DSpace at VNU: Planar Hall resistance sensor for biochip application

In comparison with the other groups, our sensor performance is highlighted with the advantages of increased stability and high signal to noise; as such, the planar Hall effect sensor’s b

phys stat sol (a) 204, No 12, 4053–4057 (2007) / DOI 10.1002/pssa.200777162

Paper

Planar Hall resistance sensor for biochip application

N T Thanh1,2, B Parvatheeswara Rao1,2, N H Duc3, and CheolGi Kim*1

1 Department of Materials Science and Engineering, Chungnam National University, 220 Gung-Dong, Yu-Seong Gu, Daejeon 305-764, Korea

2 Department of Physics, Andhra University, Visakhapatnam 530003, India

3 College of Technology, Vietnam National University, 144 Xuan Thuy, Cau Giay, Hanoi, Vietnam

Received 8 May 2007, revised 21 October 2007, accepted 30 October 2007

Published online 10 December 2007

PACS 07.07.Df, 72.25.Mk, 87.83.+a

In this work, we introduce a new type of sensor by using planar Hall effect in spin valve structure for bio-chip application due to advantage of increasing sensor sensitivity A single Dynabeads® M-280 Strepta-vidin detection has been accomplished with the sensor pattern size of 3 × 3 µm2 that was fabricated from NiFe(6.0 nm)/Cu(3.5 nm)/NiFe(3.0 nm)/IrMn(10.0 nm) spin valve structure Furthermore, it is also de-veloped to integrated arrays by including 24 sensor patterns In comparison with the other groups, our sensor performance is highlighted with the advantages of increased stability and high signal to noise; as such, the planar Hall effect sensor’s behavior has proved a possibility for detection of the biomolecule It

is also feasible to provide a vehicle for studying other molecule interactions, particular single DNA mole-cule and for the detection of binding of the streptavidin functionalized magnetic beads to sensor bound biotin Due to the simple fabrication scheme, this kind of Planar Hall effect based sensor can be easily in-tegrated into other systems for applications

© 2007 WILEY-VCH Verlag GmbH & Co KGaA, Weinheim

1 Introduction

The detection of specific interactions between two biological molecules with affinity to each other has been playing an increasingly important role when applied to biomolecule hybridization (for example: DNA-cDNA) recognition as they can be applied for genetic disease diagnostic, mutation detection or gene expression quantification, and antibody-antigen interaction (microorganism detection and biological warfare agent detection) processes [1] Such biosensors are available using many different detection methods, even though recently the advantages of magnetoresistive sensors have been cited for a portable device The magnetoresistive sensors directly provide an electric signal that can be evaluated with stan-dard electronics, and then can be produced cheaply with stanstan-dard microelectronic processing techniques

In the last few years, magnetoresistive sensors have been proposed as potential detection components

in biological devices such as high sensitive biosensors and biochips based on a magnetic labeling plat-form [1-4] Most of them were based on the giant magnetoresistive effect (GMR) which was discovered

in mid 1980s Recently the detection of paramagnetic beads for biosensor application has been reported

by use of Hall [5-6] and planar Hall sensor [7]; in which the planar Hall sensor has been of more interest due to a nano-Tesla sensitivity and the reduction of temperature drift by at least four orders of magni-tude, thus considerably improving upon the resolution and signal-to-noise of magnetoresistive sensors [1,

8, 9] Particularly, L Ejsing et al in their works have investigated the capability of planar Hall sensor for magnetic microbead detection They suggested the purposes of using IrMn/NiFe bilayers to ensure a

* Corresponding author: e-mail: cgkim@cnu.ac.kr, Phone: +82 42 821 6632, Fax: +82 42 822 3206

beads [10]

The planar Hall effect (PHE) in magnetic conductor was considered when the resistivity depends on the angle between the direction of the current density j and the magnetization M [11] For magnetization reversal of the single domain when M makes an angle θ with j, the electric field is described as follows [12]:

Ey = j(ρ// −ρ⊥)sinθcosθ (1) where ρ// and ρ┴ are the resisitivities that are parallel and perpendicular to the magnetization, respec-tively The PHE is described as longitudinal component of voltage related to Ey,which is called as PHE voltage in Eq (1) and can be revealed when anisotropy of resistivities exists On the other hand, in this sensor, the measured voltage was described as follows:

Vy = I∆Rsinθ cosθ (2) where ∆R ≡ (ρ// – ρ┴)/t with t the magnetic conductor thickness

In this report, we proposed to utilize the PHE in a spin valve structure as a single sensor and Bead Array Counter (BARC) for the single magnetic bead of Dynabeads® M-280 detection The sensor struc-ture incorporates free and pinned layers in which it was based on the resistance variation when the mag-netization orientation in free magnetic layer rotates under the change of external applied magnetic field The BARC included 24 sensors where each sensor has sizes of 3×3 µm2 The PHE sensor performances were reported to be a high signal and sensitive for Dynabeads® M-280 Streptavidin

2 Experimental procedures

For experiments, the spin valve structure of Ta(5.0 nm)/NiFe(6.0 nm) /Cu(3.5 nm)/NiFe (3.0 nm)/ IrMn(10.0 nm)/Ta(5.0 nm) for single sensor with the patterned sizes of 3.0×3.0 µm2

and sensor array were fabricated by DC magnetron sputtering system under working pressure of 1.0 mTorr on silicon dioxide wafer at room temperature During sputtering process, a uniform magnetic field of 100 Oe was applied parallel to the film plane to induce a magnetic anisotropy of ferromagnetic layers and to align the pinning direction of the antiferromagnetic IrMn layer

Fig 1 SEM images of (a) BARC and (b) a single sensor pattern for single magnetic microbead detection The BARC includes 24 single sensors with the sizes of 3 × 3 µm2

PHE sensors were prepared by lithography method into four electrode bars for each pattern The sen-sors were passivated with sputtered SiO2 layer to protect against the fluid used during experimentation

(b) (a)

phys stat sol (a) 204, No 12 (2007) 4055

Paper

The easy axis of ferromagnetic (FM) layer and the direction of current were aligned along terminal 1 and terminal 4 with sensing current of 1.0 mA, as shown in Fig 1 for the single sensor (a), and for the array (b) The BARC was fabricated by integrating 24 sensor patterns For sensor performance, the droplets of Dynabeads® M-280 solution where the size of magnetic bead is 2.8 µm in diameter were dropped on the surface of the sensor and washed by pipet-lite SL-10 The sensor signals were obtained from terminal 2 and 3 under a magnetic field of varied strength with nanovolt meter

3 Results and discussions

For the performance of single sensor and sensor array, the single sensor transfer curve is shown in Fig 2a where the PHE voltage is plotted versus an external applied magnetic field As per our calculations, a high sensitivity by about 2.5 mΩ/Oe was observed in the linear region of the profile (0–8 Oe) for bead detection This demonstrates that the planar Hall sensor patterned from spin valve structure of Ta (5.0 nm)/ NiFe(6.0 nm)/Cu(3.5)/NiFe(3.0 nm)/IrMn(10.0 nm)/Ta(5.0 nm) is sensitive to magnetic field in the sensor plane The external magnetic field dependence of the sensor signals was shown in Fig 2b, and this performance coupled with our earlier calculations for the same sensor [10] lead us to obtain an opti-mum applied magnetic field for bead detection as 6.0 Oe

Fig 2 (a) The sensor transfer curve using planar Hall effect in spin-valve structure of Ta (5.0 nm)/ NiFe (6.0 nm)/Cu (3.5)/ NiFe (3.0 nm)/IrMn (10.0 nm)/Ta (5.0 nm) with the size of 3 × 3 µm2, as a function of applied magnetic field

at a fixed current of 1 mA; and (b) external magnetic field dependence of sensor signals

For the single magnetic bead detection of M-280 streptavidin, Fig 3a shows a real-time signal of the sensor, protected by a 100 nm thick SiO2 passivation layer, under a current of 1.0 mA In this case, the observations indicate that the signals changed abruptly to zero value when the Dynabeads® M-280 Streptavidin solution was dropped on the face of the sensor This is attributed to leakage of the SiO2

layer, which is not thick enough to protect the electrodes and the sensor patterns from bead solution When the droplets of Dynabeads® M-280 Streptavidin were dropped on sensors, the solution would con-tact with electrodes and a current would pass through between them Because this solution is a conductor with cationic and anionic carriers, there the conductivity is larger than that of the sensor patterns and electrodes Hence, in this case, the current would be transferred through the droplets of Dynabeads®

M-280 Streptavidin so that the signals are reduced to zero Furthermore, under this ionic solution and an electric current through, electrochemical corrosions took place and then the electrodes are dissolved However, when the SiO2 layer was made 150 nm thick, the resulting real time profile was effective in bead detection, as shown in Fig 3b This demonstrates that the 150 nm SiO2 layer is thick enough to prevent the sensor from electrochemical corrosion The signals here were observed corresponding to the

number of beads

Table 1 Comparison of signals, noises and signal-to-noise

Type of

mag-netic beads

Signal changes (µV)

Noise (µV)

Noise/Signal (%) Bead Solution 1.43 0.42 29.57

Figure 4a shows real time profile with more stable signals for two cycles of which each one is for dropping and washing of magnetic bead solution The signals in Fig 4a were observed with continuous time by using solution of magnetic beads For the times from t = 12 s to 24 s, and 34 s to 40 s, the signal corresponding the magnetic bead solutions which are dropped on the face of the sensor; where as for the other times, the magnetic beads were simply washed away from the face of sensor However, the real time profile measured for magnetic beads after drying, as shown in Fig 4b, are significant compared to the magnetic solution with a signal magnitude of around 7.6 µV It is several times higher and more stable with a smaller noise than those measured with magnetic bead solution The comparison of signals

is shown in Table 1

Fig 3 Real-time profiles (a) for leakage of SiO2 showing an abrupt signal change with droplets of Dynabeads®

M-280 Streptavidin solution, and (b) for SiO2 thickness of 150 nm, where the sensing data is performed with nonmag-netic beads, unstable position of magnonmag-netic beads and dense magnonmag-netic beads on the top of sensing area

The profile in Fig 4a appeared some noise compared to that in Fig 4b where each cycle is around 40 seconds The more stable characteristic performance in Fig 4b was attributed to stability of magnetic beads and environment surrounding the sensor because of absence of bead motion in solution The signal change for single bead detection has been observed by around 1.2µV in magnitude It is larger by several times than that compared with other works [6, 9], in which the recent F Ejsing et al.’s publication showed a signal change for single bead detection by around 0.3 µV only

a)

b)

phys stat sol (a) 204, No 12 (2007) 4057

Paper

Fig 4 Real-time voltage changes for (a) magnetic bead solution and (b) dry Dynabeads®bead of M-280 Strepta-vidin The signal change of the dry magnetic beads reveals a higher signal by several times than that of the bead solution

4 Conclusions

The detection of Dynabeads® M-280 streptavidin applied for biosensor and biochip has been demon-strated by using planar Hall effect in spin valve structure The spin valve structure was considered as a novel method to induce a high sensitivity and uniform magnetization rotation compared to exchange biased structure for planar Hall signals, due to small interlayer coupling and very thin active sensing layer Single bead detection was obtained with 1.2 µV in magnitude, where the sensitivity of the sensor performed a value of 2.5 mΩ/Oe These, along with the advantages of more stable and high signal to noise of PHR sensor’s behavior, therefore, can be used for biomolcule recognition It is also feasible to provide a vehicle for detection and study of other molecule interactions, in particular single DNA mole-cule interaction and detection of the binding of streptavidin functionalized magnetic beads to sensor bound biotin Due to the simple fabrication scheme, this planar Hall resistance based sensor is easily to

be integrated into other bio-systems for applications

Acknowledgements This work was supported by KOSEF through ReCAMM, and MIC under project number

A1100-0601-0033; partly was supported by the State Program for Fundamental Research in Natural Sciences under Project 410.406

References

[1] P P Freitas, H A Ferreira, D L Graham et al., in: Magnetoresistive Biochips, edited by M Johnson (El-sevier, Amsterdam, 2004)

[2] M M Miller, P E Edelstein, C R Tamanaha, L.Zhong, S Bounak, L J Whitman, and R J Colton, J Magn Magn Mater 225, 138 (2001)

[3] D L Graham, H Ferreira, J Bernado, P P Freitas, and J M S Cabral, J Appl Phys 91, 7786 (2002) [4] D L Graham, H A Ferreira, P P Freitas, and J M S Cabral, Biosens Bioelectron 18, 483 (2003)

[5] P Besse, G Boero, M Demierre, V Pott, and R Popovic, Appl Phys Lett 80, 4199 (2002)

[6] L Ejsing, M.F.Hansen, A K Menon, H A Ferreira, D L Graham, and P P Fretas, Appl Phys Lett 84,

4729 (2004)

[7] A Schuhl, F Nguyen Van Dau, and J R Childress, Appl Phys Lett 66, 2751 (1995)

[8] F Nguyen Van Dau, A Schuhl, J R Childress, and M Sussiau, Sens Actuators A 53, 256 (1996)

[9] L Ejsing, M F Hansen, A K Menon, H A Ferreira, D L Graham, and P P Freitas, J Magn Magn Mater

293, 677 (2005)

[10] N.T Thanh, K.W Kim, C.O Kim, K.H Shin, and C.G Kim, J Magn Magn Mater 316, e238–e241 (2007)