Wearable Graphene-Based Electrophysiological Biosensing System for Real-Time Health Monitoring

CHAPTER

WEARABLE GRAPHENE-BASED ELECTROPHYSIOLOGICAL BIOSENSING SYSTEM FOR REAL-TIME HEALTH MONITORING

14

Numan Celik, Wamadeva Balachandran, Nadarajah Manivannan Brunel University London, Uxbridge, United Kingdom

­C HAPTER OUTLINE 1 Introduction .................................................................................................................................... 339 2 Theoretical Background of Graphene for the Purpose of ECG Monitoring ............................................ 341 2.1 General Properties of Graphene ................................................................................... 341 2.2 Graphene Production Methods ..................................................................................... 342 2.3 Raman Spectroscopy of Graphene ................................................................................ 345 3 Development and Analysis of Graphene-Based ECG Electrode ............................................................ 347 3.1 CVD Process for Graphene Coating on Ag Substrates in Development of Bio-Electrode ...... 347 3.2 Raman Spectroscopy Analysis of Proposed GN-Based ECG Electrode .............................. 348 3.3 Electrical Characteristics of Graphene-Coated Electrode ................................................. 350 3.4 ECG Measurement System and Skin-Electrode Impedance Modeling ............................... 350 4 Experimental Setup of Flexible Graphene Electrodes for ECG Monitoring ............................................ 353 4.1 Experimental Setup of Electrodes Using ECG Acquisition System ................................... 353 4.2 Development of Wearable ECG Monitoring System With Graphene-Functionalized Electrodes ............................................................................ 354 5 Results and Discussion ................................................................................................................... 355 6 Conclusion and Future Perspectives ................................................................................................. 358 References ........................................................................................................................................... 358

1 ­INTRODUCTION Electrocardiography (ECG) is one of the most significant medical tests to monitor vital signs of a person's heart activity; hence, such useful information on the heart performance and functionality can be obtained. ECG signal is characterized by various waves during one cardiac cycle. P-wave is called Graphene Bioelectronics. https://doi.org/10.1016/B978-0-12-813349-1.00014-7 © 2018 Elsevier Inc. All rights reserved.

339

340

CHAPTER 14  GRAPHENE BASED BIOSENSING SYSTEM FOR REAL-TIME HEALTH MONITORING

when contraction of the atria occurred; the Q, R, and S waves together make up a complex so that QRS complex is defined when the ventricles are depolarized; T wave of the ECG signals is associated with the return of the ventricular depolarization form to its resting electric position, mean repolarization. Periodically, monitoring of ECG signals helps with the diagnosis of the cause of chest pain and early intervention in heart attack. It is also essential to record ECG signals for the diagnosis of the cause of dizziness and breathlessness [1]. Heart rate variability (HRV) is conducted by ECG analysis becomes a noninvasive method in order to investigate the functionality of electric conduction through different parts of the heart [2]. Therefore, it is important to have a well-characterized electrode to be able to record cardiac signals including each wave in an ECG signal continuously. Traditionally, ECG signals can be recorded using electrodes attached on the various placements of the body, mostly around chest and limbs, to track any abnormal cardiac rhythms. Several attempts have been made to fabricate different types of electrodes to capture ECG signal from the surface of the body. Conventionally, the most commonly used biosensors are of the gel-type adhesive Ag/AgCl electrodes [3], and this type of electrode is simple, reliable, and cost-effective during ECG recording process. However, it is not advised to use these wet types of electrodes in long-term cardiac recording due to skin irritation and allergic reactions. Studies have showed that using adhesive gel Ag/AgCl electrodes can trigger dermal irritation, which is inconvenience of and patient discomfort, and cause potential signal degradation in long-term ECG measurements [4]. Hence, the demand for comfortable of the patient and applying electrodes with biological compatibility has resulted in considering to use of dry or gel-less ECG electrodes that eliminate the need of gel and skin preparation by several researchers [5–11]. Puurtinen [5] proposed dry textile electrodes with different sizes, and Gruetzmann [6] developed and characterized two types (adaptive and capacitive) of dry ECG electrodes. Baek [7] fabricated a flexible polymeric dry electrode for long-term ECG monitoring using elastomer polydimethylsiloxane (PDMS) layer, and Varadan [8] proposed a wearable textile-based nanobiosensor system for cardiac monitoring using a pair of nanobiosensor electrode pads, which was made of gold nanowire on flexible titanium foil. Batchelor and Casson [9] manufactured two types of electrodes, (a) ink-jet-printed capacitive and (b) tattoo-style ink-jet printed for long-term ECG monitoring. Chlaihawi [10] proposed a flexible dry electrode based on multiwalled carbon nanotube (MWCNT)/PDMS composite, and recently, Jang [11] fabricated a ferromagnetic, folded electrode composite for wireless ECG acquisition system. However, the proposed dry-based electrodes had higher skin-electrode contact impedance values than conventional type of wet Ag/AgCl electrodes due to less electric conductivity of proposed metallic layers as electroplating requires an initial conductive layer. This high impedance could affect the measurements of cardiac signals, and enough information of the heart cannot be captured from the body surface. It is crucial to develop electrodes, which should be formed with highly electrically conductive materials, by considering durability, flexibility, and comfort purposes for long-term ECG monitoring. Graphene (GN), a single-layer two-dimensional structure nanomaterial, exhibits exceptional physical, electric, and chemical properties that lead to many applications from electronics to biomedicine. The unique parameters of GN, notably the thinnest material (0.34 nm), the highest electron mobility (~200,000 cm2 V−1 s−1), the highest thermal conductivity (5300 W m−1 K−1), the highest surface-area-tovolume ratio (2630 m2/g), and the fastest moving electrons (~106 m/s) and well-advanced fabrication techniques, have created a great interest in GN-based biosensor applications [12]. Recently, Yapici developed conductive graphene (GN) textile electrodes using dip-dry-reduce process for biosignal ­acquisition in cardiac monitoring.

2 ­ THEORETICAL BACKGROUND OF GRAPHENE

341

This chapter will explore the effect of graphene (GN) for ECG monitoring within theoretical background initially. A novel dry GN-based electrode will be investigated with electric and mechanical parameters during fabrication process of GN-based electrode. The structural properties of GN will be examined by Raman spectroscopy. This chapter not only does observe the experimental results of GNbased electrodes but also will propose a chest-based wearable ECG acquisition device that consists of a wireless-enabled microcontroller unit that communicates with a smartphone to send the cardiac information wirelessly and display it continuously. Subsequently, we measured ECG signals in different positions on the body and studied feasibility of the proposed GN-based electrodes for long-term ECG measurements. Furthermore, the developed GN-based electrodes were tested on 10 different subjects, and the performance evaluation will be compared with conventional wet Ag/AgCl electrodes. At the end of this chapter, the experimental results of ECG monitoring will be demonstrated for continuous measurements, and also, skin-electrode contact impedance values will be evaluated according to frequency changes.

2 ­THEORETICAL BACKGROUND OF GRAPHENE FOR THE PURPOSE OF ECG MONITORING Properties of GN in terms of electric, mechanical, and thermal characteristics are discussed in detail in this section. A calculation of the impedance of skin-electrode model for a typical ECG measurement is made, and the effect of GN is pointed out by including its promising electric conductivity attribution in the calculation.

2.1 ­GENERAL PROPERTIES OF GRAPHENE It has been clearly indicated that nanomaterials improve the physiochemical characteristics of bulk materials such as conductivity, strength, reactivity due to their high volume surface ratio, high tensile strength, and high electron mobility attributes [13]. Among other nanomaterials, GN has received worldwide attention owing to its extraordinary physical, electronic, thermal, optical, chemical, and mechanical properties [14,15]. GN is composed of a two-dimensional (2-D) single-atom-thick layer of sp2-bonded carbon atoms arranged in a honeycomb lattice [16]. Due to the unique structure, GN is examined as main carbon allotropy to carry out other related carbon nanomaterials such as carbon nanotubes (CNT), graphite, and fullerenes, as can be seen in Fig. 1. Geim and Novoselov [15] discovered GN, the thinnest known material with a thickness of 0.35 nm, in 2004 and received the Nobel Prize in Physics for inventing this extraordinary material by a simple mechanical exfoliation. They were studying on graphite flakes at the beginning, and after applying several thinner processes into the graphite, the monolayer and bilayer of GN were produced. Since then, GN and its related derivatives have attracted the interest of various scientific fields, such as nanoelectronics [18], energy applications (e.g., supercapacitor and solar panels) [19,20], biosensing applications [21], gas sensing applications [22], and transparent flexible electronics [23] due to their physiochemical and electric-electronic properties. A summary of the key properties of GN can be classified as follows: • Graphene (GN) has the highest tensile strength of any material (is the strongest material)—Young's modulus of 1 TPa, approximately 200 times stronger than steel. • GN is the thinnest material ever tested, only in 0.34 nm thick—single atomic layer of carbon.

342

CHAPTER 14  GRAPHENE BASED BIOSENSING SYSTEM FOR REAL-TIME HEALTH MONITORING

FIG. 1 Graphene as a “building material” of fullerene, graphite, single-wall nanotubes (SWNTs), and multiwall nanotubes (MWNTs) [17].

• GN is entirely transparent—~97.7%. • Even though the thinnest material in the world, GN absorbs very high amount of light per layer—~2.3 %. • GN is the lightest continuous film (per unit area)—0.38 mg/m2. • GN is the highest surface-to-volume ratio of any material—2630 m2/g (because of this reason, GN is a quite suitable material for biosensing applications). • GN is the best conductor of electricity—it has resistivity of 10−8 Ωm (better values than any other metal, such as Au or Cu). • GN has the fastest moving electrons in any material ≅ 106 m/s (quite close to speed of light; thus, it is referred as relativistic particles). • GN has the highest electron mobility of any material—more than 200,000 cm2 V−1 s−1. • GN is the best conductor of heat—thermal conductivity of 5300 Wm−1 K−1. • GN is one of the most elastic and flexible material, because of its thin and strong characteristics. Furthermore, Table 1 compares the key properties of GN with its related carbon materials and good conductive metals.

2.2 ­GRAPHENE PRODUCTION METHODS There are a number of methods available in the making of GN, and each method has got its own strengths and weakness. In this section, four methods of GN production are discussed in detail.

2 ­ THEORETICAL BACKGROUND OF GRAPHENE

343

Table 1  Comparison of Relevant Electronic, Physical, and Thermal Properties of GN With Other Related Carbonic and Metallic Materials (Si, Cu, and Single-Wall Carbon Nanotube—SWCNT) [12] Key Parameters

Si

Cu

SWCNT

GN

DC max current density (A/cm2) Melting point (K) Electron mobility (cm2/V s) Thermal conductivity (×103 W/m K) Mean free path (nm) at room temperature

– 1687 1400 0.15 30

107 1357 – 0.385 40

>109 3800 >10,000 1.75–5.8 >103

>108 3800 >10,000 3–5 1 × 103

2.2.1 ­Micro mechanical cleavage method

Since first discovered in 2004, GN sheets have been fabricated using a number of different production methods. Geim and Novoselov followed the route in which a typical Scotch tape was used to extract thin layers of graphite from highly ordered pyrolytic graphite and then transferred these layers onto a silicon (Si) substrate following several times. This method is called as “mechanical exfoliation or micromechanical cleavage,” which is one of the procedures to synthesize GN sheets. When the cleaving process is repeated in enough times, the remains are produced on the Scotch tape and are then transferred to the Si substrate with either a 100 or 300 nm silicon oxide (SiO) layer. This two-layer thickness can be differed from other thicknesses of multilayer graphene or bulk graphite via difference in light absorption. This method of production is the best quality of GN synthesis so far, in terms of structural integrity and also relatively cheaper than other production methods [24]. Fig. 2A–D show the production of GN using micromechanical cleavage technique with “Scotch tape,” and Fig. 2E–F show atomic force microscopy (AFM) images of GN layers on SiO substrate. As can be seen in the Fig. 2, different colors refer different layers of GN in the AFM images. While a lighter purple means a monolayer GN, a darker layer of purple can be bilayer, and one-step darker area can refer trilayer or multiple layer of GN.

2.2.2 ­Epitaxial method

Around the same time of the discovery of GN layers from the thin graphite via the micromechanical cleavage method, in 2004, researchers at Georgia Institute of Technology presented that ultrathin large-scale graphite depositing on silicon carbide (SiC) wafer substracted was prepared via reduction of SiC [26], providing another method for production of high-quality mono or multiple layer of GN. This method is referred as “epitaxial growth,” and GN is produced by converting this ultrathin graphite deposited SiC to GN via sublimation of Si atoms at very high temperature (generally at ~1300°C) [27].

2.2.3 ­Chemical vapor deposition (CVD) method

Chemical vapor deposition (CVD) method can also be applied to the GN grown by epitaxial methodology due to similar technology that needs high-temperature synthesis, not only on Si substrates but also in Ni and Cu substrates. Among all other strategies to produce GN, CVD on metal substrates has become the most promising approach that has excellent advantages including best quality for largescale applications (with area larger than 1 cm2), allowing transfer of GN to other metallic substrates (including Au, Ag, and SiO2) that makes available in different kinds of applications and inexpensive production method [28]. The first GN grown by CVD method was applied in 2008 and 2009, using Ni and Cu substrates [29,30], which was followed by many research applications and publications in

344

CHAPTER 14  GRAPHENE BASED BIOSENSING SYSTEM FOR REAL-TIME HEALTH MONITORING

(A)

(B)

(C)

(D)

(E)

(F)

FIG. 2 (A) Attaching a piece of graphite to sticky tape; (B) using the sticky tape to thin out the graphite; (C) placing the thin graphite on a silicon wafer, with a surface layer of silicon dioxide; (D) removing most layers of graphite leaving behind single-layer, two-layer, or multilayer graphene (GN); (E) AFM image of low magnification; and (F) higher magnification optical micrograph just after micromechanical cleavage [25].

transition metal substrates [31]. CVD operation for GN growth on metallic substrates can be performed in vacuum with atmospheric pressure (Ar, H2, and CH4 gases) and at high temperatures around 1000°C. One of the most developing processes for producing single-layer GN using CVD is GN growth on copper (Cu). Fig. 3 shows a schematic of experimental setup of CVD operation, which is a commonly used method in order to produce high-quality single-layer GN by Cu. This method consists of a tube furnace for high-temperature heating (at around 1000°C), a quartz vacuum chamber, a vacuum and pressure control system for the growth condition adjustment, and several mass flow controllers (MFC) to provide enough carbon source (CH4) and reactant gases (Ar or H2) with necessary flow rate [32].

2.2.4 ­Other graphene production methods

Up to here, main production methods for GN are explained, and several other methods are briefly pointed out in this section. Liquid-phase exfoliation technique is another method for GN synthesis that

2 ­ THEORETICAL BACKGROUND OF GRAPHENE

Vacuum gauge

Quartz vacuum chamber High temperature tube furnace

Vacuum pump

Pressure control system

Annealing

345

Cu foil

MFC

CH4

MFC

H2

MFC

Ar

Growth

Native oxide

Cu foil

CH4/H2 at 1000°C

Coalescence of graphene domains

FIG. 3 CVD method of GN on copper foil [32].

contains a solution-based exfoliation of GN oxide (GNO) using dispersion technique [33]. In liquidphase exfoliation, GNO is mostly formed from graphite by ultrasonic bath sonication with chemically functionalized GN [34]. Reduced GN oxide (RGO) is a mostly used method, which is developed by solution suspension of GNO followed by chemical reduction, in particular electrochemical biosensor applications. The GN produced by the RGO process offers smaller sizes, more structural defects, and more functional groups than those produced by other synthesis techniques [35]. For comparison, the CVD method is mostly chosen in terms of producing high-quality GN in large-scale applications; on the other hand, the RGO is an ideal method for synthesizing small GN sheets in large-scale developments. The selectivity of methods for GN synthesizing is subjectable to the application that can require an alternative method of GN production. Fig. 4 compares the production methods in terms of price and quality of synthesizing GN sheets. There have been many production methods for GN suggested, and for an overview, the current GN synthesis techniques are listed in Table 2.

2.3 ­RAMAN SPECTROSCOPY OF GRAPHENE In graphene (GN), there are three key Raman bands: • G band (around 1582 cm−1) is a sharp band and corresponds to planar sp2-bonded carbon in GN, graphite, and CNTs. • D band (around 1350 cm−1) is a disorder band and corresponds to sp2 carbon rings (i.e., defects in sp2 carbon) and is often regarded as an indicator for defects. It is worth to note that the intensity of D band is proportional to the degree of defects in materials with sp2 carbon (graphite, CNTs, and GN). This band is quite weak in both graphite and GN, and if D band is observed in Raman spectrum, then there are defects in the carbon materials. • 2D (G′) band (around 2685 cm−1) is the final band and second order of the D band and is generally used to identify the number of GN layers. The comparison of the Raman spectra of GN and graphite (composed of millions of layers of GN stacked together) has been made in Fig. 5A and measured at 514.5 nm excitation [37]. As can be seen

346

CHAPTER 14  GRAPHENE BASED BIOSENSING SYSTEM FOR REAL-TIME HEALTH MONITORING

FIG. 4 Methods for GN synthesis. There are various methods to be chosen depending on the specific application; each one differs from one another with quality and price [25].

Table 2  A Brief Summary of GN Production Techniques [12] Synthesis Method

Brief Description

Mechanical exfoliation

• • • • • • • • • • • •

Liquid-based exfoliation Epitaxial growth

CVD growth GN

Atomic layer of GN can be seen on ~300 nm SiO2 substrates Pristine GN with the highest quality of electric properties The size and thickness are uncontrollable, thus limited practical applications Graphite powders are initially oxidized by chemical modification to be dispersed in solution Large-scale production for bulk applications, i.e., supercapacitors, composite materials Serious structural defects A conversion of SiC substrate to GN via sublimation of Si atoms on the surface Done at very high temperature (~1300°C) Accessibility is limited due to high-end equipment Most promising, inexpensive, and feasible method for single-layer GN synthesis Using transition metal (Ni, Cu, and Si) substrates Can be scaled up for large-area GN production for practical applications

in Fig. 5A, two major bands (G and 2D bands) are observed in both GN and graphite samples. While G peak was observed at 1580 cm−1, 2D (G′) band was observed at ~2700 cm−1 for GN; however, these peaks are observed differently for graphite. Furthermore, Fig. 5B illustrates the differences of 2D bands obtained from GN and graphite. There is a significant variation in the shape and intensity of the 2D peak of GN compared with graphite. According to the results, the 2D peak in graphite consists of 2D1

3 ­ DEVELOPMENT AND ANALYSIS OF GRAPHENE-BASED ECG ELECTRODE

(A)

347

(B)

FIG. 5 (A) Comparison of the Raman spectra of GN and graphite measured at 514.5 nm and (B) comparison of the 2-D peaks in GN and graphite [36].

and 2D2, around ¼ and ½ the height of the G peak, respectively. GN has a single, sharp 2D peak, about four times more intense than the G peak [37]. It is observed that these Raman spectra demonstrate the ability to differentiate between single-layer GN and graphite. Moreover, Raman spectrum can differentiate the single, double, and triple layers by looking into the positions and shapes of the G and 2D bands.

3 ­DEVELOPMENT AND ANALYSIS OF GRAPHENE-BASED ECG ELECTRODE After summarizing the theoretical analysis of GN, the development of GN grown on Ag/AgCl ECG electrode is explored using CVD technique in this section. The GN synthesis, transfer of GN onto the Ag substrate via copper, and also the quality of GN-coated electrode will be investigated in this section. Moreover, the obtained images via Raman microscopy will also be exhibited here. At the end of this section, the electric characteristics of the GN-coated ECG electrode will be revealed with regard to electric conductivity measurements using two-probe technique.

3.1 ­CVD PROCESS FOR GRAPHENE COATING ON AG SUBSTRATES IN DEVELOPMENT OF BIO-ELECTRODE In order to fabricate a GN-coated Ag/AgCl ECG electrode, CVD technique is chosen to grow GN on Ag substrates of the electrode from copper layer. This has advantages on transition into different metals such as Ni, Cu, and Si substrate and thus makes easier to transfer GN onto different substrates, and also, CVD-grown GN is inexpensive and a feasible method for single-layer GN synthesis compared with other production methods. The GN was produced utilizing hydrogen (H2) to remove oxygen from the pump system and methane (CH4) gases to grow single-layer GN on copper substrates based on the CVD procedure at

348

CHAPTER 14  GRAPHENE BASED BIOSENSING SYSTEM FOR REAL-TIME HEALTH MONITORING CVD growth

Spin coating Graphene Cu

Cu

PMMA Graphene Cu

Wet etching Rinsing PMMA Graphene DI Water

PMMA Graphene Cu

Ammonium persulphate

Drying PMMA Graphene Electrode

Solvent cleaning Graphene Electrode

FIG. 6 Schematic diagram showing CVD coating process for synthesizing GN-coated Ag electrode.

very high temperature (around 1000°C), similar to those used in the literature [38]. Fig. 6 shows the whole procedure of CVD for synthesizing GN-coated Ag ECG electrode. As can be seen from the figure, first of all, GN is grown on Cu foil via CVD process, and then, graphene-copper (GN-Cu) substrate is placed on spin coater to cover the top side of GN-Cu substrate with PMMA layer. After fitting PMMA/GN/Cu substrate together, then, this triple layer is placed into the bath of ammonium persulfate ((NH4)2S2O8) to etch away copper (Cu) from the bottom side of triple layer (PMMA/ GN/Cu) until Cu is completely etched away. After all Cu substrate is removed, the PMMA/GN layer is cleaned, and (NH4)2S2O8 residues are removed in deionized (DI) water bath several times. After the PMMA/GN layer is dried, this double layer is placed onto the target substrate (Ag-based electrode). Finally, PMMA/GN/Ag electrode substrate is placed into acetone solution to remove PMMA from the sample. Afterward, the GN/Ag electrode substrate is removed from the solution and dried. Although acetone treatment, the PMMA residues can be still on the substrate. At the end of CVD process, very thin-layer GN (around 3.7 Å ≅ 0.37 nm) is coated on top of Ag layer of the ECG electrode.

3.2 ­RAMAN SPECTROSCOPY ANALYSIS OF PROPOSED GN-BASED ECG ELECTRODE In the analysis of Raman spectroscopy, first of all, the Raman spectrum of GN grown on Cu substrates was investigated to differentiate the GN on Cu, and then, the Raman analysis of GN-coated Ag layer electrode was done to compare with that of bare Ag electrode without GN coating, as can be seen in Fig. 7. The results revealed that all three important bands (G, D, and 2D) were observed in Raman spectrum; thus, the GN-grown processes can be characterized. According to Fig. 7A, only G and 2D peaks were observed from the Raman spectra of GN-coated electrode on Cu sample. Furthermore, it can be derived that a single-layer GN is identified for the sample of GN-grown Cu after CVD process, due to high intensity ratio of I2D/IG (around 5–7). In addition, a shift in the 2D peak position of single-layer GN has been observed in the range of 2711–2723 cm−1.

3 ­ DEVELOPMENT AND ANALYSIS OF GRAPHENE-BASED ECG ELECTRODE

349

(A)

(B) FIG. 7 Raman analysis of graphene (GN) and bare Ag/AgCl electrode. (A) Raman spectrum of GN grown on Cu is shown after chemical vapor deposition (CVD) and (B) Raman spectrum of GN-coated Ag electrode and conventional dry Ag/AgCl electrode is shown for comparison.

The value of full width at half maximum varies in the range of 25–31 cm−1, which represents the singlelayer GN. Fig.  7B shows the comparison of Raman spectrum for the GN-coated Ag electrode and conventional bare Ag/AgCl electrode. It is proved that the presence of the 2D peak in the GN-coated electrode differentiates the presence of GN from the bare electrode, in comparison with red square to black square.

350

CHAPTER 14  GRAPHENE BASED BIOSENSING SYSTEM FOR REAL-TIME HEALTH MONITORING

3.3 ­ELECTRICAL CHARACTERISTICS OF GRAPHENE-COATED ELECTRODE To determine the electric property (resistance through the samples) of synthesized graphene, two-probe measurement is conducted on the samples of before and after GN coating to compare electric conductivity values. Electric conductivities are calculated from resistances and dimensions of the electrodes by following Eq. (1): σ=

H R× A

(1)

where electric conductivity (σ) is proportional to the length (H) of the electrode and proportionally inverse to the resistances (R) and area thickness (A) of the electrodes. In our experiments, both electrodes are of same length and area, which are 12 and 1 mm, respectively (the thickness of GN is negligible ~0.37 nm). The electric properties of the bare and GN-coated Ag electrodes were measured at room temperature using a digital multimeter, and the resistance values were obtained as 212.4 and 28.3 Ω for bare and GN-coated Ag electrodes, respectively. Therefore, the electric conductivity (σ) of Ag electrodes before and after GN coating was measured from the derivation of the above equation as 393 and 2976 S/m, respectively.

3.4 ­ECG MEASUREMENT SYSTEM AND SKIN-ELECTRODE IMPEDANCE MODELING 3.4.1 ­ECG measurement with 3-lead electrode setup

In order to obtain a reliable ECG signal, the electrodes are commonly used to carry out the measurement chain. As mentioned earlier, there are different types of electrodes made in several techniques such as gel, dry, noncontact, capacitive, textile, nanomaterial-based electrodes for ECG monitoring systems. If they are used incorrectly, there will be unwanted distortion orders of magnitude larger than the ECG itself. Fig. 8 introduces the typical measurement system for ECG recording that consists of two or more metal lead (generally Ag/AgCl electrodes) electrodes attached to the skin and connected to a differential amplifier. The aim of the driven-right-leg circuit is to reduce interference from the amplifier. When an ECG signal is amplified by instrumentation amplifier, it is possible to create DC common-mode interference. The driven-right-leg (DRL) connection inverts and amplifies the common-mode signal back into the right leg of the person as shown in the figure. Therefore, 60 Hz common-mode bias signal is canceled from AC power, and a better ECG signal is acquired. In order to get a clearer ECG signal, which is at the output of the instrumentation, amplifier is filtered to reduce the noise, and then, denoised ECG signal is sent to an operational amplifier that conditions the signal to better levels for the microcontroller for further signal processing algorithms of the ECG signal.

3.4.2 ­Skin-electrode interface and its equivalent circuit model

ECG signal is acquired by metallic layers of the electrodes; hence, the contact area affects the skinelectrode interface, and the impedance of skin-electrode model is strongly considered for obtained ECG signal. In this section, the impedance analysis of skin-electrode interface is made for proposed GN-coated and conventional Ag/AgCl ECG electrodes in order to compare impedances of equivalent circuit model of electrode-skin interface. This is because it is important to realize that low impedance is essential for efficient electrode performance, thus affecting ECG signals proportionally. It is worth noting that the impedance (Z) of skin-electrode interface includes elements (resistance and capacitance) that source from body tissue (the epidermis, dermis, and stratum corneum), skin-electrolyte interface,

3 ­ DEVELOPMENT AND ANALYSIS OF GRAPHENE-BASED ECG ELECTRODE

Power line

IP

220 V 60 Hz

351

RE1 Instrumentation amplifier ZA1

ZP

ZB1

A1

Differential amplifier

RE2

A3 ZA2 ZB2

RE3

ZB

Driven-right-leg

A2

Vout

RO

RF

RO ZS

Isolated common A3

– +

FIG. 8 Schematic circuit of typical ECG measurement system with DRL connection [39].

and electrode leads. To better understand the electric behavior of the skin-electrode interface and effects of its components, an equivalent circuit can be modeled to figure out the electric characteristics of this interface. Fig. 9A shows skin-electrode interface and its electric equivalent. In order to better understand the bioimpedance measurements, a simplified schematic diagram for the electrode system was used in this study, as can be seen in Fig. 9B. The impedance analysis of skin-electrode interface was performed by applying two electrodes to simplify the circuit of typical ECG measurement system in Fig. 8. In this approach, the applied two electrodes of each electrode type (GN coated and conventional Ag/AgCl) are assumed to the same (e.g., identical size and identical material, produced from the same manufacture). The electrode circuit components values for the first electrode are also assumed to be identical with the second electrode (Cd = Cd1 = Cd2, Rd = Rd1 = Rd2, and Rs = Rs1 = Rs2). Here, the capacitance Cd represents the electric charge between the electrode and skin. Rd corresponds to the resistance that occurs between the skin and ­electrode during charge transfer, and Rs represents the resistance of the electrolyte gel and skin tissue [41]. Rtissues is the resistance that is sourced by tissues. Rtissues value is small relative to the impedance of the electrode-skin interface. For a healthy human, arm's tissue is known to be almost 150 Ω [41], while the impedance of the skin-electrode interface is larger than 1 MΩ. Hence, Rtissues can be negligible for the calculation of skin-electrode impedance, and its contribution is considered in Rs.

352

CHAPTER 14  GRAPHENE BASED BIOSENSING SYSTEM FOR REAL-TIME HEALTH MONITORING

Electrode-skin interface Rd1

Body (arm)

E Zcpa

Rct

Electrode

Rs1

Gel

Rl

Cd1

Rtissues

Rd2

Elj Epidermis Cs Rsub

Ehc1

Rs2

Rs Dermis

Ehc2

Cd2

(B)

(A) FIG. 9

(A) Skin-electrode interface and electric equivalent and (B) simplified schematic diagram for the electrode system [40].

Eq. (2) describes the total impedance of the skin-electrode interface for a single electrode as a function of frequency: Z (ω ) = RS +

Rd / jCdω Rd + 1 / jCdω

= Rs +

Rd , ω = 2π f 1 + jωCd Rd

(2)

where f is the frequency (Hz). This formula will be used to compare the impedance of skin-electrode interface using GN-coated electrode to that of conventional Ag/AgCl electrode. Zg represents the impedance of skin-electrode interface using GN-coated electrode, while Zs represents the impedance of skin-electrode interface for conventional Ag/AgCl electrode. The estimated values for the electrode circuit model components (Rd, Cd, and Rs) for GN-coated and Ag/AgCl electrodes are given in Table 3. The electrode circuit model component values were estimated by performing least-mean-squares curve-fitting method using MATLAB program. Because of the superior electric conductivity characteristics of graphene (GN), GN-coated electrode would produce lower Rd and Rs values than pregelled Ag/AgCl electrodes. Furthermore, the effect of high electron mobility properties in GN sheets would increase electric charge potential. The measurements made by GN-coated electrodes resulted in having higher Cd values in comparison with conventional Ag/AgCl electrodes, as can be seen Table 3. Recording biological signals at high Cd values is translated to better biological signal quality [42]. Using the formula  (2), the impedance measurement is performed for each type of electrode using the values in Table 3. According to these values, the impedance for electrode-skin interface for GN-coated electrode is determined as (Eq. 3): Z g = Rs +

Rd 1 + j 2π fCd Rd

(3)

4 ­ EXPERIMENTAL SETUP OF FLEXIBLE GRAPHENE ELECTRODES

353

Table 3  Estimated Values for Electrode Circuit Model Components (Rd, Cd, and Rs) Electrode Type

Rd (kΩ)

Cd (nF)

Rs (Ω)

GN coated Ag/AgCl

150.86 214.62

19.1 15.8

247.4 364.3

When the circuit model component values of GN-coated electrodes were applied as in Table  3, Zg is founded as 8.72 MΩ. Similarly, when the values of pregelled Ag/AgCl electrodes were assigned as in Table  3, the impedance for electrode-skin interface for Ag/AgCl electrodes (Zs) is founded as 10.55 MΩ. It is worth to mention here that less skin-electrode impedance results better ECG signal quality with high amplitude of QRS complex. In this theoretical study, it is clearly seen that GN-coated electrodes demonstrated less impedance values than that of Ag/AgCl electrodes, which means GNcoated electrodes would demonstrate to obtain better ECG signal in experimental study. The experimental study will be deeply investigated later in Section 5.

4 ­EXPERIMENTAL SETUP OF FLEXIBLE GRAPHENE ELECTRODES FOR ECG MONITORING 4.1 ­EXPERIMENTAL SETUP OF ELECTRODES USING ECG ACQUISITION SYSTEM This section examines a platform for GN-ECG monitoring that can be applied into mobile health applications. Two types of electrodes were used in our experimental studies for ECG monitoring. As can be seen in Fig. 10A, the first type of electrode is conventional pregelled Ag/AgCl electrode, and Fig. 10B shows the second type of electrode that is the proposed GN-based ECG electrode. Both electrodes have the same sizes, which are a diameter of 24 mm and a thickness of 1 mm. An adhesive tape was attached to GN-coated electrode to be firmly attached to the skin. Ag/AgCl electrodes had an adhesive tape by the manufacture. In this study, an ECG acquisition module is developed to acquire ECG signals via chest-based three-lead ECG electrodes and to transfer obtained ECG signals wireless to a smartphone for continuously monitoring. The aim of this work is to compare the quality of acquired ECG signals using Ag/AgCl and GN-coated electrodes. Two active ECG electrodes were attached on the chest, and one reference electrode was applied on the waist to reduce common-mode power-line interference. As can be seen in Fig. 10C, the acquisition module also has a wireless connectivity using Bluetooth module that will provide a connection with a smartphone and transfer the acquired ECG data into the smartphone. Fig. 10D shows the block diagram of ECG acquisition module, which consists of three-lead ECG pins for obtaining signal, a front-end amplifier unit, analog-to-digital converter (ADC), bandpass filter, microcontroller unit (Arduino UNO ATmega328, Arduino, Italy), and wireless transmission unit. In the front-end amplifier unit, acquired raw ECG signal is amplified and filtered with a frequency ranging from 0.5 and 100 Hz. Microcontroller part is responsible to control the ADC to obtain and send ECG data to wireless transmission unit. ECG signal will be digitized by a 16 bit ADC with sampling rate of 250 Hz. Here, the wireless transmission unit contains an HC 06 Bluetooth module (HC-06, ALLNET, the United States) to send the ECG data to the smartphone. ECG acquisition module continuously operates at over 29 h with a commercial 500 mAh Li-ion battery.

354

CHAPTER 14  GRAPHENE BASED BIOSENSING SYSTEM FOR REAL-TIME HEALTH MONITORING

(A)

(C)

(B)

(D)

FIG. 10 (A) Tested pregelled Ag/AgCl electrode, (B) GN-coated electrode, (C) wireless ECG data acquisition system, and (D) block diagram of ECG signal acquisition.

4.2 ­DEVELOPMENT OF WEARABLE ECG MONITORING SYSTEM WITH GRAPHENE-FUNCTIONALIZED ELECTRODES Wearable and portable wireless electronics advanced pervasive computing and enhanced the scope of remote health monitoring systems [43]. In order to integrate the conventional ECG electrodes into wearable garments, it is essential to develop wearable ECG sensors without compromising high electric conductivity, high mechanical durability and stability against repeated use, high degree of comfort and flexibility, and large-area manufacturability. Regarding these promising attributes, graphene (GN) such a nanomaterial has these advantages to be implemented into an ECG electrode to develop a wearable, wireless ECG monitoring system. The prototype is based on ECG sensing pads, which were fabricated in 3-D printing technology, and can be built in Ag/AgCl and GN-coated electrode as the ECG biosensing unit, electronic circuitry for ECG readout as mentioned in the previous section, and wireless transmission between wearable garment and a smartphone. Fig. 11A shows the graphene-coated mechanism for ECG monitoring system as 0.37 nm length of GN layer is coated on Ag layer of the electrode to increase electric charge transfer capacity between the skin and electrode. Fig. 11B also indicates the system architecture of developed system, while three-lead ECG electrode setup was tested on a subject

5 ­ RESULTS AND DISCUSSION

355

ECG acquisition system

Electrodes

Graphene layer with 0.37 nm

Wireless transmission

Ag layer ECG data

Graphene-based electrode

Copper wire

(A)

(B)

FIG. 11 The graphene-electrode mechanism within a wearable garment: (A) The mechanism of graphene-coated electrode and (B) the schematic illustration of developed system while lead I ECG was testing on the subject including a wireless ECG data acquisition system.

with a wireless ECG data acquisition system. The proposed ECG data acquisition system was worn, and the electrodes were attached on a subject during the experimental study. The signal detected from the sensors is wirelessly transmitted to smartphone, in which a dedicated app is developed and installed for continuous monitoring and derivation of heart rate (HR) and storage of transmitted ECG signals. The stored ECG signals can also be sent to a clinical server; hence, a clinician or a doctor can monitor the patient's cardiac signals remotely.

5 ­RESULTS AND DISCUSSION Here, ECG signals, which are acquired from two different type of electrodes (adhesive Ag/AgCl and GN-coated electrodes), will be considered to compare signal quality and SNR by placing the electrodes on chest using wearable e-health module. ECG signals were separately measured from the proposed GN-based electrodes and conventional pregelled Ag/AgCl electrodes via ECG acquisition system (ehealth and Arduino) using three-lead electrode system. Initially, USB protocol was used for UART communication to observe and analyze ECG signals on PC; subsequently, the obtained ECG signals were transmitted to the smartphone to display the signals on a developed app. In order to obtain the ECG signal, lead I configuration was conducted throughout experimental testing of ECG electrodes on a 29-year-old male subject, where two active electrodes were attached on person's chest and the other driven-right-leg (DRL) electrode was attached onto the left waist of the person. The functionality of the proposed system was verified by real-time monitoring in MATLAB and transmission of obtained ECG signals to a personal computer where the signal was further denoised by applying filtering process. The recorded ECG signals from conventional Ag/AgCl and GN-based electrodes were shown in Fig. 12A and B, respectively. The recorded ECG signals demonstrate that

356

CHAPTER 14  GRAPHENE BASED BIOSENSING SYSTEM FOR REAL-TIME HEALTH MONITORING

the high performance of the GN-based electrode for measurement of real-time ECG. It is clearly seen in Fig. 12A that the recorded cardiac signals by adhesive Ag/AgCl electrodes provide high signal-tonoise ratio (SNR) (with 21.23 dB) due to pregelled approach that reduce the skin-electrode impedance and allow transferring more ECG signal from the body. In contrast, Fig.  12B shows that the recorded ECG signals via GN-based electrodes and the SNR values (25.32 dB) were higher than that of Ag/AgCl electrodes due to superior electric conductivity effect of GN. It has been already investigated in the previous section that GN-based electrodes had less impedance of skin-electrode interface and hence provided much more ECG signals than Ag/AgCl electrodes.

(A)

(B)

(C)

(D)

FIG. 12 The performance evaluation of the proposed wearable wireless ECG monitoring system: (A) the recorded ECG signals by conventional Ag/AgCl electrodes, (B) recorded ECG signals by GN-coated electrodes, and received and displayed ECG signals by a smartphone with a dedicated app written in Java using (C) conventional Ag/AgCl electrodes and (D) GN-coated electrodes.

5 ­ RESULTS AND DISCUSSION

357

The observation of signs (P-QRS-T intervals) in an ECG waveform is critical for the diagnosis of the cause of chest pain and early intervention in heart attack [1]. Therefore, it has been studied here that to identify the significant signs in the obtained ECG signal using both type of electrodes. Even though the difference between subsequent ECG recordings is small, a comparison for detection of P-QRS-T intervals can be made from recorded signals. As can be seen from Fig. 12A, P-wave and QRS complex were identified clearly using Ag/AgCl electrodes; in contrast, Fig. 12B showed that all critical signals, including T wave and U wave, can be clearly seen using GN-based electrodes. To further the performance of the wearable prototype is to compare recorded ECG signals on a smartphone in which a dedicated app that also shows heart rate (HR) simultaneously in the updated version of the app. The obtained ECG signals were transmitted from ECG acquisition module, which was worn on a subject's arm, to a smartphone for displaying recorded ECG signals. As can be seen from Fig. 12C, only R peaks were identified on a mobile app for real-time ECG monitoring; however, Fig. 12D clearly shows that nearly all critical waves of ECG signals (P-wave and QRS complex) were detected during continuous ECG measurement. As theoretically proved earlier, skin-electrode contact impedance measurement has always been of interest due to providing the reliability of the collected biopotential. Skin conductivity varies according to variations of the conditions of either the stratum corneum (whether hairy skin or not) or sweat proportions. In order to get a high-quality signal acquisition with minimal noise, the impedance measurement of skin electrode should be small and stable. To characterize the impedance of graphene-coated electrode, a measurement setup was constructed based on earlier techniques reported in the literature [44]. Here, measurements were taken with the electrode-skin contact impedance of GN-coated electrodes and compared with that of the conventional Ag/AgCl electrodes using Wayne Kerr 6500B impedance analyzer. Both the conventional Ag/AgCl electrodes and the proposed GN-coated electrodes were placed adjacent to each other on a person's forearm (between the wrist and the elbow), and both electrodes were attached so as to maintain the same distance (3 cm between them). Each measurement was carried out just after attaching the electrodes (in period of 30 s) and after removal of skin moisture. The impedance measurements were recorded in the frequency range of 20–1 kHz. According to these measurements, the change in impedances with varying frequencies was improved by graphene coating when compared with that of Baek's [7] and Meng's [44] results. The recorded measurements exhibited that the impedance values of conventional Ag/AgCl electrode range from 445.05 (at 20 Hz) to 13.82 kΩ (at 1 kHz), which are similar to that reported in the literature [7,44], and the impedance of the graphenecoated electrode varies from 65.82 (at 20 Hz) to 5.10 kΩ (at 1 kHz). The results clearly showed that graphene-coated electrode has lower skin-electrode contact impedance compared with conventional Ag/AgCl electrode, resulting in less noise and higher-quality ECG signal. To further evaluate the performance of the electrodes, ECG signals were analyzed to calculate signal-to-noise ratio (SNR) using the following equation [45]: SNR = 20 log ( S / ( S ′ − S ) )

(4)

where S is the filtered ECG signal with a frequency ranging from 0.5 to 100 Hz and S′ is defined as ECG signal without filtering. Before calculation, the power-line interference (50 Hz) was removed from both signals. SNR values were measured after signal processing steps as 21.23 and 25.32 dB for pregelled Ag/AgCl and GN-based electrodes, respectively. Results here clearly revealed that the proposed GN-based electrodes provided better signal quality of ECG signals and performance evaluation than conventional Ag/AgCl electrodes.

358

CHAPTER 14  GRAPHENE BASED BIOSENSING SYSTEM FOR REAL-TIME HEALTH MONITORING

Another key parameter for wearable ECG monitoring applications is robustness of the electrodes. The proposed GN-based electrodes were tested for 10 subjects continuously for 1 month to see the signal degradation and epidermal issues for long-term applying electrodes on the skin. At the end of the month, it has been clearly seen that there was no big difference in recorded ECG signals compared with tests at the beginning, and the P-QRS-T intervals were still distinguished with a typical ECG pattern in the recorded cardiac signal.

6 ­CONCLUSION AND FUTURE PERSPECTIVES Use of graphene (GN) for the ECG measurements is presented in this book chapter. Properties of graphene material and the methods of its production are discussed in detail with the support of many published literature. The design, development, and characterization of GN-based ECG electrodes are presented with a number of results. Graphene coating was applied to the conventional Ag/AgCl ECG electrodes using chemical vapor deposition (CVD) technique, and characterization of the developed GN-ECG electrode was performed with a set of impedance measurements and Raman spectroscopy measurements. Superior electric conductivity of graphene has shown superior performance in ECG measurements through a number of experimental results. Signal-to-noise ratio of GN-coated ECG measurements is better than that of without GN coating by 20%. Wearable app with a smartphone for realtime and continuous ECG monitoring was also implemented and presented in this chapter. There are still progressive developments need to be considered on GN-based ECG electrode measurements. First of all, the proposed GN-based ECG sensor still has PMMA residues left from transferring GN grown by chemical vapor deposition (CVD), which remains a challenge for maximizing the effect of GN in regards to ECG monitoring. New approaches can be performed for a simple cleaning method for removing PMMA residues that can reduce impairments to the electronic properties of GN-based electrodes. If a high-quality graphene-coating process on Ag substrates can be carried out, then skin-electrode impedance can be much reduced, and the proposed ECG sensor's performance can be improved further. The use of GN-based nanomaterials not only is considered for ECG monitoring applications but also brings attention to the other biosensors and derivative developments. Because of high thermal conductivity properties of GN, promising GN-based temperature sensors can be developed for wearable and health-care applications. Fabricating GN-based temperature sensors and photodetectors can result in better performance for reliable body temperature and photoplethysmography (PPG) applications due to sensitive and promising temperature coefficient resistivity and rapid light responsivity characteristics of graphene.

­REFERENCES [1] J.R. Hampton, The ECG Made Easy, Elsevier Health Sciences; Oxford, UK, 2013. [2] R. Perini, A. Veicsteinas, Heart rate variability and autonomic activity at rest and during exercise in various physiological conditions, Eur. J. Appl. Physiol. 90 (3-4) (2003) 317–325. [3] R.W. Lucky, Automatic equalization for digital communication, Bell Syst. Tech. J. 44 (4) (1965) 547–588. [4] S.P. Bingulac, On the compatibility of adaptive controllers (Published Conference Proceedings style), in: In Proc. 4th Annu. Allerton Conf. Circuits and Systems Theory, 1994, pp. 8–16.

­REFERENCES

359

[5] M.M. Puurtinen, S.M. Komulainen, P.K. Kauppinen, J.A. Malmivuo, J.A. Hyttinen, Measurement of noise and impedance of dry and wet textile electrodes, and textile electrodes with hydrogel, in: In Engineering in Medicine and Biology Society, 2006. EMBS'06. 28th Annual International Conference of the IEEE, IEEE, 2006, pp. 6012–6015. [6] A. Gruetzmann, S. Hansen, J. Müller, Novel dry electrodes for ECG monitoring, Physiol. Meas. 28 (11) (2007) 1375. [7] J.Y. Baek, J.H. An, J.M. Choi, K.S. Park, S.H. Lee, Flexible polymeric dry electrodes for the long-term monitoring of ECG, Sensors Actuators A Phys. 143 (2) (2008) 423–429. [8] V.K. Varadan, P.S. Kumar, S. Oh, H. Kwon, P. Rai, N. Banerjee, R.E. Harbaugh, e-Nanoflex sensor system: smartphone-based roaming health monitor, J. Nanotechnol. Eng. Med. 2 (1) (2011) 011016. [9] J.C. Batchelor, A.J. Casson, Inkjet printed ECG electrodes for long term biosignal monitoring in personalized and ubiquitous healthcare, in: Engineering in Medicine and Biology Society (EMBC), 2015 37th Annual International Conference of the IEEE, 2015, pp. 4013–4016. [10] A.A. Chlaihawi, B.B. Narakathu, S. Emamian, A. Eshkeiti, S.G.A. Reddy, B.J. Bazuin, M.Z. Atashbar, Development of flexible dry ECG electrodes based on MWCNT/PDMS composite, in: SENSORS, 2015, IEEE, 2015, pp. 1–4. [11] K.I. Jang, H.N. Jung, J.W. Lee, S. Xu, Y.H. Liu, Y. Ma, J.W. Jeong, Y.M. Song, J. Kim, B.H. Kim, A. Banks, Ferromagnetic, folded electrode composite as a soft interface to the skin for long‐term electrophysiological recording, Adv. Funct. Mater. 26 (40) (2016) 7281–7290. [12] N.  Celik, W.  Balachandran, N.  Manivannan, Graphene-based biosensors: methods, analysis and future perspectives, IET Circuits Devices Syst. 9 (6) (2015) 434–445. [13] O. Preining, The physical nature of very, very small particles and its impact on their behaviour, J. Aerosol Sci. 29 (5) (1998) 481–495. [14] A.A.  Balandin, S.  Ghosh, W.  Bao, I.  Calizo, D.  Teweldebrhan, F.  Miao, C.N.  Lau, Superior thermal conductivity of single-layer graphene, Nano Lett. 8 (3) (2008) 902–907. [15] K.S. Novoselov, A.K. Geim, S.V. Morozov, D. Jiang, Y. Zhang, S.V. Dubonos, I.V. Grigorieva, A.A. Firsov, Electric field effect in atomically thin carbon films, Science 306 (5696) (2004) 666–669. [16] C.E.E.  Rao, A.E.  Sood, K.E.  Subrahmanyam, A.  Govindaraj, Graphene: the new two‐dimensional nanomaterial, Angew. Chem. Int. Ed. 48 (42) (2009) 7752–7777. [17] A.K. Geim, K.S. Novoselov, The rise of graphene, Nat. Mater. 6 (3) (2007) 183–191. [18] F. Schwierz, Graphene transistors, Nat. Nanotechnol. 5 (7) (2010) 487–496. [19] M.D. Stoller, S. Park, Y. Zhu, J. An, R.S. Ruoff, Graphene-based ultracapacitors, Nano Lett. 8 (10) (2008) 3498–3502. [20] S. Casaluci, M. Gemmi, V. Pellegrini, A. Di Carlo, F. Bonaccorso, Graphene-based large area dye-sensitized solar cell modules, Nanoscale 8 (9) (2016) 5368–5378. [21] N. Celik, N. Manivannan, A. Strudwick, W. Balachandran, Graphene-enabled electrodes for electrocardiogram monitoring, Nanomaterials 6 (9) (2016) 156. [22] F. Schedin, A.K. Geim, S.V. Morozov, E.W. Hill, P. Blake, M.I. Katsnelson, K.S. Novoselov, Detection of individual gas molecules adsorbed on graphene, Nat. Mater. 6 (9) (2007) 652–655. [23] S. Bae, H. Kim, Y. Lee, X. Xu, J.S. Park, Y. Zheng, J. Balakrishnan, T. Lei, H.R. Kim, Y.I. Song, Y.J. Kim, Roll-toroll production of 30-inch graphene films for transparent electrodes, Nat. Nanotechnol. 5 (8) (2010) 574–578. [24] K.I. Bolotin, K.J. Sikes, Z. Jiang, M. Klima, G. Fudenberg, J. Hone, P. Kim, H.L. Stormer, Ultrahigh electron mobility in suspended graphene, Solid State Commun. 146 (9) (2008) 351–355. [25] K.S. Novoselov, V.I. Fal, L. Colombo, P.R. Gellert, M.G. Schwab, K. Kim, A roadmap for graphene, Nature 490 (7419) (2012) 192–200. [26] C. Berger, Z. Song, T. Li, X. Li, A.Y. Ogbazghi, R. Feng, Z. Dai, A.N. Marchenkov, E.H. Conrad, P.N. First, W.A. De Heer, Ultrathin epitaxial graphite: 2D electron gas properties and a route toward graphene-based nanoelectronics, J. Phys. Chem. 108 (2004) 19912–19916.

360

CHAPTER 14  GRAPHENE BASED BIOSENSING SYSTEM FOR REAL-TIME HEALTH MONITORING

[27] W.A. De Heer, C. Berger, X. Wu, P.N. First, E.H. Conrad, X. Li, T. Li, M. Sprinkle, J. Hass, M.L. Sadowski, M. Potemski, Epitaxial graphene, Solid State Commun. 143 (1) (2007) 92–100. [28] Y. Zhang, L. Zhang, C. Zhou, Review of chemical vapor deposition of graphene and related applications, Acc. Chem. Res. 46 (10) (2013) 2329–2339. [29] X. Li, W. Cai, J. An, S. Kim, J. Nah, D. Yang, R. Piner, A. Velamakanni, I. Jung, E. Tutuc, S.K. Banerjee, Large-area synthesis of high-quality and uniform graphene films on copper foils, Science 324 (5932) (2009) 1312–1314. [30] Q. Yu, J. Lian, S. Siriponglert, H. Li, Y.P. Chen, S.S. Pei, Graphene segregated on Ni surfaces and transferred to insulators, Appl. Phys. Lett. 93 (11) (2008) 113103. [31] P. Sutter, J.T. Sadowski, E. Sutter, Graphene on Pt (111): growth and substrate interaction, Phys. Rev. B 80 (24) (2009) 245411. [32] A. Kumar, C.H. Lee, Synthesis and biomedical applications of graphene: present and future trends, in: Advances in Graphene Science, Intech Open; Croatia, 2013, pp. 5772–5778. [33] Y. Hernandez, V. Nicolosi, M. Lotya, F.M. Blighe, Z. Sun, S. De, I.T. McGovern, B. Holland, M. Byrne, Y.K. Gun'Ko, J.J. Boland, High-yield production of graphene by liquid-phase exfoliation of graphite, Nat. Nanotechnol. 3 (9) (2008) 563–568. [34] A. Ciesielski, P. Samorì, Graphene via sonication assisted liquid-phase exfoliation, Chem. Soc. Rev. 43 (1) (2014) 381–398. [35] Y. Si, E.T. Samulski, Synthesis of water soluble graphene, Nano Lett. 8 (6) (2008) 1679–1682. [36] A.C. Ferrari, Raman spectroscopy of graphene and graphite: disorder, electron–phonon coupling, doping and nonadiabatic effects, Solid State Commun. 143 (1) (2007) 47–57. [37] A.C.  Ferrari, J.C.  Meyer, V.  Scardaci, C.  Casiraghi, M.  Lazzeri, F.  Mauri, S.  Piscanec, D.  Jiang, K.S. Novoselov, S. Roth, A.K. Geim, Raman spectrum of graphene and graphene layers, Phys. Rev. Lett. 97 (18) (2006) 187401. [38] C. Mattevi, H. Kim, M. Chhowalla, A review of chemical vapour deposition of graphene on copper, J. Mater. Chem. 21 (10) (2011) 3324–3334. [39] K.K. Kim, K.S. Park, Effective coupling impedance for power line interference in capacitive-coupled ECG measurement system, in: Information Technology and Applications in Biomedicine, 2008. ITAB 2008. International Conference on IEEE, IEEE, 2008, pp. 256–258. [40] M. Neuman, Biopotential electrodes, in: Medical Instrumentation Application and Design, third ed., Wiley, New York, 1988. [41] B.R. Eggins, Skin contact electrodes for medical applications, Analyst 118 (4) (1993) 439–442. [42] O.G. Martinsen, S. Grimnes, Bioimpedance and Bioelectricity Basics, Academic Press; Cambridge, USA, 2011. [43] N. Celik, N. Manivannan, W. Balachandran, Evaluation of a behind-the-ear ECG device for smartphone based integrated multiple smart sensor system in health applications, Int. J. Adv. Comput. Sci. Appl. 7 (7) (2016) 409–418. [44] Y. Meng, Z. Li, J. Chen, A flexible dry electrode based on APTES-anchored PDMS substrate for portable ECG acquisition system, Microsyst. Technol. 22 (8) (2016) 2027–2034. [45] J.V. Tranquillo, Biomedical signals and systems, Synthesis Lectures on Biomedical Engineering 8 (3) (2013) 1–233.