最近更新: 2015年1月21日

 

 


   Our recent research work mainly focuses on combined molecular engineering and advanced nano-lithographic techniques for fabrication of functional devices and detectors. Particularly, attempts have been made on manipulation and detection of individual DNA molecules. In that project, we have rudimentary, but substantial progresses, and there are on-going research activities along this line. In addition, we continue to work on topics that are related to the transport properties of nano-electronics, including nanowire devices and lithographically made silicon devices. In the following, we present high lights of our recent research works.

 

 

Coherent coupling between cavity photon states and 1 and 2 transmon qubits

Quantum computing is a new rising research subject in scientific areas, in which the quantum bit (qubit) is the most basic component. Different from the situation of a classical bit, the information (0 and 1) of a qubit is simultaneously stored in the form of a superposition state. They are together operated through unitary transformations []. This improves tremendously the information storage capability and computing efficiency in novel way. Therefore, in addition to investigating qubit manufacturing, qubit control and measurement becomes the most important issue in quantum computing. For the qubit control and measurement, the cavity quantum electrodynamics (cavity QED) provides a well-developed theoretical framework, which mainly deals with the behavior of the interaction between an atom and an optical cavity. Such an interaction, which is described by the Jaynes-Cummings model, can provide atomic state control mechanism by photon induced Rabi oscillation. Through it, the information about atom state can also be informed to the oscillatory behavior of the cavity, realizing the quantum nondemolition measurement. The superconducting qubit, which is built with Josephson junctions, can condense to a macroscopic quantum state in very low temperature, realizing an artificial two-level atom. For a superconducting qubit, the circuit QED is a low frequency analogue of cavity QED that can be realized by microwave circuit. In addition to qubit state control and measurement of the circuit QED, the microwave cavity can exclude the non-resonant mode in it, greatly suppressing the coupling of the qubit to the modes of free space continuous energy levels and thus reducing the relaxation and decoherence of the macroscopic quantum state of the qubit. Conventionally, half-wavelength transmission line (TL) resonator made by the coplanar waveguides is used as the Fabry-Perot cavity for circuit QED. The cavity field is coupled to the outside transmission line circuit through a capacitor at each end of the cavity. The two capacitors also provide two reflecting boundaries for the Fabry-Perot resonance.
Transmon qubits were incorporated into a superconducting coplanar waveguide structure for studying the atom-cavity coupling. The devices are made on sapphire substrates and the transmons were placed at the anti-node positions of the cavity mode. The zero-field plasma frequencies of the qubits are about 14GHz for 1-qubit and are about 9.2 and 10.1GHz for 2 qubits, and the anharmonicity, which is the charging energy of the transmons, is about 400MHz. In the far-detuned region, resonate frequencies for 1-qubit and 2-qubit cases were 5.536GHz and 5.945GHz, respectively. Right at zero-detuned, the coupling strengths was about 160~165MHz.

 


Fig. 1. Coplanar waveguide design and a transmon made at the right-end of the cavity. The waveguide is made of Nb and the transmon is made of Al. The meandering line cavity is coupled to the RF input and output signal lines through interdigital capacitors. The zero-field plasma frequency of the qubit is about 14GHz whereas the anharmonicity, which is the charging energy of the transmon, is about 400MHz.

 


Fig. 2. Transmission spectrum of the cavity coupled to single-transmon qubit. The plot on the right-hand side is a blowup view of the left plot. Green line is the cavity frequency, which in the far-detuned region resonate frequency is 5.536GHz. The two white lines are transmon plasma frequency, which can be tuned by the flux treading the transmon loop. The red curves show Rabi-splitting at small-detuned region; from the anti-crossing curves we can estimate the coupling strength to be 165MHz.

 


Fig. 3. A coplanar waveguide coupled two transmons made at the two ends of the cavity. The zero-field plasma frequencies of the two qubits are 9.2 and 10.1GHz and the charging energy is about 400MHz.

 


Fig. 4. Calculated and meausred cavity spectrums for two transmons. The zero-field transmon plasma frequencies of the two qubits are directly measured from the 2-tone measurement spectrum (bottom left panel). With these parameters and measured periods, the flux-dependent plasma frequency (red and green) curves are calculated (top left panel). The two-qubit entanglement can be seen from the crossing between the green and red curves (two right panels). Due to small quality factor of the cavity, the entanglement is barely seen.

 


Fig. 5. Calculated (top) and measured (bottom) transmission spectrums of the cavity and two-qubit system. The bottom plot shows only a narrow frequency span of the measured spectrum, and the top-inset shows a zoom-in plot of the cavity-transmon avoiding curves.

 

TOP

Resonant tunneling through discrete quantum states in stacked atomic-layered MoS2, Nano Letters (2014)

Two-dimensional materials have shown fascinating physical phenomena (Ponomarenko, L. A. et al. Nature phys. 2011; Gorbachev, R. V. et al. Nature phys. 2012) and have been used in prototype field-effect tunneling transistors (Britnell, L et al. Science 2012) as well as flexible and transparent vertical-field-effect-transistors (Artem Mishchenko et al. Nature nanotech 2013). MoS2 is a representative of two-dimensional materials that has recently attracted much attention. Few-layered MoS2 has been proven to possess properties different from those of bulk, including wide band gap and high mobility. It means that the interlayer coupling in few-layered MoS2 is crucial for development of a new type of stack electronics.
Here we report interlayer electron transport through stacked high-quality crystalline MoS2. The current-voltage characteristics exhibit a resonant peak structure, suggesting presence of discrete energy levels in the MoS2 sheets brought about by the electron wavefunction quantization in the atomic layers. More importantly, the resonant peak position shifts linearly with perpendicular magnetic fields, implying formation of Landau levels. From this linear dependence, the Fermi wavelength as well as effective mass of MoS2 sheets are obtained. These values are further confirmed by first-principal calculations, which suggests the importance role of interlayer coupling on the characters of conduction electrons. We believe such information is of significant importance for the development of few-layered MoS2 electronics.

 


Figure 1 (a) Optical microscopy image of a Si3N4 membrane chip coated with transferred trilayer MoS2. The inset is an SEM image of Device 1 showing a nanopore made on Si3N4 membrane. (b) Raman spectra for the transferred trilayer MoS2 on the membrane. (c) Schematic illustration of the nanopore device with the trilayer MoS2 sandwiched between the top and bottom electrodes. (d) A cross-sectional drawing of the trilayer MoS2 device structure at the nanopore area.

 

Table 1 The parameters summarized for three devices.

Device
#

Pore radius
(nm)

Onset magnetic field
(T)

Effective mass

Cyclotron frequency
(rad/s)

Fermi velocity
(m/s)

1

14±1

3.5±0.175

(0.047±0.002)me

(1.31±0.09)×1013

(0.183±0.018)×106

2

22±2

1.6±0.08

(0.021±0.0005)me

(1.34±0.07)×1013

(0.295±0.03)×106

3

36±3

0.5±0.025

(0.015±0.0003)me

(0.75±0.04)×1013

(0.27±0.027)×106

 


Figure 2 (a) I-Vb characteristics of Device 1 in magnetic field ranging from 0 to 9T with a step of 3T. Inset shows the same curves with a step of 1T. Each curve is shifted up by 0.2nA for clarity. (b-d) Resonant peak position as a function of magnetic field with a linear fit for Devices 1(b), 2(c) and 3(d). (e) Onset magnetic field plotted for three pore-radius, displaying a linear dependence of Bonset on 1/r. 

 


Improving Nanowire Sensing Capability by Electrical Field Alignment of Surface Probing Molecules, Nano Letters (2013)

We argue that the structure ordering of self-assembled probing molecular monolayers is essential for the reliability and sensitivity of nanowire-based field-effect sensors because it can promote the efficiency for molecular interactions as well as strengthen the molecular dipole field experienced by the nanowires. In the case of monolayers, we showed that structure ordering could be improved by means of electrical field alignment.  This technique was then employed to align multilayer complexes for nanowire sensing applications. The sensitivity we achieved for detection of hybridization between 15-base single strand DNA molecules is 0.1fM and for alcohol sensors is 0.5ppm. The reliability was confirmed by repeated tests on chips that contain multiple nanowire sensors.

 


Electric field alignment of APTES on silicon nanowire (SiNW) FETs. (a) SEM image of a SiNW-FET. The width, length and thickness of the wire are 200nm, 3mm and 40 nm, respectively, and the wire is covered by a 10 nm thick thermally oxidized layer. The scale bar is 2mm. (b) and (c) Schematic illustrations of the possible APTES molecular structures before and after alignment process.

 

Real-time detection of DNA hybridization. In (a) the target is 1pM 15A ssDNA in Tris solution, while in (c) the target is 0.1fM 15A ssDNA in PBS solution. Upon introduction of poly-15A ssDNA, a clear increase in the nanowire current was observed. In both (a) and (c) buffer was first added to ensure stability of the sensor. (b) and (d) show control experiments for (a) and (c), respectively. The bias conditions were Vds = +1.5V, Vg = -1.0V in (a) and (b) and were Vds = +1.6V, Vg = -2.5V in (c) and (d).

 

TOP

Polymer-free Patterning of Graphene at Sub-10nm Scale by Low-Energy Repetitive Electron Beam, Small (2014)

Graphene possesses extraordinary properties, including high carrier mobility, high transparency, high thermal conductivity, high flexibility and strong mechanical strength. Because of that, graphene has emerged as a material of interest for a wide variety of practical electronic applications. However, fabrication of graphene electronic devices involves various lithographic processes, which would inevitably introduce unwanted contaminations. These contaminations may dramatically degrade the extraordinary characteristics of graphene. To take full advantages of the unique properties of graphene for future applications, development of a clean patterning technique is highly desirable.
   In this work, we show that nano-patterns on both suspended and unsuspended graphene monolayers can actually be generated. Several important merits are listed below:

a. To resolve the contamination issues, we proposed and demonstrated a simple technique for
generation of graphene patterns on suspended and unsuspended graphene sheets. 

 


A polymer-free, low-energy (5-30keV) direct electron-beam patterning technique on both suspended and unsuspended graphene monolayers is demonstrated. The patterning mechanisms involve defect-induced knockout and heat-induced curling processes. Both exfoliated and chemical vapor deposited graphene layers are tested, and the minimum feature sizes achieved are 5nm and 7nm for the suspended and unsuspended graphene sheets, respectively.

 

b. In this approach, low-energy (5-30keV) repetitive electron beam bombardment is utilized to remove the unwanted graphene regions. We also discussed possible mechanisms that can account for the low threshold etching energy.
c. Using this technique, graphene patterns of various shapes can be made and the minimum feature sizes achieved are 5nm for suspended and 7nm for unsuspended graphene case.
d. In addition to the polymer-free patterning technique, we also demonstrate a polymer-free graphene transfer from copper foils that is used in CVD-growth process.
e. With this technique, the graphene devices should find many uses in electronic, optoelectronic, energy and biosensing applications.

The technique is applicable not only in research laboratories but also in semiconductor industry.

 

 

High performance phototransistors based on single crystalline PTCDA nanoparticle, Applied Physics Letters (2013)


In this work, we evaluate the mobility of individual organic semiconductor nanoparticles. The PTCDA nanoparticles are embedded in a nanopore made on a silicon nitride membrane and surrounded by a gate electrode. Thanks to the absence of grain boundary scattering, we obtained a room temperature electron mobility of 0.16cm2/Vs, which is 1~2 orders in magnitude greater than the reported values for single crystal films and is 3~4 orders greater than those of polycrystalline films. Further, because of the high mobility, the device exhibits external quantum efficiency (EQE) of 3.5×106, which is the highest reported value thus far for opto-electronic devices made of organic single crystals. The very high EQE value is attributed to the absence of deleterious effects arising from defects and grain-boundary recombination.
many uses in electronic and opto-electronic applications. More importantly, we think that the electron mobility obtained in this device is approaching the intrinsic mobility of PTCDA.

 

(a) A cross-sectional schematic illustration of the nanopore device with a gated single PTCDA nanoparticle that is sandwiched between AlxO/Al source and drain electrodes.
(b) Temperature dependence of electron mobility for devices A and B. Tcr, at which the maximum mobility appears, is around 80K for device A and around 160K for device B.
(c) An Arrhenius plot showing temperature dependence of device conductance for devices A and B. Lines of best fit (black lines) yield activation energies of 27meV and 124meV for A and B, respectively.

 

 

Focused ion beam deposition induced contamination on nanoelectronics, Nanotechnology (2015)


Focused ion beam (FIB) system has been widely used in local deposition of electrical contacts for nanowire and nanotube devices. However, due to FIBs having a Gaussian distribution, local metal contamination occurs. The side effect of deposition-induced contamination on the electrical transport of these nano-devices is important for future applications.
     Recently, a research team from Institute of Physics, Academia Sinica in Taiwan, demonstrated an experimental platform which allows correlated studies of physical structure in transmission electron microscopy and electrical characterization. Carbon nanotube is a model system and is used for this purpose. The selected single-wall CNT bundle was cut to a length of between 5~7 mm using focused electron beam irradiation, and placed atop freestanding parallel Au electrodes using a scanning electron microscope (SEM) based probe station. The resultant device is a through-hole chip comprising Au electrodes with a suspended CNT and FIB deposited contacts (as shown in left Figure).
      They found that FIB deposition of Pt on the contact regions can significantly reduce the contact resistances, but transmission electron microscope inspections reveal serious Pt contaminations near the deposition areas and on carbon nanotube. These contaminations were suspected to be the origin of the observed unstable Coulomb oscillation behavior (as shown in right Figure) in electrical measurement. This finding addresses an important issue about the future application of FIB technology on nanoelectronics

 


Left: SEM image of carbon nanotube device. The orange dots on the CNT are FIB deposition points. Enlarged portion shows nanoparticles covered with the CNT bundle after FIB deposition. Right: The observed unstable Coulomb oscillation behavior.

 


Spin-Orbit Interaction in a Single-Walled Carbon Nanotube Probed by Kondo Resonance, Carbon (2012)

 

The shifts of peaks in the presence of magnetic field parallel to the tube. (a)Schematic representation of four single-electron transitions, including two spin-flip intraorbital transitions (?,?) and two spin-conserved interorbital transitions (?,?). (b) Differential conductance as a function of bias voltage at different values of the magnetic field from 5T (top curve) to 0 (bottom curve). Successive curves are offset by 0.5 mS for clarity and the blue, red and green lines are guide to the eye. (c) Peak positions in bias voltage, extracted from (b), as a function of Bll (black dots). A linear fit of data in blue-solid line gives an average g=1.91±0.05 in Zeeman shift and 2Dso = 0.75meV. A linear fit of data in red-solid line evolves with a slope 2morb/e of 170±5meV/T. The green-solid lines are fit to spin-Zeeman with a slope gmBB. The blue, red and green dashed lines indicate the positions of other predicted high-order transitions. Because of low transition rates, these peaks are not observed in this study.

 

  In this work, we present an unprecedented experimental work that involves coexistence of three independent phenomena, namely (i) Kondo resonance, (ii) spin-orbit Interaction and (iii) spin degenerate levels in quantum dot. We report, for the first time, meticulous interplay of these three intricate concepts and employ the Kondo resonance to probe the spin-orbit interaction in a single-walled carbon nanotube quantum-dot operating inside the Coulomb blockade regime.
Recent theoretical studies of spin-orbit interactions in carbon nanotube quantum-dot [Phys.Rev.Lett 101, 246805(2008)] predicted three conductance peaks in the Kondo regime and our experimental data confirms the above prediction and could act as a benchmark for more extensive and elaborate theoretical and experimental efforts in this direction. In this area, there is an excellent experimental report [Nature 452, 448 (2008).] demonstrating the coupling of spin and orbital motion of electrons in partially suspended carbon nanotubes. In their measurement scheme that operates at high voltages outside the Coulomb blockade diamond, meticulous care was employed to rule out the possibility of electron-electron interactions, which could otherwise give rise to similar splitting in N-electron ground state. The interesting contribution of the present work to the existing standpoint is that we propose an innovative low-excitation tool to probe the spin orbit interactions in carbon nanotubes, operating well within the Coulomb blockade diamond. The single particle ground state wave function in this case does not call for the electron-electron interaction effects. Most importantly, the spin-orbit interaction is identified from two independent observations, namely, the bias voltage offset in zero magnetic field and the magnetic field offset at zero bias voltage. These consistent observations give unambiguous evidence of the spin-orbit coupling which forms the backbone of all our claims.
    The spin-orbit coupling originating from curvature effects deserves due attention and could play a critical role in the development of next generation carbon based spin electronics and the quantitative confirmation of spin-orbit coupling energy is a significant leap in that direction.

 

TOP

Effects of oxygen bonding on defective semiconducting and metallic single-walled carbon nanotube bundles, Carbon (2012)


       As a continuation of our previous works on 1D nanostructures of various materials, we wish to incorporate electric measurement with TEM inspection on the identical specimens. In this way, the TEM inspection would provide a solid base for analysis and interpretation of the transport data. To do this, the bundles of single-walled carbon nanotubes were placed on top of either thin Si3N4 membrane or completely suspended electrodes. After subjected to electron-beam bombardments in TEM, carbon atoms on the surface of the bundles were knocked out and the surface became rugged with voids. The dangling carbon bonds of the vacancies are very active and can easily adsorb oxygen molecules. In terms of the semiconducting bundles, oxygen bonding lowers the bandgap and the original p-type bundles thereby modifying them to become bi-polar. For the metallic bundles, a hysteretic bi-stable state in gate-voltage cycling is observed; this is attributed to the electrically controlled dipole field of the oxygen molecules.

 

(left) SEM image of the device. The black circular area is the through hole, and outside it is the 75nm-thick Si3N4 membrane. The bottom-right inset is a tilted view of the same device. (right) TEM images of the suspended CNT bundle. The top is a low-magnification TEM image. The main panels are high-magnification TEM images of the CNT bundle in the left, center and right regions. The enlarged red and blue rectangles show areas with missing carbon atoms knocked out by electron irradiation. The enlarged yellow rectangle shows CNT outer shell with carbon dangling bonds, leading to rugged surface.

 

(left) Bi-polar characteristics of the defective s-CNT bundle -- oxygen bonding lowers the bandgap and the original p-type bundles thereby modifying them to become bi-polar. IVg curves at different bias voltages from 3V (top) to -3V (bottom); the curves are acquired with a step of 0.6V. (right) Electrical property of O2*-adsorbed defective m-CNT bundle. A hysteretic bi-stable state in gate-voltage cycling developed, which is not observed for the bundle in N2 and He ambience. The O2, He and N2 pressure are 600 torr.

 

 

Scaling Analysis of Magnetic Field Tuned Phase Transitions in One-Dimensional Josephson Junction Arrays
( Physical Review Letters , 87, 186804, (2001))

We have studied experimentally the magnetic field induced superconductor-insulator quantum phase transition in one-dimensional arrays of small Josephson junctions. It is found that the critical magnetic field that separates the two phases corresponds to the onset of Coulomb blockade of Cooper pairs tunneling in the current-voltage characteristics. The resistance data are analyzed in the context of the superfluid-insulator transition in one dimension, and a finite temperature scaling analysis is performed to extract the critical exponents. The dynamical exponents z are determined to be close to 1, and the correlation length exponents n are found to be approximately 0.3 and 0.45 in the two groups of measured samples.

The SEM image of an 1D JJ array. The overlappingareas between “T”-shaped islands are the tunnel junctions. The scale bar at the bottom of the image is 1 m m.

The R 0 ( T ) as a function of the filling number f. At T > T cr , the array displays an f -tuned SI transition, whereas at low temperatures the R 0 ( T ) curves in the S side level off or rise.

 

 

Suppression of superconductivity by spin imbalance in ferromagnetic-
superconductor-ferromagnetic single electron transistors
( Physical Review Letters , 88, 047004 (2002))

We present here an experimental demonstration for suppression of superconductivity by spin imbalance. This effect is manifested under spin-polarized quasiparticle current injection in ferromagnet-superconductor-ferromagnet (Co/Al/Co) single electron transistors. The measured superconducting gap as a function of magnetic field reveals a dramatic decrease when the magnetizations of the two leads are misaligned. The effect of suppression increases with increasing source-drain voltage. A comparison with numerical calculations for single electron transistor in sequential tunneling regime is performed. The imbalance of spins is a nonequilibrium process. For this process to be effective, a sufficiently long spin relaxation time and a short energy relaxation time are required. Various factors that may affect this process are considered. This method may render it applicable to control superconductivity at low temperatures within low fields.

AFM and (b) magnetic force microscope (MFM) images of a sample. Because of a two-angle evaporation technique used for fabrication of the samples, there are redundant, electrically unconnected structures aside the measured device.
The inset in (a) illustrates the cross section of the island and junctions.

 

 

Single-Electron Transistors and Memory Cells with Au Colloidal Islands
( Applied Physics Letters , 81, 4595 (2002)
)

In this study, single-electron transistors and memory cells with Au colloidal islands linked by C 60 derivatives have been fabricated by hybridization of top-down advanced electron-beam lithography and bottom-up nanophased-material synthesis techniques. Low-temperature transport measurements exhibit clear Coulomb-blockade-type current-voltage characteristics and hysteretic-type gate-modulated current. The hysteresis is attributed to the presence of electrically isolated charge-storage islands. With the guidance provided by Monte Carlo simulation, we propose a circuit model and give an estimate of the sample parameters

An SEM image of the measured device; the gate electrode is not shown. The inset shows the suspended Au leads before attachment of Au particles; the scale bar is 150 nm. A three-dimensional plot of I sd vs V g and V b . The curves are measured at T =4.2 K. Note that the modulation is irregular, but the curves are reproducible, and a Coulomb-blockade region is clearly seen.

 

 

Fabrication of Two-Dimensional Arrays of CdSe Pillars Using E-Beam Lithography and Electrochemical Deposition
( Advanced Materials , 15, 49 (2003))

Two-dimensional arrays of high refractive index structures can be fabricated using a combination of e-beam lithography for pattern definition and electrochemical deposition for structure formation. The potential of this method is demonstrated for CdSe, where nanopillars, walls, and crosses are prepared. Such arrays have potential in optical device applications such as photonic crystals and waveguides.

SEM image of (a) mushroom-like pillars and (b) a continuous CdSe film formed on top of the pillars)

SEM images of (a) an array of CdSe walls, where the length of the walls is about 20 m m and cross-shaped walls, where the length is 2 m m and the height is about 400nm

Procedures for e-beam lithography. (a) spin-coating of bilayer e-beam resist (b) Exposure to focused Gaussian shaped electron beam (c) Development of the exposed pattern (d) oxygen reactive ion bean cleaning (e) Chemical electrodeposition of CdSe (f) Resist-removal (lift-off process) in acetone bath.

 

 

Controlled Placement and Electrical Contact Properties of Individual Multi-walled Carbon Nanotubes on Patterned Silicon Wafer
( Applied Physics Letters , 84, 984 (2004))

A scheme that allows on-chip growth of multiwalled carbon nanotubes at designed locations is demonstrated. The nanotubes were grown by thermal chemical vapor deposition and were contacted to nanoscaled Cr electrodes fabricated by standard e-beam lithography techniques. The contacts were found to be Ohmic with resistance values on the order of 10 3 V at room temperature. Remarkably, the contacts showed weak temperature dependence down to 40 mK and were insensitive to the magnetic field up to 5 T.

An SEM image of one of the measured MWNTs. The tube was grown from an iron catalyst ~not shown in this image! and was connected to nine Cr electrodes.

The asymptote resistance at 70 mK as a function of the perpendicularly applied field. Both two-probe and four-probe magnetoresistance decreases with increasing field. A comparison between the two curves indicated that the contact resistances are unaffected by the applied fields. The inset shows four-probe magnetoresistance of a separate tube.

 

 

Positioning of extended individual DNA molecules on electrodes by non-uniform
AC electric fields
( Nanotechnology 16 , 2738–2742 (2005) )

Recent developments in the analysis and the application of DNA often require the stretching of individual DNA molecules to specific surfaces. We propose and demonstrate a method for the positioning of unmodified extended DNA molecules. A local microscopic circular flow is created by a non-uniform AC field and utilized to stretch the λ -DNA on a gold surface or between gold electrodes. The electrical-field amplitude and frequency responses of DNA motion are studied. The method can be applied to position the DNA with accuracy on a microscopic scale while requiring no modification of the DNA for terminal binding. With a diluted DNA solution, the number of DNA molecules across the electrodes is controllable and the positioning of a single extended DNA across electrodes is achievable.

Schematic diagram illustrating the motion of DNA in an AC field. (a) A cyclic motion results from electro-osmosis flow (EOF). EOF draws DNA near the inner edges inward, and pushes them outward to the outer edges. (b) Four successive steps, a–d, show how a DNA molecule may anchor and extend in corresponding to the cyclic motion. This circular movement is mainly developed in the x z plane.

Images showing the results of positioning the DNA across electrodes. (a) Dark field image from light microscope shows the arrangement of the electrodes. The AC electric circuit is illustrated. The width of the narrowest electrodes is 400 nm. The bright areas are gold electrodes on top of the glass substrate which appears as dark background. The scale bar is 10 m m. (b) Fluorescence image shows two DNA molecules positioned on each group of electrodes. The applied voltage is 2 Vp ? p with a frequency of 200 Hz. (c) AFM image shows the area marked with the dotted rectangle in (b). The two arrows indicate the two DNA molecules across electrodes. The scale bar is 1 m m.

 

 

DNA as an Electron-Beam Sensitive Reagent for Nano-Patterning
( Advanced Materials , 18 , 1517–1520 ( 2006 ))

In this study, we propose and demonstrate a new application using DNA as an e-beam sensitive reagent for patterning. The technique allows direct electron-beam patterning of oligonucleotides. To this end, thiolated single-strand DNA was bombarded using a focused electron beam, resulting in the inhibition of hybridization to complementary strands. The degree o f inhibition as a function of the exposure dose was studied using both fluorescence-probe and Au-nano-particle labeling. Finally, for demonstration purposes gold nano-particles were used as markers to produce nano-scaled patterns. The results of which are presented in this paper. This technique has potential applications in the fabrication of DNA-based nano-structures.

Sample preparation procedures. a) Application of HS-20T solution to a gold surface. b) Immobilization of HS-20T oligonucleotides through sulfur–gold interaction. c) Electron-beam patterning on the oligonucleotide thin film. d) Hybridization of Hex-20A with HS-20T.

G old nano- particle pattern produced by utilizing DNA molecular layer as an e-beam sensitive reagent. In this approach, thiolated single-strand DNA was bombarded using a focused electron beam, resulting in the inhibition of hybridization to complementary strands , and then gold nano-particles were used as markers to reveal nano-scaled patterns . This technique has potential applications in the fabrication of DNA-based nano-structures .

 

TOP

Generation of nano-scaled DNA patterns through electro-beam induced charge trapping
( Nanotechnology 17 , 1–5 (2006) )

In this study, distinct regions of trapped charges on glass substrates created by electron beam bombardment were utilized to attract and to immobilize DNA molecules. The negatively charged DNA molecules were attracted by the positive charge layer beneath the substrate surface resulting from escape of secondary electrons. With this mechanism, we demonstrated high-precision patterning of unmodified DNA molecules, independent of the length, sequence, and number of DNA strands, and with an attachment to the glass surface strong enough to withstand vigorous washing with water. DNA patterns with the line width of 50 nm were achieved.

(a) Energy dependence of secondary electron yield quoted from [ 8 ]. (b) Equipotential curves above the substrate surface. The zero-potential curve (in yellow) divides the potential space into two regions: an inner positive potential region (red curves) and an outer negative potential region (blue curves).

(a)–(d) SEM images of uranyl acetate stained DNA patterns. The dark lines are formed by DNA molecules at (a) 500, b) 100 and (c) 50 nm line widths. (d) The characters ‘SINICA' formed by DNA at 400 nm line width. (e) AFM image of the same sample shown in (d).

 

 

Coupled single electron transistors as a differential voltage amplifier
( New Journal of Physics , 8, 300 (2006) )

We have investigated a possible application of single electron transistor (SET) devices for use as a differential voltage amplifier. The device consists of a pair of box-SET and probe-SET coupled with each other through a tunnel junction, with the gate electrodes of the two SETs acting as differential signal inputs. The voltage across the probe-SET at a fixed bias current provides information about the charge states of both the probe-SET and the box-SET, which was confirmed by simulations based on the orthodox theory of single-electron tunneling. When operated as a differential amplifier, the output probe-SET voltage signal was measured as a function of the two gate input signals. While the output signal was found to be proportional to the difference in the two input signals, it remained unchanged for input signals of the same amplitude (referred to as the common mode signal), and the common mode rejection ratio was found to be 27.5dB.

SEM image of a measured sample. The junctions areas are marked by white parallelograms and the two control gates are located outside the image.

( left ) Intensity plot of V P as a function of gate voltages V bg and V pg measured at I p =0.43nA. Bright regions indicate high V P values. ( right ) The same plot as in ( left ) but made by simulation. The dashed line and dotted line indicate the applied gate voltages for common-mode and differential-mode input signals, respectively.

 

 

Fabrication of One-Dimensional Au-Particle Electronics with DNA-mediated Charge Trapping Technique ( Advanced Functional Materials , 17, 3182 (2007) )

We report a unique approach for producing nano-scaled one-dimensional (1D) gold-particle electronic devices. In this approach, a focused electron beam was first utilized to generate a positive charge layer on a SiO 2 surface. Biotinated DNA molecules attracted by these positive charges were subsequently used to grasp Au-nanoparticles revealing the e-beam exposure single-line patterns. Due to repulsive force between Au colloidal particles, the particles in the single-line patterns were, to a large extent, orderly separated. We further develop a simple method to bridge the particles to form conductive nanowires of high or low wire resistance. While low resistance wires showed linear current-voltage characteristics with a high maximum allowed current density, the high resistance wires exhibited charging effect with clear Coulomb oscillation behavior at low temperatures. This demonstrates that the technique is capable of producing interconnects as well as single-electron-transistors, and opens up possibilities for fabrication of integrated circuits.

(a~g) SEM images of Au-nanoparticle single-line patterns. The four writings of "DNA" in (a) were made utilizing e-beam exposure line doses of, from top-right to bottom-left, 1.0, 1.5, 2.5, and 4.0 nC/cm, respectively, and the patterns in (b~g) were made with a line dose of 2.5 n C/cm. For all patterns, the center-to-center distance between the beam pulses was set to 13.6nm.

Top: SEM images of one dimension Au particle array and nanowires: (a) before bridging cycle, (b)~(c) after one bridging cycle and (d)~(f) after two bridging cycles. (g) Au particle nanowire attached between a pair of electrodes for electrical measurements . Bottom: Gate-voltage modulations of source-drain current at ramping bias voltages. The curves were taken at 6K. The zero-current Coulomb parallelogram and Coulomb oscillations are signatures of SET.

 

 

Polymer-based photonic crystals
fabricated with single-step electron beam lithography (Advanced Materials , 19, 3052 ( 2007) )

We present a simple, versatile technique for the fabrication of quasi three-dimensional suspended polymer photonic crystals (PCs) and three-dimensional multilayer polymer photonic crystals. For quasi-3D PC slabs, photonic band gap in the visible light region was evidenced from optical microscopic observation and transmission spectrum measurement. In addition, slabs with photonic band gap in both TE-like and TM-like modes for the telecommunication wavelength region were designed and fabricated. This unprecedented fabrication method utilizes only a single-step electron beam lithography process, and thus overcomes difficulties encountered by existing 3D PC techniques. This is a step forward to the realization of multifunctional PC integrated circuits. (with supplemental material)

(left) SEM image of a suspended PMMA quasi-3D PC slab with a hexagonal array of air holes. The hole radius is 260nm, and the lattice constant is 800nm. (right) Measured transmission spectrum. The calculated gaps are 454~484nm and 491~508nm.

(left) Cross-section view of a multilayer PMMA/LOR line-array structure. The structure is made using a single exposure step and layer–by-layer development processes. The PMMA and LOR layers are 200nm- and 500nm-thick, respectively. (right) SEM image of a line defect created by adjusting the exposure dose.

 

 

Cyclotron Localization in a sub - 10nm Silicon Quantum Dot Single Electron Transistor ( Applied Physics Letters , 90, 032106 (2007) )

W e have fabricated and measured a lateral Si-SET consisting of a succession of a big island and small quantum dot s. In th is device , small Coulomb oscillation wiggles, due to the big island, acted as a scale to reveal shifts in peaks of Coulomb oscillation envelopes, due to the small quantum dots, in the presence of a magnetic field. The observe d shift s in peak position are analyzed in the context of field-induced Landau level shift in dots with a soft-wall confinement potential. Furthermore, the current peak was suppressed for fields beyond a threshold value. A n explanation based on cyclotron localization at non-interacting L andau levels of the small quantum dot s is presented.

The oscillatory current modulation in gate voltage at B=0 (black curve) and B=5T (red curve) measured at 70mK for the device shown in the top-right inset. The device contains a big island connected to leads via small dots present in the nano-constrictions. The top-left inset presents IV b curves at ramping V g clearly marking the C oulomb blockade diamond ; a closer look reveals fine wiggles arising from Coulomb oscillation in the big island.

(top) Dynamics of 5 current peaks in magnetic fie l d at a bias voltage of 1.8 mV. The curves shift with field from -5T at the bottom to +5T at the top. The current peak s at zero magnetic field are clearly suppressed as seen from peak P3. Refer to the main text for the description of the origin of the shift and suppression of the current peak. The inset presents enlarged view of P5 describing the evolution of current peak position with the applied field s of 0 (black curve), 2.5T (green curve) and 5T (red curve) . Note also the small wiggles show no shift in the field. The peak position shift in (bottom) as a function of magnetic field for the 5 peaks are extracted from the raw data shown in (top) for further analysis. Peaks 2 and 3 show clear splitting in the presence of field, and the peaks shift in opposite directions .

A comparison between (left) the theoretical eigen energy spectrum in the presence of external field and (right) the experimental current peak position data . The zero field regions have multifold degeneracy and the beginning of our comparison with the theoretical curves from B>2 T and ω c /ω o > 2 is justified.

 

 

Control and detection of organosilane polarization on nanowire field-effect-transistors ( Nano Letters , 2007)

We demonstrated control and detection of UV-induced 3-Aminopropyltriethoxysilane (APTES) polarization using silicon nanowire field-effect-transistors made by top-down lithograph technology. Electric dipole moment in APTES films induced by UV-illumination was shown to produce negative effective charges. When individual dipoles were aligned with an externally applied electric field, the collective polarization can prevail over the UV-induced charges in the wires and give rise to an abnormal resistance enhancement in n -type wires. Real-time detection of hybridization of 15-mer poly-T/poly-A DNA molecules was performed, and amount of hybridization induced charges in the silicon wire was estimated. Based on these results, detection sensitivity of the wire sensors was discussed.

(a) SEM image of the SOI-based Si-wire FET device. The width, length and thickness of the wire are 100nm, 3 m m and 30 nm, respectively, and the wire is covered by a 10-nm thick thermally oxidized SiO 2 layer. The schematic drawings illustrate (b) APTES surface modification and (c) detection of DNA-hybridization .

a-c) Schematic illustration of APTES polarization on a n -type SiNW: (a) random orientation of molecules with internal dipoles; (b) because of the internal field, a high E-field helps to align APTES; (c) UV-illumination strengthens the internal dipoles. (d) UV-responses of an n-type wire before (top panel) and after (bottom panel) subjecting to an E-field as shown in (b). In the former, the resistance decreases under UV-illumination, as expected. In the latter molecular dipole field prevails, and the device shows a reversed response.

easured nanowire resistance in response to the injection of ssDNA molecules. Prior to the measurement the wires were modified with 15T ssDNA. (a) 15C ssDNA (cyan shade) and buffer (gray shade) solution were added at t=150 and 650 sec, respectively, and the wire showed no change in resistance. At t=800 sec, 15A ssDNA (yellow shade) was added and an abrupt increase in the resistance was observed. The two curves are for two SiNWs measured simultaneously. (b) The resistance of two p-type wires measured simultaneously as a function of gate voltage before addition of 15A ssDNA (dotted curve) and after hybridization (solid curve).

 

 

TOP

 

中央研究院物理研究所 台北市南港區研究院路二段128號
電話:02-2789-8955 及 02-2789-8323,02-2789-8960 (無塵室) 傳真:02-2783-4187
Institute of Physics, Academia Sinica Nankang, Taipei, Taiwan 11529 R.O.C.
(Lab) 886-2-2789-8955 and (Clean room) 886-2-2789-8323 , 886-2-2789-8960   Fax: 886-2-2783-4187

Academia Sinica