDMS valves. Automatic sample positioning can guarantee rapid classification based on synaptic characteristics with sorting throughput at a rate of 900 worms per hour and an overall sorting accuracy of 96.5 % [32]. Depending on the extracted data, the worm could be flushed to either a waste or a sorting outlet by valve actuation.

Several other techniques have been successfully implemented in high-throughput studies [61– 63]. In these systems, sorting is accomplished based on size difference in a passive, but extremely high throughput (up to 1200 worms per min) and selective manner (94 % of adults with 0.2 % larva contamination) [62].The device body contains an array of microstructured post (or filters) and a network of microfluidic channels allowing a large population of adult worms and larvae to be oriented in the desired direction.

**Immobilization.** Because of the high mobility, *C. elegans* immobilization is required for further developmental studies (e.g., neurosurgery). Manual handling and immobilization (e.g., by gluing or anesthesia) suffer from low throughput and is limited in long-term monitoring [1]. Immobilization in a microfluidic channel provides a simple well-controlled mechanism for automated handling of worms in a high-throughput manner. One of the widely used techni‐ ques is based on mechanical force. Several immobilization techniques are used to lock an individual worm against a microchannel wall in a robust and reversible manner. The first method involves microarrays of fixed-geometry clamps for *C. elegans* immobilization, which is a simple way to restrict the motion employing a single PDMS layer [36, 51, 64–67]. Using a constant pressure difference between the inlet and outlet of the device to drive fluid flow, an array of up to 128 wedge-shaped microchannels can be filled by nematodes with up to 90 % efficiency [68]. The second method includes a flexible PDMS membrane for squeezing the worm into the side of microfluidic channel under an external pressure through a control layer above the main chamber [51, 57–59, 67, 69, 70]. A third method is to trap of *C. elegans* by suction flow (Figure 1E), which is based on a vacuum-assisted restraint that aligns the worm along its axis [31, 71]. To highlight the utility of both immobilization techniques several research groups have combined suction posts with either flexible membranes [56, 65, 72, 73] or microchannel narrowing [30, 74–77] for extremely stable immobilization during microsurgery (Figure 1G).

Microfluidic devices offer advantages for both spatial and temporal control of the animal's position and microenvironment at the microscale. Based on acoustic wave in a single-layer microfluidic chip, on-chip manipulation technique permitted trapping and rotational manip‐ ulation *C. elegans* regardless of shape and physical properties in the x- or y-directions for extended periods of time without inducing physiological damage [78, 79]. By implementation of a cooling liquid supply through a control layer to lower a worm's temperature down to 4 °C, *C. elegans* can be immobilized with a throughput up to 400 worms per hour for short-term cooling (~2 s) [32, 52]. Alternatively, light-induced sublethal heat can be used to increase the worm temperature (up to 31–37 °C) for its immobilization [80].

often required for further diagnostics. For example, a passive sorting method is based on selfregulated worm distribution and loading into an array of narrowing channels [36, 50, 51] or microchambers [36, 37, 40, 52–55] with an average loading effectiveness rate of approximately 65 %. Once the worm enters the microchannel, the hydrodynamic resistance increases

When considering high-throughput manipulation, automatic classification of worms (e.g., wild-type from mutants) becomes of high relevance. Typically, sorting involves individual *C. elegans* loading and separation, for instance, according to genetic phenotype for downstream analysis [31, 32, 34, 36, 56–60]. Together with real-time rapid image extraction and data processing, media flow in the microfluidic channel is driven by a syringe and is controlled by on-chip functional components, such as PDMS valves. Automatic sample positioning can guarantee rapid classification based on synaptic characteristics with sorting throughput at a rate of 900 worms per hour and an overall sorting accuracy of 96.5 % [32]. Depending on the extracted data, the worm could be flushed to either a waste or a sorting outlet by valve

Several other techniques have been successfully implemented in high-throughput studies [61– 63]. In these systems, sorting is accomplished based on size difference in a passive, but extremely high throughput (up to 1200 worms per min) and selective manner (94 % of adults with 0.2 % larva contamination) [62].The device body contains an array of microstructured post (or filters) and a network of microfluidic channels allowing a large population of adult

**Immobilization.** Because of the high mobility, *C. elegans* immobilization is required for further developmental studies (e.g., neurosurgery). Manual handling and immobilization (e.g., by gluing or anesthesia) suffer from low throughput and is limited in long-term monitoring [1]. Immobilization in a microfluidic channel provides a simple well-controlled mechanism for automated handling of worms in a high-throughput manner. One of the widely used techni‐ ques is based on mechanical force. Several immobilization techniques are used to lock an individual worm against a microchannel wall in a robust and reversible manner. The first method involves microarrays of fixed-geometry clamps for *C. elegans* immobilization, which is a simple way to restrict the motion employing a single PDMS layer [36, 51, 64–67]. Using a constant pressure difference between the inlet and outlet of the device to drive fluid flow, an array of up to 128 wedge-shaped microchannels can be filled by nematodes with up to 90 % efficiency [68]. The second method includes a flexible PDMS membrane for squeezing the worm into the side of microfluidic channel under an external pressure through a control layer above the main chamber [51, 57–59, 67, 69, 70]. A third method is to trap of *C. elegans* by suction flow (Figure 1E), which is based on a vacuum-assisted restraint that aligns the worm along its axis [31, 71]. To highlight the utility of both immobilization techniques several research groups have combined suction posts with either flexible membranes [56, 65, 72, 73] or microchannel narrowing [30, 74–77] for extremely stable immobilization during microsurgery (Figure 1G). Microfluidic devices offer advantages for both spatial and temporal control of the animal's position and microenvironment at the microscale. Based on acoustic wave in a single-layer microfluidic chip, on-chip manipulation technique permitted trapping and rotational manip‐

dramatically, thereby locking a single worm inside the chamber.

96 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

worms and larvae to be oriented in the desired direction.

actuation.

Gases, including carbon dioxide (CO2) and nitrogen (N2), are sensed by *C. elegans* and serve as a partial and complete method to eliminate worm mobility [67, 69, 76]. By passing pure gas from a control layer into a flow layer microenvironment, *C. elegans* can be immobilized in a channel with improved sensitivity and increased resolution.

## **4. Microfluidic approach for** *C. elegans* **sensing function and behavior**

In this section, we review the use of microfluidic chips for *C. elegans* investigations under controlled physical and chemical conditions that have been advantageously used, for example, as integrated biosensors for toxicological experiments and drug screening. The two major methods of assaying are behavioral and molecular (or neuronal) studies. In behavioral studies, discussed in Section 4.1, the movement is generally analyzed by observation the animal's behavior in response to stimulation (e.g., touch, drugs, odorants, food, temperature, gases, osmolytes, or light). The key issues in neuronal studies, covered by Section 4.2, are the intracellular processes and neuronal signaling.

#### **4.1. Behavioral studies of** *C. elegans***'s responses to different stimuli**

*C. elegans* explores its surrounding environment and moves according to environmental stimuli, including temperature, chemical, electric field, and light, which are detected by 24 sensilla organs and various isolated sensor neurons [1]. Obtaining meaningful data about the mechanism of environmental sensing requires strict control over the experimental conditions. Moreover, when a high number of identical biological samples are needed to be screened, a common need and challenge of the experimental procedure is the precise manipulation of worms with an emphasis on high throughput. Microfluidics offers a straightforward solution for automation and parallelization of screening in a rapid, sensitive, and accurate manner.

The environmental cues can be applied by devices, embedded in a chip-based microfluidic system, to analyze the behavioral response of the microorganism. For example, active and automated local manipulation and chemical stimulation of the individual worms can be achieved by implementation of multilayer PDMS layers. Because of *C. elegans'* small size and its ability to grow in liquid, on-chip imaging systems and automated data processing facilitate the observation and characterization of key behavioral parameters *in vivo* with micrometer and nanometer resolution.

**Locomotion.** Owing to the precision achievable by microlithographic techniques, researchers have been able to investigate in-depth different locomotion patterns (by varying the size and

spacing of posts), applied muscular forces (by measuring the deflection of posts) (Figure 1D) and motility quantification (time-averaged kinetic power over the swimming cycle) [81–90]. In contrast to traditional experimental techniques, these systems permitted straightforward dynamic force detection of moving nematodes. Whereas the motion of the animals in the artificial soil device exhibited the same principal characteristics of the motion of crawling on agar, the constraints on motion caused by the posts better mimicked the complexity of *C. elegans'* natural environment [91]. To further study crawling behavior, a number of PDMS microfluidic devices were presented that consisted of sinusoidal channels of varying wave‐ lengths [84, 86, 87].

These devices allow researchers to manipulate the oscillating body motion of the crawling animals and investigate the biophysical and neuronal mechanisms of locomotion and pro‐ prioception. Microfluidics facilitates precise environmental control that was demonstrated by modifying the chemicals' concentration of the main chamber rapidly or and immediately observing the effect on locomotion [83]. Obviously, tracking animals through such a rapid media exchange would not be possible in a larger environment.

**Electrotaxis.***C. elegans* exhibits responsive behavior to electric fields, mediated by certain amphid sensory neurons. These neurons are sensitive to both the direction and strength of the electrical signal, and forced the animal to move toward a negatively charged pole [1]. Analysis of the nematode's electrotaxis provides a detailed model of how neurons function together to generate a behavioral response to electric fields. When microfluidic chambers are combined with electrodes to deliver electrical stimuli, both behavioral and neuronal screening can be performed, providing the chance to elucidate potential treatment for human muscular disorders. Many microfluidic systems were proposed for fully automated control of electro‐ taxis, which overcame many of the inherent problems of manual operation [49, 92–98].

Normally, worms are exposed to a uniform electric field generated by two electrodes (e.g., platinum wires) embedded in inlet and outlet reservoirs and connected to external electrical drive circuitry (Figure 2C). Exposure to direct (DC), alternating (AC), and pulsed DC electric fields in a specified range of strengths has been employed as a means of guiding nematodes in a binary manner (e.g., start and stop), for sorting, and for immobilization, aiming to provide a close look at the mechanism of neuronal signaling transduced into behavioral responses [86, 92, 93, 98–100]. Such movement-based microfluidic devices permit the differentiation of worms according to locomotive abilities and similar physiological states, for instance, to distinguish adults from larva, or healthy worms from uncoordinated, and to locate individuals defective in electric field sensing. This guiding technique allows high throughput (up to 60 worms per min) and method selectivity of 70–90 %.

The progress achieved in microfabrication technologies has made monolithic integration of electrodes into microfluidic platform possible (Figure 1F). Micropatterned electrodes on the sidewalls of microfluidic channels (i.e., without blocking optical visibility) provides a simple means of creating electrofluidic glass chips to flexibly control the movement of *C. elegans* in a sensitive and reproducible manner [101]. Placing the microelectrodes inside the microfluidic environment as close to the animal as possible allows one to create transient pores in the cell membrane, which permits the diffusion of extracellular compounds that are present in the vicinity of the pore into the interior of the cell [102]. All of these results demonstrate the potential of using active microfluidic devices as an alternative to Petri dishes for *C. elegans* assays.

**Chemicals.** Microfluidics is particularly attractive for many applications where *C. elegans* are used as integrated biosensors for toxicological experiments and drug screening. Behavioral investigations in response to chemical stimuli include real-time locomotion diagnostics of *C. elegans*. The effect on worm physiology to a variety of anesthetics, such as tricaine, muscimol, sodium azide, and levamisole [29, 55, 69, 93], odors, such as hermaphrodite-conditioned media and nicotine, and odors produced by pathogenic bacteria [55, 103–108], chemicals, such as zinc ion (Zn2+) and glucose [109], different osmolarity levels [66, 71], was successfully examined by precise chemical control in a time- and dose-dependent manner. In most cases, pre- and postexposure locomotion phenotypes are compared by a variety of parameters (e.g., average velocity, individual head swing orientations, etc.).

**Other stimuli.** The ability of integration and individual worm manipulation makes micro‐ fluidic devices attractive platforms for understanding the correlation between *C. elegans*' neuronal and behavioral responses. Based on the properties of a microfluidic device, temper‐ ature stimuli could be delivered to individual worms accurately by flexible chip design and fluidic manipulation. Behavioral mechanisms in response to temperature change is quantified in terms of an average head angle of a semi-restrained animal [74] or swimming movements of the individuals heated in a microdroplet array [107].

Although the *C. elegans* has no light-sensing organs, it modulates a response to light known as phototaxis [50, 92, 109]. To analyze *C. elegans'* sensitivity to light, wild-type and mutant nematodes are illuminated with light and their behavioral response are examined. It was experimentally demonstrated that illumination to green light is preferable for animals, while blue light triggers muscle depolarization and further body contraction.

#### **4.2. Neuronal studies**

spacing of posts), applied muscular forces (by measuring the deflection of posts) (Figure 1D) and motility quantification (time-averaged kinetic power over the swimming cycle) [81–90]. In contrast to traditional experimental techniques, these systems permitted straightforward dynamic force detection of moving nematodes. Whereas the motion of the animals in the artificial soil device exhibited the same principal characteristics of the motion of crawling on agar, the constraints on motion caused by the posts better mimicked the complexity of *C. elegans'* natural environment [91]. To further study crawling behavior, a number of PDMS microfluidic devices were presented that consisted of sinusoidal channels of varying wave‐

These devices allow researchers to manipulate the oscillating body motion of the crawling animals and investigate the biophysical and neuronal mechanisms of locomotion and pro‐ prioception. Microfluidics facilitates precise environmental control that was demonstrated by modifying the chemicals' concentration of the main chamber rapidly or and immediately observing the effect on locomotion [83]. Obviously, tracking animals through such a rapid

**Electrotaxis.***C. elegans* exhibits responsive behavior to electric fields, mediated by certain amphid sensory neurons. These neurons are sensitive to both the direction and strength of the electrical signal, and forced the animal to move toward a negatively charged pole [1]. Analysis of the nematode's electrotaxis provides a detailed model of how neurons function together to generate a behavioral response to electric fields. When microfluidic chambers are combined with electrodes to deliver electrical stimuli, both behavioral and neuronal screening can be performed, providing the chance to elucidate potential treatment for human muscular disorders. Many microfluidic systems were proposed for fully automated control of electro‐ taxis, which overcame many of the inherent problems of manual operation [49, 92–98].

Normally, worms are exposed to a uniform electric field generated by two electrodes (e.g., platinum wires) embedded in inlet and outlet reservoirs and connected to external electrical drive circuitry (Figure 2C). Exposure to direct (DC), alternating (AC), and pulsed DC electric fields in a specified range of strengths has been employed as a means of guiding nematodes in a binary manner (e.g., start and stop), for sorting, and for immobilization, aiming to provide a close look at the mechanism of neuronal signaling transduced into behavioral responses [86, 92, 93, 98–100]. Such movement-based microfluidic devices permit the differentiation of worms according to locomotive abilities and similar physiological states, for instance, to distinguish adults from larva, or healthy worms from uncoordinated, and to locate individuals defective in electric field sensing. This guiding technique allows high throughput (up to 60 worms per

The progress achieved in microfabrication technologies has made monolithic integration of electrodes into microfluidic platform possible (Figure 1F). Micropatterned electrodes on the sidewalls of microfluidic channels (i.e., without blocking optical visibility) provides a simple means of creating electrofluidic glass chips to flexibly control the movement of *C. elegans* in a sensitive and reproducible manner [101]. Placing the microelectrodes inside the microfluidic environment as close to the animal as possible allows one to create transient pores in the cell membrane, which permits the diffusion of extracellular compounds that are present in the

media exchange would not be possible in a larger environment.

98 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

min) and method selectivity of 70–90 %.

lengths [84, 86, 87].

Behavioral studies, such as physiological responses, in a whole organism population include not only movement-based analyses but also monitoring of the *C. elegans'* neuronal activity in a confined space. Coupled with microfluidic-based systems, existing neuronal recording techniques (e.g., by calcium imaging or green fluorescent protein (GFP) expression) examine neuronal responses to sensory inputs of a single animal at a time under precise environmental control. For example, it was found that immobilizing a portion of the worm can directly override rhythmic activity and may cause changes in transport parameters of the touch neuron [69, 110]. In order to explore locomotive behavior and the underlying molecular mechanism, Wang et al. monitored a subcellular distribution of the DAF-16 gene that regulates different stress responses [91]. The experiments showed an increase of DAF-16 nuclear localization, attributed to crowding stress, in a microcolumn array with intervals from 40 to 200 µm between microposts (Figure 1D). As a result, a system-level understanding of the worm's motor circuit can be obtained.

One application where microfluidics and fluorescent-based imaging open up aspects that would remain hidden from traditional laboratory techniques is drug screening. *C. elegans* can be an effective test-bed for a wide range of water-soluble chemical compounds (e.g., glycerol [30, 66, 74, 75], anticancer drugs [48], heavy metals [54], sodium chloride NaCl [58, 65, 71, 83, 111, 112], copper(II) chloride CuCl2 [66, 74], levamisole [70], manganese [102], antibiotics [104], isoamyl alcohol [113], cyanide [114], etc.). Microfluidic network manipulation allows the automation in a high-throughput manner and under reproducible experimental conditions while analysis of the nematode's chemosensitivity provides a detailed model of how neurons function together to generate behavioral response. For example, neurotransmitters and hormones, such as 1-methyl-4-phenylpyridinium (MPP+), 6-hydroxy dopamine (6-OHDA), and rotenone, have widespread effects as chemical regulators for coordinating physiological activity throughout the body of both nematodes and humans [1]. The microfluidic-based experiments proved that MPP +, 6-OHDA, and rotenone induce mobility defects in the animal (i.e., significant reduction in speed) after treatment and was potentially neurotoxic for dopaminergic neurons [43, 51, 112].

Due to PDMS microfluidic devices, much progress has been made to overcome the limitations of precise chemical control. The effect of ageing on physiological properties of the ASH chemosensory neuron can be characterized and quantified by the direct delivery of a chemical odor to the nose of *C. elegans* [30]. To emphasize the influence of different anesthetics on subcellular activity, a microfluidic platform was used for studying the contribution of vesicle transport to synaptic growth [70]. As a result, imaging of subcellular processes, such as presynaptic vesicle transport, intraflagellar transport (IFT), dendritic transport, and migration of neuroblasts during early developmental stages of the nematode, has become feasible. Moni‐ toring of neuron activity (e.g., ASH neurons) with respect to osmotic gradient, can access the pattern-generating activity (e.g., individual head swing orientations) of the chemosensory circuit [66].

Another field where polymer-based fabrication techniques have already demonstrated themselves, is in investigations of gas sensing in nematodes [69, 76, 115–117]. In order to understand how oxygen level variation causes behavioral and physiological changes, freely moving adult animals were subjected to a gas-phase oxygen gradient. Experiments showed that specific soluble guanylate cyclase homologues (GCY-31, GCY-33, GCY-35, and GCY-36), located in URX, AQR, and PQR sensory neurons, activate hypoxia or hyperoxia avoidance [115, 116].

For many applications, such as characterizing stochastic neural responses, it should be beneficial to increase experimental throughput at the expense of image resolution. Microflui‐ dics promotes simultaneous recording of calcium transients in individual neurons from multiple animals (up to 20), and increases experimental throughput [82, 118]. Thus, a system‐ atic characterization of chemosensory neuron responses to multiple odors, odor concentra‐ tions, and temporal patterns, as well as responses to pharmacological manipulation can be performed.

The described experiments benefit enormously from the use of microfluidic technologies. The precise handling and chemical mixing of chemicals and neurotoxins in nanoliter volume droplets tremendously decreases reagent consumption and reaction time. The combination of brightfield imaging, fluorescent imaging, and microfluidics allows *in vivo* observation of biomolecules and automated analysis of protein aggregation phenomena in *C. elegans* for amyotrophic lateral sclerosis (ALS) at unprecedented resolution [119]. The level of precision that researchers have already achieved demonstrates the potential for the dissection of neuronal function and toxin-induced neurodegeneration *in vivo*.

#### *4.2.1. Intracellular studying techniques*

One application where microfluidics and fluorescent-based imaging open up aspects that would remain hidden from traditional laboratory techniques is drug screening. *C. elegans* can be an effective test-bed for a wide range of water-soluble chemical compounds (e.g., glycerol [30, 66, 74, 75], anticancer drugs [48], heavy metals [54], sodium chloride NaCl [58, 65, 71, 83, 111, 112], copper(II) chloride CuCl2 [66, 74], levamisole [70], manganese [102], antibiotics [104], isoamyl alcohol [113], cyanide [114], etc.). Microfluidic network manipulation allows the automation in a high-throughput manner and under reproducible experimental conditions while analysis of the nematode's chemosensitivity provides a detailed model of how neurons function together to generate behavioral response. For example, neurotransmitters and hormones, such as 1-methyl-4-phenylpyridinium (MPP+), 6-hydroxy dopamine (6-OHDA), and rotenone, have widespread effects as chemical regulators for coordinating physiological activity throughout the body of both nematodes and humans [1]. The microfluidic-based experiments proved that MPP +, 6-OHDA, and rotenone induce mobility defects in the animal (i.e., significant reduction in speed) after treatment and was potentially neurotoxic for

100 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

Due to PDMS microfluidic devices, much progress has been made to overcome the limitations of precise chemical control. The effect of ageing on physiological properties of the ASH chemosensory neuron can be characterized and quantified by the direct delivery of a chemical odor to the nose of *C. elegans* [30]. To emphasize the influence of different anesthetics on subcellular activity, a microfluidic platform was used for studying the contribution of vesicle transport to synaptic growth [70]. As a result, imaging of subcellular processes, such as presynaptic vesicle transport, intraflagellar transport (IFT), dendritic transport, and migration of neuroblasts during early developmental stages of the nematode, has become feasible. Moni‐ toring of neuron activity (e.g., ASH neurons) with respect to osmotic gradient, can access the pattern-generating activity (e.g., individual head swing orientations) of the chemosensory

Another field where polymer-based fabrication techniques have already demonstrated themselves, is in investigations of gas sensing in nematodes [69, 76, 115–117]. In order to understand how oxygen level variation causes behavioral and physiological changes, freely moving adult animals were subjected to a gas-phase oxygen gradient. Experiments showed that specific soluble guanylate cyclase homologues (GCY-31, GCY-33, GCY-35, and GCY-36), located in URX, AQR, and PQR sensory neurons, activate hypoxia or hyperoxia avoidance

For many applications, such as characterizing stochastic neural responses, it should be beneficial to increase experimental throughput at the expense of image resolution. Microflui‐ dics promotes simultaneous recording of calcium transients in individual neurons from multiple animals (up to 20), and increases experimental throughput [82, 118]. Thus, a system‐ atic characterization of chemosensory neuron responses to multiple odors, odor concentra‐ tions, and temporal patterns, as well as responses to pharmacological manipulation can be

dopaminergic neurons [43, 51, 112].

circuit [66].

[115, 116].

performed.

Several other techniques for studying and characterization of intracellular processes, including dielectrophoresis and electrochemical impedance spectroscopy, have been adopted by researchers for whole-animal drug screening (Figure 1F) [77, 92, 120, 121]. These methods use a noninvasive electrophysiological readout of neuromuscular function and can provide highquality neurogenetic and neuropharmacological data on nematodes. Automatic real-time monitoring and parallelization (up to 8 worms simultaneously) with throughput of up to 12 worms per hour facilitate the rapid neuroactive drug screening, e.g., effects of drugs on neurons, as well as on muscles [77, 121].

### *4.2.2. Microsurgery and microinjection*

In combination with microfluidics and optical image analysis systems, microsurgery and microinjection are employed for *in vivo* neuronal regeneration and cell-to-cell communication studies [52, 73, 80, 122–127]. Because the nervous system is described in great detail, the role of an individual neuron can be directly studied with laser ablation experiments (Figure 1G). Using a laser nanobeam in the UV wavelength region, fluorescent-labeled whole cell ablation is possible and a single synapse removal can be achieved [124–127]. Afterward, the resulting phenotypes (e.g., degeneration and regeneration) can be examined *in vivo*. Advances in optical imaging and microfluidic methods support this procedure. The capabilities of current chipbased systems are sufficient to perform precise animal manipulations, required for high immobilization stability of the worm, and complex image-based assaying with high through‐ put (up to 200 animals per hour with a success rate of 89 %) [52, 122]. This provides approxi‐ mately one order of magnitude improvement over manually performed axotomies (when considering study of a single population) and gives an opportunity to perform genetic screening in a reasonable timeframe to identify the molecular mechanisms involved in nerve regeneration and degeneration.

The *in vivo* injection of chemical materials that have significant implications in genetics, drug discovery, and other biological applications is another way to study the mechanisms under‐ lying intercellular communication in *C. elegans* (Figure 1H). Using a single needle tip of the micromanipulator, localized chemical stimulation can be delivered to a single intestinal cell of the immobilized worms [123, 128].

## **5. Conclusions and perspectives**

The advances in microfabrication technologies have demonstrated the potential of using active lab-on-a-chip (LoC) devices as an alternative to microwell plates for worm-based assays. LoC technology offers a straightforward solution to all of the problems during manual manipula‐ tion. Complex three-dimensional (3D) microenvironments have been created, where a whole population of worms is cultured and analyzed in a reproducible way. Currently available microfluidic-based systems are capable of recording from sensory neurons in animals *in vivo*, whose neuronal responses could be correlated with behavior. Microsurgery and microinjection allow the investigation of many processes, including the role of individual neurons in neuronal networks and in cell-to-cell interaction. Obviously, this is pushing forward fundamental studies in biology and biochemistry.

The use of fluorescence-based techniques and microfluidics to image the activity of specific neurons requires that animals be labeled either chemically or genetically. However, for monitoring certain biological processes, fluorescent labeling might be inconvenient or may interfere with normal behavior. Moreover, many dynamic phenomena of motile samples might be missed during impedance spectroscopy, microsurgery, and microinjection because of the long-term immobilization required for subcellular-level stabilization of *C. elegans*. This makes monitoring of actual metabolic activity impossible.

Several other approaches can be used to study the neuronal and metabolic activity of a biological system. For example, nuclear magnetic resonance imaging (MRI) and nuclear magnetic resonance spectroscopy (NMR) are two of the most information-rich methods that provide a unique opportunity to link morphological, functional, and chemically specific spectroscopic information from small volume (e.g., µl) samples (Figure 1I). MRI and NMR uses strong time-varying radio frequency (RF) fields to generate a weak specific RF response from a certain tissue type [129]. Because the technology is noninvasive and only nonionizing radiation is absorbed and emitted, it might be especially suitable for the study of *C. elegans* in the identification and quantification of metabolites (intermediate products of metabolism) within the metabolic pathway *in vivo* [130–136].

In many of the reviewed research articles, the easy integration of microfluidic control and detection modules was a key factor in helping to link *in vitro* and *ex vivo* experimental investigations. The ability of *C. elegans* tracking in real time (i.e., with minimal latency) for further diagnostic applications could provide a close look at the cellular, molecular, and genetic levels. Consequently, an understanding of the underlying molecular mechanisms in multicel‐ lular model organisms would provide a unique opportunity to unthread analogous and complex biological processes in humans. This certainly will promote more automated and higher throughput applications in the future.

## **Acknowledgements**

We gratefully acknowledge financial support from the European Research Council (ERC) (contract number 290586 from 1.07.2012), which funded this work.

## **Author details**

**5. Conclusions and perspectives**

102 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

studies in biology and biochemistry.

makes monitoring of actual metabolic activity impossible.

within the metabolic pathway *in vivo* [130–136].

higher throughput applications in the future.

**Acknowledgements**

The advances in microfabrication technologies have demonstrated the potential of using active lab-on-a-chip (LoC) devices as an alternative to microwell plates for worm-based assays. LoC technology offers a straightforward solution to all of the problems during manual manipula‐ tion. Complex three-dimensional (3D) microenvironments have been created, where a whole population of worms is cultured and analyzed in a reproducible way. Currently available microfluidic-based systems are capable of recording from sensory neurons in animals *in vivo*, whose neuronal responses could be correlated with behavior. Microsurgery and microinjection allow the investigation of many processes, including the role of individual neurons in neuronal networks and in cell-to-cell interaction. Obviously, this is pushing forward fundamental

The use of fluorescence-based techniques and microfluidics to image the activity of specific neurons requires that animals be labeled either chemically or genetically. However, for monitoring certain biological processes, fluorescent labeling might be inconvenient or may interfere with normal behavior. Moreover, many dynamic phenomena of motile samples might be missed during impedance spectroscopy, microsurgery, and microinjection because of the long-term immobilization required for subcellular-level stabilization of *C. elegans*. This

Several other approaches can be used to study the neuronal and metabolic activity of a biological system. For example, nuclear magnetic resonance imaging (MRI) and nuclear magnetic resonance spectroscopy (NMR) are two of the most information-rich methods that provide a unique opportunity to link morphological, functional, and chemically specific spectroscopic information from small volume (e.g., µl) samples (Figure 1I). MRI and NMR uses strong time-varying radio frequency (RF) fields to generate a weak specific RF response from a certain tissue type [129]. Because the technology is noninvasive and only nonionizing radiation is absorbed and emitted, it might be especially suitable for the study of *C. elegans* in the identification and quantification of metabolites (intermediate products of metabolism)

In many of the reviewed research articles, the easy integration of microfluidic control and detection modules was a key factor in helping to link *in vitro* and *ex vivo* experimental investigations. The ability of *C. elegans* tracking in real time (i.e., with minimal latency) for further diagnostic applications could provide a close look at the cellular, molecular, and genetic levels. Consequently, an understanding of the underlying molecular mechanisms in multicel‐ lular model organisms would provide a unique opportunity to unthread analogous and complex biological processes in humans. This certainly will promote more automated and

We gratefully acknowledge financial support from the European Research Council (ERC)

(contract number 290586 from 1.07.2012), which funded this work.

Natalia A. Bakhtina1,2\*, Neil MacKinnon2 and Jan G. Korvink2

\*Address all correspondence to: natalia.bakhtina@kit.edu

1 Laboratory for Simulation, IMTEK – Department of Microsystems Engineering, University of Freiburg, Freiburg, Germany

2 Korvink's Group, Institute of Microstructure Technology (IMT), Karlsruhe Institute of Technology (KIT), Germany

## **References**


[25] Lange D, Storment CW, Conley CA, Kovacs GTA. A microfluidic shadow imaging system for the study of the nematode *Caenorhabditis elegans* in space. Sens. Actuat. B 2005, 107: 904–914. DOI: 10.1016/j.snb.2004.12.039.

[11] Jian-Ping J, Fan Y, Xin-Chun L, Yan-Yan Y, Zuan-Guang C. Advances on biomedical research in *Caenorhabditis elegans* based on microfluidic device. Prog. Biochem. Bio‐

[12] Shi W, Wen H, Lin B, Qin J. Microfluidic platform for the study of *Caenorhabditis* ele‐

[13] Yanik MF, Rohde CB, Pardo-Martin C. Technologies for micromanipulating, imag‐ ing, and phenotyping small invertebrates and vertebrates. Annu. Rev. Biomed. Eng.

[14] Xu X, Kim SK. The early bird catches the worm: new technologies for the *Caenorhab‐ ditis elegans* toolkit. Nat. Rev. Genet. 2011, 12: 793–801. DOI: 10.1038/nrg3050.

[15] Wlodkowic D, Khoshmanesh K, Akagi J, Williamsand DE, Cooper JM. Wormometryon-a-chip: innovative technologies for in situ analysis of small multicellular organ‐

[16] San-Miguel A, Lu H. Microfluidics as a tool for C. elegans research, in WormBook, ed. The C. elegans Research Community, 2013. DOI: 10.1895/wormbook.1.162.1. [17] Rezai P, Salam S, Selvaganapathy PR, Gupta BP. Microfluidic systems to study the biology of human diseases and identify potential therapeutic targets in C. elegans, in

[18] Hui W, JianHua Q. Analysis of *Caenorhabditis elegans* in microfluidic devices. Sci. Chi‐

[19] Wlodkowic D, Khoshmanesh K, Akagi J, Williams DE, Cooper JM. Wormometry-ona-chip: innovative technologies for in-situ analysis of small multicellular organisms.

[20] Gurkan UA, Moon S, Geckil H, Xu F, Wang S, Lu TJ, Demirci U. Miniaturized lens‐ less imaging systems for cell and microorganism visualization in point-of-care test‐

[21] Zhu H, Isikman SO, Mudanyali O, Greenbaum A, Ozcan A. Optical imaging techni‐ ques for point-of-care diagnostics. Lab Chip 2013, 13: 51–67. DOI: 10.1039/c2lc40864c.

[22] Wu J, Zheng G, Lee LM. Optical imaging techniques in microfluidics and their appli‐

[23] Heng X, Erickson D, Baugh LR, Yaqoob Z, Sternberg PW, Psaltisa D, Yang C. Opto‐ fluidic microscopy — a method for implementing a high resolution optical micro‐

[24] Cui X, Lee LM, Heng X, Zhong W, Sternberg PW, Psaltis D, Yang C. Lensless highresolution on-chip optofluidic microscopes for *Caenorhabditis elegans* and cell imag‐ ing. Proc. Natl. Acad. Sci. U. S. A. 2008, 105: 10670–10675. DOI: 10.1073/pnas.

gans. Top. Curr. Chem. 2011, 304: 323–338. DOI: 10.1007/128\_2011\_145.

2011, 13: 185–217. DOI: 10.1146/annurev-bioeng-071910-124703.

isms. Cytometry Part A 2011, 79: 799–813. DOI: 10.1002/cyto.a.21070.

Integrated Microsystems, ed. Iniewski K, CRC Press, 581–608; 2011.

na: Chem. 2012, 55: 484–493. DOI: 10.1007/s11426-012-4541-x.

Cytometry, Part A 2011, 79: 799–813. DOI: 10.1002/cyto.a.21070.

ing. Biotechnol. J. 2011, 6: 138–149. DOI: 10.1002/biot.201000427.

cations. Lab Chip 2012, 12: 3566–3575. DOI: 10.1039/c2lc40517b.

0804612105.

scope on a chip. Lab Chip 2006, 6: 1274–1276. DOI: 10.1039/B604676B.

phys. 2011, 38: 877–883. DOI: 10.3724/SP.J.1206.2011.00079.

104 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences


[49] Wang X, Hu R, Ge A, Hu L, Wang S, Feng X, Du W, Liu B. Highly efficient microflui‐ dic sorting device for synchronizing developmental stages of *C. elegans* based on de‐ flecting electrotaxis. Lab Chip 2015, 15: 2513–2521. DOI: 10.1039/c5lc00354g.

[37] Krajniak J, Lu H. Long-term high-resolution imaging and culture of *C. elegans* in chip-gel hybrid microfluidic device for developmental studies. Lab Chip 2010, 10:

[38] Kim N, Dempsey CM, Zoval JV, Sze J, Madou MJ. Automated microfluidic compact disc (CD) cultivation system of *Caenorhabditis elegans*. Sens. Actuat. B 2007, 122: 511–

[39] Bringmann H. Agarose hydrogel microcompartments for imaging sleep- and wakelike behavior and nervous system development in *Caenorhabditis elegans* larvae. J.

[40] Xian B, Shen J, Chen W, Sun N, Qiao N, Jiang D, Yu T, Men Y, Pang Z, Kaeberlein M, Huang Y, Han JD. WormFarm: a quantitative control and measurement device to‐ ward automated *Caenorhabditis elegans* aging analysis. Aging Cell 2013, 12: 398–409.

[41] Jung J, Nakajima M, Tajima H, Huang Q, Fukuda T J. A microfluidic device for the continuous culture and analysis of *Caenorhabditis elegans* in a toxic aqueous environ‐ ment. Micromech. Microeng. 2013, 23: 085008. DOI: 10.1088/0960-1317/23/8/085008.

[42] Aubry G, Zhan M, Lu H. Hydrogel-droplet microfluidic platform for high-resolution imaging and sorting of early larval *Caenorhabditis elegans*. Lab Chip 2015, 15: 1424–

[43] Shi W, Qin J, Ye N, Lin B. Droplet-based microfluidic system for individual *Caeno‐ rhabditis elegans* assay. Lab Chip 2008, 8: 1432–1435. DOI: 10.1039/b808753a.

[44] Clausell-Tormos J, Lieber D, Baret J, El-Harrak A, Miller OJ, Frenz L, Blouwolff J, Humphry KJ, Koester S, Duan H, Holtze C, Weitz DA, Griffiths AD, Merten CA. Droplet-based microfluidic platforms for the encapsulation and screening of mam‐ malian cells and multicellular organisms. Chem. Biol. 2008, 15: 427–437. DOI:

[45] Uppaluri S, Brangwynne CP. A size threshold governs *Caenorhabditis elegans* devel‐ opmental progression. Proc. R. Soc. B 282: 20151283. DOI:10.1098/rspb.2015.1283. [46] Turek M, Besseling J, Bringmann H. Agarose microchambers for long-term calcium imaging of Caenorhabditis elegans. J. Vis. Exp. 2015, 100: e52742 (8 pp.), DOI:

[47] Turek M, Besseling J, Spies J, Koenig S, Bringmann H. Sleep-active neuron specifica‐ tion and sleep induction require FLP-11 neuropeptides to systemically induce sleep.

[48] Cornaglia M, Mouchiroud L, Marette A, Narasimhan S, Lehnert T, Jovaisaite V, Au‐ werx J, Gijs MAM. An automated microfluidic platform for *C. elegans* embryo array‐ ing, phenotyping, and long-term live imaging. Scientific Reports 2015, 5:10192 (13

eLife 2016, 5: e12499 (18 pp.). DOI: 10.7554/eLife.12499.

Neurosci. Methods 2011, 201: 78–88. DOI: 10.1016/j.jneumeth.2011.07.013.

1862–1868. DOI: 10.1039/c001986k.

106 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

518. DOI: 10.1016/j.snb.2006.06.026.

DOI: 10.1111/acel.12063.

1431. DOI: 10.1039/c4lc01384k.

10.1016/j.chembiol.2008.04.004.

pp.). DOI: 10.1038/srep10192.

10.3791/52742.


[74] Wang J, Feng X, Du W, Liu B. Microfluidic worm-chip for *in vivo* analysis of neuronal activity upon dynamic chemical stimulations. Anal. Chim. Acta. 2011, 701: 23–28. DOI: 10.1016/j.aca.2011.06.007.

[62] Solvas XC, Geier FM, Leori AM, Bundy JG, Edela JB, Mello AJ. High-throughput age synchronisation of *Caenorhabditis elegans*. Chem. Commun. 2011, 47: 9801–9803. DOI:

[63] Dong L, Cornaglia M, Lehnert T, Gijs MAM. Versatile size-dependent sorting of *C. elegans* nematodes and embryos using a tunable microfluidic filter structure. Lab

[64] Lee H, Kim SA, Coakley S, Mugno P, Hammarlund M, Hilliard MA, Lu H. A multichannel device for high-density targetselective stimulation and long-term monitoring of cells and subcellular features in *C. elegans*. Lab Chip 2014, 14: 4513–4522. DOI:

[65] Wang Y, Wang J, Du W, Feng XJ, Liu B. Identification of the neuronal effects of etha‐ nol on *C. elegans* by *in vivo* fluorescence imaging on a microfluidic chip. Anal. Bioa‐

[66] Chronis N, Zimmer M, Bargmann C. Microfluidics for *in vivo* imaging of neuronal and behavioral activity in *Caenorhabditis elegans*. Nat. Methods 2007, 4: 727–731. DOI:

[67] Chokshi TV, Ben-Yakar A, Chronis N. CO2 and compressive immobilization of *C. ele‐*

[68] Hulme SE, Shevkoplyas SS, Apfeld J, Fontana W, Whitesides GM. A microfabricated array of clamps for immobilizing and imaging C. elegans. Lab Chip 2007, 7: 1515–

[69] Mondal S, Ahlawat S, Rau K, Venkataraman V, Koushika SP. Imaging *in vivo* neuro‐ nal transport in genetic model organisms using microfluidic devices. Traffic 2011, 12:

[70] Mondal S, Ahlawat S, Rau K, Venkataraman V, Koushika SP. Imaging *in vivo* neuro‐ nal transport in genetic model organisms using microfluidic devices. Traffic 2011, 12:

[71] McCormick KE, Gaertner BE, Sottile M, Phillips PC, Lockery SR. Microfluidic Devi‐ ces for analysis of spatial orientation behaviors in semi-restrained *Caenorhabditis ele‐*

[72] Gilleland CL, Rohde CB, Zeng F, Yanik MF. Microfluidic immobilization of physio‐ logically active *Caenorhabditis elegans*. Nat. Protoc. 2010, 5: 1888–1902. DOI: 10.1038/

[73] Rohde CB, Gilleland C, Samara C, Norton S, Haggarty S, Yanik MF. Microfluidic *in vivo* screen identifies compounds enhancing neuronal regeneration. Conf. Proc. IEEE

Eng. Med. Biol. Soc. 2009, 5950–5952. DOI: 10.1109/IEMBS.2009.5334771.

*gans*. PLoS One 2011, 6: e25710. DOI: 10.1371/journal.pone.0025710.

nal. Chem. 2011, 399: 3475–3481. DOI: 10.1007/s00216-010-4148-z.

*gans* on-chip. Lab Chip 2009, 9: 151–157. DOI: 10.1039/b807345g.

Chip 2016, 16: 574–585. DOI: 10.1039/C5LC01328C.

108 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

10.1039/c1cc14076k.

10.1039/c4lc00789a.

10.1038/nmeth1075.

nprot.2010.143.

1523. DOI: 10.1039/b707861g.

372–385. DOI: 10.1111/j.1600-0854.2010.01157.x.

372–385. DOI: 10.1111/j.1600-0854.2010.01157.x.


[98] Tong J, Rezai P, Salam S, Selvaganapathy PR, Gupta BP. Microfluidic-based electro‐ taxis for on-demand quantitative analysis of *Caenorhabditis elegans'* locomotion. J. Vis. Exp. 2013, 75: e50226. DOI: 10.3791/50226.

[85] Johari S, Nock V, Alkaisi MM, Wang W. On-chip analysis of *C. elegans* muscular forces and locomotion patterns in microstructured environments. Lab Chip 2013, 13:

[86] Han B, Kim D, Ko UH, Shin JH. A sorting strategy for *C. elegans* based on size-de‐ pendent motility and electrotaxis in a micro-structured channel. Lab Chip 2012, 12:

[87] Parashar A, Lycke R, Carr JA, Pandey S. Amplitude-modulated sinusoidal micro‐ channels for observing adaptability in *C. elegans* locomotion. Biomicrofluidics 2011, 5:

[88] Doll JC, Harjee N, Klejwa N, Kwon R, Coulthard SM, Petzold B, Goodman MB, Pruitt BL. SU-8 force sensing pillar arrays for biological measurements. Lab Chip 2009, 9:

[89] Oliver CR, Gourgou E, Bazopoulou D, Chronis N, Hart AJ. On-demand isolation and manipulation of C. elegans by In Vitro maskless photopatterning. PLoS One 2016, 11:

[90] Kuo WJ, Sie YS, Chuang HS. Characterizations of kinetic power and propulsion of the nematode *Caenorhabditis elegans* based on a micro-particle image velocimetry sys‐

[91] Wang X, Tang L, Xia Y, Hu L, Feng X, Du W, Liu B. Stress response of Caenorhabdi‐ tis elegans induced by space crowding in a micro-column array chip. Integr. Biol.

[92] Chuang H, Raizen D, Lamb A, Dabbish N, Bau H. Dielectrophoresis of *Caenorhabditis*

[93] Carr JA, Parashar A, Gibson R, Robertson AP, Martin RJ, Pandey S. A microfluidic platform for high-sensitivity, real-time drug screening on C. elegans and parasitic

[94] Rezai P, Salam S, Selvaganapathy PR, Gupta BP. Effect of pulse direct current signals on electrotactic movement of nematodes *Caenorhabditis elegans* and *Caenorhabditis*

[95] Rezai P, Siddiqui A, Salam S, Selvaganapathy PR, Gupta BP. Behavior of *Caenorhabdi‐ tis elegans* in alternating electric field and its application to their localization and con‐

[96] Rezai P, Siddiqui A, Selvaganapathy PR, Gupta BP. Electrical sorting of *Caenorhabdi‐*

[97] Rezai P, Siddiqui A, Selvaganapathy PR, Gupta BP. Electrotaxis of Caenorhabditis el‐ egans in a microfluidic environment. Lab Chip 2010, 10: 220–226. DOI: 10.1039/

1699–1707. DOI: 10.1039/c3lc41403e.

110 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

4128–4134. DOI: 10.1039/C2LC40209B.

024112–024119. DOI: 10.1063/1.3604391.

e0145935 (16 pp.). DOI:10.1371/journal.pone.0145935.

2013, 5: 728–737. DOI: 10.1039/C3IB20289E.

B917486A.

tem. Biomicrofluidics 2014, 8: 024116. DOI: 10.1063/1.4872061.

*elegans*. Lab Chip 2011, 11: 599–604. DOI: 10.1039/C0LC00532K.

nematodes. Lab Chip 2011, 11: 2385–2396. DOI: 10.1039/C1LC20170K.

*briggsae*. Biomicrofluidics 2011, 5: 044116. DOI: 10.1063/1.3665224.

trol. Appl. Phys. Lett. 2010, 96: 153702 (3 pp.). DOI: 10.1063/1.3383223.

*tis elegans*. Lab Chip 2012, 12: 1831–1840. DOI: 10.1039/c2lc20967e.

1449–1454. DOI: 10.1039/b818622g.


*gans* forward locomotion. Neuron 2012, 76: 750–761. DOI: 10.1016/j.neuron. 2012.08.039.


[122] Gokce SK, Guo SX, Ghorashian N, Everett WN, Jarrell T, et al. (2014) A fully auto‐ mated microfluidic femtosecond laser axotomy platform for nerve regeneration stud‐ ies in *C. elegans*. PLoS One 2014, 9: e113917 (28 pp.). DOI: 10.1371/journal.pone. 0113917.

*gans* forward locomotion. Neuron 2012, 76: 750–761. DOI: 10.1016/j.neuron.

[111] Hwang H, Kim E, Kim SH, Park S. A sensitive C. elegans chemotaxis assay using mi‐ crofluidic device generating a linear gradient of chemoeffectors. Bull. Korean Chem.

[112] Salam S, Ansari A, Amon S, Rezai P, Selvaganapathy PR, Mishra RK, Gupta BP. A microfluidic phenotype analysis system reveals function of sensory and dopaminer‐ gic neuron signaling in *C. elegans* electrotactic swimming behavior. Worm 2013, 2:

[113] Chalasani SH, Chronis N, Tsunozaki M, Gray JM, Ramot D, Goodman MB, Barg‐ mann CI. Dissecting a circuit for olfactory behaviour in *Caenorhabditis elegans*. Nature

[114] Saldanha JN, Parashar A, Pandey S, Powell-Coffman JA. Multiparameter behavioral analyses provide insights to mechanisms of cyanide resistance in *Caenorhabditis ele‐*

[115] Gray JM, Karow DS, Lu H, Chang AJ, Chang JS, Ellis RE, Marletta MA, Bargmann CI. Oxygen sensation and social feeding mediated by a *C. elegans* guanylate cyclase ho‐

[116] Zimmer M, Gray JM, Pokala N, Chang AJ, Karow DS, Marletta MA, Hudson ML, Morton DB, Chronis N, Bargmann CI. Neurons detect increases and decreases in oxygen levels using distinct guanylate cyclases. Neuron 2009, 61: 865–879. DOI:

[117] Santos SI, Mathew M, Loza-Alvarez P. Real time imaging of femtosecond laser in‐ duced nano-neurosurgery dynamics in *C. elegans*. Opt. Express 2010, 18: 364–377.

[118] Buonanno M, Garty G, Grad M, Gendrel M, Hobert O, Brenner DJ. Microbeam irradi‐ ation of *C. elegans nematode* in microfluidic channels. Radiat. Environ. Biophys. 2013,

[119] Cornaglia M, Krishnamani G, Mouchiroud L, Sorrentino V, Lehnert T, Auwerx J, Gijs MAM. Automated longitudinal monitoring of in vivo protein aggregation in neuro‐ degenerative disease *C. elegans* models. Mol. Neurodegener 2016, 11: 1–13. DOI:

[120] Hu C, O'Connor V, Holden-Dye L, Morgan H. Conf. Proc. MicroTAS 2013, pp. 1441–

[121] Hu C, Dillon J, Kearn J, Murray C, O'Connor V, Holden-Dye L, Morgan H. Neuro‐ Chip: a microfluidic electrophysiological device for genetic and chemical biology screening of *Caenorhabditis elegans* adult and larvae. PLoS One 2013, 8: e64297. DOI:

*gans*. Toxicol. Sci. 2013, 135: 156–168. DOI: 10.1093/toxsci/kft138.

mologue. Nature 2004, 430: 317–322. DOI: 10.1038/nature02714.

Soc. 2015, 36: 1096–1099. DOI: 10.1002/bkcs.10201.

112 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

e24558. DOI: 10.4161/worm.24558.

10.1016/j.neuron.2009.02.013.

DOI: 10.1364/OE.18.000364.

10.1186/s13024-016-0083-6.

1443. DOI: 10.13140/2.1.1565.0244.

10.1371/journal.pone.0064297.

52: 531–537. DOI 10.1007/s00411-013-0485-6.

2007, 450: 63–70. DOI: 10.1038/nature06292.

2012.08.039.


## **Imaging and Spectroscopy**

[134] Kalfe A, Telfah A, Lambert J, Hergenröder R. Looking into living cell systems: planar waveguide microfluidic NMR detector for in vitro metabolomics of tumor spheroids.

[135] Meier RC, Höfflin J, Badilita V, Wallrabe U, Korvink JG. Microfluidic integration of wirebonded microcoils for on-chip applications in nuclear magnetic resonance. J. Mi‐ cromech. Microeng. 2014, 24: 045021 (12 pp.). DOI: 10.1088/0960-1317/24/4/045021.

[136] Wong A, Li X, Molin L, Solari F, Elena-Herrmann B, Sakellariou D. µhigh resolutionmagic-angle spinning NMR spectroscopy for metabolic phenotyping of *Caenorhabdi‐*

Anal. Chem. 2015, 87: 7402–7410. DOI: 10.1021/acs.analchem.5b01603.

114 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

*tis elegans*. Anal. Chem. 2014, 86: 6064–6070. DOI: 10.1021/ac501208z.

## **Microfluidics for Ultrafast Spectroscopy**

## Adrien A. P. Chauvet

Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/64428

#### **Abstract**

Ultrafast laser technologies became one of the essential tool in the characterization of mo‐ lecular compounds. Being comprised of spectroscopists, laser scientists, chemists and bi‐ ologists, the "ultrafast community" is often disconnected and consequently unaware of the developments in microfluidic systems. The challenges of studying limited amount of precious liquid sample by means of ultrafast spectroscopy remains silent and, while no commercial systems are available, each research group is developing its own "homemade" options. This chapter will therefore contribute in filling up the gap that exist be‐ tween the two communities, that of the ultrafast spectroscopy and that of microfluidics by revealing the importance of this analytical tool as well as the advantages of applying microfluidic technics to it. In this goal, the chapter will focus of the recently developed microfluidic flow-cell. With a minimal volume of about 250 µL, the flow-cell enables the study of precious protein complexes that are simply not available in larger quantities. The multiple advantages of the microfluidic flow-cell will be illustrated by the analysis of the cytochrome *bc*1. In particular, the study will describe how the capabilities of the mi‐ crofluidic flow-cell enabled the resolution of the ultrafast electronic and nuclear dynam‐ ics of specific embedded chromophores.

**Keywords:** Microfluidics, Ultrafast Spectroscopy, Liquid Sample

## **1. Introduction**

The aim of this chapter is to address the gap that exists between two research communities: ultrafast spectroscopy and microfluidics. Indeed, the development of pulsed laser systems in the last few decades has ushered in new techniques in ultrafast spectroscopy. These techniques have opened new doors for the study of fundamental photo-chemical and photo-physical behavior of a variety of photosynthetic protein complexes.[1] For methodological reasons, i.e. samples being rare and the use of highly specialized equipment, there existed a pressing need to apply microfluidics systems in ultrafast spectroscopy, two fields that unfortunately

© 2016 The Author(s). Licensee InTech. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

developed separately and whose researchers are rarely knowledgeable in both areas. For the microfluidic readers who may be unfamiliar with this literature, I introduce, in the first part of this chapter, the basic concepts of ultrafast pump-probe spectroscopy. In so doing, I highlight the relevance of this technique in gaining understanding protein dynamics and therefore biological properties and functions. It is in fact due to the laborious procedures that are involved in the purification processes of these proteins complexes that most biological samples are only available in sub-milliliters quantities. In the second part of this chapter, I will therefore expose some of the most common solutions that are in use in order to manipulate liquid samples in ultrafast spectroscopy. Finally, in the third part of this chapter, I describe one of the latest applications to the field of ultrafast spectroscopy—what is now called microfluidics—as it handles micro-liters volumes of a given sample. I illustrate the benefits of the application of such a microfluidic system through an analysis the of the cytochrome *bc*1. I conclude with a discussion on the areas of the present microfluidic flow-cell in need of further research and investigation.

## **2. The need for microfluidics in ultrafast spectroscopy**

In this section I will introduce the field of ultrafast spectroscopy to which the microfluidics systems will be applied. I describe the basics of the technique and its relevance to the current fundamental research efforts in biochemistry and biophysics. In the last section of this chapter, I cover the practical challenges that emerge from such studies, which justify the need for microfluidics.

#### **2.1. Ultrafast transient absorption spectroscopy**

Transient absorption is a spectroscopic technique whose aim is to resolve the relaxation dynamics of an excited molecule "simply" by looking at its spectral modifications. The technique involves two principal light beam: a pump and a probe. While the role of the pump is to promote the molecule to a particular excited state, the probe is used to "look" at the state in which this molecule is in. The fact that these light beams are not continuous but are strains of pulses allows one to excite the molecule for a brief instant (the duration of the pulse) and to probe its state at a later time. The light pulses are first produced in a cavity (oscillator) that is built around a doped crystal such as titanium doped sapphire crystals (Ti:S) as shown in Figure 1.[2]

The crystal, when excited, serves as a photon tank that will amplify any pulses passing through, via stimulated emission. The cavity then enters into a "mode locked" state as soon as the enclosed pulse strains are in resonance with the dimension of the cavity itself. The energy trapped within the cavity builds up until the pulses are intense enough to leak through one of the cavity end-mirror. Typical Ti:S cavity produces strains of ~30 fs pulses centered around ~800 nm at 80 MHz repetition rate; each pulse being about tenth of nano-Joules. The extracted pulses are then amplified in a similar fashion by passing in a second cavity comprised of a second Ti:S crystal.[2] The frequency is however decreased to few kHz in order to reach few

#### Microfluidics for Ultrafast Spectroscopy http://dx.doi.org/10.5772/64428 119

**Figure 1.** Ti:S crystal cavity.

developed separately and whose researchers are rarely knowledgeable in both areas. For the microfluidic readers who may be unfamiliar with this literature, I introduce, in the first part of this chapter, the basic concepts of ultrafast pump-probe spectroscopy. In so doing, I highlight the relevance of this technique in gaining understanding protein dynamics and therefore biological properties and functions. It is in fact due to the laborious procedures that are involved in the purification processes of these proteins complexes that most biological samples are only available in sub-milliliters quantities. In the second part of this chapter, I will therefore expose some of the most common solutions that are in use in order to manipulate liquid samples in ultrafast spectroscopy. Finally, in the third part of this chapter, I describe one of the latest applications to the field of ultrafast spectroscopy—what is now called microfluidics—as it handles micro-liters volumes of a given sample. I illustrate the benefits of the application of such a microfluidic system through an analysis the of the cytochrome *bc*1. I conclude with a discussion on the areas of the present microfluidic flow-cell in need of further

In this section I will introduce the field of ultrafast spectroscopy to which the microfluidics systems will be applied. I describe the basics of the technique and its relevance to the current fundamental research efforts in biochemistry and biophysics. In the last section of this chapter, I cover the practical challenges that emerge from such studies, which justify the need for

Transient absorption is a spectroscopic technique whose aim is to resolve the relaxation dynamics of an excited molecule "simply" by looking at its spectral modifications. The technique involves two principal light beam: a pump and a probe. While the role of the pump is to promote the molecule to a particular excited state, the probe is used to "look" at the state in which this molecule is in. The fact that these light beams are not continuous but are strains of pulses allows one to excite the molecule for a brief instant (the duration of the pulse) and to probe its state at a later time. The light pulses are first produced in a cavity (oscillator) that is built around a doped crystal such as titanium doped sapphire crystals (Ti:S) as shown in Figure

The crystal, when excited, serves as a photon tank that will amplify any pulses passing through, via stimulated emission. The cavity then enters into a "mode locked" state as soon as the enclosed pulse strains are in resonance with the dimension of the cavity itself. The energy trapped within the cavity builds up until the pulses are intense enough to leak through one of the cavity end-mirror. Typical Ti:S cavity produces strains of ~30 fs pulses centered around ~800 nm at 80 MHz repetition rate; each pulse being about tenth of nano-Joules. The extracted pulses are then amplified in a similar fashion by passing in a second cavity comprised of a second Ti:S crystal.[2] The frequency is however decreased to few kHz in order to reach few

**2. The need for microfluidics in ultrafast spectroscopy**

118 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

**2.1. Ultrafast transient absorption spectroscopy**

research and investigation.

microfluidics.

1.[2]

milli-Joules per pulses. It is these amplified pulses that are then split into pump and probe pulses.

**Figure 2.** Pump-probe experimental scheme.

As illustrated in Figure 2, varying the time delay that separate both pump and probe enables to follow the evolution of the transient excited states over time. This delay is introduced by physically increasing the path of one of the pump or probe arm of the setup via a precision stage. In a typical configuration, a 30 cm stage provides with a temporal window of about 4 ns (round-trip). The time resolution with which we can follow the molecular dynamics is then given by the duration of the light pulses themselves. A regular transient absorption setup is today capable of producing amplified strains of ~40 fs pulse with a kHz repetition rate. Such femto-to-nano-second time window corresponds to the dynamics of energy and electron transfer within and between molecules as well as local structural modifications.[3] Further‐ more, the high repetition rate and the development in matters of laser stability and detection system gives the ability to monitor the absorption changes of a single molecule out of a thousand (corresponding to changes of ~10-4 OD) in less than a second of accumulation time. [4] With such setups, it is for example possible to trigger the charge separation in the Photo‐ system I protein complex and to follow the liberated electron as it progresses from one side of the protein to the other.[4] Another example is the monitoring of the heme-ligand dissociation and rebinding dynamics that results from the absorption of a photon such as it is the case of various types of cytochromes.[5, 6]

The use of non-linear optics, such as in the famous and various kind of optical parametric amplifiers (OPA, Non-collinear OPA and multi-pass OPA)[7] as well as the different pulse shaping devices (in either transmission or reflection),[8] gives the ability to tune both the pump and the probe to the desire wavelength, therefore enabling to excite and to follow a particular molecular transition. For example, in the study of the bacterial reaction centers complexes, being comprise of multiple types of pigments that are spectrally distinct, careful tuning of the pump allows to preferably excite one pigment while living the others in their fundamental state.[9]

It is therefore out of the development in laser technology and specifically in tunable table-top pulsed lasers that the field of ultrafast spectroscopy came to know the success it knows today.

#### **2.2. Studying biological samples**

As implied in the previous section, transient spectroscopy is best suited for the study of compounds that have distinct spectral feature. Fortunately, most organic compounds are made of either aromatic amino acids or incorporates chromophores within their protein structure, each having distinct spectral features. The technics consequently became in the past decade a common analytical tool for biologists and chemists. Ultrafast transient spectroscopy is indeed used for a broad range of investigations: being sensitive to changes in absorption spectrum of the proteins, it is possible to collect data on local conformational deformations, electronic transitions and (low) vibrational modes of oscillation within molecules, intra and inter molecular energy and electron transfers, etc.[10] In the field of solar energy conversion for example, which is one of today's essential topic in our energy savvy societies, this technique allowed to better understand the conversion processes from light to consumable energy. In particular, the study of photosynthesis showed how the specific arrangement of pigment within larger protein structure either favor the absorption and passing of the photon energy, as it takes place in antennae systems, or favors the generation of a charge separated state, as it happens in photosynthetic reaction centers, [4, 9] which results in the liberation of a high energy, and therefore usable, electron. Other chromophores can also serve as electron docking sites and electron carriers.[6] The knowledge gained from such studies is then applied for medical and industrial purposes and used to optimize specific molecular reactions. From these examples, it is possible to understand why the study of biological samples by means of ultrafast transient spectroscopy, among other spectroscopic techniques, became and remains one of the standard analytical tool for the fundamental understanding of a broad range of molecular dynamics.

### **2.3. The challenges of the application**

more, the high repetition rate and the development in matters of laser stability and detection system gives the ability to monitor the absorption changes of a single molecule out of a thousand (corresponding to changes of ~10-4 OD) in less than a second of accumulation time. [4] With such setups, it is for example possible to trigger the charge separation in the Photo‐ system I protein complex and to follow the liberated electron as it progresses from one side of the protein to the other.[4] Another example is the monitoring of the heme-ligand dissociation and rebinding dynamics that results from the absorption of a photon such as it is the case of

120 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

The use of non-linear optics, such as in the famous and various kind of optical parametric amplifiers (OPA, Non-collinear OPA and multi-pass OPA)[7] as well as the different pulse shaping devices (in either transmission or reflection),[8] gives the ability to tune both the pump and the probe to the desire wavelength, therefore enabling to excite and to follow a particular molecular transition. For example, in the study of the bacterial reaction centers complexes, being comprise of multiple types of pigments that are spectrally distinct, careful tuning of the pump allows to preferably excite one pigment while living the others in their fundamental

It is therefore out of the development in laser technology and specifically in tunable table-top pulsed lasers that the field of ultrafast spectroscopy came to know the success it knows today.

As implied in the previous section, transient spectroscopy is best suited for the study of compounds that have distinct spectral feature. Fortunately, most organic compounds are made of either aromatic amino acids or incorporates chromophores within their protein structure, each having distinct spectral features. The technics consequently became in the past decade a common analytical tool for biologists and chemists. Ultrafast transient spectroscopy is indeed used for a broad range of investigations: being sensitive to changes in absorption spectrum of the proteins, it is possible to collect data on local conformational deformations, electronic transitions and (low) vibrational modes of oscillation within molecules, intra and inter molecular energy and electron transfers, etc.[10] In the field of solar energy conversion for example, which is one of today's essential topic in our energy savvy societies, this technique allowed to better understand the conversion processes from light to consumable energy. In particular, the study of photosynthesis showed how the specific arrangement of pigment within larger protein structure either favor the absorption and passing of the photon energy, as it takes place in antennae systems, or favors the generation of a charge separated state, as it happens in photosynthetic reaction centers, [4, 9] which results in the liberation of a high energy, and therefore usable, electron. Other chromophores can also serve as electron docking sites and electron carriers.[6] The knowledge gained from such studies is then applied for medical and industrial purposes and used to optimize specific molecular reactions. From these examples, it is possible to understand why the study of biological samples by means of ultrafast transient spectroscopy, among other spectroscopic techniques, became and remains one of the standard analytical tool for the fundamental understanding of a broad range of molecular

various types of cytochromes.[5, 6]

**2.2. Studying biological samples**

state.[9]

dynamics.

As discussed above, ultrafast transient spectroscopy is suitable for the study of biological samples. These samples are usually made of purified and solvated proteins. Once a particular molecule is excited, the deposited energy ultimately dissipates into the solvent (so long as the changes are not permanent). It is to remember that the typical repetition rate of the pulses are in the order of the kHz and at this rate the sample is excited about every milliseconds. The risk is that the photo-induced molecular modifications live for a time that is comparable, which will then results in a rapid saturation of the sample. In other words, the excited molecule might not have sufficient time to relax to its fundamental state before the coming of the next light pulse. Saturation thus takes place and as the molecules gets overexcited: they are unable to release the deposited energy quickly enough and end up by "burning". In order to avoid such consequences, the sample is usually flown in front of the laser beams. The condition is that the flux is high enough so that the sample is refreshed for each laser pulse.

The second major constraint is related to the quantity of the sample available. The sample being made of purified proteins, it is then concentrated in order to reach an optical density that is suitable for spectroscopic analysis. Ideally, transient absorption spectroscopy requires an optical density of about 0.6 for the transition of interest, which in the case of heme protein corresponds to a concentration of hundreds of micro molars. The samples are consequently limited in terms of volumes and sub-milliliters quantities already requires months-long of successive growth of the organisms and protein purification cycles.[11, 12]

The third constraint concerns the susceptibility of the sample in respect to its solvent and atmosphere in which it is enclosed. Indeed some biological samples are hydrophobic and require to be dissolved in various chemicals in order to avoid aggregation and the subsequent scattering of the light pulses, such as it is the case for solvated porphyrins. Many samples are also sensitive to oxygen and therefore require the atmosphere to be controlled. For example, myoglobin is able to effectively bind a variety of diatomic molecule. It however has a high affinity for oxygen, so much that it is not possible to study its deoxygenated state unless in anaerobic conditions.[13]

In conclusion, ultrafast transient spectroscopy is today one of the standard analytical tool for whoever desire to study the local structural, electronic and vibrational photo-induced dynamics. In particular, solvated chromophores and chromophore-containing proteins are well suited for the techniques as they can be specifically targeted via their absorption band. However, these liquid samples are often limited in quantities and have to be flown in order to avoid any photo-damages. From these challenges arise the need for microfluidics so as to flow the limited sample volumes. Furthermore, in order to perform ultrafast spectroscopic meas‐ urement, the probe pulses must pass through the sample, therefore through the flow-cell in which it is enclosed. The cell consequently requires adequate windows that do not alter the signal-to-noise ratio nor the temporal and spectral resolution of the apparatus. Additionally, due to the properties and high sensitivity of certain samples, the cell must be resistant to the solvent while providing control of the atmosphere. In such cases, the sample must be hermet‐ ically confined within the microfluidic system which therefore has to also play the role of an anaerobic chamber.

## **3. Most common solutions available**

In this section I will review different techniques that are commonly employed to flow the sample in front of the laser beam. I will discuss the advantages and inconveniences of each in terms of their compatibilities with the requirements of ultrafast spectroscopic laser systems.

### **3.1. Flow cell**

The main idea is to flow the sample in between two transparent plates. These windows are usually made of quartz so as to permit the broadband (near UV-Visible-near IR) beams to pass through. This type of cell allows for small path lengths and thin windows, down to 0.02 mm such as the one shown in Figure 3, therefore reducing scattering of the excitation beam through the quartz. The thin windows also have the advantage to minimally alter the pulse duration (limited group velocity dispersion), therefore allowing for an optimal time resolution. Furthermore, the cell is steady and its stability allows for optimal signal to noise levels.

**Figure 3.** Flow cell from Starna Cell [14]

The quartz cells, by themselves, are commercially available.[14] However, due to their size, these cells already enclose about one milliliter of sample. Furthermore they must be connected to a pump, typically a peristaltic or flow-through pump, in order to generate the flow. Altogether, the flow cell system requires few tenth of milliliters and is consequently not adequate for precious samples that are simply not available in such quantities.

#### **3.2. Liquid micro-jet**

The term micro here comes from the diameter of the jet itself, which produces a couple of centimeters-long of regular flow. The advantage of expelling the sample as a jet is to remove the constraints of having it to pass through windows, i.e. avoiding any additional absorption, scattering and group velocity dispersion. It therefore allows pump probe experiment in all spectral region including UV and X-ray region. Associated with an adequate sample collector, as shown in Figure 4, it is also applicable in vacuum chambers and becomes suitable for photoelectron spectroscopy.[15]

**Figure 4.** Micro-jet implemented for photo-electron spectroscopy. Courtesy of José Ojeda, EPFL.

The inconvenience of having to produce a constant flow rate is that the sample must pass through a sophisticated HPLC pump,[15-17] which consequently requires sample volumes that are larger than our targeted sub-milliliter. Note also that while passing in either the air or in vacuum, the sample's solvent is subjected to evaporation. In such conditions, recycling of the sample results in a change of the sample's concentration and temperature over the course of the experiment.[16] Furthermore, the high speed at which the sample go through the nozzle induces charging of the liquid and or of the nozzle, which might alter the measurement.[17]

#### **3.3. Spinning /moving cell**

**3. Most common solutions available**

122 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

**3.1. Flow cell**

**Figure 3.** Flow cell from Starna Cell [14]

**3.2. Liquid micro-jet**

electron spectroscopy.[15]

In this section I will review different techniques that are commonly employed to flow the sample in front of the laser beam. I will discuss the advantages and inconveniences of each in terms of their compatibilities with the requirements of ultrafast spectroscopic laser systems.

The main idea is to flow the sample in between two transparent plates. These windows are usually made of quartz so as to permit the broadband (near UV-Visible-near IR) beams to pass through. This type of cell allows for small path lengths and thin windows, down to 0.02 mm such as the one shown in Figure 3, therefore reducing scattering of the excitation beam through the quartz. The thin windows also have the advantage to minimally alter the pulse duration (limited group velocity dispersion), therefore allowing for an optimal time resolution. Furthermore, the cell is steady and its stability allows for optimal signal to noise levels.

The quartz cells, by themselves, are commercially available.[14] However, due to their size, these cells already enclose about one milliliter of sample. Furthermore they must be connected to a pump, typically a peristaltic or flow-through pump, in order to generate the flow. Altogether, the flow cell system requires few tenth of milliliters and is consequently not

The term micro here comes from the diameter of the jet itself, which produces a couple of centimeters-long of regular flow. The advantage of expelling the sample as a jet is to remove the constraints of having it to pass through windows, i.e. avoiding any additional absorption, scattering and group velocity dispersion. It therefore allows pump probe experiment in all spectral region including UV and X-ray region. Associated with an adequate sample collector, as shown in Figure 4, it is also applicable in vacuum chambers and becomes suitable for photo-

adequate for precious samples that are simply not available in such quantities.

The idea behind the spinning-cell is similar to that of the flow cell as the sample is housed between two circular glass plates that are spaced by the desirable optical path-length.[18] While rotating, the sample either creates a rim at the edge of the cell, or at lower speed, the solution remains at the bottom of the cell and is constantly mixed due to friction with the glass as shown in Figure 5. The advantage is that it typically requires minimal amounts of sample (~0.3 mL) as well as to provide control over the initial atmosphere in which the sample in enclosed since the sample is hermetically confined.

However, because the rotation of the glass plate causes the excitation beam to sweep a large surface, the cleanness of the plates is directly related to the noise. It is consequently burden‐ some to clean. Furthermore, the fact of having a moving piece of glass in which the beam is focused renders the alignment of the cell crucial, and any slight asymmetry of the glass plates has consequences on the transmitted probe beam. Also, since the glass plates are typically few cm in diameter, the required minimal thickness of the glass lowers the time resolution. Another inconvenience is that once the cell is set it does not allow access to the enclosed sample and each experiments therefore requires its own sample preparation.

In conclusion, the proposed solutions used to flow the sample in front of the laser beam have each distinct advantages and inconveniences, and none ideally respond to all the requirement, i.e. is suitable for sub-milliliter sample volumes; provides a close atmosphere; grant access to the sample once closed; affecting minimally the signal-to-noise ratio and resolution of the

**Figure 5.** Spinning cell used if few-kHz spectroscopy setups.

apparatus. Through these few examples I hope to have convinced the reader of the need for a development in the application of microfluidic system to the field of ultrafast spectroscopy.

## **4. Recent improvement and application: the microfluidic flow cell**

In this section I will describe the recently developed microfluidic flow-cell in order to illustrate the importance and advantages of applying microfluidic systems to the field of ultrafast spectroscopy. I will show that recent improvements in the field of microfluidics have the capabilities to solve the previously mentioned constraint all at once. I will thus discuss the advantages of the system in light of the other technics. Following the technical properties I will illustrate the flow-cell's effectiveness through a study done on rare *bc*-cytochrome and conclude with an appeal for further development.

#### **4.1. The microfluidic flow-cell**

The microfluidic flow-cell, [19] as illustrated in Figure 6, is composed of three main elements that are connected via flexible tubing of 1-mm diameter:

*The decantation chamber* is a home-made polymer cylindrical chamber as shown in Figure 6. The 0.5-mm diameter inlet and outlet are at the bottom of the chamber in order to minimize turbulences that are created at high flow rates. The chamber requires a minimal amount of ~50 µL of liquid sample in order to have a continuous flow between the inlet and outlet. Any excess of sample fills up the chamber and allows the bubbles that might be enclosed in the closed circuit to rise to the surface. The bubbles are naturally trapped by the chamber while passing through at low flow-speed. At high flow-speed however, larger sample volume are required in order to avoid the suction of air due to the liquid's turbulences. The top of the chamber is threaded to fit a standard septum screw cap. This allows for the addition of chemicals to the enclosed solution while keeping the confined atmosphere protected and avoiding evaporation of the solvents.

*The capillary window* is made of a square quartz silica capillary bought from Composite Metal Services Ltd (CMS). It has a path-length of 0.5 mm with 0.1-mm thin walls. Knowing that the focus of the laser beam is about 100 um in diameter, the window can easily be set within the beam path. The capillary is fixed at the center of a xyz-rotation mount that allows for fine adjustment of the angle between the incident beam and the window.

*The turbisc pump* is a design from CSEM.[20] In short, the flow is created by direct friction between a grooved barrel and the liquid. The inner volume that the pump contains is about 100 µL only. Because the housing and the seal are respectively made out of Polyetherimide and of Polyetheretherketone, the pump is relatively resistant to chemicals.

apparatus. Through these few examples I hope to have convinced the reader of the need for a development in the application of microfluidic system to the field of ultrafast spectroscopy.

In this section I will describe the recently developed microfluidic flow-cell in order to illustrate the importance and advantages of applying microfluidic systems to the field of ultrafast spectroscopy. I will show that recent improvements in the field of microfluidics have the capabilities to solve the previously mentioned constraint all at once. I will thus discuss the advantages of the system in light of the other technics. Following the technical properties I will illustrate the flow-cell's effectiveness through a study done on rare *bc*-cytochrome and

The microfluidic flow-cell, [19] as illustrated in Figure 6, is composed of three main elements

*The decantation chamber* is a home-made polymer cylindrical chamber as shown in Figure 6. The 0.5-mm diameter inlet and outlet are at the bottom of the chamber in order to minimize turbulences that are created at high flow rates. The chamber requires a minimal amount of ~50 µL of liquid sample in order to have a continuous flow between the inlet and outlet. Any excess of sample fills up the chamber and allows the bubbles that might be enclosed in the closed circuit to rise to the surface. The bubbles are naturally trapped by the chamber while passing through at low flow-speed. At high flow-speed however, larger sample volume are required in order to avoid the suction of air due to the liquid's turbulences. The top of the chamber is threaded to fit a standard septum screw cap. This allows for the addition of chemicals to the enclosed solution while keeping the confined atmosphere protected and avoiding evaporation

**4. Recent improvement and application: the microfluidic flow cell**

conclude with an appeal for further development.

**Figure 5.** Spinning cell used if few-kHz spectroscopy setups.

124 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

that are connected via flexible tubing of 1-mm diameter:

**4.1. The microfluidic flow-cell**

of the solvents.

**Figure 6.** (A) Overall view of the assembled system, with zoom on the capillary junctions. (B) Side view of the bubble chamber and (C) of the capillary window. Reproduced from Ref. [19].

When the pump, the chamber and the capillary are connected, the minimal volume of sample needed for good working conditions is about 250 µL only. This configuration includes a total tubing (1-mm inner diameter) length of ~6 cm and a sufficient amount of sample in the cuvette to avoid the suction of bubbles due to turbulences with a minimal flow of ~0.1 ml/sec. The flow is proportional to the voltage applied to the pump and under the same configuration the maximum flow rate was measured to be ~0.36 mL/sec, as shown in Figure 7.

**Figure 7.** Measured flow-rate (dots) and its best fit (line, 2nd order polynomial) while the pump is connected to ~6 cm of 1-mm diameter tubing and ~1-cm long 0.5x0.5 mm2 square capillary. In order to ensure fresh sample at each laser shot, while assuming a typical laser focus size of 100 µm, the expected maximum repetition rate of excitation is indicated as a reference only (right axis).

While assuming a typical laser focus diameter of 100 µm within the 0.5x0.5 mm2 square capillary, the flow, when assumed to be uniform, is expected to effectively refresh the sample within the laser focus for each laser shot at an excitation rate of up to ~14 kHz. In practice, it is to remember that the flow is impeded on the edges of the capillary and consequently the value of 14 kHz has to be taken as an upper limit only. Taking into account that the inner volume of the pump is only about 100 µL, it represents one of the best (if not the actual best) compromise between flow-rate and required volume. The assembled microfluidic flow cell, in working conditions is shown in Figure 8.

**Figure 8.** Microfluidic flow-cell in action

#### **4.2. Example of application: the study of cythochrome** *bc***1 complexes**

In order to illustrate the applicability of the microfluidic flow-cell as well as some of its advantages I present here a unique analysis, that of the Cytochrome (cyt) *bc*1 complex. [6] The cyt *bc*<sup>1</sup> complex is a key player in mitochondrial and bacterial respiratory chains.[21] It is the main actor in the protonmotive Q cycle and results in the formation of a proton gradient across the membrane via a series of embedded hemes as shown in Figure 9.[22-24] The generated potential gradient serves as the driving force for the synthesis of ATP, the universal energy transporter in living organisms. The understanding of the protein complex is therefore of primary importance. However the sample is rare and mainly because of the limited quantity available after each purification process, the heme dynamics were until then never studied by means of ultrafast spectroscopy.

**Figure 7.** Measured flow-rate (dots) and its best fit (line, 2nd order polynomial) while the pump is connected to ~6 cm of 1-mm diameter tubing and ~1-cm long 0.5x0.5 mm2 square capillary. In order to ensure fresh sample at each laser shot, while assuming a typical laser focus size of 100 µm, the expected maximum repetition rate of excitation is indicated as

126 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

While assuming a typical laser focus diameter of 100 µm within the 0.5x0.5 mm2 square capillary, the flow, when assumed to be uniform, is expected to effectively refresh the sample within the laser focus for each laser shot at an excitation rate of up to ~14 kHz. In practice, it is to remember that the flow is impeded on the edges of the capillary and consequently the value of 14 kHz has to be taken as an upper limit only. Taking into account that the inner volume of the pump is only about 100 µL, it represents one of the best (if not the actual best) compromise between flow-rate and required volume. The assembled microfluidic flow cell, in

a reference only (right axis).

working conditions is shown in Figure 8.

**Figure 8.** Microfluidic flow-cell in action

**4.2. Example of application: the study of cythochrome** *bc***1 complexes**

In order to illustrate the applicability of the microfluidic flow-cell as well as some of its advantages I present here a unique analysis, that of the Cytochrome (cyt) *bc*1 complex. [6] The

**Figure 9.** (Left) Structure of the *bc*<sup>1</sup> complex[25] with the protein backbone being shaded for clear visualization of the key actors in the proton-coupled-electron mechanisms. (Right) Static absorption spectrum of the sequential reduction and oxidation of the cyt *bc*1 dimer complex: after pre-reduction by ubiquinol and before (red, *c*1-hemes reduced) and right after (blue, *b*- and *c*1-hemes reduced) the addition of dithionite, and at the end, after oxidation of the complexes by oxygen (green, *c*1-hemes reduced). The 523-nm excitation pump is indicated as reference. Reproduced from Ref. [6] with permission from the Royal Society of Chemistry.

Thanks tothemicrofluidic flow-cell,andmoreparticularlytotheaccess itprovides tothe sample via the septum, the reduction and oxidation of the hemes of interests could be controlled chemically. All data were therefore taken from the same sample preparation within the same experiment, in the same experimental conditions. The different signals that emerge from the reduced states of the hemes could then be directly compared: either the *c*1-hemes can be exclusively reduced, or both *c*1- and *b*-hemes can be reduced together. Direct subtraction of the two data set enabled the extraction of the sole signal form the *b*-hemes, as shown in Figure 10.

The dynamics of each heme type could be analysed separately by means of singular value decomposition and global fitting as described in detail elsewhere.[6] The resulting decay associated spectra, shown in Figure 11, revealed the clear differences between the *b*- and *c*1 hemes behaviours within the *bc*<sup>1</sup> protein complex: while the *c*1-hemes undergo photo-dissoci‐ ation of their axial ligand as a result of ultrafast laser excitation, the *b*-hemes were shown to undergo photo-oxidation with a high (> 0.4) quantum yield that is beyond all expectations.

**Figure 10.** (A) Spectra at selected time delays while both *b*- and *c*1-hemes are reduced. (B) Spectra at selected time de‐ lays while only the c1-hemes are left reduced. (C) Difference (A-B) corresponding solely to the signal of the ferrous bheme, as discussed in the text. Note that the vertical scales below and above 515 nm differs by a factor of two. Reproduced from Ref. [6] with permission from the Royal Society of Chemistry.

**Figure 11.** Photo-dissociation of the *c*1-heme (left) and photo-oxidation of the b-heme (right) upon light excitation of the cyt *bc*1 complex. Reproduced from Ref. [6] with permission from the Royal Society of Chemistry.

The *b*-heme's high electronic reactivity makes sense in the light of cyt *bc*1 having to efficiently fulfil its role in the Q-cycle: it favours the reduction and oxidization of the ubiquinone and ubiquinol, respectively.[22] The *b*-hemes have to efficiently "process" the electrons, which demand them to easily loose or gain electron. Similar electronic-reactivity would in fact be counterproductive in soluble cytochromes as they would less efficiently keep their electrons from being scavenged by other solutes. The hydrophilic environment of the *bc*1 core on the other hand preserves the *b*-hemes from unwanted solvated electron carrier and their high electronic reactivity is then an advantage. In contrast to the high electronic reactivity of the *b*hemes, the high photo-dissociation quantum yield of the *c*1-hemes can be understood as being an efficient "heat sink" that protects the reduced state of the heme against light excitations.

Overall, this study illustrate that, even though the *b*-hemes in cyt *bc*1 and in other cyt *b* have similar ligation to their protein backbone; specific structural constraints and amino-acid arrangements result in clearly different responses, and therefore functions. While cytochromes were known to serve only as electron carriers, this study demonstrates that with the appro‐ priate environment, light-induced charge separation can readily be initiated within single heme structures. The use of the microfluidic flow-cell therefore not only enabled the study of this rare protein complex, but allowed to shine light on the relevance of local heme-bonding and structural environment in initiating larger chemical reactions. This particular case study is only one example of how the field of ultrafast spectroscopy can benefit from the application of microfluidics technologies.

#### **4.3. Call for development**

**Figure 11.** Photo-dissociation of the *c*1-heme (left) and photo-oxidation of the b-heme (right) upon light excitation of

**Figure 10.** (A) Spectra at selected time delays while both *b*- and *c*1-hemes are reduced. (B) Spectra at selected time de‐ lays while only the c1-hemes are left reduced. (C) Difference (A-B) corresponding solely to the signal of the ferrous bheme, as discussed in the text. Note that the vertical scales below and above 515 nm differs by a factor of two.

Reproduced from Ref. [6] with permission from the Royal Society of Chemistry.

128 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

the cyt *bc*1 complex. Reproduced from Ref. [6] with permission from the Royal Society of Chemistry.

As I mentioned, the described microfluidic flow-cell, beside the numerous advantages it provides in respect to the other systems commonly employed, is certainly not perfect, which leaves room for improvement.

For example, in terms of sample volumes, most of the liquid is contained in the pump that is used to generate the flow and in the tubing. Miniaturization of both would allow to use even smaller sample volumes. Already the small turbisc pump that is employed uses a newly developed friction based technologies[20] that is able to flow even viscous samples. The samples studied in ultrafast spectroscopy are however water-like and, not being in need of the actual viscous sample capabilities of the pump. I can therefore imagine that a simplified version of the pump would be sufficient and require even smaller volumes. Concerning the actual 1 mm diameter tubing, that links the pump with the capillary, it could be replaced by other microfluidics technologies that uses micro-channels that are specifically designed for spectro‐ scopy.[26] The goal through these proposed improvements is to reduce the amount of liquid used, keeping in mind that reducing further the diameter of the channels might impede the overall flow rate.

Another field of improvement that I see is that of the control over the inner atmosphere of the cell in which the sample is enclosed. As I mentioned, the actual pumping is done by friction between a grooved barrel and the sample. The spinning motion is however done via a shaft and under high pressure difference between the inner and outer part of the cell, either the sample can leak through or air can be sucked in via the shaft's joints. In order to palliate to this issue, I can imagine that the shaft would be replaced, not by another mechanical interaction, but remote electromagnetical interaction so as to preserve the inner atmosphere of the sample. Such systems are already implemented for applications in biomedical for example.[27] Furthermore, the microfluidic flow-cell uses epoxy beads in order to hermetically fit the square capillary in the cylindrical tubing. Better seal could be achieved if the capillary themselves were to be made with initial beads on each sides such as it is the case for the larger commercially available flow-cells.[14]

At last, I would like to precise that, due to friction between the sample and the capillary, the capillary does not provide with a flow that is homogeneous. Sample that is the closest to the window has consequently lower flow rate and might not be refreshed for each laser pulse, being therefore subjected to photo-damage. One solution would be to employ the newly developed Electro-Osmotic flow systems that are used to generate homogeneous flows.[28]

In conclusion, the microfluidic flow-cell that I propose is specifically designed to fits the requirement imposed by the ultrafast spectroscopy of quantity-limited and sensitive sample, while it remains cost effective and easy to use. As an illustration of the flow-cell's effectiveness, I showed how its implementation enabled the study of the cytochrome *bc*1. More precisely, I was able to resolve the ultrafast electronic and nuclear motions that precedes some of the larger physiological function of the protein. The microfluidic flow-cell not only facilitates but also opens the door to the study of a whole range of samples that cannot be purified in large quantities such as it is the case for most organic compound that are extracted from living organisms.

The advantages of this microfluidic flow-cell over other pre-existing solutions are clear; as clear as there is room for development.

## **5. General conclusion**

The implementation of the microfluidic flow-cell to the field of ultrafast spectroscopy can be considered as one of the first attempts to bridge two communities, i.e. that of microfluidics and that of ultrafast spectroscopy. In this chapter I have first described the technique of ultrafast spectroscopy in order to show its requirements in terms of sample as well as the importance of such analytical tool when applied to the study of biological protein complexes. In particular, ultrafast transient spectroscopy became one of the essential approach for whoever desires to understand the local electronic and nuclear modifications that are at the origin of the larger physiological functions of proteins. I then exposed the advantages and inconveniences of the different techniques that are commonly used in order to flow liquid samples in front of the laser beams. The aim of this discussion is to better appreciate how the application of microfluidics technologies is able to responds to the challenges raised by the technique. In this aim, the recently developed microfluidic flow-cell is adequate as it requires only about 250 µL while generating flow rate that are suitable for high repetition rate laser systems. Its steady window and decantation chamber allow for an optimal time and spectral resolution. By providing direct access to the sample while running a single experiment and monitoring chemical changes in "real time", the microfluidic system enables studies that were otherwise not possible. The advantages of microfluidics over other usual systems are numer‐ ous but as it represents only one of the first attempt, there is ample room for improvement. In this regard, one of the goal of this chapter is to serve as an initial step in an effort to bridge the microfluidics community with that of ultrafast spectroscopy in order to foster new ideas, new applications and new perspectives.

## **Acknowledgements**

issue, I can imagine that the shaft would be replaced, not by another mechanical interaction, but remote electromagnetical interaction so as to preserve the inner atmosphere of the sample. Such systems are already implemented for applications in biomedical for example.[27] Furthermore, the microfluidic flow-cell uses epoxy beads in order to hermetically fit the square capillary in the cylindrical tubing. Better seal could be achieved if the capillary themselves were to be made with initial beads on each sides such as it is the case for the larger commercially

130 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

At last, I would like to precise that, due to friction between the sample and the capillary, the capillary does not provide with a flow that is homogeneous. Sample that is the closest to the window has consequently lower flow rate and might not be refreshed for each laser pulse, being therefore subjected to photo-damage. One solution would be to employ the newly developed Electro-Osmotic flow systems that are used to generate homogeneous flows.[28] In conclusion, the microfluidic flow-cell that I propose is specifically designed to fits the requirement imposed by the ultrafast spectroscopy of quantity-limited and sensitive sample, while it remains cost effective and easy to use. As an illustration of the flow-cell's effectiveness, I showed how its implementation enabled the study of the cytochrome *bc*1. More precisely, I was able to resolve the ultrafast electronic and nuclear motions that precedes some of the larger physiological function of the protein. The microfluidic flow-cell not only facilitates but also opens the door to the study of a whole range of samples that cannot be purified in large quantities such as it is the case for most organic compound that are extracted from living

The advantages of this microfluidic flow-cell over other pre-existing solutions are clear; as clear

The implementation of the microfluidic flow-cell to the field of ultrafast spectroscopy can be considered as one of the first attempts to bridge two communities, i.e. that of microfluidics and that of ultrafast spectroscopy. In this chapter I have first described the technique of ultrafast spectroscopy in order to show its requirements in terms of sample as well as the importance of such analytical tool when applied to the study of biological protein complexes. In particular, ultrafast transient spectroscopy became one of the essential approach for whoever desires to understand the local electronic and nuclear modifications that are at the origin of the larger physiological functions of proteins. I then exposed the advantages and inconveniences of the different techniques that are commonly used in order to flow liquid samples in front of the laser beams. The aim of this discussion is to better appreciate how the application of microfluidics technologies is able to responds to the challenges raised by the technique. In this aim, the recently developed microfluidic flow-cell is adequate as it requires only about 250 µL while generating flow rate that are suitable for high repetition rate laser systems. Its steady window and decantation chamber allow for an optimal time and spectral resolution. By providing direct access to the sample while running a single experiment and

available flow-cells.[14]

organisms.

as there is room for development.

**5. General conclusion**

Thanks to Professor S. Savikhin (Purdue University, USA) as well as to Professor M. Chergui (EPFL, Switzerland) from whose laboratory of Ultrafast Spectroscopy the different pictures are taken. The described microfluidic flow cell as well as the cyt *bc*1's study have been funded by the Swiss NSF via the NCCR:MUST, by the FP7 Marie Curie COFUND, by the Excellence Initiative of the German Federal and State Governments (EXC 294, BIOSS) and by the Deutsche Forschungsgemeinschaft (RTG 1976).

## **Author details**

Adrien A. P. Chauvet\*

Address all correspondence to: adrien.chauvet@unige.ch

GAP-Biophotonics, Geneva University, Geneva, Switzerland

## **References**


[21] Berry EA, Guergova-Kuras M, Huang L, Crofts AR. Sructure and Function of Cyto‐ chrome bc Complexes. Annu Rev Biochem. 2000;69:1005-75.

[6] Chauvet AAP, Al Haddad A, Kao W-C, van Mourik F, Hunte C, Chergui M. Photoinduced dynamics of the heme centers in cytochrome *bc*1. Phys Chem Chem Phys.

[7] Cerullo G, De Silvestri S. Ultrafast Optical Parametric Amplifiers. Rev Sci Instrum.

[8] Weiner AM. Femtosecond Pulse Shaping Using Spatial Light Modulators. Rev Sci

[9] Chauvet A, Sarrou J, Lin S, Romberger SP, Golbeck JH, Savikhin S, et al. Temporal and Spectral Characterization of the Photosynthetic Reaction Center from *Heliobacte‐*

[10] Rosspeintner A, Bernhard Lang B, Vauthey E. Ultrafast Photochemistry in Liquids.

[11] Mitra S. Sample Preparation Techniques in Analytical Chemistry. Winefordner JD,

[12] Berg JM, Tymoczko JL, Stryer L. The Purification of Proteins is an Essential First Step in Understanding their Function. In: Moran S, Hadler GL, Zimmerman P, editors. Bi‐

[13] Monni R, Al Haddad A, van Mourik F, Auböck G, Chergui M. Tryptophan-to-Heme

[14] Starna Cells. Linear Flow Cells, Type 48 series http://www.starnacells.com/d\_cells\_s/

[15] Arrell CA, Ojeda J, Sabbar M, Okell WA, Witting T, Siegel T, et al. A Simple Electron Time-of-Flight Spectrometer for Ultrafast Vacuum Ultraviolet Photoelectron Spectro‐

[16] A. K. Charge Transfer to Solvent Dynamics in Iodide Aqueous Solution Studied at

[17] Duffin AM, Saykally RJ. Electrokinetic Power Generation from Liquid Water Micro‐

[18] Savikhin S, Wells T, Song P-S, Struve WS. Ultrafast Pump-Probe Spectroscopy of Na‐

[19] Chauvet A, Tibiletti T, Caffarri S, Chergui M. A microfluidic flow-cell for the study of the ultrafast dynamics of biological systems. Rev Sci Instrum. 2014;85: 103118.

[20] Lisibach A, Casartelli E, Schmid N, editors. Flow Investigation in a Disk Micropump. ASME 2010 3rd Joint US-European Fluids Engineering Summer Meeting and 8th In‐ ternational Conference on Nanochannels, Microchannels, and Minichannels; 2010;

Electron Transfer in Ferrous Myoglobins. PNAS. 2015;112(18):5602-06.

scopy of Liquid Solutions. Rev Sci Instrum. 2014;85(103117).

tive Etiolated Oat Phytochrome. Biochemistry. 1993;32:7512-8.

Ionization Threshold: Freie Universität Berlin; 2015.

jets. J Phys Chem. 2008;112:17018-22.

Montreal, Quebec, Canada.

2014.

2003;74(1).

Instrum. 2000;71(1929).

*rium modesticaldum*. Photosynth Res. 2013;116:1-9.

132 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

ochemistry. 5th edition ed: W H Freeman; 2002.

Annu Rev Phys Chem. 2013;64:247-71.

editor: John Wiley & sons, INC.; 2003.

flow/linear/T048.html2015.


## **Flow-Scanning Microfluidic Imaging**

Nicolas Pégard, Chien-Hung Lu, Marton Toth, Monica Driscoll and Jason Fleischer

Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/64707

#### **Abstract**

The advantages of microfluidics for fast analysis of microscopic suspensions have led to the commercial development of flow cytometers. In this chapter, we propose new micro‐ scopy methods that combine controlled motion of micro-organisms in a laminar micro‐ fluidic flow, optics, and computation. We propose three new imaging modalities. We first introduce a flow-based version of structured illumination microscopy, where the necessa‐ ry phase shifts are no longer obtained by controlled displacement of the illumination pat‐ tern but by flowing the sample itself. Then, we propose a three-dimensional (3D) deconvolution microscopy method with a microfluidic device for continuous acquisition of gradually defocused images. Finally, we introduce a microfluidic device for phasespace image acquisition, and computational methods for the reconstruction of either phase of intensity, in 3D. The imaging modalities we introduce all retain the benefits of fluid systems for noninvasive bioimaging. The proposed devices can easily be integrated on existing microscopes as a modified microscope slide, or on flow cytometers, and aquatic imagers with minor adjustments. Alternative on-chip implementations are also possible, with lens-free devices, and near-field optical and microfluidic elements directly assembled on the surface of a CCD (Charge-Coupled Device) or CMOS (Complementary metal–oxide–semiconductor) chip.

**Keywords:** lab on-a-chip, microscopy, computational imaging, optofluidics, cytometry, structured illumination, deconvolution, light field, tomography

## **1. Introduction**

#### **1.1. Microfluidic structured illumination microscopy**

The resolution of an optical imaging system is subject to the diffraction limit, which for a fixed wavelength is governed by the numerical aperture (NA) of the system. A popular technique

© 2016 The Author(s). Licensee InTech. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

to go beyond these limits is structured illumination (SI) [1‒7], in which a known illumination pattern (usually periodic) is projected onto a sample. Spectral beating of this pattern with the object modes folds high-resolution information into lower spatial frequencies (Moiré patterns) that can be detected by the imaging device. Deconvolving this photonic aliasing can improve resolution by a factor of two [1‒6]. Greater improvements are possible using nonlinearity [7] or with successive applications of structured illumination with higher spatial frequency patterns. Alternatively, structured illumination may be viewed as a type of diffraction, in which the signal is shifted in the spectral frequency domain by an amount equal to the applied grating period (see Figure 1). This imaging technique, however, requires the acquisition of several raw images (at least three) with a series of precise displacements of the illumination pattern in order to remove phase ambiguity. Previous SI systems relied on mechanical moving parts (e.g., piezoelectric actuators) [4] or on a spatial light modulator (SLM) [8] to perform the shift. These methods add complexity to the imaging system and can significantly reduce the image acquisition speed. Further, mechanical movement is subject to vibration error and artifacts, while SLMs are limited by their pixel size.

**Figure 1.** The principle of structured illumination microscopy is to increase imaging resolution beyond the limit of a microscope objective. An optical signal carries a given distribution of spatial frequencies (red curve), but the range of frequencies that can be collected is limited by the numerical aperture (NA) of the optical system (dashed blue line). Illuminating the sample with structured light that is patterned with a spatial frequency *k*<sup>g</sup> shifts the distribution and brings higher spatial frequencies back into the NA-limited imaging domain.

Separate from the structured illumination approach, fluidic imaging systems for improved resolution have been developed. In microfluidic microscopes and aquatic imagers, these systems have received renewed attention with the development of integrated optofluidic devices, which are lensless imagers that place flowing samples directly over a detector [9‒11]. Among their advantages is simple and low-cost object manipulation, with little or no sample preparation. In most devices, the flow is used only to provide object throughput (e.g., to measure gas kinetics [12], live cells [13], or two-phase flow [14]). However, the flow can be used as an additional degree of freedom for imaging. For example, recent work has used fluid transport as a scanning mechanism to enhance resolution, either by using small holes before the detector [9] or by taking multiple frames with subpixel displacements [10]. In the former method, the small apertures limit the amount of light captured and greatly reduce the effective recording area. In the latter method, resolution is limited by the camera frame rate (vs. flow speed) and edge effects from pixels. In all cases, the illumination was kept as uniform as possible.

Here, we combine a steady illumination pattern and use the fluid flow to provide the necessary scanning that shifts the phase of the illumination pattern with respect to the object. From a flow perspective, the instant wavenumber (*k*-space) shift gives improved spatial resolution at greater speeds than subpixel methods, with better use of the camera's dynamic range. The combined scheme thus retains all the benefits of fluidics, including high sample throughput and object sorting [15], while enabling easy integration with existing microscopes, flow cytometers, and aquatic imaging systems.

#### **1.2. Reconstruction algorithm for flow-based structured illumination**

to go beyond these limits is structured illumination (SI) [1‒7], in which a known illumination pattern (usually periodic) is projected onto a sample. Spectral beating of this pattern with the object modes folds high-resolution information into lower spatial frequencies (Moiré patterns) that can be detected by the imaging device. Deconvolving this photonic aliasing can improve resolution by a factor of two [1‒6]. Greater improvements are possible using nonlinearity [7] or with successive applications of structured illumination with higher spatial frequency patterns. Alternatively, structured illumination may be viewed as a type of diffraction, in which the signal is shifted in the spectral frequency domain by an amount equal to the applied grating period (see Figure 1). This imaging technique, however, requires the acquisition of several raw images (at least three) with a series of precise displacements of the illumination pattern in order to remove phase ambiguity. Previous SI systems relied on mechanical moving parts (e.g., piezoelectric actuators) [4] or on a spatial light modulator (SLM) [8] to perform the shift. These methods add complexity to the imaging system and can significantly reduce the image acquisition speed. Further, mechanical movement is subject to vibration error and

**Figure 1.** The principle of structured illumination microscopy is to increase imaging resolution beyond the limit of a microscope objective. An optical signal carries a given distribution of spatial frequencies (red curve), but the range of frequencies that can be collected is limited by the numerical aperture (NA) of the optical system (dashed blue line). Illuminating the sample with structured light that is patterned with a spatial frequency *k*<sup>g</sup> shifts the distribution and

Separate from the structured illumination approach, fluidic imaging systems for improved resolution have been developed. In microfluidic microscopes and aquatic imagers, these systems have received renewed attention with the development of integrated optofluidic devices, which are lensless imagers that place flowing samples directly over a detector [9‒11]. Among their advantages is simple and low-cost object manipulation, with little or no sample preparation. In most devices, the flow is used only to provide object throughput (e.g., to measure gas kinetics [12], live cells [13], or two-phase flow [14]). However, the flow can be used as an additional degree of freedom for imaging. For example, recent work has used fluid transport as a scanning mechanism to enhance resolution, either by using small holes before the detector [9] or by taking multiple frames with subpixel displacements [10]. In the former method, the small apertures limit the amount of light captured and greatly reduce the effective recording area. In the latter method, resolution is limited by the camera frame rate (vs. flow speed) and edge effects from pixels. In all cases, the illumination was kept as uniform as

artifacts, while SLMs are limited by their pixel size.

136 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

brings higher spatial frequencies back into the NA-limited imaging domain.

possible.

In standard applications of SI, the illumination is simply a periodic pattern which is displaced (phase-shifted) by a convenient amount, such as a quarter wavelength (*π*/2 in phase). In a fluidic system, however, the phase shift between frames will depend on the flow velocity and the camera frame rate. For that reason, we first consider the problem of an arbitrary phase shift of the illumination pattern between two consecutive frames. Let *A* be the object to be imaged and *P* be the point-spread function. The *p*th recorded image is given by

$$I\_p\left(\mathbf{x}\right) = \int P\left(\mu\right) A\left(\mathbf{x} + \mu\right) M\left(\mathbf{x} + \mu - p\delta\right) d\mu\tag{1}$$

where we allow for a non-ideal displacement, *δ,* due to fluid flow, and where *M* is the structured illumination pattern. To generate a fixed illumination pattern, we consider the 0th and ±1st order mode of a diffraction grating so that the structured illumination pattern *<sup>M</sup>* (*x*)= <sup>|</sup> *<sup>g</sup>* <sup>+</sup> <sup>1</sup> <sup>2</sup> *<sup>e</sup>iβ*⋅*<sup>x</sup>* <sup>+</sup> <sup>1</sup> <sup>2</sup> *<sup>e</sup>* <sup>−</sup>*iβ*⋅*<sup>x</sup>* <sup>|</sup> <sup>2</sup> <sup>=</sup> (*<sup>g</sup>* <sup>2</sup> <sup>+</sup> <sup>1</sup> <sup>2</sup> ) <sup>+</sup> <sup>2</sup>*g*cos(*<sup>β</sup>* <sup>⋅</sup> *<sup>x</sup>*)+ <sup>1</sup> <sup>2</sup> cos(2*β* ⋅ *x*), where *g* is the ratio of the grating efficiency between the 0th and ±1st-order diffraction. Here, we neglect the last term, assuming that the resolving power of the imaging system is not enough to detect the Moiré patterns caused by the double frequency 2*β*. In the following derivation, we use the form *<sup>M</sup>* (*x*)= *<sup>U</sup>* <sup>+</sup> 2cos(*<sup>β</sup>* <sup>⋅</sup> *<sup>x</sup>*), where *<sup>U</sup>* <sup>=</sup> *<sup>g</sup>* <sup>+</sup> <sup>1</sup> <sup>2</sup>*<sup>g</sup>* .

Fourier transforming the convolution product in Eq. (1) gives

$$
\tilde{X}\_p(k) = \tilde{P}(k) \Big[ \mathcal{U}\tilde{A}(k) + \gamma^p \tilde{A}\_\rho(k) + \gamma^{-p} \tilde{A}\_{-\rho}(k) \Big] \tag{2}
$$

where *X*˜ *<sup>p</sup>*(*k*) = *eipδ*⋅*<sup>k</sup> <sup>I</sup>* ˜ *<sup>p</sup>*(*k*), *γ* = *eipδ*⋅*<sup>β</sup>* , and *A*˜ <sup>±</sup>*β*(*k*)= *<sup>A</sup>*˜(*<sup>k</sup>* <sup>∓</sup> *<sup>β</sup>*). For simplicity, we use the optical transfer function *P*˜(*k*)=1 for ∥*<sup>k</sup>* <sup>∥</sup> <sup>&</sup>lt;*k*0 and *P*˜(*k*)=0 elsewhere, where *k*<sup>0</sup> is the maximum spatial frequency that can be resolved. Using three consecutive frames, we deduce

$$
\begin{bmatrix}
\tilde{X}\_{p+1} - \tilde{X}\_p \\
\tilde{X}\_p - \tilde{X}\_{p-1}
\end{bmatrix} = \tilde{P}\begin{pmatrix}
k \end{pmatrix} \cdot \begin{pmatrix}
\gamma^p & \gamma^{-p-1} \\
\gamma^{p-1} & \gamma^{-p}
\end{pmatrix} \begin{bmatrix}
\tilde{A}\_\beta\begin{pmatrix}k\\k\end{pmatrix} \\
\tilde{A}\_{-\beta}\begin{pmatrix}k\end{pmatrix}
\end{bmatrix} \tag{3}
$$

which can be inverted to reconstruct the object field with extended resolution *AE* :

$$A\_E\left(\mathbf{x}\right) = \mathcal{R}\begin{bmatrix} I\_{p+1}\left(\mathbf{x} + \boldsymbol{\delta}\right) \\ I\_p\left(\mathbf{x}\right) \\ I\_{p-1}\left(\mathbf{x} - \boldsymbol{\delta}\right) \end{bmatrix} \tag{4}$$

where

$$\mathcal{R} = \frac{\mathcal{Y}}{(\mathcal{Y}-1)^2} \begin{bmatrix} \frac{\mathcal{Y}^{1-p}}{\mathcal{Y}+1} e^{-i\boldsymbol{\beta}\cdot\boldsymbol{x}} + \frac{\mathcal{Y}^p}{\mathcal{Y}+1} e^{i\boldsymbol{\beta}\cdot\boldsymbol{x}} - \frac{1}{\mathcal{U}} \\\\ -\mathcal{Y}^{-p} e^{-i\boldsymbol{\beta}\cdot\boldsymbol{x}} - \mathcal{Y}^p e^{i\boldsymbol{\beta}\cdot\boldsymbol{x}} + \frac{\mathcal{Y}^2+1}{\mathcal{Y}\mathcal{U}} \\\\ \frac{\mathcal{Y}^{-p}}{\mathcal{Y}+1} e^{-i\boldsymbol{\beta}\cdot\boldsymbol{x}} + \frac{\mathcal{Y}^{1+p}}{\mathcal{Y}+1} e^{i\boldsymbol{\beta}\cdot\boldsymbol{x}} - \frac{1}{\mathcal{U}} \end{bmatrix} \tag{5}$$

In Section 4, we provide a detailed derivation of this result.

**Figure 2.** A 500 µm wide, 50 µm deep fluidic channel is at the focal plane of a 20× objective lens. The imaging system has low numerical aperture (NA = 0.1). The structured illumination source is a steady sinusoidal profile (2.78 µm stripes) orthogonal to the flow direction. Images are recorded on a CCD camera at a frame rate of 15 frames/s (fps).

#### **1.3. Experimental setup**

The experimental setup is shown in Figure 2. The microfluidic channel is a 500 µm wide, 50 µm deep groove etched on a glass slide. To generate the structured light, a 532 nm continuous laser is patterned using a transmission grating and then demagnified to reduce the fringe spacing. This light then illuminates the channel with a steady sinusoidal pattern: 2.78 µm stripes oriented orthogonal to the fluid flow direction. While the SI technique will work for any imaging system, including lensless ones, we place the channel at the focal plane of an optical microscope. As a compromise between magnification and field of view, e.g. for water analysis, we use a 20× optical objective. The objective is part of a 4*f* imaging configuration with an aperture located at the confocal plane. The resulting value of the numerical aperture (NA = 0.1) corresponds to a resolution limit of approximately 4 µm.

which can be inverted to reconstruct the object field with extended resolution *AE* :

*E p*

1


g

g

g

= - -+ -

g

g

2

( 1)

g

In Section 4, we provide a detailed derivation of this result.

g

*Ax Ix*

( ) ( ) ( )

1

*<sup>t</sup> p p i x i x*

> b

d

d

2

g

 g

 b

 b

+ - + + ë û

1

*U*

1

1

*U*

*U*

R (5)

+

R (4)

1

é ù <sup>+</sup> ê ú <sup>=</sup> ê ú

+

*I x*

*p*

*p*

1

1 1

b

b

*e e*

 g

 g


é ù ê ú + - + +

*p p ix ix*

 g

*e e*


1 1

**Figure 2.** A 500 µm wide, 50 µm deep fluidic channel is at the focal plane of a 20× objective lens. The imaging system has low numerical aperture (NA = 0.1). The structured illumination source is a steady sinusoidal profile (2.78 µm stripes) orthogonal to the flow direction. Images are recorded on a CCD camera at a frame rate of 15 frames/s (fps).

The experimental setup is shown in Figure 2. The microfluidic channel is a 500 µm wide, 50 µm deep groove etched on a glass slide. To generate the structured light, a 532 nm continuous laser is patterned using a transmission grating and then demagnified to reduce the fringe spacing. This light then illuminates the channel with a steady sinusoidal pattern: 2.78 µm

*e e*

 g

 g

*p p i x i x*


b


*I x*

ê ú - ê ú ë û

( )

138 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

where

**1.3. Experimental setup**

As a proof of principle, we flow a suspension of yeast particles in glycerol through the microfluidic channel. Multiple images are recorded by a CCD camera (pixel size 9.9 µm) at a constant frame rate (15 fps). Figure 3(a)-(c) shows three consecutive frames. It is clear that different features of the object are revealed as it flows past the stationary illumination pattern.

**Figure 3.** Three consecutive frames of two yeast particles under structured illumination. The constant phase shift (*δφ* = 0.28*π*) of the illumination pattern between consecutive frames shows the evolution of the recorded Moiré pattern.

For numerical reconstruction, we first assume that particles flowing along the center of the microfluidic channel undergo a negligible amount of rotation during three consecutive frames time (Poiseuille-type laminar flow with negligible on-axis velocity shear). The displacement of the object between two consecutive frames, *δ*, is then determined by the maximization of the cross-correlation function:

$$C\left(\mu\right) = -\left[\left(I\_1\left(\mathbf{x}\right) - I\_2\left(\mathbf{x} - \mu\right)\right)^2 d\mathbf{x}\right] \tag{6}$$

For the conditions here, we obtain *δ* =0.8 µm, which corresponds to a flow velocity of 12 µm / s, and a phase shift of *δϕ* =0.28*π* between consecutive frames. We note that when the spatial frequency of the illumination pattern |*β* | <2*k*0, it is necessary to consider the overlap among the modes *A*˜(*k*), *A*˜ *<sup>β</sup>*(*<sup>k</sup>* <sup>+</sup> *<sup>β</sup>*) and *A*˜ <sup>−</sup>*β*(*k* −*β*). We conduct an average reconstruction among all overlaps to evaluate the corresponding spatial frequencies.

Experimental results are shown in Figure 4. The yeast image without structured illumination is shown in Figure 4(a), and numerical reconstruction of the yeast particles using Eq. (4) is shown in Figure 4(c). It is clear that SI provides greater visibility and reveals more details than the uniform illumination image, even for a 1D illumination pattern. The amount of improve‐

**Figure 4.** Flow-scanning structured illumination reconstruction of yeast cells. (a) Reference image with uniform illumi‐ nation (without SI). (b) Reconstructed yeast particles using SI, calculated with the three consecutive images of Figure 3(a)‒(c) and Eq. (4). (c) Fourier spectrum of (a), log scale (d) Fourier spectrum of (b), log scale. (e and f) Line profiles of the cross-sections shown in (a and b) and (c and d), respectively.

ment can be quantified using the visibility *V* = (*I*max − *I*min)*/*(*I*max + *I*min), where *V* = 0.15 corresponds to the Rayleigh resolution criterion. Figure 4(e) shows cross-sections of the intensity along the line connecting the two particles. For uniform illumination, the image of the left yeast particle is well below the Rayleigh limit (*V*left = 0*.*07) while the right particle is barely visible (*V*right = 0*.*18). In contrast, the respective visibilities increase to 0*.*48 and 0*.*50 with structured illumination. The results shown in Figure 4(e) indicate that the left and right yeast particles are approxi‐ mately 2 and 3 µm in diameter, respectively, with a center-of-mass separation of 4 µm. These measurements would not be possible using only uniform illumination, which is limited to the bare system resolution (4 µm). This ability to discriminate using a single criterion, such as Rayleigh criterion, is necessary for many applications, e.g., automated identification and classification.

Another metric of improvement follows from a measurement of the spatial frequency spec‐ trum. Figure 4(b) and (d) shows the magnitude of Fourier transform of Figure 4(a) and (c), respectively (in log scale). Compared to the uniform illumination case of Figure 4(b), Figure 4(d) displays many more spatial frequencies along the flow direction. According to the line profiles in Figure 4(f), structured illumination provides twice as many spatial frequencies along the illumination pattern direction, *kx*, in agreement with linear theory [1‒5] and the *x*space observations. In terms of *k*-space area, we measure a coverage ratio of 2*.*3, corresponding to an equivalent radius ratio (=*k*extended*/k*0) of 1*.*5. This is a significant improvement considering that only a 1D illumination pattern was used.

#### **1.4. Structured illumination with non-ideal phase shifts**

Structured illumination for super-resolution microscopy has been extensively developed. To date, all techniques have however relied on precisely controlled phase shifts of the illumination grating. Here, we consider the case of microfluidic structured illumination, where the illumi‐ nation pattern remains steady, and where the phase shifts are induced by displacing the object at constant speed in a microfluidic channel above a fixed illumination grating. This approach does not allow precise control of the inter frame phase shifts and requires a more complex reconstruction technique. For this, we derive this solution and show simulation results for proof of concept. We define the following notations.


$$\stackrel{\bullet}{\quad} \text{\textquotedblleft} \text{\textquotedblright} \text{\textquotedblleft} \text{\textquotedblright} \text{\textquotedblright} \text{\textquotedblright} \text{\textquotedblleft} \text{\textquotedblleft} \text{\textquotedblright} \text{\textquotedblleft} \text{\textquotedblleft} \text{\textquotedblright} \text{\textquotedblleft} \text{\textquotedblleft} \text{\textquotedblleft} \text{\textquotedblleft} \text{\textquotedblleft} \text{\textquotedblleft} \text{\textquotedblleft} \text{\textquotedblleft} \text{\textquotedblleft} \text{\textquotedblleft} \text{\textquotedblleft} \text{\textquotedblleft} \text{\textquotedblleft} \text{\textquotedblright} \text{\textquotedblright} \text{\textquotedblleft} \text{\textquotedblleft} \text{\textquotedblleft} \text{\textquotedblleft} \text{\textquotedblleft} \text{\textquotedblleft} \text{\textquotedblleft} \text{\textquotedblleft} \text{\textquotedblleft} \text{\textquotedblleft} \text{\textquotedblleft} \text{\textquotedblleft} \text{\textquotedblleft} \text{\textquotedblleft} \text{\textquotedblleft} \text{\textquotedblleft} \text{\textquotedblleft} \text{\textquotedblleft} \text{\textquotedblleft} \text{\textquotedblleft} \text{\textquotedblright}} \text{\textquotedblleft} \text{\textquotedblleft} \text{\textquotedblleft} \text{\textquotedblleft} \text{\textquotedblleft} \text{\textquotedblleft} \text{\textquotedblleft} \text{\textquotedblleft} \text{\textquotedblleft} \text{\textquotedblleft} \text{\textquotedblleft} \text{\textquotedblleft} \text{\textquotedblleft} \text{\textquotedblleft} \text{\textquotedblleft} \text{\textquotedblleft} \text{\textquotedblleft} \text{\textquotedblright}} \text{\textquotedbl$$

**•** " . " is the scalar multiplication.

ment can be quantified using the visibility *V* = (*I*max − *I*min)*/*(*I*max + *I*min), where *V* = 0.15 corresponds to the Rayleigh resolution criterion. Figure 4(e) shows cross-sections of the intensity along the line connecting the two particles. For uniform illumination, the image of the left yeast particle is well below the Rayleigh limit (*V*left = 0*.*07) while the right particle is barely visible (*V*right = 0*.*18).

the cross-sections shown in (a and b) and (c and d), respectively.

140 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

**Figure 4.** Flow-scanning structured illumination reconstruction of yeast cells. (a) Reference image with uniform illumi‐ nation (without SI). (b) Reconstructed yeast particles using SI, calculated with the three consecutive images of Figure 3(a)‒(c) and Eq. (4). (c) Fourier spectrum of (a), log scale (d) Fourier spectrum of (b), log scale. (e and f) Line profiles of


**•** " *k* " represents a wavevector in the spatial frequency space (Fourier space).

Structured Illumination enables the reconstruction of extended resolution beyond the numer‐ ical aperture of a given imaging device. By illuminating the sample with a known high definition pattern, it is possible to wrap high spatial frequencies into visible large-scale Moiré patterns. Structured illumination is a well-known imaging technique [16, 17]. Current algo‐ rithms rely on pattern phase shifting by displacement of the illumination source. Here, instead, OFM microscopy leads us to displace the sample. Furthermore, because flow measurement is extremely difficult in a fluid channel, the phase shift between frames is not well controlled. In this paragraph, we solve the problem of extended resolution reconstruction for an arbitrary object displacement *δ* between frames.

Let *v* be the speed of the flow, and *T* be the time between frames, so that *δ* =*vT* is the displace‐ ment of the sample between two frames (assumed to be constant). Let *β* be the wavevector of the illumination pattern. Assuming constant flow speed, and constant image rate, we may write that the *p* th recorded signal is given by

$$I\_p\left(\mathbf{x}\right) = PSP\left(\mathbf{x}\right) \otimes \mathcal{A}\left(\mathbf{x} - p\delta\right) \mathcal{M}\left(\mathbf{x}\right) \tag{7}$$

where

$$\mathcal{LM}\left(\mathbf{x}\right) = \mathcal{U} + e^{i\boldsymbol{\beta}\cdot\mathbf{x}} + e^{-i\boldsymbol{\beta}\cdot\mathbf{x}} \tag{8}$$

In *k* space, this corresponds to

$$\tilde{I}\_{\rho}\begin{pmatrix}k\end{pmatrix} = \text{OTF}\begin{pmatrix}k\end{pmatrix} \cdot \left(\mathcal{F}\begin{bmatrix}\mathcal{A}\begin{pmatrix}\mathbf{x}-p\boldsymbol{\delta}\end{pmatrix}\end{pmatrix} \otimes \mathcal{F}\begin{bmatrix}\mathcal{M}\begin{pmatrix}\mathbf{x}\end{pmatrix}\end{bmatrix}\right) \tag{9}$$

where

$$\mathcal{F}\left[\mathcal{M}\left(\mathbf{x}\right)\right] = \mathcal{U}\delta\left(k\right) + \delta\left(k-\beta\right) + \delta\left(k+\beta\right) \tag{10}$$

and

$$\mathcal{F}\left[\mathcal{A}\left(\mathbf{x} - p\delta\right)\right] = e^{-ip\delta \cdot k} \tilde{\mathcal{A}}(k) \tag{11}$$

Defining *X*˜ *<sup>p</sup>*(*k*)= *I* ˜ *<sup>p</sup>*(*k*)*eipδ*⋅*<sup>k</sup>* and *γ* =*eiδ*⋅*<sup>β</sup>* Eq. (9) becomes

$$\tilde{X}\_p(k) = \text{OTF}(k) \cdot \left( \iota L\tilde{\mathcal{A}}(k) + \gamma^p \tilde{\mathcal{A}}(k-\beta) + \gamma^{-p} \tilde{\mathcal{A}}(k+\beta) \right) \tag{12}$$

$$
\tilde{X}\_{p+1}\left(k\right) - \tilde{X}\_p\left(k\right) = \text{OTF}\left(k\right) \cdot \left(\mathcal{Y}^p\left(\mathcal{Y} - 1\right) \tilde{\mathcal{A}}\left(k - \beta\right) + \mathcal{Y}^{-p}\left(\frac{1}{\mathcal{Y}} - 1\right) \tilde{\mathcal{A}}\left(k + \beta\right)\right)
$$

Assuming ∀*k*, |*k* | <*k*0, so that OTF(*k*)=1, we deduce that ∀ *p* ∈ℤ, and ∀ |*k* | <*k*<sup>0</sup>

$$
\tilde{\mathcal{A}}\left(k-\beta\right) = \frac{\mathcal{Y}^{1-p}}{\left(\mathcal{Y}-1\right)^{2}\left(\mathcal{Y}+1\right)} \left[\mathcal{Y}\left(\tilde{\mathcal{X}}\_{p+1}\left(k\right)-\tilde{\mathcal{X}}\_{p}\left(k\right)\right)-\left(\tilde{\mathcal{X}}\_{p}\left(k\right)-\tilde{\mathcal{X}}\_{p-1}\left(k\right)\right)\right]
$$

$$
\tilde{\mathcal{A}}\left(k+\beta\right) = \frac{\mathcal{Y}^{1+p}}{\left(\mathcal{Y}-1\right)^{2}\left(\mathcal{Y}+1\right)} \left[-\mathcal{Y}\left(\tilde{\mathcal{X}}\_{p}\left(k\right)-\tilde{\mathcal{X}}\_{p-1}\left(k\right)\right)+\left(\tilde{\mathcal{X}}\_{p+1}\left(k\right)-\tilde{\mathcal{X}}\_{p}\left(k\right)\right)\right]
$$

Writing *Jp*(*x*)= *Ip*(*x* + *pδ*), this becomes

**•** " *k* " represents a wavevector in the spatial frequency space (Fourier space).

142 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

object displacement *δ* between frames.

In *k* space, this corresponds to

where

where

and

Defining *X*˜

*<sup>p</sup>*(*k*)= *I*

˜ *<sup>p</sup>*(*k*)*eipδ*⋅*<sup>k</sup>*

write that the *p* th recorded signal is given by

Structured Illumination enables the reconstruction of extended resolution beyond the numer‐ ical aperture of a given imaging device. By illuminating the sample with a known high definition pattern, it is possible to wrap high spatial frequencies into visible large-scale Moiré patterns. Structured illumination is a well-known imaging technique [16, 17]. Current algo‐ rithms rely on pattern phase shifting by displacement of the illumination source. Here, instead, OFM microscopy leads us to displace the sample. Furthermore, because flow measurement is extremely difficult in a fluid channel, the phase shift between frames is not well controlled. In this paragraph, we solve the problem of extended resolution reconstruction for an arbitrary

Let *v* be the speed of the flow, and *T* be the time between frames, so that *δ* =*vT* is the displace‐ ment of the sample between two frames (assumed to be constant). Let *β* be the wavevector of the illumination pattern. Assuming constant flow speed, and constant image rate, we may

d

 b

d) ( ) ) = × -Ä é ùé ù

> bd

ë ûë û % FA FM (9)

 b( ) ( ) ( ) ë û F M (10)


 b

× -× M =+ + (8)

) ( ) (7)

*I x PSP x x p x <sup>p</sup>* ( ) = Ä- ( ) A M (

( ) *ix ix x Ue e* b

*Ik k xp x <sup>p</sup>* ( ) OTF( ) ( (

é ù (*x Uk k k* ) = + -+ + d

d

( ) OTF( ) ( ( ) ( ) ( )) *p p Xk k U k k k <sup>p</sup>* g

and *γ* =*eiδ*⋅*<sup>β</sup>*

 d

( ) ( ) *ip k xp e k* d

Eq. (9) becomes

 bg


$$\tilde{\mathcal{A}}\left(k+\beta\right) = \frac{\gamma^{1-p}}{\left(\gamma-1\right)^{2}\left(\gamma+1\right)} \mathcal{F}\left[\gamma I\_{p+1}\left(\mathbf{x}\right) - \left(\gamma+1\right)I\_{p}\left(\mathbf{x}\right) + I\_{p-1}\left(\mathbf{x}\right)\right] \tag{13}$$

$$\tilde{\mathcal{A}}\left(k-\beta\right) = \frac{\gamma^{1+p}}{(\gamma-1)^2\left(\gamma+1\right)} \mathcal{F}\left[J\_{p+1}\left(\mathbf{x}\right) - \left(\gamma+1\right)J\_p\left(\mathbf{x}\right) + \gamma J\_{p-1}\left(\mathbf{x}\right)\right] \tag{14}$$

$$\tilde{\mathcal{A}}\left(k\right) = \frac{\mathcal{Y}}{\mathcal{U}\left(\mathcal{Y}-1\right)^{2}} \mathcal{F}\left[-J\_{p+1}\left(\mathbf{x}\right) + \frac{\mathcal{Y}^{2}+1}{\mathcal{Y}} J\_{p}\left(\mathbf{x}\right) - J\_{p-1}\left(\mathbf{x}\right)\right] \tag{15}$$

where *δ* and *β* are experimental parameters that are either known or derived from the measurements. The time lapse between recorded frames, as well as the speed of the fluid, should be identical. By recording a series of images *Ip*, *p* ∈ 1, *N* , we may use Eqs. (13)–(15) in order to reconstruct the extended *k*-space.

Accordingly, the extended resolution AEXT becomes

$$
\widetilde{\mathcal{A}\_{\text{ext}}}\left(k\right) = \tilde{\mathcal{A}}\left(k+\beta\right) + \tilde{\mathcal{A}}\left(k\right) + \tilde{\mathcal{A}}\left(k-\beta\right) \tag{16}
$$

$$\begin{split} \widehat{\mathcal{A}\_{\text{Ext}}}\left(k\right) &= \frac{\mathcal{Y}}{\left(\mathcal{Y}-1\right)^{2}} \mathcal{F} \left[ \left(\frac{\mathcal{Y}}{\mathcal{Y}+1}a\_{+} + \frac{1}{\mathcal{Y}+1}a\_{-} - \frac{1}{\mathcal{U}}\right) l\_{p+1}\left(\mathbf{x}\right) \\ &- \left(a\_{+} + a\_{-} - \frac{\mathcal{Y}^{2}+1}{\mathcal{Y}} \frac{1}{\mathcal{U}}\right) l\_{p}\left(\mathbf{x}\right) \\ &+ \left(\frac{1}{\mathcal{Y}+1}a\_{+} + \frac{\mathcal{Y}}{\mathcal{Y}+1}a\_{-} - \frac{1}{\mathcal{U}}\right) l\_{p-1}\left(\mathbf{x}\right) \right] \end{split} \tag{17}$$

where *α*<sup>+</sup> <sup>=</sup>*<sup>γ</sup>* <sup>−</sup> *<sup>p</sup> e <sup>β</sup><sup>x</sup>* , and *α*<sup>−</sup> <sup>=</sup>*<sup>γ</sup> <sup>p</sup> e* <sup>−</sup>*β<sup>x</sup>* . Inverting the Fourier transform gives the result in Eqs. (4) and (5).

## **2. 3D microfluidic microscopy**

In this section, we present two applications of microfluidic flow for 3D microscopy. The first method uses a microfluidic channel that is tilted along the optical axis. We record several progressively defocused images of the flowing sample as it passes across the focal plane. The resulting focal stack is then processed using a Wiener deconvolution algorithm to generate three-dimensional images. Experimental results are shown on flowing yeast cells and reveal precise surface profile information. The second method is a 3D tomography device that combines a light source providing patterned illumination through a slit aperture, a microflui‐ dic channel, and a Fourier lens for simultaneous acquisition of multiple perspective angles in the phase-space domain. 3D absorption is retrieved using standard back-projection algo‐ rithms, here a limited-domain inverse Radon transform. Simultaneously, 3D differential phase contrast images are obtained by computational refocusing and asymmetric comparison of complementary illumination angles. We have implemented the technique on a compact glass slide. We demonstrate non invasive 3D phase contrast and absorption imaging capabilities on live, freely swimming *C. elegans*.

The microfluidic channel eliminates the need for a precise translation stage to control the extra degree of freedom required to acquire 3D images on a 2D sensor. Here, either with defocusing or flow scanning. In addition, high sample throughput in an insulated, nontoxic, liquid environment perfectly fits the usual requirements for bio-compatibility.

#### **2.1. 3D microfluidic microscopy using a tilted channel**

In the simplest description of an imaging system, with a fixed lens, the imaging condition ensures sharp images when an object is located at a particular depth called focal plane. Here, by tilting the channel [18], samples can descend through this plane, so that different crosssections will sequentially come into focus.

#### *2.1.1. Experimental setup*

The experimental device is presented in Figure 5. The microfluidic channel (500 µm width, 50 µm depth) is etched on a glass slide and placed under a standard wide-field microscope. The slide is tilted by a 15° angle with respect to the optical axis of the microscope objective. The tilt angle, *α*, is chosen to represent a good compromise between magnification and axial defocus‐ ing range. The channel is illuminated with incoherent white light, and the CMOS camera records a 25× magnified image at a constant frame rate, here: 30 fps. We demonstrate the principle with a suspension of yeast cells in glycerol flowing through the microfluidic channel.

The frame rate and the flow speed are selected to allow the acquisition of 100 consecutive frames as each sample passes from one end of the observation window to the other. A constant flow is maintained using a fixed pressure difference (a 50 cm hydrostatic water column) between the channel input and output. Exposure is adjusted to minimize the flow-induced blur, here below the resolution limit of the imaging device. With shallow channel depth, high kinematic fluid viscosity yields a Hagen‒Poiseuille-type laminar flow [19]. Also, the concen‐ tration of particles is lower than 250 µl−1 allowing for easy separation and minimal interactions between flowing particles.

**Figure 5.** (a) The experimental device is a microfluidic channel placed under the objective of a wide field microscope with a tilt angle of 15° along the optical axis. The sample is carried by the laminar flow along the channel at constant velocity with a static pressure difference. We record video data with a CMOS camera operating at a constant frame rate. (b) Samples, seen as they flow along the channel axis and pass across the focal plane.

#### *2.1.2. Controlled motion and rotation in a laminar liquid flow*

where *α*<sup>+</sup> <sup>=</sup>*<sup>γ</sup>* <sup>−</sup> *<sup>p</sup>*

and (5).

*e <sup>β</sup><sup>x</sup>*

**2. 3D microfluidic microscopy**

live, freely swimming *C. elegans*.

, and *α*<sup>−</sup> <sup>=</sup>*<sup>γ</sup> <sup>p</sup>*

*e* <sup>−</sup>*β<sup>x</sup>*

144 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

In this section, we present two applications of microfluidic flow for 3D microscopy. The first method uses a microfluidic channel that is tilted along the optical axis. We record several progressively defocused images of the flowing sample as it passes across the focal plane. The resulting focal stack is then processed using a Wiener deconvolution algorithm to generate three-dimensional images. Experimental results are shown on flowing yeast cells and reveal precise surface profile information. The second method is a 3D tomography device that combines a light source providing patterned illumination through a slit aperture, a microflui‐ dic channel, and a Fourier lens for simultaneous acquisition of multiple perspective angles in the phase-space domain. 3D absorption is retrieved using standard back-projection algo‐ rithms, here a limited-domain inverse Radon transform. Simultaneously, 3D differential phase contrast images are obtained by computational refocusing and asymmetric comparison of complementary illumination angles. We have implemented the technique on a compact glass slide. We demonstrate non invasive 3D phase contrast and absorption imaging capabilities on

The microfluidic channel eliminates the need for a precise translation stage to control the extra degree of freedom required to acquire 3D images on a 2D sensor. Here, either with defocusing or flow scanning. In addition, high sample throughput in an insulated, nontoxic, liquid

In the simplest description of an imaging system, with a fixed lens, the imaging condition ensures sharp images when an object is located at a particular depth called focal plane. Here, by tilting the channel [18], samples can descend through this plane, so that different cross-

The experimental device is presented in Figure 5. The microfluidic channel (500 µm width, 50 µm depth) is etched on a glass slide and placed under a standard wide-field microscope. The slide is tilted by a 15° angle with respect to the optical axis of the microscope objective. The tilt angle, *α*, is chosen to represent a good compromise between magnification and axial defocus‐ ing range. The channel is illuminated with incoherent white light, and the CMOS camera records a 25× magnified image at a constant frame rate, here: 30 fps. We demonstrate the principle with a suspension of yeast cells in glycerol flowing through the microfluidic channel. The frame rate and the flow speed are selected to allow the acquisition of 100 consecutive frames as each sample passes from one end of the observation window to the other. A constant

environment perfectly fits the usual requirements for bio-compatibility.

**2.1. 3D microfluidic microscopy using a tilted channel**

sections will sequentially come into focus.

*2.1.1. Experimental setup*

. Inverting the Fourier transform gives the result in Eqs. (4)

The velocity distribution of the flow (see Figure 6) is parabolic. Particles in suspension flow at a constant velocity along the channel axis. However, except for the central axis of the channel, they also experience shear-induced rotation. Here, the acquisition of accurate focal stacks relies on the absence of rotation (or its compensation), and our setup has been designed to minimize the effects of shear in all directions. Along the channel axis *u*, ∇*v* ⋅*u* =0, and the absence of rotation is a property of the laminar flow. Along the *y*-axis, we choose to only observe samples flowing in the middle part of a wide channel. Similarly, along the *z*-axis rotating objects are excluded by considering only particles flowing at the highest velocity in the middle of the channel, where the shear effects cancel. In practice, rotation can be minimized by injecting particles in the center of the channel with microfluidic injection on a separate channel. We note, however, that object rotation may be useful in other contexts, e.g., for multiple viewpoints, and is easily accessible by changing the injection point or imaging different parts of the flow.

**Figure 6.** Flow velocity and shear in a laminar flow for a rectangular channel section. Small particles propagate along the streamlines of the flow, and shear effects resulting from the liquid-channel edge interface induce rotations for sam‐ ples propagating away from the central axis.

#### *2.1.3. Extraction of gradually defocused images*

Focal stacks are generated by tracking samples flowing into the channel, as shown in Figure 7. The background noise is subtracted from the signal, and the zero value of the signal (in gray) corresponds to the nominal transparency of the free-running fluid. Letting *In*(*x,y*) be the intensity of the *n*th recorded frame. The focal stack, *S*, is given by

$$S\left(x, y, z\_0 + n\left|\delta\right|\sin(\alpha)\right) = I\_n\left(x + n\left|\delta\right|, y\right) \tag{18}$$

where *v* is the flow velocity, *T* , is the frame recording period, and *δ* =*Tv*, is the object displace‐ ment along the channel axis between two frames.

#### *2.1.4. Focal stack alignment using a defocusing invariant*

The extraction of accurate focal stacks is based on aligning successive views of the moving sample as it passes across the focal plane. We experimentally estimate this displacement using the fluid flow velocity and the frame rate, but we also correct for possible position errors using a simple method of particle tracking detailed based on an optical property of defocused images. We compute the center of mass of an image and show that in the case of defocused images of 3D objects with symmetric point-spread functions, the center of mass does not depend on the amount of defocusing [20] (see Figure 8).

**Figure 7.** We record focal stacks by observing samples in-motion passing through the tilted microfluidic channel and across the focal plane. With constant flow velocity and frame rate, we record ≈ 100 progressively defocused frames along the *z*-axis. We digitally track the object with an algorithm based on defocusing invariant properties of the center of gravity of the image. This allows perfect vertical alignment of the focal stack. (a) We measure iso-intensity contours of yeast cells through focus (≈36 µm). Insets show images of cells at selected levels. (b) Normalized intensity display of the focal stack with background subtracted.

#### *2.1.5. Wiener deconvolution*

however, that object rotation may be useful in other contexts, e.g., for multiple viewpoints, and is easily accessible by changing the injection point or imaging different parts of the flow.

146 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

**Figure 6.** Flow velocity and shear in a laminar flow for a rectangular channel section. Small particles propagate along the streamlines of the flow, and shear effects resulting from the liquid-channel edge interface induce rotations for sam‐

Focal stacks are generated by tracking samples flowing into the channel, as shown in Figure 7. The background noise is subtracted from the signal, and the zero value of the signal (in gray) corresponds to the nominal transparency of the free-running fluid. Letting *In*(*x,y*) be the

where *v* is the flow velocity, *T* , is the frame recording period, and *δ* =*Tv*, is the object displace‐

The extraction of accurate focal stacks is based on aligning successive views of the moving sample as it passes across the focal plane. We experimentally estimate this displacement using the fluid flow velocity and the frame rate, but we also correct for possible position errors using a simple method of particle tracking detailed based on an optical property of defocused images. We compute the center of mass of an image and show that in the case of defocused images of 3D objects with symmetric point-spread functions, the center of mass does not depend on the

 d

) *<sup>n</sup>* ( , ) (18)

*Sxyz n I x n y* ( , , sin ( ) <sup>0</sup> + =+ da

ples propagating away from the central axis.

*2.1.3. Extraction of gradually defocused images*

ment along the channel axis between two frames.

amount of defocusing [20] (see Figure 8).

*2.1.4. Focal stack alignment using a defocusing invariant*

intensity of the *n*th recorded frame. The focal stack, *S*, is given by

The reconstruction of the volume absorption distribution of the object, *O*, relies on solving the well-known deconvolution problem using the experimentally measured focal stack, *S*, and point-spread function, *P* in 3D space. *O* satisfies the volume integral

$$S(r) = \left[\mathcal{O} \otimes P\right](r) = \left[\mathcal{O}(r+r')P(r')dr'\right] \tag{19}$$

**Figure 8.** Samples are observed as they flow along the *ux*axis and pass across the focal plane. For each acquired image, we compute the center of mass, or first moment, of the image data (red cross). This quantity is a defocusing invariant and can be used to correct for motion correction and imperfections of the laminar flow.

where *r* = *x*.*x* + *y*.*y* + *z*.*z*. Ideally, the optical transfer function, ℱ *P* , is positive and the solution for Eq. (8) is given by

$$O \approx \mathcal{F}^{-1}\left[\frac{\mathcal{F}\left[\mathcal{S}\right]}{\mathcal{F}\left[\mathcal{P}\right]}\right] \tag{20}$$

Unfortunately, this solution is known to be extremely sensitive to experimental noise [21]. several noise-reducing techniques exist to facilitate object recovery when the point-spread function is not known, e.g., maximum likelihood estimation [22] and blind deconvolution [23]. In these methods, which in general are computationally complex, the point-spread function is guessed instead of measured. Adaptive measures [24] are particularly suited to flowing objects, and engineered point-spread functions [25], can be used as well. In this experiment, we choose to measure the point-spread function of the microscope directly by applying the same focal stack acquisition procedure introduced above to a suspension of submicron-sized reference particles. For this calibration, we use a sparse suspension of 800 nm dyed polystyrene beads in glycerol. Flow velocity, frame rate, and exposure conditions are identical as those selected for imaging the samples. We record a focal stack and track one of the flowing reference particles, align the defocused images using the defocusing invariant *Gn*, and check that the field of observation is clear of other flowing objects. The resulting stack, centered in the window of computation and normalized to a unitary absorption, represents the point-spread function of the microfluidic microscope for this particular tilt angle. While not done here, these reference particles can be embedded in the flow with the samples, as real-time reference points for changing conditions and/or shear compensation. Even with a known, precisely measured point-spread function, the inversion (Eq. (20)) is sensitive to zeros and noise in the measure‐ ment. To compensate for this, we use a Wiener deconvolution filter [26]. In this approach, we return to Eq. (19) and consider explicitly an additive noise term, *N*, which we assume to be independent from the signal.

Eq. (19) becomes

$$S(r) = \left[\bigcirc \otimes P\right](r) + N(r) \tag{21}$$

The Wiener filter finds the best deconvolution operator, *D*, so that the retrieved object *OR* :

$$\mathcal{O}\_{\mathbb{R}}\left(r\right) = \left[\left.D \otimes S\right]\right](r) \tag{22}$$

minimizes the RMS reconstruction error, *E*, given by

$$E = \left| \mathbf{O} - \mathbf{O}\_{\mathbb{R}} \right|^2 \tag{23}$$

Here, the Wiener solution, representing an optimal compromise between noise and resolution, is therefore given by

$$O\_{\mathbb{R}} \approx \mathcal{F}^{-1}\left[\frac{\mathcal{F}\left[\mathbb{S}\right]}{\mathcal{F}\left[\mathbb{P}\right] + \epsilon}\right] \tag{24}$$

where = <sup>|</sup> *<sup>N</sup>* <sup>|</sup> <sup>2</sup> <sup>|</sup> *<sup>S</sup>* <sup>|</sup> <sup>2</sup> =1.2 <sup>10</sup>−<sup>3</sup> , is a regularization constant corresponding to the inverse value of the signal-to-noise ratio. Here, the signal intensity is normalized to 1 and the root-mean-square value of the noise (background intensity) is measured and time-averaged in an empty area near the flowing object.

#### *2.1.6. Results*

where *r* = *x*.*x* + *y*.*y* + *z*.*z*. Ideally, the optical transfer function, ℱ *P* , is positive and the solution

**Figure 8.** Samples are observed as they flow along the *ux*axis and pass across the focal plane. For each acquired image, we compute the center of mass, or first moment, of the image data (red cross). This quantity is a defocusing invariant

<sup>1</sup> *S*

*P* - é ù é ù ë û » ê ú é ù ê ú ë û ë û F

Unfortunately, this solution is known to be extremely sensitive to experimental noise [21]. several noise-reducing techniques exist to facilitate object recovery when the point-spread function is not known, e.g., maximum likelihood estimation [22] and blind deconvolution [23]. In these methods, which in general are computationally complex, the point-spread function is guessed instead of measured. Adaptive measures [24] are particularly suited to flowing objects, and engineered point-spread functions [25], can be used as well. In this experiment, we choose to measure the point-spread function of the microscope directly by applying the same focal stack acquisition procedure introduced above to a suspension of submicron-sized reference particles. For this calibration, we use a sparse suspension of 800 nm dyed polystyrene beads in glycerol. Flow velocity, frame rate, and exposure conditions are identical as those selected for imaging the samples. We record a focal stack and track one of the flowing reference particles, align the defocused images using the defocusing invariant *Gn*, and check that the field of observation is clear of other flowing objects. The resulting stack, centered in the window of computation and normalized to a unitary absorption, represents the point-spread function of the microfluidic microscope for this particular tilt angle. While not done here, these reference particles can be embedded in the flow with the samples, as real-time reference points for changing conditions and/or shear compensation. Even with a known, precisely measured point-spread function, the inversion (Eq. (20)) is sensitive to zeros and noise in the measure‐ ment. To compensate for this, we use a Wiener deconvolution filter [26]. In this approach, we return to Eq. (19) and consider explicitly an additive noise term, *N*, which we assume to be

<sup>F</sup> (20)

*Sr O P r Nr* ( ) =Ä + é ù( ) ( ) ë û (21)

*O*

and can be used to correct for motion correction and imperfections of the laminar flow.

148 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

F

for Eq. (8) is given by

independent from the signal.

Eq. (19) becomes

Experimental results on flowing yeast cells are shown in Figure 9. The focal stack, *S*, and the point-spread function, *P*, are each processed using Eq. (24). We compute iso-level surface contours from the retrieved three-dimensional data. Figure 9(a) shows a these contours and their projection view along the optical axis, revealing small-scale surface features (≈1−2 µm), which are clearly resolved (though smoothed somewhat by the regularization process). These features most likely represent early-stage budding of the yeast cells, though other factors can also contribute to their specific morphology [27]. In Figure 9(b), we show contours from the side. All the cells lie in the same vertical plane (a result of the controlled injection), and each cell has flat side walls. This deformation is common in flowing cells [28] and is often used as a diagnostic tool [29]. Many details that are hidden in standard imaging using 2D projections, such as cell orientation, 3D shape, and surface roughness, are readily apparent in the volume images here.

#### **2.2. Microfluidic flow-scanning tomography**

In the previous section, we demonstrated 3D surface topology of flowing objects. In this section, we present methods for full-volume tomography [30]. Depending on the respective size of the microfluidic channel and the flowing objects, two methods for the acquisition of tomographic optical projections are shown in Figure 10.

**Figure 9.** The 3D structure of the object is digitally reconstructed using Eq. (24). (a) An iso-level surface shows subcel‐ lular structures at the surface of the cellular membrane. (b) A 3D view of the aggregated yeast cells flowing along the channel axis (*u*).

The method proposed in Figure 10(a) can be implemented by placing a channel in the optical path of an optical microscope. Here, we will present in greater details the lens-based, flowscanning tomography technique shown in Figure 10(b).

The phase-space distribution is a 4D space that describes an entire light field, with spatial position {*x,y*} and phase information {*kx,ky*} (wave/ray propagation direction). 4D light-field acquisition requires more advanced recording methods than traditional imaging, such as

**Figure 10.** There are two methods for 3D microfluidic tomography. (a) By observing objects flowing at a focal plane located near the upper (or lower) face of the channel, laminar flow-shear effects induce rotation of the objects suspend‐ ed in the flowing fluid. (b) For larger samples, we introduce a phase-space flow-scanning methods that simultaneously records multiple optical projections along a broad range of angles, slice by slice, as the samples flow across an optical slit.

scanning Fourier windows [31], wave front sensors [32], or light-field cameras [33]. Here, we demonstrate a technique that enables the acquisition of a 3D subspace {*x,y,kx*} of the 4D light field. We then extract 3D images showing phase and intensity information in two separate tomograms using computational imaging methods. We propose a microfluidic device that combines an illumination source, patterned with a slit aperture along the (*y*) axis, and a cylindrical lens to collect the light passing through the sample. When placed into an optical microscope, the device enables the simultaneous acquisition of multiple views of the slit aperture for a broad range of perspective angles. This {*y,kx*} image is recorded by the video camera while the object flows through the microfluidic channel past the slit aperture. Motion therefore provides line-by-line scanning along the (*x*) axis.

This new approach has several advantages. Each frame records data that are relative to a specific slice through the object. Consequently, there is very little redundancy between the information contained in two distinctive frames. This means that sampling of the optical signal is very effective. In addition, as we show in the experimental results, line-by-line acquisition is very robust to sample motion. This is a great advantage for in-vivo imaging applications.

#### *2.2.1. Experimental setup*

The method proposed in Figure 10(a) can be implemented by placing a channel in the optical path of an optical microscope. Here, we will present in greater details the lens-based, flow-

**Figure 9.** The 3D structure of the object is digitally reconstructed using Eq. (24). (a) An iso-level surface shows subcel‐ lular structures at the surface of the cellular membrane. (b) A 3D view of the aggregated yeast cells flowing along the

The phase-space distribution is a 4D space that describes an entire light field, with spatial position {*x,y*} and phase information {*kx,ky*} (wave/ray propagation direction). 4D light-field acquisition requires more advanced recording methods than traditional imaging, such as

scanning tomography technique shown in Figure 10(b).

150 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

channel axis (*u*).

The experimental setup is shown in Figure 11. A white light-emitting diode (LED) light source provides a uniform illumination given by *I*(*x*, *y*, *θx*, *θy*) = *I*0, which is restricted with a slit aperture *I*(| *x* | >0.5 µm)=0. The stage is positioned at the focalplane of an optical microscope, in this case a *f* =2 mm objective with a 6 mm working distance and numerical aperture (NA=0.70). In between, we place a cylindrical lens at a focal distance ( *f* '=2 mm) from the aperture. With this 1D optical Fourier transform, we record the angular spectrum *kx* for each point *y* illuminated along the slit axis (*y*). The result is a continuous range of perspective angles for projection tomography, given by *θ*MAX =2arcsin ( NA *<sup>n</sup>* ), where *n* is the refractive index of the flowing fluid. In the experiments below, we use a buffer solution for *in vivo* experiments with *C. elegans*, with (*n* ≈1.33).

**Figure 11.** Experimental device. A microfluidic channel is fitted with a 1 µm wide slit aperture along the *y*-axis that provides a static, cylindrical illumination pattern (in green). A cylindrical lens in a Fourier imaging configuration con‐ verts the transmitted light beam into a phase-space image {*y*,*kx*}. All components are assembled onto a standard micro‐ scope slide to be used directly in a microscope.

We adjust the flow speed and frame rate so that |*δ<sup>x</sup>* | ≈2 µm.

In Figure 12, we show the optical path from source to detector. In the (*y,z*) plane, the objective creates an image of the 100 µm wide microfluidic channel cross-section (with a 50× magnifi‐ cation) onto the camera. The optical resolution is limited by the numerical aperture of the microscope objective and given by

$$R\_y = \frac{\lambda}{2\text{NA}} \approx 0.4 \text{ }\mu\text{m} \tag{25}$$

where *λ* is the imaging wavelength (here, we use *λ* =500 nm. nm at the center of the led spectrum). In the (*x*, *z*) plane, the Fourier lens separates the continuous range of perspective views of the slit along the other axis of the camera. With an adjustable slit aperture, we reduce the numerical aperture to NA*<sup>x</sup>* =0.15. The optical resolution limit along the slit axis is given by

(NA=0.70). In between, we place a cylindrical lens at a focal distance ( *f* '=2 mm) from the aperture. With this 1D optical Fourier transform, we record the angular spectrum *kx* for each point *y* illuminated along the slit axis (*y*). The result is a continuous range of perspective angles

flowing fluid. In the experiments below, we use a buffer solution for *in vivo* experiments with

**Figure 11.** Experimental device. A microfluidic channel is fitted with a 1 µm wide slit aperture along the *y*-axis that provides a static, cylindrical illumination pattern (in green). A cylindrical lens in a Fourier imaging configuration con‐ verts the transmitted light beam into a phase-space image {*y*,*kx*}. All components are assembled onto a standard micro‐

In Figure 12, we show the optical path from source to detector. In the (*y,z*) plane, the objective creates an image of the 100 µm wide microfluidic channel cross-section (with a 50× magnifi‐ cation) onto the camera. The optical resolution is limited by the numerical aperture of the

0.4 m

where *λ* is the imaging wavelength (here, we use *λ* =500 nm. nm at the center of the led spectrum). In the (*x*, *z*) plane, the Fourier lens separates the continuous range of perspective views of the slit along the other axis of the camera. With an adjustable slit aperture, we reduce the numerical aperture to NA*<sup>x</sup>* =0.15. The optical resolution limit along the slit axis is given by

m

(25)

2NA

l= »

*R y* NA

*<sup>n</sup>* ), where *n* is the refractive index of the

for projection tomography, given by *θ*MAX =2arcsin (

152 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

*C. elegans*, with (*n* ≈1.33).

scope slide to be used directly in a microscope.

microscope objective and given by

We adjust the flow speed and frame rate so that |*δ<sup>x</sup>* | ≈2 µm.

**Figure 12.** (a) The microfluidic slide is placed at the focal plane of an optical microscope, with a video camera for data acquisition. (b) A secondary slit aperture is used to adjust the depth of field of the acquired optical projections by re‐ ducing the effective numerical aperture to NA*<sup>x</sup>* = 0.15 along the slit axis. This is a required tradeoff between focal depth and resolution. (c) Each acquired frame contains a continuous range of optical projections of the slit aperture. The depth of field is given by *Dz*. The resolution *Rz* along the *z*-axis corresponds to the size of the domain intersecting all optical projections in the angular range given by *θ*MAX.

$$R\_{\chi} = \frac{\lambda}{2\text{NA}\_{\chi}} \approx 1.7 \text{ } \mu\text{m} \tag{26}$$

The resolution along the z*-*axis corresponds to the geometrical limitations of the projection area shown in Figure 12 (c) and is given by

$$R\_z = \frac{\delta \mathbf{x}}{\tan(\theta\_{\text{MAX}} / \text{2})} \approx 3.2 \text{ } \mu \text{m} \tag{27}$$

The associated depth of field, required to acquire 3D information for stack reconstruction, is given by *Dz* <sup>=</sup> <sup>2</sup>*<sup>λ</sup>* NA*<sup>x</sup>* <sup>2</sup> ≈50 *µ*m. In the simplest case demonstrated here, the limitation of the NA along this axis is a required tradeoff between depth of field and resolution:

$$D\_z = \frac{8}{\lambda} R\_x^2 \tag{28}$$

We note, however, that additional computational methods are possible to overcome this relation [34].

#### *2.2.2. 3D absorption tomography*

The first imaging modality of the device is 3D absorption tomography. As the sample flows in the channel, we record *N* consecutive frames:

$$I\_n(k\_{\ge}, y), n = 1, \ldots \text{ N} \tag{29}$$

The raw data are then reassembled into the angular projection domain *Pθ*, given by

$$P\_{\theta} \left( x, y \right) = I\_{\imath} \left( \frac{2\pi}{\mathcal{L}} \sin \theta, y \right) \tag{30}$$

where

$$m = \left[\frac{\text{x}\cos\theta}{\text{VT}}\right] \tag{31}$$

*Pθ* contains (*x* − *y*) views of the flowing sample for different values of the projection angle |*θ<sup>x</sup>* | <*θ*MAX / 2. Because each frame contains simultaneous perspective views of the slit aperture, all projected views are already aligned and therefore marginally affected by longrange sample motion. Reconstruction of the 3D structure *S*(*x*, *y*, *z*) is directly derived from data with tomographic back-projection algorithms, here an inverse Radon transform, by applying the Fourier Slice Theorem to *P<sup>θ</sup>* ^ =ℱ*<sup>x</sup> Pθ* :

$$\widehat{P}\_{\theta}\left(k,y\right) = \hat{S}\left(k\cos(\theta), y, k\sin(\theta)\right) \tag{32}$$

Finally, we retrieve the 3D structure:

$$S\left(\mathbf{x}\_{\prime}\boldsymbol{y}\_{\prime}\boldsymbol{z}\right) = \mathcal{F}\_{\boldsymbol{x},\boldsymbol{z}}^{-1}\Big[\hat{\boldsymbol{S}}\big] \tag{33}$$

A proof-of-principle experiment is shown in Figure 13. We used a translation stage to displace the microscopic sample at a constant speed along the *x*-axis, and a prepared microscope slide

**Figure 13.** Experimental proof of principle showing the 3D imaging capabilities of the tomographic microscopy device on a prepared microscope slide. A translation stage provides the constant speed displacement along the *x*-axis during data acquisition. Tomographic views are shown at two different levels (a) in green and (b) in blue color maps. (c) The image as it would have been observed in a white light microscope is compared to (d), an overlapping image of (a) and (b) showing layer separation capability with a 6 µm depth difference.

with well-known biomaterial (Elodea leaf slice) to calibrate the phase-space microscopy setup. The 3D structure of cell walls was retrieved and depth information was obtained with enough spatial resolution to distinguish two cellular layers with a 6 µm depth difference (Figure 13(a) and (b)). Such separation is not possible with conventional 2D microscopy (Figure 13(c) and (d)).

#### *2.2.3. 3D differential phase contrast tomography*

The associated depth of field, required to acquire 3D information for stack reconstruction, is

We note, however, that additional computational methods are possible to overcome this

The first imaging modality of the device is 3D absorption tomography. As the sample flows

The raw data are then reassembled into the angular projection domain *Pθ*, given by

( ) <sup>2</sup> , sin , *<sup>n</sup> P xy I y*

p

æ ö <sup>=</sup> ç ÷

l

cos VT *x*

*Pθ* contains (*x* − *y*) views of the flowing sample for different values of the projection angle |*θ<sup>x</sup>* | <*θ*MAX / 2. Because each frame contains simultaneous perspective views of the slit aperture, all projected views are already aligned and therefore marginally affected by longrange sample motion. Reconstruction of the 3D structure *S*(*x*, *y*, *z*) is directly derived from data with tomographic back-projection algorithms, here an inverse Radon transform, by

=ℱ*<sup>x</sup> Pθ* :

q

A proof-of-principle experiment is shown in Figure 13. We used a translation stage to displace the microscopic sample at a constant speed along the *x*-axis, and a prepared microscope slide

 q

µ( , cos ( ), , sin ) ( ) <sup>ˆ</sup> *P ky Sk yk* ( )

( ) <sup>1</sup> , , <sup>ˆ</sup> , *x z Sxyz S* - <sup>=</sup> é ù

é ù q= ê ú

*n*

^

q

q

<sup>8</sup> <sup>2</sup> *D R z x* l

along this axis is a required tradeoff between depth of field and resolution:

154 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

<sup>2</sup> ≈50 *µ*m. In the simplest case demonstrated here, the limitation of the NA

<sup>=</sup> (28)

*I ky n N n x* ( , , 1,, ) = (29)

è ø (30)

ë û (31)

(32)

ë û <sup>F</sup> (33)

given by *Dz* <sup>=</sup> <sup>2</sup>*<sup>λ</sup>*

relation [34].

where

NA*<sup>x</sup>*

*2.2.2. 3D absorption tomography*

in the channel, we record *N* consecutive frames:

applying the Fourier Slice Theorem to *P<sup>θ</sup>*

Finally, we retrieve the 3D structure:

q=

> The second imaging modality of the device is 3D differential phase-contrast (DPC) tomogra‐ phy. We consider the angular projection domain *Pθ* defined previously, and first digitally refocus our data by gradually shifting all perspective views as if the intersection of all projection directions was displaced along the optical axis by a depth *z* from the center of the channel [35]. The virtually defocused views at depth, *z*, are given by

$$P\_{\theta}^{z}\left(\mathbf{x},\boldsymbol{y}\right) = P\_{\theta}\left(\mathbf{x} + z \tan\left|\theta,\boldsymbol{y}\right)\right.\tag{34}$$

An individual DPC image [36] at focal depth *z* is then given by

$$
\Delta\phi\_z\left(\mathbf{x},\mathbf{y}\right) = \frac{I\_L^z\left(\mathbf{x},\mathbf{y}\right) - I\_R^z\left(\mathbf{x},\mathbf{y}\right)}{I\_L^z\left(\mathbf{x},\mathbf{y}\right) + I\_R^z\left(\mathbf{x},\mathbf{y}\right)}\tag{35}
$$

where

$$I\_L^z \left( \mathbf{x}, \boldsymbol{y} \right) = \bigcap\_{\theta = -\theta\_{\rm Max}}^0 P\_\theta^z \left( \mathbf{x}, \boldsymbol{y} \right) \tag{36}$$

and

$$I\_{\mathcal{R}}^z \left( \mathbf{x}, \boldsymbol{y} \right) = \bigcap\_{\theta=0}^{\theta\_{\text{MAX}}} P\_{\theta}^z \left( \mathbf{x}, \boldsymbol{y} \right) \tag{37}$$

The resulting DPC tomogram is given by

$$
\Delta\phi\left(\mathbf{x}, y, \mathbf{z}\right) = \Delta\phi\_z\left(\mathbf{x}, y\right) \tag{38}
$$

Because DPC (∆*φ*) and intensity *S* are simply different methods of processing the raw data (*In*), the two imaging modalities are perfectly registered. This alignment also provides a strong foundation for other methods of phase retrieval, e.g., for ambiguities in reconstruction [37], and does not suffer from possible artifacts present in coherent methods, e.g., speckle and sensitivity to interference jitter. In addition, phase measurement can also be made quantitative with suitable calibration, and it is possible to combine phase and intensity reconstructions to compute the full (complex) refractive index profile [38].

#### *2.2.4. Results on C. elegans*

Now we demonstrate 3D imaging of a live *C. elegans* nematode freely swimming in the microfluidic channel. An XX-hermaphrodite is raised at room temperature to the adult stage of development, using standard techniques [39], and placed into a water-based liquid environment with balanced electrolytes (M9 buffer solution). The motion of the nematode along the (*y,z*) directions is limited by the boundaries of the microfluidic channel but can also be accounted for by preprocessing data, and cancelled with digital frame alignment techni‐ ques.

The phase-space scanning microscope setup used in Figure 13 was modified to integrate a 100 µm deep, 100 µm wide microfluidic channel for *C. elegans* 3D tomography. A low flow speed, 10 nl.s−1, was chosen to provide the best compromise between the available frame rate of the camera and desired resolution.

( , ) ( tan , ) *<sup>z</sup> P xy P x z y*

( ) ( ) ( )

( ) ( ) MAX

( ) ( ) MAX 0 , , *z z RI xy P xy* q

q=

Δ ,, Δ ,

f

compute the full (complex) refractive index profile [38].

q

 f

Because DPC (∆*φ*) and intensity *S* are simply different methods of processing the raw data (*In*), the two imaging modalities are perfectly registered. This alignment also provides a strong foundation for other methods of phase retrieval, e.g., for ambiguities in reconstruction [37], and does not suffer from possible artifacts present in coherent methods, e.g., speckle and sensitivity to interference jitter. In addition, phase measurement can also be made quantitative with suitable calibration, and it is possible to combine phase and intensity reconstructions to

Now we demonstrate 3D imaging of a live *C. elegans* nematode freely swimming in the microfluidic channel. An XX-hermaphrodite is raised at room temperature to the adult stage of development, using standard techniques [39], and placed into a water-based liquid environment with balanced electrolytes (M9 buffer solution). The motion of the nematode along the (*y,z*) directions is limited by the boundaries of the microfluidic channel but can also be accounted for by preprocessing data, and cancelled with digital frame alignment techni‐

q q=-

q

0 , , *z z LI xy P xy*

*z z L R*

*I xy I xy x y I xy I xy*

*L R*

, , Δ , , ,

*z z z*

( ) ( )


<sup>=</sup> ò (36)

<sup>=</sup> ò (37)

(*xyz xy* ) = *<sup>z</sup>* ( ) (38)

q

(34)

 q= +

q

156 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

An individual DPC image [36] at focal depth *z* is then given by

f

The resulting DPC tomogram is given by

*2.2.4. Results on C. elegans*

ques.

where

and

**Figure 14.** Experimental results on live, adult, awake, wild type *C. elegans* nematodes. We display absorption (*S*) and differential phase contrast (∆*φ*) tomographic images for two different depths ((a) *z* = 0 µm, and (b) *z* = 22 µm). Tomo‐ graphic data show the precise 3D location of the reproductive system with eggs (a), and of the digestive system (intes‐ tine), the cuticle and oblique somatic muscle fibers 22 µm above (b). In a conventional white light microscopy device (c), these internal features overlap on the same image and it is nearly impossible to identify and locate them.

Experimental results for 3D amplitude and phase contrast tomography are shown in Figure 14. Figure 14(a) and (b) shows digital slices of the retrieved 3D tomogram at two *z*-levels of interest (*z* = 0 µm and *z* = 22 µm). At each depth level, two images on the nematode show absorption, from optical projection tomography (*S*), and difference phase-contrast (DPC) tomography (∆*φ*). In the reference frame at *z* = 0 µm, absorption tomographic slices (*S*) show the pharynx and its two bulbs on the left side (head), and the reproductive system, with a view of the eggs in the center part of the body. At a different depth (*z* = 22 µm), the digestive system, with distal gonad and the intestine, is clearly apparent.

Differential phase contrast is well suited for the observation of interfaces between tissue layers with different refractive indices. With nearly transparent live roundworms, the DPC tomo‐ graphic slices (∆*φ*) enable the observation and localization of a few eggs at a time at each depth level. Eggs that could not be observed clearly with absorption images only at the level of the digestive system (*z* = 22 µm).

We show a conventional optical microscopy image of the worm in Figure 14(c) to compare our technology with more conventional imaging methods. Here, because all the structures previously identified now overlap in a single image, it is much harder to identify them from only one perspective view. In addition, it is also impossible to find their respective positions along the optical axis.

## **3. Conclusion**

In conclusion, we have presented several microfluidic microscopy methods that combine a liquid channel, optical instrumentation, and computational imaging.

These technologies inherit the advantages of microfluidic channels. Samples move with the flow in a biocompatible fluid and are guided precisely to the observation window for optical imaging. High sample throughput allows both large data sets for population studies as well as repeated imaging of the same sample for longitudinal studies (e.g., to track development/ aging or to evaluate drug delivery and response). Microfluidics also provides a pathway for object sorting and fully automated imaging with little to no sample preparation.

By using the flow as a degree of freedom for imaging, we allow the imaging sensor to capture a more diverse data set than is possible with static samples. Beyond simple multiplicity of images, multiple illumination orientations, shifts, and perspectives are possible. Computa‐ tional analysis then leverages the image diversity into improvements in resolution, quantita‐ tive measurement of surface structure, and even 3D tomographic imaging of phase and absorption. Remarkably, the flow also allows for a certain amount of self-error correction, as integrating the known properties of cell transport in a laminar flow enables computational adjustment for imperfections of the microfluidic channel, such as pinching and fabrication defects, as well as variations in the flow velocity.

These microfluidic microscopy methods can be either scaled down in size for individual cells or scaled up for larger animals. They can operate on their own or be integrated easily with existing devices, such as flow cytometers, microscopes, and imaging systems, e.g., with a modified microscope slide. Likewise, they can be implemented with or without lenses, enabling a variety of miniaturized, on-chip forms. And finally, they can coexist with other modalities of imaging, such as spectroscopy and (photo-) acoustic sampling, for the acquisition of higher-dimensional data cubes.

## **Author details**

Nicolas Pégard1 , Chien-Hung Lu1 , Marton Toth1 , Monica Driscoll1 and Jason Fleischer2\*

\*Address all correspondence to: jasonf@princeton.edu

1 Department of Electrical Engineering, Princeton University, Olden Street, Princeton, USA

2 Department of Molecular Biology and Biochemistry Nelson Biological Labs, Rutgers University, Piscataway, USA

## **References**

only one perspective view. In addition, it is also impossible to find their respective positions

In conclusion, we have presented several microfluidic microscopy methods that combine a

These technologies inherit the advantages of microfluidic channels. Samples move with the flow in a biocompatible fluid and are guided precisely to the observation window for optical imaging. High sample throughput allows both large data sets for population studies as well as repeated imaging of the same sample for longitudinal studies (e.g., to track development/ aging or to evaluate drug delivery and response). Microfluidics also provides a pathway for

By using the flow as a degree of freedom for imaging, we allow the imaging sensor to capture a more diverse data set than is possible with static samples. Beyond simple multiplicity of images, multiple illumination orientations, shifts, and perspectives are possible. Computa‐ tional analysis then leverages the image diversity into improvements in resolution, quantita‐ tive measurement of surface structure, and even 3D tomographic imaging of phase and absorption. Remarkably, the flow also allows for a certain amount of self-error correction, as integrating the known properties of cell transport in a laminar flow enables computational adjustment for imperfections of the microfluidic channel, such as pinching and fabrication

These microfluidic microscopy methods can be either scaled down in size for individual cells or scaled up for larger animals. They can operate on their own or be integrated easily with existing devices, such as flow cytometers, microscopes, and imaging systems, e.g., with a modified microscope slide. Likewise, they can be implemented with or without lenses, enabling a variety of miniaturized, on-chip forms. And finally, they can coexist with other modalities of imaging, such as spectroscopy and (photo-) acoustic sampling, for the acquisition

, Marton Toth1

1 Department of Electrical Engineering, Princeton University, Olden Street, Princeton, USA

2 Department of Molecular Biology and Biochemistry Nelson Biological Labs, Rutgers

, Monica Driscoll1

and Jason Fleischer2\*

object sorting and fully automated imaging with little to no sample preparation.

liquid channel, optical instrumentation, and computational imaging.

158 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

defects, as well as variations in the flow velocity.

, Chien-Hung Lu1

\*Address all correspondence to: jasonf@princeton.edu

of higher-dimensional data cubes.

University, Piscataway, USA

**Author details**

Nicolas Pégard1

along the optical axis.

**3. Conclusion**


*ceedings of the National Academy of Sciences of the United States of America*, 2005;102(52): 19015–19020.

[28] Abkarian M, Faivre M, Horton R, Smistrup K, Best-Popescu CA, Stone HA. Cellularscale hydrodynamics. *Biomedical Materials*, 2008;3(3):034011.

[13] Fiolka R, Shao L, Hesper Rego E, Davidson MW, Gustafsson MGL. Time-lapse twocolor 3d imaging of live cells with doubled resolution using structured illumination. *Proceedings of the National Academy of Sciences of the United States of America*,

[14] Kristensson E, Berrocal E, Richter M, Pettersson S-G, Ald´en M. High-speed struc‐ tured planar laser illumination for contrast improvement of two-phase flow images.

[15] Kim P, Abkarian M, Stone HA. Hierarchical folding of elastic membranes under biax‐

[16] Gustafsson MGL. Surpassing the lateral resolution limit by a factor of two using structured illumination microscopy. *Journal of Microscopy*, 2000;198(2):82–87.

[17] Gustafsson MGL. Nonlinear structured-illumination microscopy: wide-field fluores‐ cence imaging with theoretically unlimited resolution. *Proceedings of the National*

[18] Pégard NC, Fleischer JW. 3D deconvolution microfluidic microscopy using a tilted

[19] Stone HA, Kim S. Microfluidics: basic issues, applications, and challenges. *AIChE*

[20] Reed Teague M. Deterministic phase retrieval: a green's function solution. *JOSA*,

[21] Tikhonov AN, Goncharsky AV, Stepanov VV, Yagola AG. *Numerical methods for the*

[22] Richardson WH. Bayesian-based iterative method of image restoration. *JOSA*,

[23] Chan TF, Wong C-K. Total variation blind deconvolution.*IEEE Transactions on Image*

[24] Dong W, Zhang L, Shi G, Wu X. Image deburring and super-resolution by adaptive sparse domain selection and adaptive regularization. *IEEE Transactions on Image Proc‐*

[25] Quirin S, Prasanna Pavani SR, Piestun R. Optimal 3D single molecule localization for super resolution microscopy with aberrations and engineered point spread func‐ tions. *Proceedings of the National Academy of Sciences of the United States of America*,

[26] Chatwin CR, Wang RK. *Frequency domain filtering strategies for hybrid optical informa‐*

[27] Ohya Y, Sese J, Yukawa M, Sano F, Nakatani Y, Saito TL, Saka A, Fukuda T, Ishihara S, Oka S, et al. High-dimensional and large-scale phenotyping of yeast mutants. *Pro‐*

*Academy of Sciences of the United States of America*, 2005;102(37):13081–13086.

ial compressive stress. *Nature Materials*, 2011;10(12):952–957.

channel, *Journal of Biomedical Optics*, 2013;18:040503.

*solution of ill-posed problems*, vol. 328. Springer; 1995.

*tion processing*. Research Studies Press Ltd.; 1996.

2012;109(14):5311–5315.

*Optics Letters*, 2008;33(23):2752–2754.

160 Advances in Microfluidics - New Applications in Biology, Energy, and Materials Sciences

*Journal*, 2004;47(6):1250–1254.

*Processing*, 1998;7(3):370–375.

*essing*, 2011;20(7):1838–1857.

2012;109(3):675– 679.

1983;73(11):1434–1441.

1972;62(1):55–59.

