**5. Biofunctionalized magnetic bead detection for state of the art lab-on-achip**

Ever since the report of Baselt *et al.* on a magnetoresistive-based biochip with magnetic la‐ bels instead of fluorescent labels [3], the magnetic biochip has been extensively investigated as an advanced tool for sensitive detection of low bio-target concentration in body fluids for early diagnostics. Obviously, the focus in these investigations lies in development of a high sensitive magnetic field sensor that is optimized for magnetic label detection, and therefore different magnetoresistive sensing approaches, including the one that has just been descri‐ bed above i.e., hybrid AMR and PHR ring sensor, were adopted subsequently for this pur‐ pose. All these magnetic biosensors detect the stray field of magnetic particles that are bound to biological molecules. Since the biological environment is normally non-magnetic, the possibility of false signals being detected is negligible. In addition, the properties of magnetic particles are also stable over time and they may also be manipulated via magnetic forces, which can be produced by current lines that are fabricated into the chip itself. The advantages of magnetic labeling techniques have ultimately led the researchers to intensify their efforts in developing modern technologies for on-chip integration of micro- and nano‐ scale magnetics with molecular biology with a final goal of realizing highly sensitive, fast, reliable, cost-effective, portable and easy-to-use biomolecular sensor, the so called *magnetic lab-on-a-chip*.

#### **5.1. Magnetic beads**

**Figure 18.** The calculation and experimental results of PHR and AMR output voltage components versus r/ω ratio of the ring. The insets show schematics of a ring junction with defined r and ω, and a representative PHE voltage profile

**Figure 19.** Experimental results of the sensitivity versus *r*/ω ratio of the rings using a Ta(5)/Ru(1)/NiCo(10)/ IrMn(10)Ru(1)/Ta(5) (nm) bilayer thin film and trilayer thin film Ta(3)/NiFe(10)/Cu(0.12)/IrMn(10) /Ta(3) (nm).

of ring sensor for *r* = 150 μm, ω = 20 μm.

222 State of the Art in Biosensors - General Aspects

Superparamagnetic nanoparticles coated with Streptavidin make ideal labels in bio-applica‐ tions using magnetic sensors, because they can be readily magnetized to large magnetic mo‐ ments. Most of our experiments were carried out with Dynabeads®M-280, which are composed of ultra small Fe2O3 nanopaticles embedded in a polymer matrix and the Strepta‐ vidin was conjugateed with the surface of the beads. The magnetization curve of the mag‐ netic beads is shown in Fig. 20 [27, 39].

**Figure 20.** Magnetization curve of Dynabeads M-280 Streptavidin. This is supported by Dynal company.

When the magnetic bead appears on the sensor surface under an external magnetic field the magnetic field strength produced by a single bead can be estimated as [12, 14]

$$H = \frac{MR}{3r^3} (\Re(\hat{M} \cdot \hat{r})\hat{r} - \hat{M})\tag{10}$$

where *M*, *M* ^ are the magnitude and unit vector of magnetization. *<sup>R</sup>* is the bead radius, and *r, r* ^ are the magnitude and unit vactor of the distance from the center of the bead to observa‐ tion point as shown in Fig. 21.

**Figure 21.** Schematic of a bead with the radius *R* placed above the sensor, *r* is the distance from the center of the bead to the observation point, *z*<sup>o</sup> is the vertical distance from the center of the bead to the sensor, ρ is the distance in the sensor plane from the center of the bead to the observation point.

Assuming that the applied field is in *x* direction in a polar coordinate system then Eq. (10) can be rewritten as:

$$H\_{\mathbf{x}} = H \stackrel{\wedge}{\mathbf{x}} = \frac{M \, R}{3r^3} \text{(3} \sin^2 \theta \cos^2 \varphi - 1\text{)}\tag{11}$$

with *x* ^*r* ^=sin*<sup>θ</sup>*cos*φ* and *<sup>x</sup>* ^*<sup>M</sup>* =1when converting from polar coordinate system to spherical coordinate system.

Substituting sin*<sup>θ</sup>* <sup>=</sup> *<sup>ρ</sup> r* and *r* = *ρ* <sup>2</sup> + *z* <sup>2</sup> into Eq. (11), *H*x can be rewritten as

$$H\_{\chi} = \frac{MR}{3} \frac{3\rho^2 \cos^2 \phi - (\rho^2 + z\_{\text{o}}^2)}{(\rho^2 + z\_{\text{o}}^2)^5} \tag{12}$$

Following the Eq. (12), stray field of magnetic bead reaches maximum when *ρ* = 0, in this case (*H*x) max <sup>=</sup> <sup>−</sup> *<sup>M</sup> <sup>R</sup>* <sup>3</sup> *z*o <sup>3</sup> . This field reaches maximum at a right angle to the magnetization of the bead (*r* ≡ *z*o in Fig. 21) and decreases at other points on the sensor plane. The effective field of a bead influences the sensor, *H*x is integrated over a general sensor area, A.

$$
\langle H\_{\mathbf{x}} \rangle = \frac{1}{A} \Big[ H\_{\mathbf{x}} dA \tag{13}
$$

If the sensor geometry is considered as a circle, the effective field of a bead can be calculated from Eq. (13) as

$$\{H\_{\chi}\} = -\frac{M \, R}{3z\_{\text{o}}^3} \frac{1}{\left(1 + \frac{\rho\_{\text{s}}^2}{z\_{\text{o}}^2}\right)^{3/2}} \tag{14}$$

Here *ρs* is the radius of the circular ring sensor

**Figure 20.** Magnetization curve of Dynabeads M-280 Streptavidin. This is supported by Dynal company.

magnetic field strength produced by a single bead can be estimated as [12, 14]

<sup>3</sup>*<sup>r</sup>* <sup>3</sup> (3(*<sup>M</sup>*

*<sup>H</sup>* <sup>=</sup> *<sup>M</sup> <sup>R</sup>* <sup>3</sup>

where *M*, *M*

tion point as shown in Fig. 21.

224 State of the Art in Biosensors - General Aspects

*r*

When the magnetic bead appears on the sensor surface under an external magnetic field the

^ <sup>⋅</sup> *<sup>r</sup>* ^)*r* ^−*<sup>M</sup>*

^ are the magnitude and unit vactor of the distance from the center of the bead to observa‐

**Figure 21.** Schematic of a bead with the radius *R* placed above the sensor, *r* is the distance from the center of the bead to the observation point, *z*<sup>o</sup> is the vertical distance from the center of the bead to the sensor, ρ is the distance in

the sensor plane from the center of the bead to the observation point.

^ are the magnitude and unit vector of magnetization. *<sup>R</sup>* is the bead radius, and *r,*

^ ) (10)

And if the sensor geometry is quadrate the effective field is given by

$$\{H\_{\chi}\} = -\frac{M}{3z\_{\text{o}}^3} \frac{1}{(1+\frac{\omega^2}{4z\_{\text{o}}^2})(1+\frac{\omega^2}{2z\_{\text{o}}^2})} \tag{15}$$

here *ω* is the width of the cross-junction sensor

It is revealed from Eq. (14) and Eq. (15) that the field effect to the sensor is very much de‐ pending on the size of the sensor, , it is proportional to the invert cube of radius of circular sensor or of the width of a quadratic sensor ( <sup>1</sup> *<sup>ρ</sup>* <sup>3</sup> or 1 *<sup>ω</sup>* <sup>3</sup> ).

#### **5.2. Biofunctionalization of the beads**

It is known that the biotin-streptavidin is one of the strongest non-covalent biological inter‐ action systems having a dissociation constant, '*K*d', in the order of 4 × 10-14 M leading to the strength and specificity of the interaction to be one of the most widely used affinity pairs in molecular, immunological and cellular assays [40]. Usually in most assays, streptavidin is coupled to a solid phase such as a magnetic bead, or a biosensor chip, while biotin is cou‐ pled to the biomarker of interest, often a nucleic acid or antibody. Taking advantages of magnetic labels and specific ligand-receptor interactions of the biomolecules one can manip‐ ulate, separate and detect specific biomolecules.

**Figure 22.** Procedure for the immobilization of fluorescent labeled biotin on the Streptavidin coated dynabeads measured by confocal optical microscope.

To demonstrate the translocation of streptavidin-biotin magnetic labels using the micro system, we have chosen the commercially available streptavidin coated magnetic beads (Dynabead® M-280) of 2.8 *µ*m size to bind with fluorescent labelled biotin. Atto 520 is a new label with high molecular absorption (110.000) and quantum yield (0.90) as well as sufficient stokes shift (excitation maximum 520 nm, emission maximum 524 nm). Due to an insignificant triplet formation rate it is well suited for single molecule detection appli‐ cations. In this experiment, Atto 520 biotin is attached on the streptavidin coated magnet‐ ic beads and observed the fluorescence signal through confocal microscope. In order to attach the Atto 520 biotin on streptavidin coated magnetic labels, we have taken, 5 *µ*l of streptavidin coated magnetic labels (Dynabead® M-280) mixed with 0.1 M PBS buffer sol‐ ution (90 *µ*l) with pH of 7, and 5 *µ*l of fluorescent label biotin (chemical concentration of fluorescent label was 1 mg/200 *µ*l in EtOH) also added to the previous mixing solution and continuously stirring the solution for 2 hours at room temperature for the reaction completion.

Fig. 22 provides the direct evidence of protein immobilization which was obtained by im‐ mobilizing green fluorescent protein (GFP) and observed the fluorescence through confocal laser microscopy.

#### **5.3. Sensor size and bead detection capability**

It is revealed from Eq. (14) and Eq. (15) that the field effect to the sensor is very much de‐ pending on the size of the sensor, , it is proportional to the invert cube of radius of circular

It is known that the biotin-streptavidin is one of the strongest non-covalent biological inter‐ action systems having a dissociation constant, '*K*d', in the order of 4 × 10-14 M leading to the strength and specificity of the interaction to be one of the most widely used affinity pairs in molecular, immunological and cellular assays [40]. Usually in most assays, streptavidin is coupled to a solid phase such as a magnetic bead, or a biosensor chip, while biotin is cou‐ pled to the biomarker of interest, often a nucleic acid or antibody. Taking advantages of magnetic labels and specific ligand-receptor interactions of the biomolecules one can manip‐

**Figure 22.** Procedure for the immobilization of fluorescent labeled biotin on the Streptavidin coated dynabeads

To demonstrate the translocation of streptavidin-biotin magnetic labels using the micro system, we have chosen the commercially available streptavidin coated magnetic beads (Dynabead® M-280) of 2.8 *µ*m size to bind with fluorescent labelled biotin. Atto 520 is a new label with high molecular absorption (110.000) and quantum yield (0.90) as well as sufficient stokes shift (excitation maximum 520 nm, emission maximum 524 nm). Due to

*<sup>ρ</sup>* <sup>3</sup> or

1 *<sup>ω</sup>* <sup>3</sup> ).

sensor or of the width of a quadratic sensor ( <sup>1</sup>

ulate, separate and detect specific biomolecules.

measured by confocal optical microscope.

**5.2. Biofunctionalization of the beads**

226 State of the Art in Biosensors - General Aspects

For the micro-bead detection using a PHE sensor, it is noted that the magnetization of the magnetic sphere is purely a dipole at the center of the sphere with a magnetic field at a dis‐ tance identified by the dipole field from Eq. (15). The stray field of a single bead on the sen‐ sor surface could be crudely calculated by [41]

$$H\_{\text{bed}} \approx -\frac{\chi V}{4\pi r^3} H \tag{16}$$

where *V* is the volume of magnetic bead, *χ*is the volume susceptibility of magnetic beads. This stray field is in the opposite direction to the applied field, thus it reduces the effective field on the sensor surface. Under the experiment conditions, the stray field of N beads on the sensor surface reduced the sensor output signal as follows:

$$V\_{\rm head} = V\_{\rm PHR}(H\_{\rm eff}) \approx V\_{\rm PHR} \{ 1 - \rm N H\_{\rm head} \} \approx V\_{\rm PHR} + \Delta V\_{\rm head} \tag{17}$$

where *H*eff is the effective field on the sensor surface, *S* is the sensor sensitivity of PHR sensor. The voltage signal, *ΔV*bead generated by the magnetic bead themselves can be ex‐ pressed as

$$
\Delta V\_{\text{bead}} = V\_{\text{bead}} - V\_{\text{PHR}} \approx V\_{\text{PHR}} \left( \frac{\text{N} \chi V}{4 \pi r^3} H \right) \tag{18}
$$

By substituting the value *χ* =0.13 [39] and *r* =1.55 µm (the distance including the radius of Dynabeads® M-280 and the thickness of passivated SiO and Ta layers) into Eq. (17), the stray field of single bead is estimated to be *H*bead ~ 0.03 *H* under the applied field. The num‐ ber of bead separately placed on the sensor surface can be calculated using the PHR sensor.

Fig. 23 shows the *V*PHR and the *V*bead in the functions of magnetic field with number of bead N=1, 5, and 10, respectively. It is clearly shown that the beads on the sensor surface modify the PHR signal due to the small stray field compared with applied magnetic field.

In the PHR sensor, the *V*PHR can be used the reference signal. The difference voltage *ΔV*bead between the *V*PHR and *V*bead can be estimated, which is shown in Fig. 24.

The pure bead signal *ΔV*bead is small compared with *V*PHR. However, the *ΔV*bead changes with the applied magnetic field and show maximum and minimum values at special mag‐ netic field, which is due to the PHR sensor performance. Therefore, the bead detection capa‐ bility can be determined at the maximum and minimum *ΔV*bead. If we set the applied field at the maximum or minimum value of *ΔV*bead, we can detect the magnetic bead with high signal voltage.

**Figure 23.** *V*PHR without bead (black solid line) and *V*bead with bead (N=1, 5 and 10) by using the F/AF bilayers in the functions of applied magnetic field *H*.

**Figure 24.** Calculation of Δ*V*bead of the PHR sensor with N=1, 5 and 10.

#### *5.3.1. Multi-bead detection*

Fig. 23 shows the *V*PHR and the *V*bead in the functions of magnetic field with number of bead N=1, 5, and 10, respectively. It is clearly shown that the beads on the sensor surface modify

In the PHR sensor, the *V*PHR can be used the reference signal. The difference voltage *ΔV*bead

The pure bead signal *ΔV*bead is small compared with *V*PHR. However, the *ΔV*bead changes with the applied magnetic field and show maximum and minimum values at special mag‐ netic field, which is due to the PHR sensor performance. Therefore, the bead detection capa‐ bility can be determined at the maximum and minimum *ΔV*bead. If we set the applied field at the maximum or minimum value of *ΔV*bead, we can detect the magnetic bead with high

**Figure 23.** *V*PHR without bead (black solid line) and *V*bead with bead (N=1, 5 and 10) by using the F/AF bilayers in the

the PHR signal due to the small stray field compared with applied magnetic field.

between the *V*PHR and *V*bead can be estimated, which is shown in Fig. 24.

signal voltage.

228 State of the Art in Biosensors - General Aspects

functions of applied magnetic field *H*.

We performed the magnetic bead detection using PHR sensor using Ta(3)/NiFe(16)/Cu(1.2)/ NiFe(2)/IrMn(10)/Ta(3) (nm) to demonstrate the feasibility of magnetic bead detection for bio applications. The diluted 0.1 % magnetic bead solution streptavidin coated Dynabeads® M-280 is used for bead drop and wash experiments on the sensor surface. The real-time pro‐ file measurements of the PHE voltage for magnetic bead detection are carried out in the op‐ timum conditions, that is, in an applied magnetic field of 7 Oe and with a sensing current of 1 mA. The results are illustrated in Fig. 25 for three consecutive cycles. The lower state rep‐ resents the signal change in sensor output voltage after dropping the magnetic bead solution on the sensor surface whereas the higher state represents the sensor output voltage after washing magnetic beads from the sensor surface. Total output signal annuls in three consec‐ utive cycles were found to be about 7.1 *µ*V, 16 *µ*V and 21.8*µ*V for the first step and 11.3 *µ*V and 16.7 *µ*V in the second step of the second and third cycles, respectively. It is clearly shown from the figure that for the first cycle, the signal changed by one-step and the signal was further changed into two steps in the second and third cycles.

This two step-type profile is due to the aggregation process of the magnetic beads on the sensor surface. The aggregation of the magnetic beads occurs at the drying stage. That is, after dropping the bead solution on the sensor surface, it needs some time to dry. The first step changes of the signals are assumed to be due to the viscous flow motion for stabiliza‐ tion as well as the Brownian motion of the beads. When the solution dries, the beads rear‐ range. During this time, some beads aggregate and become clusters on the sensor surface.

**Figure 25.** Real-time profile of PHR sensor under an applied magnetic field of 7 Oe with the sensing current of 1 mA

This lessens the total stray field on the sensor surface and hence, the second step in the sec‐ ond and third cycles was observed in the real-time profile.

In the process of analyzing the micro-bead detection using PHE sensor, it is noted that the direction of magnetic field *H* and the stray field of magnetic bead on the sensor surface *H*bead (Eq. (16)) are oppositely aligned, and thus the effective field on the sensor surface is re‐ duced.

Thus, a rough estimate of the number of magnetic particles on the sensor surface in this identical experiment based on the reduced stray field and sensor output signal can be ex‐ pressed from Eq. (18) by rephrasing it again here for better clarity:

$$
\Delta V\_{\text{bead}} = V\_{\text{PHR}} \left( \frac{\text{N} \chi V}{4 \pi r^3} H \right)
$$

By substituting the value *χ* = 0.13 and *r* = 1.55 *µ*m (the distance including the radius of Dyna‐ beads® M-280 and the thickness of passivated SiO2 and Ta layers) into Eq. (16), the stray field of single bead is estimated to be 2.2×10-2 Oe under the applied field of 7 Oe. Theoretically, with the sensor sensitivity *S* = 7.6 *µ*V/Oe and the sensing current *I* = 1 mA, the number of beads separately placed on the sensor surface can be calculated in the first step of the three cycles by using Eq. (18), which are estimated to be about 4, 10 and 13 beads, respectively.

These estimated results strengthen our explanation. It is clearly shown in the first cycle, the number of beads on the sensor surface is estimated to be small, and the distance among beads on the sensor junction is far enough to avoid the effect from the rearrangement of beads during the drying stage. In the second and third cycles, the number of magnetic beads on the sensor junction is larger; they easily aggregate to become clusters under applied mag‐ netic field due to short bead-bead distance

#### *5.3.2. Single bead detection*

This lessens the total stray field on the sensor surface and hence, the second step in the sec‐

**Figure 25.** Real-time profile of PHR sensor under an applied magnetic field of 7 Oe with the sensing current of 1 mA

In the process of analyzing the micro-bead detection using PHE sensor, it is noted that the direction of magnetic field *H* and the stray field of magnetic bead on the sensor surface *H*bead (Eq. (16)) are oppositely aligned, and thus the effective field on the sensor surface is re‐

Thus, a rough estimate of the number of magnetic particles on the sensor surface in this identical experiment based on the reduced stray field and sensor output signal can be ex‐

By substituting the value *χ* = 0.13 and *r* = 1.55 *µ*m (the distance including the radius of Dyna‐ beads® M-280 and the thickness of passivated SiO2 and Ta layers) into Eq. (16), the stray field of single bead is estimated to be 2.2×10-2 Oe under the applied field of 7 Oe. Theoretically, with the sensor sensitivity *S* = 7.6 *µ*V/Oe and the sensing current *I* = 1 mA, the number of beads separately placed on the sensor surface can be calculated in the first step of the three cycles by using Eq. (18), which are estimated to be about 4, 10 and 13 beads, respectively.

These estimated results strengthen our explanation. It is clearly shown in the first cycle, the number of beads on the sensor surface is estimated to be small, and the distance among beads on the sensor junction is far enough to avoid the effect from the rearrangement of beads during the drying stage. In the second and third cycles, the number of magnetic beads

ond and third cycles was observed in the real-time profile.

pressed from Eq. (18) by rephrasing it again here for better clarity:

duced.

*<sup>Δ</sup>V*bead <sup>=</sup>*V*PHR( <sup>N</sup>*χ<sup>V</sup>*

<sup>4</sup>*π<sup>r</sup>* <sup>3</sup> *<sup>H</sup>* )

230 State of the Art in Biosensors - General Aspects

We performed single magnetic bead detection experiments on several kinds of sensor struc‐ tures such as spin-valve and bilayer exchange biased thin films [27, 42 - 44], and the repre‐ sentative results are being presented here. For the purpose of performing single micro-bead detection, the PHR sensor with the junction size of 3 *µ*m × 3 *µ*m was fabricated using Ta(5)/ NiFe(16)/Cu(1.2)/NiFe(2)/IrMn(15)/Ta(5) (nm). This is the optimized spin-valve thin film for the PHR sensor in our lab. A droplet of 0.1 % dilute solution of the Dynabeads® M-280 was introduced on the surface of the sensor. A single micro-bead was isolated and positioned on the center of the sensor junction by using a micro magnetic needle which is known as a tweezer method. The magnetic needle was prepared by using a soft magnetic micro wire, the wire is magnetized by attaching a permanent magnet to one end of the wire, the single magnetic bead is attracted with the other end due to the magnetic field of the wire and it is dragged and positioned to the center of the sensor junction. It is noteworthy that the mag‐ netic bead is attracted by the magnetic force; this force is strong enough to compensate the Brownian motion during the experiment. The experiment was carried out under the obser‐ vation of an optical microscope. When the solution dried, the bead was fixed on to the sen‐ sor surface.

Since the magnetic properties of the MR as well as the PHE response to the magnetic field are described in the previous section, the results of single bead detection using 3 *µ*m × 3 *µ*m PHR sensor will be discussed here.

The SEM image of a single bead on the center of the sensor junction is shown in the Fig. 26(a). The voltage profiles of the PHE sensor in the absence and presence of a single microbead are presented in Fig. 26(b) by black circle and red rectangle ones, respectively. It is shown from the figure that in the increasing region of the PHE voltage profile (in the field ranging from 0 Oe to 10.6 Oe), the *V*PHE(*H*) is decreased when the magnetic bead exists on the sensor surface and vice versa for the decreasing region of the PHE voltage profile (at the fields exceeding 10.6 Oe).

For understanding the role of a single micro-bead detection using a PHE sensor, we consider the voltage drop by stray field of a single magnetic bead. The calculation method is the same as deduced for Eq. (18). And when considering that the magnetic bead is located on the cen‐ ter of sensor junction, the stray field affects the PHE voltage as follows:

$$V\_{\rm stray} = I \times S \times \left(1 - k \frac{\chi V\_{\rm dead}}{4\pi \varepsilon^3} \right) \times H\_{\rm app} \tag{19}$$

where *V*stray denotes the voltage change due to the stray field of magnetic bead, *S* = ∂*V*PHE <sup>∂</sup>*H* is the sensitivity of the sensor at instantaneous applied fields.

**Figure 26.** (a) The SEM image of the sensor junction in the presence of a single micro-bead, (b) the theoretical and experimental PHE voltage profiles in the absence and presence of a micro-bead, (b-1) enlarged picture of the increas‐ ing PHE voltage region at the field range of 4.75-6.74 Oe and (b-2) enlarged picture of the PHE voltage profiles around the maximum PHE voltages [44].

By substituting *χ* = 0.13 [39] in Eq. (19) with active fraction of *k* = 0.62 and *z* = 1.55 *µ*m (along with 150 nm thick SiO2 passivation layer and 1.4 *µ*m of magnetic bead radius), the PHE volt‐ age is calculated at instant applied fields for the presence of a micro-bead. The solid lines in the Fig. 26(b) illustrate the calculated profiles for the case of absence and presence of a mi‐ cro-bead, respectively. These calculated results are in good agreement with the experimental results.

By comparing the PHE voltage profiles in the absence and presence of a micro-bead, one can find that (*i*) at low magnetic fields, the PHE voltage increases with the field increase, *i.e*. the sensitivity of the sensor is positive. In this case, the presence of magnetic bead lessens its PHE voltage as illustrated in Fig. 26(b-1). (*ii*) In the presence of the magnetic bead, the maxi‐ mum PHE voltage shifts to a higher field with an amount of *H*bead as presented in Eq. (16); at

**Figure 27.** Voltage change of the PHR sensor versus applied field when a single magnetic bead appear on the sensor surface.

about 10 Oe this stray field strength is approximately 0.43 Oe. (*iii*) At higher applied fields (> 10 Oe), the PHE voltage decreases with the field increase, *i.e*. the sensitivity of the sensor is negative. In this case, the presence of magnetic bead increases the PHE voltage with an amount of *k χV*bead <sup>4</sup>*π<sup>z</sup>* <sup>3</sup> *I* ⋅*S* ⋅*H*app. This is clearly evident in the Fig. 26(b-2) and thus the PHE sig‐ nal satisfies Eq. (19).

By substituting *χ* = 0.13 [39] in Eq. (19) with active fraction of *k* = 0.62 and *z* = 1.55 *µ*m (along with 150 nm thick SiO2 passivation layer and 1.4 *µ*m of magnetic bead radius), the PHE volt‐ age is calculated at instant applied fields for the presence of a micro-bead. The solid lines in the Fig. 26(b) illustrate the calculated profiles for the case of absence and presence of a mi‐ cro-bead, respectively. These calculated results are in good agreement with the experimental

**Figure 26.** (a) The SEM image of the sensor junction in the presence of a single micro-bead, (b) the theoretical and experimental PHE voltage profiles in the absence and presence of a micro-bead, (b-1) enlarged picture of the increas‐ ing PHE voltage region at the field range of 4.75-6.74 Oe and (b-2) enlarged picture of the PHE voltage profiles

By comparing the PHE voltage profiles in the absence and presence of a micro-bead, one can find that (*i*) at low magnetic fields, the PHE voltage increases with the field increase, *i.e*. the sensitivity of the sensor is positive. In this case, the presence of magnetic bead lessens its PHE voltage as illustrated in Fig. 26(b-1). (*ii*) In the presence of the magnetic bead, the maxi‐ mum PHE voltage shifts to a higher field with an amount of *H*bead as presented in Eq. (16); at

results.

around the maximum PHE voltages [44].

232 State of the Art in Biosensors - General Aspects

In particular, at low field range, a very good linear and large change of the PHE voltage al‐ ways occur, so this field range is usually chosen to demonstrate the feasibility of the digital detection of the magnetic beads [10-14]. In our approach for this sensor, the signal change versus the applied field is extracted from PHE voltage curves in the presence and absence of magnetic bead, the result is drawn in Fig. 27, the maximum change of *V*PHE(*H*) about 1.14 *µ*V can be obtained at the applied field ~ 5.6 Oe. This calculated result satisfies Eq. (19). Further, Fig 26(b) shows that there is a very good agreement between the single bead measurement data and the theoretical curves. There is only a very small noise scatter of experiment data from the fitting curve, this is the evidence showing that the fabricated PHE sensor has high SNR. Therefore, the PHE sensor has advantages for more accurate detection of the small stray fields of magnetic beads.

This simple calculation is suitable for the effect of a single bead on the center of small size sensor junction. When the area of the sensor junction is larger than the area of mag‐ netic beads, the calculation must be considered the effect of the magnetic bead from dif‐ ferent positions of sensor junction and the contribution of nearby beads or chains of beads on the sensor. In such a case the output signal changes negative for the bead in‐ side of the sensor junction and changes positive for the beads outside of the junction. Moreover, the signal change does not depend on the number of magnetic beads propor‐ tionally. This was studied systematically and was reported by P.P. Freitas *et al*.,[23], L. Ejsing et al., [9] and Damsgaard *et al*., [45].

#### **5.4. Integration of magnetic sensors/microfluidic channels**

In this part, we design and optimize the planar Hall ring sensor for detecting the hydro‐ dynamic magnetic labels. Once the magnetic labels appear on one arm of the ring sensor, the resistance of the sensor will be changed, the role of resistance change obey the Wheatstone bridge circuit geometry hence the sensor is very sensitive to detect the mag‐ netic labels.

Planar Hall ring sensor was fabricated by photolithography technique. Sensor material Ta(3)/NiFe(10)/IrMn(10)/Ta(3) (nm) was fabricated by using a DC sputtering system with the based pressure of 7×10-8 Torr. The field sensitivity of the ring sensor based on the bilayer thin film was found to be about 0.3 mV.Oe-1. The sensor was integrated with a microfluidic channel, which can produce the laminar flow of the magnetic labels (beads and/or tags) in the specific arms of the ring sensor by hydrodynamic flow focusing technique. This magnet‐ ic platform can detect even a single magnetic bead of 2.8 *µ*m motion in real time by the measurement system with a sampling rate of 5 kHz.

The schematic representing the integrated magnetic platform is shown in Fig. 28. In magnetic bead separation experiments initially the magnetic beads with different sizes are injected into the main stream of the microfluidic channel with certain fluidic flow rate. Then the beads are gathered at the weir in the fluid channel and then sorted according to the attractive force exert‐ ed on the magnetic bead by the magnetic elements/magnetic pathways. Therefore, the labeled magnetic beads of same kind will attract to one of the magnetic pathways in the sub channel. The weir at the entrance of the sub-channels opposes the beads temporarily for magnetic beads whose magnetization is insufficient to be attracted by the magnetic elements. But, the beads whose magnetization is sufficient to be attracted by the poles of the saturated ellipses due to the external rotating magnetic field can overcome the weir.

After successful separation of the magnetic beads of different sizes we wish to adopt two types of different sensing techniques such as an array of PHR biosensors and multi-seg‐ mented nanowires. The planar array of PHR sensor can detect magnetic beads with micron size only. But in case of nanometer size magnetic beads, we wish to use simple read out technique of multi-segmented nanowires. We are also planning to combine magnetic path‐ way method with the microwire and coil method.


**Figure 28.** Schematic represents the magnetic platform integrating an array of planar Hall ring sensors and a micro‐ fluidic channel.
