silicon-based novel bio-sensing platforms at the...

SILICON-BASED NOVEL BIO-SENSING PLATFORMSAT THE MICRO AND NANO SCALE

YALING LIU Department of Mechanical and Aerospace Engineering

The University of Texas at Arlington Arlington, Texas, 76019, USA

SAMIR M. IQBAL Department of Electrical Engineering

Nanotechnology Research and Teaching Facility University of Texas at Arlington Arlington, Texas, 76019, USA

E-mail: [email protected]

Microfabrication traditionally has been limited to silicon as demonstrated by the multi-billion dollar industry. The field is now understandably developing into a combination of physics, chemistry and biology in very non-traditional approaches to develop functional devices. Some examples are on-chip hybridization of deoxyribonucleic acid (DNA) and optical detection, DNA/protein/cell detection at the micro scale for diagnostics in medical, military, and food safety applications, etc. One area where silicon based nanotechnology can have an immediate and far reaching impact is sensing biological molecules and their processes, especially in genomics and proteomics. The promise of nanofabrication lies in the special ability to fabricate devices not only down to the scales of bacteria and viruses, but even down to the scales of the smallest biological entities, such as DNA!

Nanotechnology and BioMEMS will have a significant impact on medicine and biology in the areas of single cell detection, diagnosis and combating disease, providing specificity of drug delivery for therapy, and avoiding time consuming steps to provide faster results and solutions to the patient. Integration of biology and silicon at the micro and nano scale offers tremendous opportunities for solving important problems in biology and medicine and enables a wide range of applications in diagnostics, therapeutics, and tissue engineering. This paper reviews state of the art in silicon-based bioMEMS and bionanotechnology as well as the future challenges and opportunities. A range of silicon devices integrating micro-systems engineering with biology are discussed with focus on rapid detection of biological entities and development of point of care devices using electrical or mechanical phenomena at the micro and nano scale.

ECS Transactions, 16 (15) 25-45 (2009)10.1149/1.3104598 © The Electrochemical Society

25Downloaded 23 Mar 2009 to 129.107.47.90. Redistribution subject to ECS license or copyright; see http://www.ecsdl.org/terms_use.jsp

Evolution of Bio and Nano

The field of nanotechnology has seen unprecedented growth and development in last few years. It has been even termed as a “disruptive technology”, a technology that can sustainably overturn the existing dominant technological paradigms in the act of its evolution (1). The unusual chemical and physical properties of nanostructures have huge potential of transforming many aspects of modern life. There are two important subfields (emerged in last few years) that have brought life-scientists and engineers very close to each other in their quest of understanding life and finding cures: nano-biotechnology and bio-nanotechnology. These areas hold some of the greatest promises, drawing up contributions from chemistry, physics, biology, materials science, medicine and engineering. A long list of high-impact research is an evidence of the revolutionary advancement that is coming about at the confluence of biology and nanotechnology. The research efforts have focused on both directions i.e. the development of new synthesis methods of nanostructures, mediated by or built upon biological components (DNA, proteins, viruses, and cells); or development of nanostructures for biological uses, to sense, regulate or disrupt biological processes. Integration of biology and silicon at the micro and nano scale offers tremendous opportunities for solving important problems in biology and medicine and enables a wide range of applications in diagnostics, therapeutics, and tissue engineering.

Micro/Nano-Biosensors and Lab-on-a-Chip

The key characteristics of a silicon based detection modality for biological and chemical detection include accuracy, low cost for mass production, fast response, and ease of use. The techniques in currency rely on time consuming sample preparation, use of fluorescent tags, and expensive equipment that require special expertise. The modifications like tagging can change the thermodynamic properties of the molecular interactions and in some cases un-naturally stabilize/destabilize the DNA duplex and change the melting temperature (Tm) significantly (2,3). Additionally, expensive fluorescent microscopes are needed to visualize the data and normalization of the data to references remains problematic. Furthermore, the hybridization of the probes and target molecules is a diffusion-limited process requiring long-incubation time as the target must travel to the arrayed probe on the surface of the chip. The microarray technology is well adapted with vast expertise in sample preparation and handling of assays but the fluorophores are also known to have great effects on the stability of the duplexes as a function of the sequence itself (4). In recent years, several silicon-based nano-devices and nano-objects have been reported for chemical and biological sensing, including carbon nanotubes (5), cantilevers (6), nanowires (7), and field effect transistors (8). Silicon based porous materials have also been used for bio-sensing (9).

In this review, we focus on solid-state nanopores and their applications in biologicallyimportant detection, dielectrophoresis filters and traps for separating biological entities, and nano-mechanical cantilever sensors for detection of viruses.

ECS Transactions, 16 (15) 25-45 (2009)


Nanopore Sensors for DNA Detection

The concept of DNA detection with a single biological or solid-state pore has been investigated in recent years. Numerous studies have explored the biophysical properties of simple DNA translocation through the nanopores. While there have been reports of engineering selectivity in biological nanopores, devising means to impart selectivity to the solid-state nanopores remains a challenge. The biological nanopores, referred to as ion-channels in the biological community, are crucially important parts of living organisms; providing paths for selective transport of ions or macromolecules in cells and organelles. And such selective transport occurs due to the binding of the diffusing particles to the channel (10). These ion-channels have inspired the extension of the solid-state nanopores as selective detectors of translocating species. Design of such sensitive sensors, that can identify even single-base mismatch in the target molecules, is a major step towards cheap and disposable personalized therapeutics, opening avenues to novel devices for sequencing, detection of single nucleotide polymorphism, expression analysis, etc. from very few copies of the target DNA molecules. Such a label-free electrical detection design is also adapted towards bigger entities like proteins and other macromolecules at extremely low concentrations. Some of the recent reports mimic the selectivity that is characteristic of natural ion-channels that are present at the surfaces of cells or nuclei. There are some very interesting physical phenomena in such selective channels that result into facilitated transport of biomolecules. The knowledge of the chemical interactions and kinetics of the transported entities can provide useful means in areas like drug-discovery and tissue engineering. The functionalization of synthetic nanopore channels can also be used to explore the nature of cationic selectivity in the regulation of cellular processes. Solid-state nanopores have emerged as tools for single molecule manipulation, characterization and chemical analysis.

Biological Nanopores

Kasianowicz et al. pioneered the use of a single 2.6 nm diameter -hemolysin ( -HL)channel (11). -HL is a protein toxin from the bacteria Staphylococcus aureus. A solvent-free bilayer membrane of diphytanoyl phosphatidylcholine (an artificial lipid bilayer membrane) was formed across a 0.1 mm diameter orifice. The orifice was in a Teflon partition, separating two buffer filled compartments. When -HL was added to a compartment it reconstituted into the lipid bilayer making a channel. A single channel usually formed within 5 minutes. A patch clamp amplifier was used to convert current to voltage. In the absence of DNA, applying a potential of 120 mV resulted in baseline current that was free of transients. Following the addition of DNA to the cis side of the channel, numerous short-lived current blockades occurred. They used these channels for the current fluctuation detection when a single DNA traversed and passed the pore. The duration for which there were transient decreases in ionic current were called translocation times. The translocation times were proportional to the DNA lengths. Typically, the magnitude of the blockades was large (the current was reduced by 85–100%) and, depending on polymer length, the blockades lasted from several hundred to several thousand microseconds. For a given polymer size, the total number of transient blockades was directly proportional to polymer molar concentration. Virtually no blockades were observed in the absence of DNA. It was further shown that DNA length, one of the DNA characteristics, was directly proportional to the mean lifetime of the peaks in the signals. This realized the determination of lengths of individual RNA or



DNA chains using single channel measurements. Applying the same principle towards a chemical sensor, Hagan Bayley and co-workers detected organic molecules of as low as 100 molecular mass using -HL pores (12). The innovative step in their study was to equip the -HL nanopore with an internal, non-covalently bound molecular “adapter” which mediated channel blocking by the analyte, thus sensitizing the nanopore to specific organic and chemical species. They used -cyclodextrin ( CD) as an “adapter” that fitted comfortably inside the pore and presented a hydrophobic cavity suitable for binding a variety of organic analytes. CD, a doughnut-like molecule made of seven sugar units, diffused into the -HL channel, partially obstructing its water-filled pore. The authors showed that cyclodextrin molecule was exquisitely sensitive to different guest molecules. Thus, when bound to cyclodextrin, members of the adamantane family of petroleum derivatives could be distinguished, as could the members of the group of tricyclic pharmaceuticals that included imipramine and promethazine. These guest molecules each bound to -HL resident CD for milliseconds and made their presence known by altering the electrical conductivity of the -HL pore. A single sensing element of this sort could be used to analyze a mixture of organic molecules with different binding characteristics, so that the pore could be in a way “programmed” for range of sensing functions.

Meller et al. used the statistical data derived from the patterns of events to show that nucleotides of different sequences could be distinguished from each other (13). Six polymers with the same lengths but different sequences were measured, and results were then statistically analyzed. The translocation events were characterized by the peaks in the blockade-current histograms, the translocation duration, and the temporal dispersion of translocation duration. Differences between these parameters were then used to distinguish between poly(dA)100 and poly(dC)100 DNA molecules. The translocation times varied among the polymers even if they had the equal lengths, thus different sequences could be discriminated in a mixture. The authors also discovered a strong T–2

dependence of the translocation duration between 15 °C and 50 °C. Above this temperature, the structure of the pore was not stable and measurements could not be performed. The temperature dependence provided another parameter to control the movement of the DNA through the -HL nanopores.

Bayley and co-workers again demonstrated covalent attachment of single-stranded DNA (ss-DNA) within the lumen of the modified -HL pores, what can be termed as a “DNA–nanopore” (14). The ss-DNA molecule was attached on the cis side of the -HLchannel. The binding of target ss-DNA molecules to the tethered DNA strand caused changes in the ionic current flowing through the -HL nanopore. On the basis of DNA duplex lifetimes, the DNA–nanopores were able to discriminate between individual DNA strands up to 30 nucleotides in length differing by a single base. The mean event lifetimes were obtained by lifetime histogram analysis and it was seen that the type of mismatched base pair influenced the lifetime, and that the mismatch had the most dramatic effect when it was positioned in the middle of the oligonucleotide. Using an array of DNA tethered -HL pores they sequenced a complete codon in an individual DNA strand. However, they did not look into any change in the translocation time of a DNA sequence complementary to the tethered DNA as compared to the translocation of the same through an un-tethered pore. Some recent work has focused on characterizing physical parameters that govern the analyte translocation through the ion-channels, like biasing voltage or salinity of the buffer solutions (15).



Solid-State Nanopores

Researchers have been making “point contacts” since as far back as 1977. These were used mainly for material science, especially the ballistic transport studies. The schematics of these structures were very much like the nanopores, as these stand today, but the aim was quite different (16,17). The design of current solid-state nanopore was inspired by one such work by Gibrov et al. (17) and these are the next most ideal replacement for biological nanopores owing to several established and foreseen advantages: First, as for other chemical sensors, sensitive electronic circuitry and photonic sensing capabilities can be integrated directly into a pore–membrane system. Secondly, simultaneous and automated analysis of hundreds of arrays of different channels can potentially be achieved with such an integrated system. Next, they are more robust to withstand wide range of temperature, analyte solutions properties, environments and chemical treatments that might be required for the target detection and to eliminate interference. And finally, these can be customized to fit in practical biosensors. These properties have heightened the interest in solid-state nanopores, with the particular attention as progenitors of rapid and cheap next-generation DNA sequencing “machines”.

Fig. 1 (a) The schematic shows how a 60 nm pore was made, and (b) shows the whole system of feedback-controlled ion milling process (18). Reprinted by permission from Macmillan

Publishers Ltd: Nature Materials, copyright (2003)

In an earlier report on solid-state nanopores, Li et al. utilized a feedback-controlledion-beam sculpting process to make a nanopore in a silicon nitride membrane (18). Fig. 1 shows the schematic diagram of their process. They made a bowl-shaped cavity in a silicon nitride membrane and removed material from the other side of this membrane using Ar ion-beam sputtering. As soon as the ion-sculpted side reached the bottom of the cavity (seen as the top side of the cavity, as cavity opening is facing down shown in Fig. 1 (a)) a hole around 60 nm was opened. Continuous ion-beam exposure reduced the hole to 1.8 nm diameter & ultimately closed it. The closing of hole was indirectly seen as the drop in transmitted ion counting rate, and directly imaged in transmission electron



microscope (TEM). The two processes of ion-beam erosion and atomic flow of matter were suggested to be occurring simultaneously. They characterized the various parameters that affected the pore formation e.g. the temperature, ion-dose, and ion-beam exposure time. This gave ability to engineer the process flow parameters to consistently fabricate pores of intended diameters. They postulated two mechanisms for the pore size reduction: Surface matter moving due to reduced viscosity to relax the stress caused by implantation or the creation of adatoms on the surface by incident ions that could diffuse to close the pore. They used a 5 nm diameter pore with 500 base-pair (bp) double-stranded DNA (ds-DNA) just like the studies with biological nanopores. After applying a voltage bias that would draw the negatively charged DNA molecules through the nanopore, blockages of the ionic current were observed, similar to those observed for ss-DNA translocating through the -HL channel in a lipid bilayer. The same group subsequently carried out comparative translocation measurements on 3000 and 10000 bp ds-DNA with 3 nm and 10 nm pores (19). They correlated translocation times to DNA lengths, under various bias conditions. Longer DNA translocation was noted to be more complex because of folding. However, the electrophoretical velocity of shorter 3000 bp DNA was found to be 1.6 cm/s. They noted that most measurements were done with 10 nm pores as 3 nm pores were too small to let a folded 10000 bp molecule to pass, thus comparisons done with 3 nm pores would have been unrealistic.

Fig. 2 (a) TEM image of the nanopore channel in the oxidized silicon membrane. (b) Schematic of the measurement setup showing the chip clamped between Teflon blocks. Reprinted with

permission from (20). Copyright 2004 American Chemical Society.

Dekker and co-workers reported the fabrication of a nanopore with electron-beam lithography (EBL) and TEM techniques (21). Their method realized an in-situ observation of pore size while the size was precisely controlled at the nanometer scale. Bashir and co-workers reported a similar approach, although developed independently (20). They fabricated 50-60 nm long 4-5 nm diameter nanopore channels (NPC), in micromachined Si membranes. The NPCs were fabricated in double-side polished SOI wafers. They used EBL to initially define the pore and with standard solid-state processing fabricated 3-4 nm nanopores. The pore was examined visually in a TEM while its diameter shrank to the desired size. The pore shrinkage occurred under an electron beam (e-beam) of high energy. However, their mechanism was completely different from Li’s process. In this process, the e-beam locally fluidized the oxide around the pore, and hence oxide flowed towards the pore and then filled it (22). Figure 2(a)

(a) (b)



shows a final TEM image of a typical NPC (20). Each NPC, fabricated as shown in the Figure 2, was formed in a 3 mm2 die that was about 0.5 mm thick. The nanopore die was sandwiched between two silicone rings and placed in a finely milled pocket within Teflon blocks separating the chambers. The Teflon blocks were clamped together, as shown in Figure 2(b).

Fig. 3 The signal showing translocation of the 200 bp ds-DNA. Typical pulses are shown. The inset shows one pulse in more detail. Reprinted with permission from (20). Copyright 2004

American Chemical Society.

Electrophysiology measurements were performed using an Axon 200B amplifier in the resistive feedback mode, and a CV201 head stage. The nanopore and the head stage were housed inside a grounded Faraday cage to reduce the environmental noise. The amplifier signal was passed through a 300 Hz low-pass filter and fed to a Digidata 1322A board on computer with PClamp 9.0 (Molecular Devices, Sunnyvale, CA) software for data acquisition. A 200 bp fragment from the human CRISP-3 gene was PCR amplified using cDNA from a prostate cancer specimen and 0.1 M KCl, 2mM Tris (pH 8.5) buffer was used for measurements. The nanopore surface was passivated with 2-propanol to avoid bubble formation on contact to the buffer solution. Once the baseline current stabilized, purified DNA at a final concentration of 0.3 g/ml was applied to one chamber. Ag/AgCl electrodes were used to make the ionic current measurements. The steady-state ionic currents in the range of 55-70 pA were measured through the nanopore channel at 200 mV. As the ds-DNA passed through the pore, it modified the bulk and the interface currents resulting in typical pulses as shown in Figure 3. The current increased unlike previous studies, where the current was always shown to decrease. The results were explained by the charge on the DNA which can be detected in the nanopore channel due to an inherent charge amplification and transduction of the DNA charge into ionic



current in the mobile surface charge, similar to a field effect transistor. The equation of the ionic current through the solid-state nanopores can be expressed as (23):

pulse DNA Counterions BlockI = I + I - I [1]

where Icounterions stems from the condensed counterions on DNA, IBlock is the blockage that comes from the mechanical reduction of pore area during DNA translocation and IDNA is the current from DNA itself. The sign notations indicate whether the current component enhances (+) or reduces (-) the pulse, resulting in contributions to upward or downward direction of the pulses, respectively.

The nanopore technology has attracted more attention by a number of research groups in recent years. The research has focused in variety of directions; towards improving the structures, reducing the time to fabricate, fine tuning various parameters like noise, studying the effects of changing voltages, studying force kinetics, exploring conformation changes of molecules and most importantly towards use of nanopores in investigating various biophysical properties of DNA.

Chen et al. employed atomic layer deposition (ALD) of alumina to controllably shrink the pore size to required nanometer dimensions. The ALD was shown to reduce the noise of the pore and change the surface properties with the passivation effects of deposited alumina film (24). The same group compared ds-DNA translocation through nanopores of different channel lengths; nanopore in a thin membrane and nanochannels through a thicker membrane. They showed that DNA of varying lengths (3, 10 and 48.5 kbp) all traveled at same velocity through both types of the nanopores (25). They showed DNA electrophoretic mobility to be independent of the strength of the applied electric field and the length of the DNA. The nanopores have also been used in applications to fabricate point electrodes with lateral dimensions in the range 15-200 nm (26). The nanopores were made in insulated silicon membrane followed by 5 nm/300 nm Cr/Au evaporation on one side of the device. The metal thickness was sufficient enough to close the pore resulting in a controlled conical nanometer sized Au electrode protruding on the other side of the nanopore. The diffusion-limited current with stepped-current voltammetry was measured quantitatively showing current dependence on electrode size taking into account the known geometry of the electrodes. In another study, the effects of imaging beam of a TEM through nanopores with various geometries have also been studied (27). It was shown that the pores that were smaller than a certain critical size shrank while the larger ones expanded, on the same wafer, agreeing with the hypothesis that surface-tension effects drove the modifications. The chemical composition in the pore region was also analyzed showing that contamination growth was not the underlying mechanism of pore closure. To study the pore formation after electron beam deformation, electron energy loss spectroscopy and high-resolution electron microscopy studies were also reported (28). The material composition was studied extensively during the pore shrinkage, pore expansion and pore closure, concluding that the pore formation occurred by the removal of material at the entrance and exit side of the electron beam. The pore formed had wedge like edges and contained Si, O, and N.

Heins et al. used a conically shaped nanopore with small end diameter of ~4.5 nm to even detect porphyrin molecules, as against detection of particles or DNA (29). They fabricated the nanopores by anisotropic etching of the single-ion-irradiated polymer



films. They applied high trans-membrane potentials without membrane rupture and noted the translocation events increased with voltage. In another approach to make nanopores, Ho et al. used tightly focused e-beam to make nanopores in 10 nm thick silicon nitride membranes (30). They measured the ionic conductance as a function of various conditions and found it to be much larger than the bulk conductivity in case of dilute bath concentrations. The ionic conductance was found comparable with or less than the bulk for high bath concentrations. These were explained in terms of the Debye length which was larger than the pore radius for the former case. They also reported consistent multiscale simulations using molecular dynamics of the ion transport through the pores. They explained the ion transport by the coupled Poisson–Nernst–Planck and the Stokes equations solved self-consistently for the ion concentration, velocity and electrical potential. The results suggested the presence of fixed negative charges on the pore walls, thus the ion proximity to the pore walls reduced the ion mobility. In most experiments, the DNA translocates the nanopores at as high speeds as 27-30 bases/µs (31). It was reported that the translocation speed could be reduced by controlling the electrolyte temperature, salt concentration, viscosity, and the bias voltage across the nanopores (32). The speed was reduced to 3 bases/µs for 3 kbp ds-DNA through a 4-8 nm diameter silicon nitride pore. This significantly reduced the bandwidth requirements on electronic sensing system. However, slowing the DNA also slowed conducting ions which decreased the current blockage signal. In another study, it was reported that most of the translocation events followed a rule of constant event charge deficit. This conclusion was derived from the observation that molecules of the same length produced current traces whose areas were same. The current trace areas were calculated from the time integral of the deviation from the baseline current. The experiments compared the distribution of molecular event durations and blockade at various pH values, conclusively demonstrating a large number of long denatured ss-DNA translocated in stretched out conformation (33). This showed an ability to detect hybridization from the translocation behavior.

Heng et al. investigated the mechanical properties of DNA using translocation of single molecules through the nanopore (34). They used pores with radii from 0.5 to 1.5 nm in a 10 nm thick Si3N4 membranes. They also carried out molecular dynamics simulations to explore the stretching and bending of DNA during translocation through such small pores. They reported a pore dimension- and bias-dependent threshold for translocation. Simulating a 1 nm pore, they calculated forces on DNA to be in the range of 1-300 pN through it and observed the threshold for ds-DNA translocation conditioned by its mechanical properties. Heins et al., using the conically shaped synthetic polymer membrane nanopores, investigated the effect of the crown ether (18C6) on the measured ion current (35). Adding 18C6 to one side of electrolyte rectified the ionic current flowing through the molecule sized nanopore (~1.5 nm dia). They modeled the results based on the formation of a junction potential at the membrane solution interface. Keyser et al. used the ionic current through a solid-state nanopore to attempt accurate measurements of optical intensity and temperature distributions in aqueous solutions without using any optical elements. They profiled a laser in three-dimension by creating topographic maps of a 5 nm pore conductance at different z-positions. The pore conductance changed with the heating of laser as laser was scanned at different z levels (up to 20 K temperature change was measured at the center of the laser focus). They found a linear dependence of the ionic current on the incident laser power by measuring the laser-induced heat (36). Timp and co-workers reported the permeation capabilities



through various sizes of nanopores by changing the electric fields (37). The electromechanical properties of DNA were studied to define threshold of nanopore sizes that ss- or ds-DNA could permeate under various electric fields. The effect of pH on such threshold was also investigated. Their group has been more involved with simulation studies of DNA in very small (0.5-2 nm) nanopores (38,39). In the wake of the low salt current enhancement studies reported by Bashir’s group, Smeets et al.reported the effects of salt concentration on ion transport and DNA translocation using 10 nm diameter nanopores (40). They changed the salt concentration by 6 orders from 1 M to 1 µM, noting 3 order magnitude decrease in the ionic conductance, deviating from linear bulk behavior. They carried out 16.5 µm long ds-DNA translocation experiments, observing downward or upward translocation pulses as a function of ion concentration. They, like Bashir and co-workers had explained (20), described a model attributing current decreases due to partial blocking of the pore and current increases to the motion of the counter-ions screening the charge of the DNA backbone. They concluded a threshold KCl concentration of 370±40 mM where two competing effects cancelled out. A new method of fabricating nanopores was reported where a free standing Si3N4membrane was drilled to open a nanopore using tightly focused e-beam. Using same nanopores, fabrication of electrodes with radii 2 nm was also reported by the same group (41). They filled the nanopores with Au making conically shaped nanoelectrodes. These nanoelectrodes would have higher sensitivity, lower detection limits and small sample requirement. Metal nanoelectrodes of such high temporal resolution can be used for single-molecule detection, to probe neurophysiologic signals, as biomolecular sensors in medical diagnostics, and for real-time monitoring of cell exocytosis. Many more studies have appeared from Dekker’s group measuring the force on a single DNA while translocating through a nanopore by using optical tweezers, and ionic transport in electrochemical reactions (42-44). They integrated optical tweezers technique to the DNA translocation through nanopore system. They showed slowing and arresting of the DNA mid-way of its passage through the pore and carried out force measurements. They reported effective charge of 0.50±0.05 electrons per base-pair which is equivalent to 25% of the bare DNA charge. In the ionic transport studies they fabricated Au nanoelectrodes of 2-150 nm radii through the nanopores and investigated the conductance across the double layer via electrochemical reactions at the Au electrodes. They reported non-equilibrium distribution and transport of the ionic species caused by extremely sharp concentration gradients of the nanoscale electrodes. These findings deviated from the predictions based on the continuum model and need for advanced models to describe local high ionic fluxes was highlighted.

Bashir and co-workers reported two major findings in the realm of nanopores in mid of 2006; one was the fabrication of nanopores using conventional field-emission scanning electron microscope (FESEM) instead of TEM (45), and the other was to use nanopores as sensors of DNA counterion current and saturation (23). In the first report of FESEM based fabrication of nanopores, the initial pore fabrication was done using standard micromachining to diameters between 50 nm and 200 nm, followed by in situ FESEM controlled pore shrinking to sub 10 nm diameters. The irradiation from the electron source of the FESEM showed different shrinking mechanism possibly due to the effects of surface defects under radiolysis and subsequent motion of silicon atoms to the pore periphery. The process was reliable and reproducible, opening roadways for the parallel fabrication nanopores in an array format. The latter report investigated the ionic current fluctuations associated with the translocation of 1681 bp long DNA as a function of the



applied electric field. The pulse direction (upward or downward) was investigated to develop a model of DNA counterion ionic current and dipole saturation based on a well-cited work by Manning (46). These experiments showed that at low concentrations, the electrophoretic bias can reverse the pulse directions from upwards to downwards indicating competition between the mechanical blocking current and the current arising from the condensed counterions on DNA backbone (Equation 1), which saturated at high electric field. The experiments provided a direct mean to explore the charge polarization and dipole saturation at the single molecule level.

The dynamics of DNA in nanopores has become a crucial topic toward the development of lab-on-chip devices for biomolecular analysis. However, the interaction between DNA and nanopore is still not well understood due to the small size of DNA/nanopore and dynamic translocation process. Studies on electrophoretic transportation of DNA in nanochannels have revealed significant contribution of DNA-channel surface interaction, which leads to a diffusion rate much lower than that predicted by traditional diffusion theory (47). Moreover, various chemical modifications are applied on nanopore surface for improved signal yield.

From Bare Nanopores to Surface-Modified or Functionalized Nanopores

Notwithstanding these many reports on “bare” nanopores, a number of important potential applications can still be developed especially for genome sequencing. There are diverse scientific challenges with the most important being selectivity to detect single base.

Fig. 4 (a) Schematic of a functionalized nanopore (First appeared in (48)). The chemical modification reduced the nanopore diameter as shown. (b) Example of current traces for PC

(blue/left) and MM (red/right) DNA data, shown side by side for clarity. (c) TEM micrograph of a nanopore at 1 million times magnification. Scale bar is 20 nm.

It is important to slow down DNA not only to gather more information with reasonable bandwidth requirements on the measurement systems but also to detect the individual nucleotide, which is probably the ultimate enabling step in a sequencing application, all the while maintaining high signal-to-noise ratio. A wide variety of nano-

(a) (b)

(c)



channels have been researched in recent years, with varying degrees of success, for their potential applications in molecule separation, stretching, etc. These devices possess distinct features, with size control in the range of a few nanometers. Now is the opportunity to modify their surfaces bio-chemically for DNA sequencing and specific gene/biomarker detection.

Functionalized Nanopores for Selective Detection

A number of papers have attempted coating conical pores and membranes to impart selectivity, e.g. conical gold nanopores, made in 12-µm thick poly(ethylene terephthalate) membranes (with one side opening of 600 nm and narrow opening of 5–9 nm), have been modified with biotin, protein G and anti-ricin IgG (49). In another study, a conical pore of 650 nm diameter has been used for selective detection of the icosahedral chlorella virus, Parsamecium bursaria (50). The molecular recognition is confined to the narrow opening of these conical pores that would probably limit the lowest detectable sample concentration. Solid-state nanopore channels with DNA selectivity have been reported recently where the nanopores were coated with hairpin loop DNA molecules (Fig. 4(a), 4(c)) (48). This coating made the nanoproes selective towards short lengths of 'target' ss-DNA, transporting perfect complementary (PC) molecules faster than mismatched ones (MM). Even a single base mismatch between the probe and the target resulted in longer translocation pulses and a significantly reduced number of translocation events. The pulses for PC ss-DNA were narrower and deeper than those for mismatched ones (Fig 4(b)), indicating facilitated (and faster) transport of PC-DNA while interacting with surface bound hairpin molecules. Such and similar functionalization schemes can be used for a variety of ligand-receptor combinations of significant importance, and the solid-state functionalized nanopore can serve as next generation tools with nanopores functionalized with specific probe can be used as detector of specific nucleotide(s)/biomarkers. In our recent work, we have established critical dependences between properties like surface roughness/charges, pore size, electric bias across the nanopores, and translocation kinetics using models that link atomistic DNA-nanopore interaction to meso-scale particle dynamics (51).

Importance of DNA-Nanochannel Interaction

Due to the small size and compact confinement, the DNA-nanochannel surface interaction largely influences the translocation process. It is generally known that DNA transport in microchannels could be treated similar to free diffusion in gel. However, for nanochannel or nanopores with diameter under 10 nm, DNA-channel surface interactions become important. In this process, a DNA molecule is confined in a channel with size smaller than its persistence length. The DNA interacts with the channel surface via van der Waals force or with coated complimentary DNA hairpins via specific hybridization. To enable rational designs of the nanopore-based sequencing device or nanochannel-based molecular carriers, it is crucial to understand the underlying molecular mechanisms that govern the DNA-nanochannel interaction.

There are various factors that may contribute to surface interaction, such as nanopore sizes, topology, roughness, DNA length, pH level, electric field strength, surface property, etc. In particular, the size, electric field strength effect and surface property effect can add interaction-specific effects. This can be also applied to protein detection,



as some recent work has shown specific, irreversible and diffusion-limited detection of streptavidin, IgG and ricin with single conical Au nanopores fabricated in poly(ethylene terephthalate) membranes (49). Protein detection, especially focused on for disease biomarkers, can be efficiently realized with greater performance efficiency by using single biofunctionalized nanopores integrated in microfluidics systems and utilizing electrical-field-driven mass-transport conditions (52).

Using DEP Force for Biomolecule/Bacterial/Cell Concentration on a Chip.

Important biological functions like transport, detection, isolation and manipulation of nano and biomaterials in MEMS devices are drawing more attention in recent years. Various mechanisms can be used to capture nano and biomaterials. Among them, electric forces, with tunable amplitude and easy experimental setup, have been successfully applied to manipulate different biomolecules and cells. An applied electric field will introduce several major effects such as dielectrophoretic force (DEP), electroosmotic flow, and electrothermal effect. The DEP force arises from induced dipole moments on a particle immersed in a non-uniform electric field. The electroosmosis flow is driven by the electrostatic force applied to the charged double layer on the electrode surfaces. The time-averaged DEP force on a particle is given by the effective dipole moment theory (53,54):

2DEP 1 fRe{ } ( )K EF [2]

where is a geometric parameter that depends on the particle shape and size; 1 is the real part of the permittivity of the medium; fK is the polarization factor that depends on the complex permittivities of both the particle and the medium; E is the root-mean-square magnitude of the local electric field. For a spherical particle with radius a , 32 a and

2 1 2 1( ) ( 2 )fK (called the Clausius-Mossotti factor).

The DEP force has been used to either attract or repel particles according to the difference in permittivities. Ramos et al. (55) has reviewed the various factors that contribute to motion of particles in a suspension when subjected to an AC electric field. Liu et al. (56-59) have studied the dielectrophoretic assembly of nanowires, viruses, and the electrodeformation of cells through coupled electrokinetics and immersed finite element method (57,60). Besides attraction or repulsion, the manipulation of cells or bacteria has been of greater interest. For example, live cells could be differentiated from dead cells by tuning the frequency of an AC electric field (61). Zimmermann et al. used a high-amplitude AC electric field to manipulate mammalian cells (62). Recently, more precise and integrated methods have been proposed to detect cells and bacteria. Bacterial cells have been concentrated on a chip using DEP (63). A microfabricated electrode array is used to trap cells and bacteria within microfluidic chips. When cells are transported across an array of microelectrodes, the positive DEP force traps cells onto the edges of the microelectrodes. The required voltage to capture different particles such as polystyrene beads, yeast cells, spores and bacteria has been characterized through an experimental protocol. A finite element model considering DEP force, hydrodynamic drag force, sedimentation force, and hydrodynamic lifting force is proposed to predict the DEP holding force for both positive and negative dielectrophoretic traps. For 2.4 µm



polystyrene beads with conductivity of 2x10-4 S/m, the calculated DEP forces magnitude ranges from about 8x10-15 N to 2.4 x10-14 N with a voltage of 20 Vpp applied on to the electrodes. The predictions of the model agree well with the experimental measurement.

Fig. 5 Illustration of using DEP force to trap beads and bacteria in a flow channel. With permission from Rashid Bashir.

Sjober et al. have developed a microfluidic bioprocessor for rapid detection of bacterial metabolism (64). The bioprocess is composed of particle separation by DEP deviation, particle concentration, and impedance detection, as illustrated in Fig. 5. The bacterial cells from a dilute sample are concentrated by DEP by factors on the order of 104 to 105. The concentration process eliminates the long culturing steps to amplify bacterial population, thus enables detection of bacteria in extremely small volumes. The metabolic activity of the concentrated cells is measured through the impedance change of the medium. Live cells usually release ionic metabolites, which change the impedance of a pair of electrodes immersed in the culture medium, commonly called “Impedance Microbiology”. Detection results with dilute L. Monocytogenes shows significant time reduction compared to conventional methods (65).

Cantilever for Molecular Detection

Microcantilevers have been extensively used as ultrasensitive biosensors for biomolecule detection. There are two modes of microcantilever operation. In static mode, surface stress generated by the binding molecule induces deflection of the cantilever, which is used to detect receptor–ligand binding (66), DNA hybridization (66), and antigen–antibody binding (67). In dynamic mode, shift of the cantilever’s resonant frequency is used to detect added mass of the binding molecules. In both modes, the deflection measurement is important. Various methods has been proposed to measure the deflection of cantilever beams such as optical reflection (68), capacitance (69), piezoelectric effect (70), and piezoresistance (71,72).

Static Mode Operation

When operated in static mode, the microcantilever is functionalized with receptors at only one side of the surface. The receptors bind with a particular type of analyte in the solution, thus generate surface stress which bends the cantilever. The deflection of the cantilever beam at the free end is given as (73):

DetectionC hamber

Electrodes

FlowFlow

Beadsand

bacteria

DetectionC hamber

Electrodes

FlowFlow

Beadsand

bacteria



2

1 214 ( )l v

zt E

[3]

where 1 and 2 are the surface stresses changes on the top and bottom of the cantilever; v is Poisson’s ratio, E is Young’s modulus, l is the length, and t is the thickness of the cantilever beam.

From Eq. (3), the deflection of the cantilever increases with decreased thickness and increased length, which drives the fabrication of thinner and longer cantilever for higher detection sensitivity.

(a) (b)

Fig. 6 (a) Hydrogel patterned microcantilever (b) Bending data versus pH for hydrogels patterned microcantilever. Solid diamonds are for the increasing pH path, whereas hollow diamonds are for the decreasing pH path. Reprinted with permission from reference (74).

Copyright 2002, American Institute of Physics.

Based on this principle, Bashir et al. fabricated a pH microsensor, as shown in Fig. 6 (74). Intelligent hydrogels were coated on the surface of the microcantilever. The hydrogels contain acidic pendant groups which are sensitive to environmental conditions, i.e. these ionize if the pH of the environment is above the acid group’s characteristic pKa.For a microcantilever with a thickness of 0.8 µm and coated with a 2.5 µm thick polymer, a sensitivity of 18.3 µm/pH unit is demonstrated. For a deflection detection resolution of 1 nm which can be easily achieved by optical laser, the translated sensitivity is 5 105 pH/nm. This ultrasensitivity is two orders of magnitude higher than that offered by other microscale techniques such as scanning probe potentiometer (75).

Dynamic Mode Operation

In dynamic mode operation, the microcantilever’s resonant frequency shifts when biomolecules attach to the microcantilever surface. The resonant frequency of an unloaded cantilever is:

01

2k

fm

[4]



where k is the spring constant of the microcantilever and m is the cantilever mass. When a biomolecule or particle with mass m is loaded onto the cantilever, the resonant frequency decreases to

11

2k

fm m

. [5]

thus

2 2 21 0

1 14

km

f f [6]

The detection of a single virus particle using the resonant frequency shift principle was reported by Bashir and co-workers (76). Microfabricated cantilevers in the range of 4-5 µm length, 1-2 µm width and 20-30 nm in thickness were used in the experiments, with measured spring constant around 0.005-0.01 N/m. As shown in Fig. 7, vaccinia virus was used in the study. The frequency change of the cantilever beam after virus attachment was measured by a laser Doppler vibrometer under ambient conditions. A linear function between resonant frequency shift and effective number of attached virus particles was calibrated. A resonant frequency shift of 60 kHz after the addition of a single virus particle was reported, with measured average dry mass of 9.5 fg (consistent with the range of expected mass of 5-8 fg).

(a) (b)

Fig. 7 Detection of a single vaccinia virus by a microcantilever. Reprinted with permission from reference (76). Copyright 2004, American Institute of Physics

The sensitivity of the cantilever resonant frequency shift based biosensor is determined by 00.5 / cf m , where cm is the mass of the cantilever. Cantilever with smaller size are more sensitive, which motivates scaling biosensors to nanoscale. Gupta et al.have fabricated nanomechanical biosensor cantilevers and showed a complex response to the attachment of protein molecules (77). Three different receptor attachment schemes were used to functionalize the cantilevers (Fig. 8): Composite protein layer which

f0 = 1.27 MHz

f1 = 1.21 MHz Q ~ 5 k = 0.006 N/m

f=60kHz



captured viruses (Fig. 8(a)); Control BSA layer coated cantilever which did not capture virus particles (Fig. 8(b)), and; Antibody-coated cantilever without BSA which captured a large number of particles/aggregates non-selectively (Fig. 8(c)). For scheme 1, the attachment of proteins was found to either decrease or increase the resonant frequency, which was explained by a size-specific modification of diffusion and attachment kinetics of biomolecules on the cantilever. Longer cantilevers were found to have higher protein density due to the competitive protein attachment among the adsorbing sensor surfaces. Such cantilever length dependent protein density needs to be considered when designing cantilever based nanomechanical sensors.

Fig. 8 Three protein attachment schemes and corresponding images of the functionalized cantilevers(77). Copyright by The National Academy of Sciences of the United States of

America.

Conclusions and Future Outlook

The key characteristics of silicon-based detection modalities are (a) accuracy, down to single molecule/cell sensitivity without altering the molecular interaction dynamics that come from label tagging, (b) low cost from parallel mass production using modern fabrication techniques, (c) fast and real-time response from the change in electrical characteristics, (d) ease of use with the portability of on-board measurement circuit and



algorithms, and (e) multiplexing from reusable and analyte-specific chips that can be custom made for various targets. Silicon-based processing has significant advantages over other sensor technologies, especially for the characteristics (b), (c) and (d) above.

The solid-state silicon based devices are thus having a great impact in bio-sensing and detection. The promise of selective, cheap and rapid detection of the biological entities of interest with electrical sensing modalities has resulted in many new ways of looking at disease and will lead to many more ultra-sensitive point-of-care bio-sensors. Such devices not only provide means to characterize biological processes at the most fundamental scales but also an opportunity to perform biochemical analysis with minimal sample volumes.

Acknowledgements

Y. Liu acknowledges funding from University of Texas at Arlington Research Enhancement Program. S. M. Iqbal acknowledges funding from Metroplex Research Consortium for Electronic Devices and Materials (MRCEDM). Authors would like to thank Rashid Bashir whose insights and directions helped accomplish this paper. Special thanks to Yi-Shao Liu (UIUC), Oguz Elibol (Intel Corporation), Kidong Park (UIUC), Demir Akin (Stanford University), Bala Murali Venkatesan (UIUC), Bobby Reddy Jr. (UIUC), and Hung Chang (Arizona State University), for useful discussions.

References

1. S. T. Walsh, Technological Forecasting and Social Change, 71, 161-185 (2004). 2. K. M. Ririe, R. P. Rasmussen and C. T. Wittwer, Analytical Biochemistry, 245,

154-160 (1997). 3. B. G. Moreira, Y. You, M. A. Behlke and R. Owczarzy, Biochemical and

Biophysical Research Communications, 327, 473-484 (2005). 4. E. E. Merkina and K. R. Fox, Biophys. J., 89, 365-373 (2005). 5. K. Maehashi, T. Katsura, K. Kerman, Y. Takamura, K. Matsumoto and E.

Tamiya, Anal. Chem., 79, 782-787 (2007). 6. A. K. Gupta, P. R. Nair, D. Akin, M. R. Ladisch, S. Broyles, M. A. Alam and R.

Bashir, Proceedings of the National Academy of Sciences, 103, 13362-13367 (2006).

7. F. Patolsky, B. P. Timko, G. Zheng and C. M. Lieber, MRS Bulletin, 32 (2007). 8. J. Fritz, E. B. Cooper, S. Gaudet, P. K. Sorger and S. R. Manalis, Proceedings of

the National Academy of Sciences, 99, 14142-14146 (2002). 9. L. De Stefano, L. Moretti, A. Lamberti, O. Longo, M. Rocchia, A. M. Rossi, P.

Arcari and I. Rendina, Nanotechnology, IEEE Transactions on, 3, 49-54 (2004). 10. W. R. Bauer and W. Nadler, Proceedings of the National Academy of Sciences,

103, 11446-11451 (2006). 11. J. J. Kasianowicz, E. Brandin, D. Branton and D. W. Deamer, Proceedings of the

National Academy of Sciences, 93, 13770-13773 (1996). 12. L. Q. Gu, O. Braha, S. Conlan, S. Cheley and H. Bayley, Nature, 398, 686-690

(1999).13. A. Meller, L. Nivon, E. Brandin, J. Golovchenko and D. Branton, Proceedings of

the National Academy of Sciences, 97, 1079-1084 (2000). 14. S. Howorka, S. Cheley and H. Bayley, Nature Biotechnol, 19, 636-639 (2001). 15. Q. Zhao, D. A. Jayawardhana and X. Guan, Biophysical Journal, 94, 1267 (2008).



16. K. S. Ralls, R. A. Buhrman and R. C. Tiberio, Appl. Phys. Lett., 55, 2459 (1989). 17. N. N. Gribov, S. Theeuwen, J. Caro and S. Radelaar, Microelectronic

engineering, 35, 317-320 (1997). 18. J. Li, D. Stein, C. McMullan, D. Branton, M. J. Aziz and J. A. Golovchenko,

Nature, 412, 166-169 (2001). 19. J. L. Li, M. Gershow, D. Stein, E. Brandin and J. A. Golovchenko, Nature Mater.,

2, 611-615 (2003). 20. H. Chang, F. Kosari, G. Andreadakis, M. A. Alam, G. Vasmatzis and R. Bashir,

Nano Lett., 4, 1551-1556 (2004). 21. A. J. Storm, Single Molecule Experiments On DNA With Novel Silicon

Nanostructures, I O S Press, Incorporated (2004). 22. H. Chang, F. Kosari, G. Andreadakis, G. Vasmatzis, E. Basgall, A. H. King and

R. Bashir. “Towards integrated micro-machined silicon-based nanopores for characterization of DNA”; Hilton Head MEMS conference, Hilton Head, South Carolina, (2004).

23. H. Chang, B. M. Venkatesan, S. M. Iqbal, G. Andreadakis, F. Kosari, G. Vasmatzis, D. Peroulis and R. Bashir, Biomedical Microdevices, 8, 263-269 (2006).

24. P. Chen, T. Mitsui, D. B. Farmer, J. Golovchenko, R. G. Gordon and D. Branton, Nano Lett., 4, 1333-1338 (2004).

25. P. Chen, J. Gu, E. Brandin, Y. R. Kim, Q. Wang and D. Branton, NANOLETTERS, 4, 2293-2298 (2004).

26. S. G. Lemay, D. M. vandenBroek, A. J. Storm, D. Krapf, R. M. M. Smeets, H. A. Heering and C. Dekker, Anal. Chem., 77, 1911-1915 (2005).

27. A. J. Storm, J. H. Chen, X. S. Ling, H. W. Zandbergen and C. Dekker, J. Appl. Phys., 98, 014307 (2005).

28. M. Y. Wu, D. Krapf, M. Zandbergen, H. Zandbergen and P. E. Batson, Appl.Phys. Lett., 87, 113106 (2005).

29. E. A. Heins, Z. S. Siwy, L. A. Baker and C. R. Martin, Nano Lett., 5, 1824-1829 (2005).

30. C. Ho, R. Qiao, J. B. Heng, A. Chatterjee, R. J. Timp, N. R. Aluru and G. Timp, Proceedings of the National Academy of Sciences, 102, 10445 (2005).

31. P. Chen, J. Gu, E. Brandin, Y. R. Kim, Q. Wang and D. Branton, Nano Lett., 4,2293-2298 (2004).

32. D. Fologea, J. Uplinger, B. Thomas, D. S. McNabb and J. L. Li, Nano Lett., 5,1734-1737 (2005).

33. D. Fologea, M. Gershow, B. Ledden, D. S. McNabb, J. A. Golovchenko and J. Li, Nano Lett., 5, 1905-1909 (2005).

34. J. B. Heng, A. Aksimentiev, C. Ho, P. Marks, Y. V. Grinkova, S. Sligar, K. Schulten and G. Timp, Nano Lett., 5, 1883–1888 (2005).

35. E. A. Heins, L. A. Baker, Z. S. Siwy, M. O. Mota and C. R. Martin, Journal of Physical Chemistry B, 109, 18400 (2005).

36. U. F. Keyser, D. Krapf, B. N. Koeleman, R. M. M. Smeets, N. H. Dekker and C. Dekker, Nano Lett., 5, 2253-2256 (2005).

37. J. B. Heng, A. Aksimentiev, C. Ho, P. Marks, Y. V. Grinkova, S. Sligar, K. Schulten and G. Timp, Biophysical Journal, 90, 1098-1106 (2006).

38. A. Aksimentiev, J. B. Heng, G. Timp and K. Schulten, Biophysical Journal, 87,2086-2097 (2004).



39. J. B. Heng, C. Ho, T. Kim, R. Timp, A. Aksimentiev, Y. V. Grinkova, S. Sligar, K. Schulten and G. Timp, Biophysical Journal, 87, 2905-2911 (2004).

40. R. M. M. Smeets, U. F. Keyser, D. Krapf, M. Y. Wu, N. H. Dekker and C. Dekker, Nano Lett., 6, 89-95 (2006).

41. D. Krapf, M. Y. Wu, R. M. M. Smeets, H. W. Zandbergen, C. Dekker and S. G. Lemay, Nano Lett., 6, 105-109 (2006).

42. U. F. Keyser, B. N. Koeleman, S. van Dorp, D. Krapf, R. M. M. Smeets, S. G. Lemay, N. H. Dekker and C. Dekker, Nat Phys, 2, 473-477 (2006).

43. U. F. Keyser, J. van der Does, C. Dekker and N. H. Dekker, Review of Scientific Instruments, 77, 105105 (2006).

44. D. Krapf, B. M. Quinn, M. Y. Wu, H. W. Zandbergen, C. Dekker and S. G. Lemay, Nano Lett., 6, 2531-2535 (2006).

45. H. Chang, S. M. Iqbal, E. A. Stach, A. H. King, N. J. Zaluzec and R. Bashir, Appl.Phys. Lett., 88, 103109 (2006).

46. G. S. Manning, Q Rev Biophys, 11, 179-246 (1978). 47. J. D. Cross, E. A. Strychalski and H. G. Craighead, Journal of Applied Physics,

102, - (2007). 48. S. M. Iqbal, D. Akin and R. Bashir, Nat Nano, 2, 243-248 (2007). 49. Z. Siwy, L. Troffin, P. Kohli, L. A. Baker, C. Trautmann and C. R. Martin, J. Am.

Chem. Soc, 127, 5000-5001 (2005). 50. Jeffrey D. Uram, K. K. Alan , J. Hunt and M. Mayer, Small, 2, 967-972 (2006). 51. Y. Liu and S. M. Iqbal. “A Mesoscale Model for Molecular Interaction in

Functionalized Nanopores”; ASME International Mechanical Engineering Congress 2008, Boston, MA, (2008).

52. R. E. Gyurcsányi, Trends in Analytical Chemistry, (2008). 53. T. B. Jones, Electromechanics of particles, Cambridge University Press, New

York (1995). 54. H. A. Pohl, Dielectrophoresis, Cambridge University Press, Cambridge (1978). 55. A. Ramos, H. Morgan, N. G. Green and A. Castellanos, Journal of Physics D-

Applied Physics, 31, 2338-2353 (1998). 56. Y. L. Liu, J. H. Chung, W. K. Liu and R. S. Ruoff, Journal of Physical Chemistry

B, 110, 14098-14106 (2006). 57. W. K. Liu, Y. L. Liu, D. Farrell, L. Zhang, X. S. Wang, Y. Fukui, N. Patankar, Y.

J. Zhang, C. Bajaj, J. Lee, J. H. Hong, X. Y. Chen and H. Y. Hsu, ComputerMethods in Applied Mechanics and Engineering, 195, 1722-1749 (2006).

58. Y. Liu, W. K. Liu, T. Belytschko, N. Patankar, A. C. To, A. Kopacz and J. H. Chung, International Journal for Numerical Methods in Engineering, 71, 379-405 (2007).

59. Y. L. Liu, K. Oh, J. G. Bai, C. L. Chang, W. Yeo, J. H. Chung, K. H. Lee and W. K. Liu, Computer Methods in Applied Mechanics and Engineering, 197, 2156-2172 (2008).

60. L. Zhang, A. Gerstenberger, X. D. Wang and W. K. Liu, Computer Methods in Applied Mechanics and Engineering, 193, 2051-2067 (2004).

61. H. A. Pohl and I. Hawk, Science, 152, 647 (1966). 62. U. Zimmermann, U. Friedrich, H. Mussauer, P. Gessner, K. Hamel and V.

Sukhoruhov, Ieee Transactions on Plasma Science, 28, 72-82 (2000). 63. H. B. Li, Y. N. Zheng, D. Akin and R. Bashir, Journal of Microelectromechanical

Systems, 14, 103-112 (2005).



64. R. Gomez-Sjoberg, D. T. Morisette and R. Bashir, Journal of Microelectromechanical Systems, 14, 829-838 (2005).

65. P. P. Banada, Y. S. Liu, L. Yang, R. Bashir and A. K. Bhunia, InternationalJournal of Food Microbiology, 111, 12-20 (2006).

66. J. Fritz, M. K. Baller, H. P. Lang, H. Rothuizen, P. Vettiger, E. Meyer, H. J. Guntherodt, C. Gerber and J. K. Gimzewski, Science, 288, 316-318 (2000).

67. G. H. Wu, R. H. Datar, K. M. Hansen, T. Thundat, R. J. Cote and A. Majumdar, Nature Biotechnology, 19, 856-860 (2001).

68. G. Meyer and N. M. Amer, Applied Physics Letters, 53, 1045-1047 (1988). 69. Z. J. Davis, G. Abadal, O. Kuhn, O. Hansen, F. Grey and A. Boisen, Journal of

Vacuum Science & Technology B, 18, 612-616 (2000). 70. S. Watanabe and T. Fujii, Review of Scientific Instruments, 67, 3898-3903 (1996). 71. C. Ziegler, Analytical and Bioanalytical Chemistry, 379, 946-959 (2004). 72. N. V. Lavrik, M. J. Sepaniak and P. G. Datskos, Review of Scientific Instruments,

75, 2229-2253 (2004). 73. H. J. Butt, Journal of Colloid and Interface Science, 180, 251-260 (1996). 74. R. Bashir, J. Z. Hilt, O. Elibol, A. Gupta and N. A. Peppas, Applied Physics

Letters, 81, 3091-3093 (2002). 75. S. R. Manalis, E. B. Cooper, P. F. Indermuhle, P. Kernen, P. Wagner, D. G.

Hafeman, S. C. Minne and C. F. Quate, Applied Physics Letters, 76, 1072-1074 (2000).

76. A. Gupta, D. Akin and R. Bashir, Applied Physics Letters, 84, 1976-1978 (2004). 77. A. K. Gupta, P. R. Nair, D. Akin, M. R. Ladisch, S. Broyles, M. A. Alam and R.

Bashir, Proceedings of the National Academy of Sciences of the United States of America, 103, 13362-13367 (2006).



silicon-based novel bio-sensing platforms at the...

Documents