A wireless biosensor using microfabricated phage-interfaced magnetoelastic particles

Available online at www.sciencedirect.com

Sensors and Actuators A 144 (2008) 38–47

A wireless biosensor using microfabricated phage-interfaced magnetoelastic particles Michael L. Johnson a,∗ , Jiehui Wan a , Shichu Huang a , Zhongyang Cheng a , Valery A. Petrenko b , Dong-Joo Kim a , I.-Hsuan Chen c , James M. Barbaree c , Jong Wook Hong a , Bryan A. Chin a a

Materials Research & Education Center, Auburn University, 275 Wilmore Labs, Auburn, AL 36849, USA b Department of Pathobiology, Auburn University, Auburn, AL 36849, USA c Department of Biological Sciences, Auburn University, Auburn, AL 36849, USA Received 7 May 2007; received in revised form 17 October 2007; accepted 27 December 2007 Available online 10 January 2008

Abstract A micro-scale, free standing, wireless biosensor has been developed using magnetoelastic particles composed of an amorphous iron–boron binary alloy. Upon the application of an external magnetic field, these particles exhibit a characteristic resonance frequency, determined by their size and mass, due to the phenomena of magnetoelasticity. The particles are produced using the microelectronic fabrication techniques of photolithography and physical vapor deposition (sputtering). The biosensor is formed by coating the magnetoelastic particle with a thin layer of gold and immobilizing a biomolecular recognition element (bacteriophage) on the surfaces. Bacteriophage genetically engineered to bind Bacillus anthracis spores was used in this set of experiments as the detection probe. Once these targeted spores come into contact with the biosensor, the phage will bind selectively with only that pathogen, thereby increasing the particle’s mass and causing a shift in the resonance frequency. Due to the magnetic nature of the sensing platform, this resonance frequency shift may be detected remotely by a wireless scanning device, presenting a distinct advantage over other techniques. A good correlation between the actual number of spores bound to the sensors and the calculated attached mass, based upon resonance frequency shifts, was obtained from the experiments. © 2008 Elsevier B.V. All rights reserved. Keywords: Magnetoelastic; Magnetostrictive; Biosensor; Sputtering; Bacteriophage; Bacillus anthracis; Spores

1. Introduction As part of the ongoing effort to secure the safety of our food supply, as well as to guard against the possibility of bioterrorism, much research has been focused on developing detection techniques that are not only accurate, fast, and inexpensive, but can also offer detection remotely of sealed containers and packages. Towards these goals, immunoassay-based, label-free biosensing systems have proven to be among the most promising methods for microorganism detection [1–3]. Immunoassay-based biosensors, such as acoustic wave biosensors, are composed of a transducer, which converts biological signals of target–receptor interaction into electrical or other measurable signals, and a bioprobe that possesses affinity for a target pathogen. Various types

∗

Corresponding author. Tel.: +1 334 728 0282; fax: +1 334 844 3400. E-mail address: [email protected] (M.L. Johnson).

0924-4247/$ – see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.sna.2007.12.028

of acoustic wave devices have been fabricated and used for biological detection in combination with immobilization techniques of biological recognition probes [4,5]. These acoustic wave sensor platforms provide the capability of real-time detection and several other advantages, such as high sensitivity, simplicity, and low cost [3,6,7]. Pathogen detection at very low concentrations, down to the ultimate goal of single cell or spore detection, has been a key challenge in the biosensing area. Substantial progress has been made in the application of silicon-based and piezoelectric-based cantilevers and microcantilevers due to their high sensitivities [8]. The ability to detect about 200 bacterial cells [9] and 60 fungal spores [10] using functionalized cantilevers has been reported. Additional studies have shown that as few as 50 spores can be detected in water by cantilevers using thermal noise as the excitation source [11]. Compared with traditional microcantilevers, magnetoelastic-based microcantilevers have been demonstrated to have a better overall sensitivity and a quality

M.L. Johnson et al. / Sensors and Actuators A 144 (2008) 38–47

factor (Q-value) due to the material’s properties and wireless actuation mode [12,13]. The potential of a free-standing, magnetoelastic strip as the basis of a remote sensor has been demonstrated for the detection of changes in various physical properties [14], as well as the presence of specific biological species [15–17]. This method offers real-time detection of targeted pathogens in various liquid media and, since it is wireless, no direct contact between the sensor and the measuring device is required. A key issue with this technique, however, has been sensitivity. As is the case with microcantilevers, sensitivity will increase as the size of the sensor is decreased [13]. In order to directly detect the presence of a small number of spores or cells (e.g. on the order of 102 or less), the sensor must theoretically have a maximum dimension in the range of 102 microns. Traditionally, such sensor platforms have been fabricated by the mechanical polishing and dicing of commercially available amorphous alloy ribbons, which are then coated with a gold layer for biocompatibility [18]. This method has led to problems in dimensional repeatability (e.g. edge defects) and surface quality (e.g. poor gold layer adhesion). Additionally, it has proven to be extremely difficult to make samples of the required size with a mechanical dicing technique. The goal of this research is to solve the above problems by employing techniques found in microelectronic fabrication (specifically photolithography and physical vapor deposition) to produce magnetoelastic sensor platforms of desired size and shape with excellent dimensional control and surface quality. Functionality of the sensors in detection of Bacillus anthracis spores was explored using a landscape phage recognition interface. 2. Materials and methods 2.1. Principle of detection The phenomenon of magnetoelasticity occurs when, upon the application of an external magnetic field, a material experiences a shape change due to the superposition of its internal magnetic moments due to spin–orbital coupling [19]. If this field is alternating (“AC”) and the piece of material is planar (i.e. its thickness is much less than its length and width), then the fundamental resonance frequency of the longitudinal vibration caused by magnetostriction is described by Liang et al. [20]:  1 E f0 = (1) ρ(1 − ν) 2L where E, ρ, ν, and L represent the elastic modulus, density, Poisson’s ratio, and the length, respectively. Additionally, this resonance frequency will be affected by the sensor’s mass load as described by Grimes et al. [14]: f = −

f0 m 2 M

(2)

where f is the change in resonance frequency, m is the change in the sensor’s mass due to the attached load, and M is the original mass of the sensor. Note that the negative sign indicates that the

39

frequency will shift to a lower value upon an increase in the sensor’s mass load. Also, this relation assumes a load that is evenly distributed over the surface of the sensor. The principle by which the property of magnetoelasticity may be utilized as a biosensing platform is shown schematically in Fig. 1. First, the magnetoelastic particle (MEP) is coated with a gold layer to enable biocompatibility with protein molecules such that they may be attached onto the surface. Next, a biomolecular recognition element (either bacteriophage or antibody) is immobilized onto the gold surface, enabling the capture of a single targeted pathogenic species. At this point, the fundamental resonance frequency is measured in order to completely characterize the biosensor prior to exposure to target containing solutions. This is accomplished by applying an AC magnetic field to set up resonance, as well as a DC biasing magnetic field for amplification of the signal. Then, when the sensor is exposed to the targeted bacteria or spores, they bind to the sensor, thereby increasing its mass and lowering its resonance frequency. Also, based on the magnitude of the frequency shift, the approximate number of attached spores or cells may be calculated using Eq. (2) (assuming an approximate mass of each spore or cell). 2.2. Magnetoelastic platform material An amorphous binary alloy of iron–boron was chosen as the basis for the sensor platform material for several reasons. Primarily, it is a much-studied magnetoelastic system that easily forms the amorphous phase with boron content equal to or greater than the eutectic composition of 16–17 at.% [21,22]. Also, as is the case with other metallic glasses, its soft magnetic properties, moderately high level of magnetostriction (λs ), and low magneto-crystalline anisotropy combine to result in a very high magneto–mechanical coupling efficiency (k33 ), making it better suited to high frequency sensor applications, as opposed to the giant magnetostrictive compounds, which are better suited for application in mechanical stress sensors and high-power acoustic devices [23,24]. Finally, and practically, since it is only a two-component system, fabrication is relatively simple and it is much easier to achieve the desired composition using physical vapor deposition (or “sputtering”) than would be the case with a multi-component system, such as those commonly employed in commercially available amorphous metal ribbons. 2.3. Sensor platform fabrication The sensor platform fabrication process is shown schematically in Fig. 2. Fabrication of the sensor platform begins by coating a 100 mm plain silicon test wafer with a layer of chromium, and then gold, each at a thickness of 30–40 nm. This is accomplished using a Denton Vacuum Discovery-18TM magnetron sputtering system, which employs three cathodes (each holding a 3-in. diameter disk-shaped target) aimed off-axis at a circular, rotating substrate platform, along with DC and RF power supplies. The gold layer is needed to adhere the next deposited film (also gold) to the wafer, while the chromium merely serves to act as a bond between the silicon and the

40

M.L. Johnson et al. / Sensors and Actuators A 144 (2008) 38–47

Fig. 1. Diagram of the operating principle of a magnetoelastic biosensor. HAC and HDC are the alternating and biasing magnetic fields, respectively. The resonance frequency after attachment of the pathogen is fload , such that Δf = f0 − fload .

Fig. 2. Diagram of the magnetoelastic particle fabrication process.

gold. Next, a layer of photoresist (STR-1045, from Rohm & Haas Electronic materials, LLC) is applied to the gold surface of the wafer by spin coating such that the resultant thickness is at least twice that of the desired magnetoelastic film to be

deposited later. This photoresist is then UV exposed using a positive mask comprised of evenly spaced rectangles, which are the desired length and width of the magnetoelastic particles. The wafer is then developed in a 2:1 solution of de-ionized

M.L. Johnson et al. / Sensors and Actuators A 144 (2008) 38–47

41

Fig. 3. SEM micrographs of a gold coated Fe79 B21 film (a) before and (b) after annealing at 200 ◦ C for 3 h.

water and AZ-400 K developer (from AZ Electronic Materials USA Corp.) for 1–3 min (depending on thickness), rinsed, dried using pressurized dry nitrogen, and then inspected for pattern integrity (using an optical microscope) and thickness (using an Alpha-Step profilometer, from Tencor Instruments). The magnetoelastic film is then deposited onto the patterned wafer using the same sputtering system as before. First, the wafer is loaded into the deposition chamber, along with a gold, iron, and boron target for each of the three cathodes, and then the chamber is pumped down to 7 × 10−7 Torr in order to minimize residual oxygen in the film. Next, a gold layer is deposited onto the patterned wafer to a thickness of about 30–40 nm. The purpose of this layer is threefold: (a) it serves to bond the magnetoelastic film to the exposed gold surface of the wafer where there is no photoresist, (b) it will form a protective layer on the bottom of the particles which will prevent oxidation of the particles in aqueous environments, and (c) it will provide a bioactive surface on the particles to which biological species may be easily immobilized. The magnetoelastic layer is formed by co-depositing iron (DC) and boron (RF) simultaneously using a dual-cathode method. This method differs somewhat from the usual procedure

for co-sputtering iron and boron, which typically involves using a specially made composite target that is mounted on either a single DC cathode [25,26] or a single RF cathode [27–29]. The advantage here is that the power of each cathode can be tuned separately such that the film has the desired composition at a reasonable deposition rate. For the other method, a new target must be manufactured each time until the desired composition is achieved. Thickness of this film depends on process conditions, and is generally limited by the thickness of the photoresist layer, but highly magnetostrictive films of up to about 7 ␮m have been obtained using this dual-cathode method. Finally, another gold layer, using the same processing conditions as before, is applied on top of the iron–boron film, such that the magnetostrictive particles will be completely enclosed in gold. Once deposition is complete, lift-off of the particles is accomplished by rinsing the wafer in acetone, which dissolves the photoresist. This liberates the particles, which were merely the part of the film that grew on top of the photoresist “islands”, while the surrounding “waste” film remains adhered to the wafer, which can then be discarded. The loose particles are collected in a beaker (in the solution of acetone), and are rinsed and decanted

Fig. 4. Diagram of the characterization method for the determination of resonance frequency. All symbols are as previously defined, with the addition of iAC to represent the alternating current in the coil.

42

M.L. Johnson et al. / Sensors and Actuators A 144 (2008) 38–47

Fig. 5. Examples of magnetostrictive particles that have been produced using the techniques described herein: (a) 50 ␮m × 6 ␮m × 3 ␮m, (b) 500 ␮m × 100 ␮m × 5 ␮m, (c) 250 ␮m × 50 ␮m × 2 ␮m, and (d) the surface of a wafer after particle lift-off; the rectangular “holes” are where the photoresist was dissolved to release the particles.

several times with fresh acetone to ensure complete removal of all photoresist residues. Then, they are rinsed several times with methanol, and finally stored in fresh methanol in a sealed glass vial. Prior to use, the particles were annealed to correct the defects that exist in the as-sputtered microstructure. This takes place in a vacuum oven set at 200 ◦ C for 3 h. This temperature was chosen because it is high enough such that the as-sputtered defects may be healed within a reasonable time, yet it is still far below the Curie temperature of 374 ◦ C [30], as well as the recrystallization temperature, which may vary between 390 and 460 ◦ C, depending on exact boron content [23,31]. Fig. 3 shows the surface microstructure of the as-sputtered and annealed samples of an Fe79 B21 film; notice the healing of the microcracks. 2.4. Biological testing procedure The fabrication of the sensor is completed by immobilizing bacteriophage on the gold surface, thereby enabling the specific binding of the targeted pathogenic species. In this research, the biorecognition interface was formed by filamentous landscape phage specifically binding B. anthracis (Sterne) spores [32]. A detailed discussion of how bacteriophages may be advantageously used in place of traditional antibodies for the detection of specific microbial species can be found in Petrenko et al. [33–35]. Phage is immobilized onto the gold surface of the magnetostrictive particles by immersing them individually in PCR tubes containing the phage, diluted in 1× TBS solution to a concentration of 5 × 1011 vir/ml (23 ␮g/ml) [5,36]. Uni-

form coverage of the phage on the particle surface is ensured by placing the particles and the phage solution in PCR tubes, and then rotating in a laboratory mixer. Afterwards, the phage solution is decanted off the particles, which are then subsequently rinsed in distilled water, allowed to dry in air, and then transferred to clean PCR tubes. The biosensors are now ready for testing in solutions containing known concentrations of anthrax spores. It should be noted that all experiments conducted to investigate the performance of the magnetoelastic biosensors used B. anthracis Sterne spores, which is the nonpathogenic vaccine strain. The Sterne strain is identical to the pathogenic strain of B. anthracis except for the number of internal plasmids contained within the spore. Binding characteristics of the Sterne strain spores are anticipated to be identical to those of pathogenic spores because the outer membranes of both strains are identical. Upon the completion of sensor fabrication, its ability to detect the targeted species is tested by loading B. anthracis spore solution (diluted to concentrations ranging from 1 × 107 to 1 × 108 spores/ml) into each PCR tube, which is then rotated using a laboratory mixer as before. Afterwards, the spore solution is decanted and each biosensor is again rinsed with distilled water. Upon drying in air, each particle is transferred to a Petri dish, into which one drop (∼50 ␮l) of osmium tetroxide (OsO4 , in 4% aqueous solution) is placed and allowed to incubate, covered, for 1 h. Exposure to OsO4 vapor kills and fixes the spores, enabling re-measurement of the sensors’ resonance frequency, as well as the observation of actual spore binding by microscopy.

M.L. Johnson et al. / Sensors and Actuators A 144 (2008) 38–47

Fig. 6. XPS analysis used to determine a composition of Fe79 B21 in a sample of the waste film from the deposition run that produced the particle shown in Fig. 5(b).

43

Fig. 7. Response of a typical 500 ␮m × 100 ␮m × 5 ␮m MEP in air, before and after exposure to spores.

3. Results and discussion

2.5. Characterization of sensor resonance

3.1. Initial characterization of sensor platforms

In order to determine the resonance frequency of a sensor, it is first placed into a plastic pipette tip (approx. 0.3/0.2 mm OD/ID), which is then inserted into a glass tube (approx. 0.9/0.4 mm OD/ID), around which a copper coil is wrapped with enough length such that the particle can be completely contained within the coil. The two leads of the coil are then connected to a network analyzer (HP-8751A, by Hewlett–Packard, Inc.) through an S-parameter test set, which is operated in S11 reflection mode. In this configuration, an AC signal is applied across the coil, and then the reflected signal is compared to a reference signal across a selected range of frequencies. In this manner, frequency-dependant loss, phase angle, or impedance may be determined for the circuit. At the resonant frequency of the magnetoelastic sensor, the magnetic field generated by the coil will be transformed into mechanical energy (vibration), causing a sharp change in the reflected power of the circuit at that particular frequency, which is visible as a “valley” in the network analyzer output. This frequency is approximated by Eq. (1), which can be expressed as a constant over length since elastic modulus, density, and Poisson’s ratio are all material properties. Based on experimental data for these sensors, Eq. (1) simplifies to:

As can be seen from Fig. 5, practically any size particle can be made using this technique, limited only by the minimum feature size of the mask and photoresist used in the microelectronic fabrication procedure. Fig. 6 shows the results of an XPS analysis of the waste film from the wafer that produced the sample shown in Fig. 5(b), which yielded a composition of 79 at.% iron and 21 at.% boron. The result of the resonance frequency measurement for a typical 500 ␮m × 100 ␮m × 5 ␮m sensor platform (like the one shown in Fig. 5(b)), prior to biological testing, is shown in Fig. 7. In air, at an optimum DC bias field of 50 G, this sample had a quality (Q) factor of 656, which was calculated by using the following approximation:

f0 =

2 × 103 m/s L

(3)

where f0 is the resonance frequency in Hz and L is the length (longest dimension) of the particle in meters. For signal amplification, a bar magnet external to the coil is used to provide a DC biasing magnetic field. For each sensor size and composition, there is an optimum biasing field required for maximum amplitude. The field is increased slowly to amplify the signal; however, if it is increased further, the amplitude will decrease again due to dampening of the response. In order to measure this field for experimental repeatability, a DC magnetometer probe was placed as close as possible to the coil as shown in Fig. 4, which schematically depicts the measurement method.

Q=

f0 s

(4)

where s is defined as the spread of the signal trace at√a distance from the baseline equal to the amplitude divided by 2. 3.2. Biological testing results Phage was immobilized onto the gold surface of the above sensor platform, which was afterwards exposed to B. anthracis Sterne spores as per the procedure detailed in Section 2.4. Fig. 7 shows the shift in resonance frequency for this specific sensor due to the attachment of spores. As expected, there is a decrease in both resonance frequency and amplitude due to the increase in non-magnetoelastic mass. The resonance frequency has been shifted by 200 kHz, which, using Eq. (2) (with M = 1.72 ␮g and f0 = 4.135 MHz), would indicate a mass increase of 166 ng. An exact count of the number of spores attached to the surface of this sensor was performed using scanning electron microscopy (see Fig. 8), and was found to be 6795 spores. This procedure was repeated for a total of 9 sensors, with the data shown in Table 1. The average mass per spore for all sensors is calcu-

44

M.L. Johnson et al. / Sensors and Actuators A 144 (2008) 38–47

Fig. 8. SEM micrographs of a gold-coated, 500 ␮m × 100 ␮m × 5 ␮m, Fe79 B21 particle after loading with phage and spores. Attachment of the spores to the gold surface by the filamentous phage can be seen in (c).

Table 1 Resonance frequency shifts and spore count data Sensor

Initial f0 (MHz)

Final f0 (MHz)

f (kHz)

m (ng)

Spore count

Mass/spore (pg)

1 2 3 4 5 6 7 8 9

4.093 4.151 4.056 4.135 3.998 4.108 4.158 4.049 4.048

3.853 3.932 3.948 3.936 3.994 4.056 4.150 4.042 4.046

240 219 108 199 4 52 8 7 2

201.9 181.7 91.7 165.7 3.4 43.6 6.6 6.0 1.7

8355 7920 4605 6795 1270 2805 1490 875 440

24.2 22.9 19.9 24.4 2.7 15.5 4.4 6.9 3.9

Avg.

4.088

13.9

S.D.

0.054

9.4

Table 2 Sensor detection limits based on platform size and f resolution L (␮m)

w (␮m)

t (␮m)

Volume (␮m3 )

Mass (ng)

f0 (MHz)

f (Hz)

m (pg)

Number of spores

1000 1000 500 500 500 500 250 250 100 100

200 200 100 100 100 100 50 50 20 20

15 15 15 15 5 5 3 3 2 2

3,000,000 3,000,000 750,000 750,000 250,000 250,000 37,500 37,500 4,000 4,000

23,700 23,700 5,925 5,925 1,975 1,975 296 296 32 32

2.000 2.000 4.000 4.000 4.000 4.000 8.000 8.000 20.000 20.000

10,000 1000 10,000 1,000 10,000 1,000 10,000 1,000 10,000 1,000

237,000 23,700 29,625 2,963 9,875 988 741 74 32 3

118,500 11,850 14,813 1,481 4,938 494 370 37 16 1

M.L. Johnson et al. / Sensors and Actuators A 144 (2008) 38–47

Fig. 9. Correlation between the calculated attached mass (based on observed Δf and Eq. (2)) and spore count for the group of sensors in Table 1.

lated to be about 14 pg, which is much greater than the value of ∼2 pg/spore as is usually reported [11]. This is due primarily to the osmium tetroxide treatment, since OsO4 replaces the water in the exposed spores, effectively increasing their mass. Fig. 9 shows the correlation between the calculated attached mass (based on the observed resonance frequency shift) and the actual number of attached spores. Any error is largely due to the varying amounts of background material (e.g. salts from the buffer solutions that were not completely rinsed away) remaining on the sensors in addition to the spores after drying. In general, with larger amounts of bound spores, there will also be a larger amount of background material. This leads to a greater exaggeration of the mass/spore number for the sensors with higher amounts of spores attached to their surfaces. 3.3. Sensitivity improvement The sensitivity of these sensors may be improved primarily in four ways: (1) The particle size may be reduced, (2) the Q-factor of the sensor platforms may be increased, (3) the measurement technique may be improved, and (4) the measurements may be performed in a liquid system. A decrease in the sensor size will reduce the initial sensor mass and increase the resonance frequency, which will enable the detection of smaller amounts of bound mass (i.e. fewer spores). An increase in the Q-factor of the particles (irrespective of any decrease in size) will result in better resolution, enabling the detection of smaller resonance frequency shifts and, hence, the detection of smaller amounts of bound mass. This inherent sensitivity of the particles may be improved mainly through refinements to the fabrication techniques, including photoresist patterning, deposition conditions, and the annealing treatment. Improvements to the measurement technique involve coil configuration and optimization of the network analyzer test parameters. More sensitive coils, especially, will improve signal amplitude and stability, which will, in turn, lead to higher effective Q-values and greater stand-off distances between the coil and the sensor. Lastly, if the measurements

45

are performed in a flowing liquid system, then resonance frequency shifts may be detected in real-time as spores attach to the sensor. This will eliminate the additional mass load on the sensors due to dried background material and the osmium tetroxide treatment. Table 2 shows how a reduction in particle size, as well as an improvement in the resolution of the frequency shift due to mass loading, may be combined to improve the spore detection limit. As can be seen, it is not only important to reduce the size of the sensor platform, but also to improve the sensitivity by increasing the Q-factor. The sizes listed with thicknesses of 15 ␮m are representative of the particles that have been fabricated in the past from commercially available amorphous metal ribbons by polishing and dicing. The 500 ␮m × 100 ␮m × 15 ␮m size represents the smallest particle that has been able to be fabricated using these mechanical methods. Smaller sizes have all been produced using microelectronic fabrication techniques. From the example given in Table 2, it can be seen that the ultimate goal of single spore detection may be approached if a frequency shift of 1 kHz can be resolved using a sensor with a size of 100 ␮m × 20 ␮m × 2 ␮m (assuming an average spore mass of about 2 pg). Note that L, w, and t represent the length (longest dimension), width, and thickness (shortest dimension) of the particles, respectively. 4. Conclusions The potential of producing magnetoelastic platforms for use as biosensors via standard microelectronic fabrication techniques has been demonstrated. Gold-coated, magnetoelastic particles with a composition of Fe79 B21 and with dimensions in the range of 2–500 ␮m were fabricated by sputtering onto a chromium and gold-coated silicon wafer that had been previously patterned with photoresist. Unlike the method of mechanically dicing and polishing commercially available amorphous metal ribbons, this technique holds the promise of making available sensor platforms of sufficiently small size and high sensitivity such that the detection of ultra-low spore/cell counts may be realized. This is crucial in the effort to achieve the ultimate goal of detecting a single spore or cell. In addition to decreasing the sensor size and improving the sensitivity, future work will focus on the detection of spores in a liquid environment. Real-time detection while the sensor is immersed in liquid will also facilitate more accurate correlation of the actual number of bound spores with the observed frequency shift. This will eliminate the effect of the additional mass added by the dried background material and the osmium tetroxide treatment required for handling the sensor post-exposure to biological agents. Acknowledgments This research has been carried out as part of the ongoing efforts of the Auburn University Detection & Food Safety (AUDFS) Center to improve the safety of the U.S. food supply chain, and was funded by CSREES under grant USDA20053439415674A.

46

M.L. Johnson et al. / Sensors and Actuators A 144 (2008) 38–47

References [1] R.D. Vaughan, C.K. O’Sullivan, G.G. Guilbault, Development of a quartz crystal microbalance (QCM) immunosensor for the detection of Listeria monocytogenes, Enzyme Microb. Technol. 29 (2001) 635–638. [2] S.T. Pathirana, J.M. Barbaree, B.A. Chin, M.G. Hartell, W.C. Neely, V.J. Vodyanoy, Rapid and sensitive biosensor for Salmonella, Biosens. Bioelectron. 15 (2000) 135–141. [3] P. Leonard, S. Hearty, J. Brennan, L. Dunne, J. Quinn, T. Chakraborty, R. O’Kennedy, Advances in biosensors for detection of pathogens in food and water, Enzyme Microb. Technol. 32 (2003) 3–13. [4] C.A. Bailey, B. Fiebor, W. Yen, V.J. Vodyanoy, R.W. Cernosek, B.A. Chin, Thickness shear mode (TSM) resonators used for biosensing, Proc. SPIE Int. Soc. Opt. Eng. 4575 (2001) 138–149. [5] E.V. Olsen, I.B. Sorokulova, V.A. Petrenko, I.-H. Chen, J.M. Barbaree, V.J. Vodyanoy, Affinity-selected filamentous bacteriophage as a probe for acoustic wave biodetectors of Salmonella typhimurium, Biosens. Bioelectron. 21 (2006) 1434–1442. [6] R. Raiteri, M. Grattarola, R. Berger, Micromechanics senses biomolecules, Mater. Today 5 (2002) 22–29. [7] O. Tamarin, C. Dejous, D. Rebiere, J. Pistre, S. Comeau, D. Moynet, J. Bezian, Study of acoustic Love wave devices for real time bacteriophage detection, Sens. Actuators, B Chem. 91 (2003) 275–284. [8] B. Ilic, D. Czaplewski, M. Zalautdinov, H.G. Craighead, P. Neuzil, C. Campagnolo, C. Batt, Single cell detection with micromechanical oscillators, J. Vac. Sci. Technol. B 19 (2001) 2825–2828. [9] K. Gfeller, N. Nugaeva, M. Hegner, Micromechanical oscillators as rapid biosensor for the detection of active growth of Escherichia coli, Biosens. Bioelectron. 21 (2005) 528–533. [10] N. Nugaeva, K. Gfeller, N. Backmann, H. Lang, M. Duggelin, M. Hegner, Micromechanical cantilever array sensors for selective fungal immobilization and fast growth detection, Biosens. Bioelectron. 21 (2005) 849–856. [11] A.P. Davila, J. Jang, A.K. Gupta, T. Walter, A. Aronson, R. Bashir, Microresonator mass sensors for detection of Bacillus anthracis Sterne spores in air and water, Biosens. Bioelectron. 22 (2007) 3028–3035. [12] S. Li, L. Orona, Z. Li, Z.-Y Cheng, Biosensor based on magnetostrictive microcantilevers, Appl. Phys. Lett. 88 (2006) 073507. [13] Z.Y. Cheng, Applications of smart materials in the development of high performance biosensors, Mater. Res. Soc. Symp. Proc. 888 (2006) 185–195. [14] C.A. Grimes, K.G. Ong, K. Loiselle, P.G. Stoyanov, D. Kouzoudis, Y. Liu, C. Tong, F. Tefiku, Magnetoelastic sensors for remote query environmental monitoring, Smart Mater. Struct. 8 (1999) 639–646. [15] K.G. Ong, K. Zeng, C. Ruan, C.A. Grimes, A wireless remote-query microsensor for simultaneous quantification of multiple bioagents, Proc. IEEE Sens. 2 (2004) 999–1002. [16] R.S. Lakshmanan, J. Hu, R. Guntupalli, J. Wan, S. Huang, H. Yang, V.A. Petrenko, J.M. Barbaree, B.A. Chin, Detection of Salmonella typhimurium using phage-based magnetostrictive sensor, SPIE Chem. Biol. Sens. 6218 (2006) 246–254. [17] R. Guntupalli, J. Hu, R.S. Lakshmanan, T.S. Huang, J.M. Barbaree, B.A. Chin, A magnetoelastic resonance biosensor immobilized with polyclonal antibody for the detection of Salmonella typhimurium, Biosens. Bioelectron. 22 (2007) 1474–1479. [18] J. Wan, H. Shu, S. Huang, B. Fiebor, I.-H. Chen, V.A. Petrenko, B.A. Chin, Phage-based magnetoelastic wireless biosensors for detecting Bacillus anthracis spores, IEEE Sens. J. 7 (2007) 470–477. [19] R. Wu, V.I. Gavrilenko, A.J. Freeman, in: M.R.J. Gibbs (Ed.), Modern Trends in Magnetostriction Study and Application, Kluwer Academic Publishers, Boston, 2000, pp. 1–12, Ch. 1. [20] C. Liang, S. Morshed, B.C. Prorok, Correction for longitudinal mode vibration in thin slender beams, Appl. Phys. Lett. 90 (2007)221912. [21] W.G. Moffatt, Handbook of Binary Phase Diagrams, General Electric Co., Schenectady, 1976. [22] T. Van Rompaey, K.C. Hari Kumar, P. Wollants, Thermodynamic optimization of the B–Fe system, J. Alloy. Compd. 334 (2002) 173–181. [23] G.E. Fish, Soft magnetic materials, Proc. IEEE 78 (1990) 947–972. [24] G. Engdahl (Ed.), Handbook of Giant Magnetostrictive Materials, Academic Press, San Diego, 2000.

[25] H.-S. Shin, J.-W. Hong, T. Jang, E.-P. Yoon, I. Kim, G.-Y. Yeum, J.-W. Park, Effect of microstructures on the coercivity of Fe1−x Bx (0 ≤ x ≤ 0.2) films prepared by dc magnetron sputtering, Vacuum 67 (2002) 185–189. [26] J.F. Pierson, C. Rousselot, Amorphous Fe–B–N films deposited by reactive sputtering of a FeB target, Surf. Coat. Technol. 180–181 (2004) 44–48. [27] N.A. Blum, K. Moorjani, T.O. Poehler, F.G. Satkiewicz, Mossbauer investigation of sputtered ferromagnetic amorphous Fex B100−x films, J. Appl. Phys. 53 (1982) 2074–2076. [28] T. Fujii, T. Kumura, T. Miyama, M. Inoue, Magnetoelastic wave excitation properties of amorphous iron–boron films prepared by RF-high rate sputtering, Jpn. J. Appl. Phys. 22 (1983) 79–87. [29] I. Carabias, A. Martinez, M.A. Garcia, E. Pina, J.M. Gonzalez, A. Hernando, P. Crespo, Magnetostrictive thin films prepared by RF sputtering, J. Magn. Magn. Mater. 290–291 (2005) 823–825. [30] E. du Tremolet de Lacheisserie, Magnetostriction: Theory and Applications of Magnetoelasticity, CRC Press, Boca Raton, 1993. [31] N. Cowlam, G.E. Carr, Magnetic and structural properties of Fe–B binary metallic glasses I: variation of magnetic moment with composition, J. Phys. F. Met. Phys. 15 (1985) 1109–1116. [32] J.R. Brigati, D.D. Williams, I.B. Sorokulova, V. Nanduri, I.-H. Chen, C.L. Turnbough, V.A. Petrenko, Diagnostic probes for Bacillus anthracis spores selected from a landscape phage library, Clin. Chem. 50 (2004) 1899–1906. [33] V.A. Petrenko, G.P. Smith, X. Gong, T. Quinn, A library of organic landscapes on filamentous phage, Methods Enzymol. 9 (1996) 797–801. [34] V.A. Petrenko, V.J. Vodyanoy, Phage display for detection of biological threat agents, J. Microbiol. Methods 53 (2003) 253–262. [35] V.A. Petrenko, J.R. Brigati, in: J.M. Van Emon (Ed.), Immunoassay and Other Bioanalytical Techniques, CRC Press, Boca Raton, 2007, pp. 187–216. [36] V. Nanduri, I.B. Sorokulova, A.M. Samoylov, A.L. Simonian, V.A. Petrenko, V.J. Vodyanoy, Phage as a molecular recognition element in biosensors immobilized by physical adsorption, Biosens. Bioelectron. 22 (2007) 986–992.

Biographies Michael L. Johnson received a bachelors degree in ceramic engineering in 1990 from The Georgia Institute of Technology, Atlanta, GA, USA. He then worked for Corning, Inc. in the fiber optic waveguide division at facilities located in Wilmington, NC and Corning, NY, USA. While there, he served in various roles, including process engineering, specials manufacturing, process development, international technology transfer, and pilot plant operations. Since 2004, he has been pursuing a PhD degree in materials science and engineering at Auburn University, Auburn, AL, USA. His interests include biosensors, magnetostrictive materials, metallic glasses, PVD, CVD, microelectronic fabrication, and process development. Jiehui Wan received her BS (1999) and MS (2002) degrees in automatic control from Shanghai Jiao Tong University, Shanghai, China. She is currently pursuing PhD degree in materials science and engineering at Auburn University, Auburn, Alabama, USA. Her fields of interest include smart materials and structures, MEMS biosensors, process development and testing, and microfluidic chip development for cell/spore detection. Shichu Huang received her BS in materials science and engineering from Fuzhou University, Fuzhou, China. She is currently a PhD degree student in materials engineering at Auburn University, Auburn, Alabama, USA. Her fields of interest include magnetoelastic materials, optimization of filamentous phage as a biorecognition element for biosensors, microelectronic fabrication, process development, and multiple sensors for the simultaneous detection of bacteria and spores. Zhongyang (Z.-Y.) Cheng is an associate professor of the Materials Research and Education Center and the Department of Mechanical Engineering at Auburn University. He received his BS in applied physics, MS in electronic materials and components, and PhD in electronic engineering from Xian Jiaotong University, China. Afterwards, he went to Heinrich-Hertz Institute, Germany, as a visiting scientist for one year and then to University of Puerto Rico as a postdoctoral research associate for one and half year. Prior to joining Auburn

M.L. Johnson et al. / Sensors and Actuators A 144 (2008) 38–47 University in the summer of 2002, he was a research associate at Penn State for four years. His research interests include various smart materials (ferroelectric/piezoelectric/dielectric ceramics and polymers, magnetostrictive alloys, ceramic-polymer composites) and their applications in actuators, transducers, and sensors, especially for use in biosensors. Valery Petrenko graduated from Moscow State University (1972), and received a PhD degree from the Institute of Organic Chemistry, Moscow, Russia (1976), and a DSc degree from Moscow State University, Russia (1988). He worked as junior and senior scientist (1977–1982), laboratory head (1982–1985), scientific deputy director (1985–1989), director of institute, scientific deputy general director, and professor (1989–1993) in the Scientific Association “Vector” (Novosibirsk, Russia). He was awarded Honor Ranks of Senior Scientist in bioorganic chemistry and professor in molecular biology from the Supreme Attestation Committee of the Government of the USSR (1984, 1992). In 1993, he joined the faculty of University of Missouri-Columbia as visiting and research professor, and, in 2000, the faculty of Auburn University as professor. Dr. Petrenko is recipient (PI) of grants from ARO and NIH, and the Pfizer Animal Health Award for Research Excellence (2006). He is author of 91 peer reviewed journal articles and 4 book chapters. His research interests include monitoring and detection of biological threats, diagnosis of infectious and cancer diseases, and tumor targeting. He is a member of American Chemical Society and Phi Zeta Honor Society of Veterinary Medicine. Dong-Joo Kim received BS and MS degrees in ceramic engineering from Yonsei University in 1993 and 1995, respectively. He was awarded a PhD from North Carolina State University in 2001 for work on piezoelectric thin films. Following this, he worked as a postdoctoral researcher at Argonne National Laboratory and joined Auburn University in 2003 as an assistant professor. Dr. Kim’s interests include investigation of smart materials, such as piezoelectric and magnetoelastic materials, for functional applications.

47

I-Hsuan Chen received her BS in microbiology from Soochow University at Taipei, Taiwan in 1997 and a MS degree in microbiology in 2001 from Auburn University, Auburn, Alabama, USA. She is currently a research associate in Dr. James Barbaree’s group in the Department of Biological Sciences, Auburn University. Current research interests are focused on the development and evolutionary improvement of diagnostic probes against B. anthracis and Salmonella. James M. Barbaree is currently a professor and Chairman of the Department of Biological Sciences at Auburn University. He is the associate director for Auburn University Detection and Food Safety Center (AUDFS) at Auburn University. He has worked with the CDC for about 20 years. His research interests include microchip sensor development for rapid bacteria detection, molecular subtyping and detection of foodborne/waterborne bacterial pathogens, and Legionella pneumophila quorum sensing. J.W. Hong is currently an assistant professor of materials engineering in the Department of Mechanical Engineering at Auburn University. His research interests include single cell/molecule detection, BioMEMS/NEMS, biomaterials, automated high-throughput systems for systems biology, and on-chip nanoparticles synthesis. He received his PhD in chemical engineering from the University of Tokyo, Japan. Before joining Auburn, he was a postdoctoral scholar at the T.J. Watson, Sr. Laboratories of Applied Physics of the California Institute of Technology, Pasadena, CA. Bryan A. Chin is a professor of materials science and engineering at Auburn University, as well as chair of the program, and the director of AUDFS (Auburn University Detection and Food Safety Center). He is one of 50 US scientists inducted into the Russian Academy of Engineering Sciences. He has received numerous national awards and is the author of over 200 journal articles, 150 scientific reports, and chapters in various major reference books.