arXiv subjects

E. D. Williams

Publications and source records attributed to E. D. Williams.

At least 19 recordsLinked to original sources

arXiv2011

Tunable Kondo Effect in Graphene with Defects

Graphene is a model system for the study of electrons confined to a strictly two-dimensional layer1 and a large number of electronic phenomena have been demonstrated in graphene, from the fractional2, 3 quantum Hall effect to superconductivity4. However, the coupling of conduction electrons to local magnetic moments5, 6, a central problem of condensed matter physics, has not been realized in graphene, and, given carbon's lack of d or f electrons, magnetism in graphene would seem unlikely. Nonetheless, magnetism in graphitic carbon in the absence of transition-metal elements has been reported7-10, with explanations ranging from lattice defects11 to edge structures12, 13 to negative curvature regions of the graphene sheet14. Recent experiments suggest that correlated defects in highly-ordered pyrolytic graphite (HOPG) induced by proton irradiation9 or native to grain boundaries7, can give rise to ferromagnetism. Here we show that point defects (vacancies) in graphene15 are local moments which interact strongly with the conduction electrons through the Kondo effect6, 16-18 providing strong evidence that defects in graphene are indeed magnetic. The Kondo temperature TK is tunable with carrier density from 30-90 K; the high TK is a direct consequence of strong coupling of defects to conduction electrons in a Dirac material18. The results indicate that defect engineering in graphene could be used to generate and control carrier-mediated magnetism, and realize all-carbon spintronic devices. Furthermore, graphene should be an ideal system in which to probe Kondo physics in a widely tunable electron system.

cond-mat.mes-hall↗

arXiv2010

Transfer Printing Approach to All-Carbon Nanoelectronics

Transfer printing methods are used to pattern and assemble monolithic carbon nanotube (CNT) thin-film transistors on large-area transparent, flexible substrates. Airbrushed CNT thin-films with sheet resistance 1kOhmsquare^{-1} at 80% transparency were used as electrodes, and high quality chemical vapor deposition (CVD)-grown CNT networks were used as the semiconductor component. Transfer printing was used to pre-pattern and assemble thin film transistors on polyethylene terephthalate (PET) substrates which incorporated Al_{2}O_{3}/poly-methylmethacrylate (PMMA) dielectric bi-layer. CNT-based ambipolar devices exhibit field-effect mobility in range 1 - 33 cm^{2}/Vs and on/off ratio ~10^{3}, comparable to the control devices fabricated using Au as the electrode material.

cond-mat.mtrl-sci↗

arXiv2010

Optimizing Transistor Performance of Percolating Carbon Nanotube Networks

In percolating networks of mixed metallic and semiconducting CNTs, there is a tradeoff between high on-current (dense networks) and high on/off ratio (sparse networks in which the metallic CNT fraction is not percolating). Experiments on devices in a transistor configuration and Monte Carlo simulations were performed to determine the scaling behavior of device resistivity as a function of channel length (L) for CNT density p in the range 0.04 - 1.29 CNT/μm^2 in the on- and off-states. Optimized devices with field-effect mobility up to 50 cm^2/Vs at on/off ratio > 10^3 were obtained at W = 50 μm, L > 70 μm for p = 0.54 - 0.81 CNTs/μm^2.

cond-mat.mtrl-sci↗

arXiv2010

Controlled Growth, Patterning and Placement of Carbon Nanotube Thin Films

Controlled growth, patterning and placement of carbon nanotube (CNT) thin films for electronic applications are demonstrated. The density of CNT films is controlled by optimizing the feed gas composition as well as the concentration of growth catalyst in a chemical vapor deposition process. Densities of CNTs ranging from 0.02 CNTs/μm^2 to 1.29 CNTs/μm^2 are obtained. The resulting pristine CNT thin films are then successfully patterned using either pre-growth or post-growth techniques. By developing a layered photoresist process that is compatible with ferric nitrate catalyst, significant improvements over popular pre-growth patterning methods are obtained. Limitations of traditional post-growth patterning methods are circumvented by selective transfer printing of CNTs with either thermoplastic or metallic stamps. Resulting as-grown patterns of CNT thin films have edge roughness (< 1 μm) and resolution (< 5 μm) comparable to standard photolithography. Bottom gate CNT thin film devices are fabricated with field-effect mobilities up to 20 cm^2/Vs and on/off ratios of the order of 10^3. The patterning and transfer printing methods discussed here have a potential to be generalized to include other nanomaterials in new device configurations.

cond-mat.mtrl-sci↗

arXiv2009

Facile fabrication of suspended as-grown carbon nanotube devices

A simple scalable scheme is reported for fabricating suspended carbon nanotube field effect transistors (CNT-FETs) without exposing pristine as-grown carbon nanotubes to subsequent chemical processing. Versatility and ease of the technique is demonstrated by controlling the density of suspended nanotubes and reproducing devices multiple times on the same electrode set. Suspending the carbon nanotubes results in ambipolar transport behavior with negligible hysteresis. The Hooges constant of the suspended CNT-FETs (2.6 x 10-3) is about 20 times lower than for control CNT-FETs on SiO2 (5.6 x 10-2).

cond-mat.mes-hall↗

arXiv2009

Defect scattering in graphene

Irradiation of graphene on SiO2 by 500 eV Ne and He ions creates defects that cause intervalley scattering as evident from a significant Raman D band intensity. The defect scattering gives a conductivity proportional to charge carrier density, with mobility decreasing as the inverse of the ion dose. The mobility decrease is four times larger than for a similar concentration of singly charged impurities. The minimum conductivity decreases proportional to the mobility to values lower than 4e^2/(pi*h), the minimum theoretical value for graphene free of intervalley scattering. Defected graphene shows a diverging resistivity at low temperature, indicating insulating behavior. The results are best explained by ion-induced formation of lattice defects that result in mid-gap states.

cond-mat.mtrl-sci↗

arXiv2008

Diffusive Charge Transport in Graphene on SiO2

We review our recent work on the physical mechanisms limiting the mobility of graphene on SiO2. We have used intentional addition of charged scattering impurities and systematic variation of the dielectric environment to differentiate the effects of charged impurities and short-range scatterers. The results show that charged impurities indeed lead to a conductivity linear in density in graphene, with a scattering magnitude that agrees quantitatively with theoretical estimates [1]; increased dielectric screening reduces scattering from charged impurities, but increases scattering from short-range scatterers [2]. We evaluate the effects of the corrugations (ripples) of graphene on SiO2 on transport by measuring the height-height correlation function. The results show that the corrugations cannot mimic long-range (charged impurity) scattering effects, and have too small an amplitude-to-wavelength ratio to significantly affect the observed mobility via short-range scattering [3, 4]. Temperature-dependent measurements show that longitudinal acoustic phonons in graphene produce a resistivity linear in temperature and independent of carrier density [5]; at higher temperatures, polar optical phonons of the SiO2 substrate give rise to an activated, carrier density-dependent resistivity [5]. Together the results paint a complete picture of charge carrier transport in graphene on SiO2 in the diffusive regime.

cond-mat.mtrl-sci↗

arXiv2008

Pentacene islands grown on ultra-thin SiO2

Ultra-thin oxide (UTO) films were grown on Si(111) in ultrahigh vacuum at room temperature and characterized by scanning tunneling microscopy. The ultra-thin oxide films were then used as substrates for room temperature growth of pentacene. The apparent height of the first layer is 1.57 +/- 0.05 nm, indicating standing up pentacene grains in the thin-film phase were formed. Pentacene is molecularly resolved in the second and subsequent molecular layers. The measured in-plane unit cell for the pentacene (001) plane (ab plane) is a=0.76+/-0.01 nm, b=0.59+/-0.01 nm, and gamma=87.5+/-0.4 degrees. The films are unperturbed by the UTO's short-range spatial variation in tunneling probability, and reduce its corresponding effective roughness and correlation exponent with increasing thickness. The pentacene surface morphology follows that of the UTO substrate, preserving step structure, the long range surface rms roughness of ~0.1 nm, and the structural correlation exponent of ~1.

cond-mat.mtrl-sci↗

arXiv2008

Atomic Structure of Graphene on SiO2

We employ scanning probe microscopy to reveal atomic structures and nanoscale morphology of graphene-based electronic devices (i.e. a graphene sheet supported by an insulating silicon dioxide substrate) for the first time. Atomic resolution STM images reveal the presence of a strong spatially dependent perturbation, which breaks the hexagonal lattice symmetry of the graphitic lattice. Structural corrugations of the graphene sheet partially conform to the underlying silicon oxide substrate. These effects are obscured or modified on graphene devices processed with normal lithographic methods, as they are covered with a layer of photoresist residue. We enable our experiments by a novel cleaning process to produce atomically-clean graphene sheets.

cond-mat.mtrl-sci↗

arXiv2008

Spatial First-passage Statistics of Al/Si(111)-(root3 x root3) Step Fluctuations

Spatial step edge fluctuations on a multi-component surface of Al/Si(111)-(root3 x root3) were measured via scanning tunneling microscopy over a temperature range of 720K-1070K, for step lengths of L = 65-160 nm. Even though the time scale of fluctuations of steps on this surface varies by orders of magnitude over the indicated temperature ranges, measured first-passage spatial persistence and survival probabilities are temperature independent. The power law functional form for spatial persistence probabilities is confirmed and the symmetric spatial persistence exponent is measured to be theta = 0.498 +/- 0.062 in agreement with the theoretical prediction theta = 1/2. The survival probability is found to scale directly with y/L, where y is the distance along the step edge. The form of the survival probabilities agree quantitatively with the theoretical prediction, which yields exponential decay in the limit of small y/L. The decay constant is found experimentally to be ys/L= 0.076 +/- 0.033 for y/L <= 0.2.

cond-mat.stat-mech↗

arXiv2008

Percolative Effects on Noise in Pentacene Transistors

The 1/f noise in pentacene thin film transistors has been measured as a function of device thickness from well above the effective conduction channel thickness to only two conducting layers. Over the entire thickness range, the spectral noise form is 1/f, and the noise parameter varies as (gate voltage)-1, confirming that the noise is due to mobility fluctuations, even in the thinnest films. Hooge's parameter varies as an inverse power-law with conductivity for all film thicknesses. The magnitude and transport characteristics of the spectral noise are well explained in terms of percolative effects arising from the grain boundary structure.

cond-mat.mtrl-sci↗

arXiv2008

Effect of impurities on pentacene island nucleation

Pentacenequinone (PnQ) impurities have been introduced into a pentacene source material in a controlled manner to quantify the relative effects of the impurity content on grain boundary structure and thin film nucleation. Atomic force microscopy (AFM) has been employed to directly characterize films grown using 0.0-7.5% PnQ by weight in the source material. Analysis of the distribution of capture zones areas of submonolayer islands as a function of impurity content shows that for large PnQ content the critical nucleus size for forming a Pn island is smaller than for low PnQ content. This result indicates a favorable energy for formation of Pn-PnQ complexes, which in turn suggests that the primary effect of PnQ on Pn mobility may arise from homogeneous distribution of PnQ defects.

cond-mat.mtrl-sci↗

arXiv2008

Tuning the effective fine structure constant in graphene: opposing effects of dielectric screening on short- and long-range potential scattering

We reduce the dimensionless interaction strength in graphene by adding a water overlayer in ultra-high vacuum, thereby increasing dielectric screening. The mobility limited by long-range impurity scattering is increased over 30 percent, due to the background dielectric constant enhancement leading to reduced interaction of electrons with charged impurities. However, the carrier-density-independent conductivity due to short range impurities is decreased by almost 40 percent, due to reduced screening of the impurity potential by conduction electrons. The minimum conductivity is nearly unchanged, due to canceling contributions from the electron/hole puddle density and long-range impurity mobility. Experimental data are compared with theoretical predictions with excellent agreement.

cond-mat.mtrl-sci↗

arXiv2007

Charged Impurity Scattering in Graphene

Since the experimental realization of graphene1, extensive theoretical work has focused on short-range disorder2-5, ''ripples''6, 7, or charged impurities2, 3, 8-13 to explain the conductivity as a function of carrier density sigma_(n)[1,14-18], and its minimum value sigma_min near twice the conductance quantum 4e2/h[14, 15, 19, 20]. Here we vary the density of charged impurities nimp on clean graphene21 by deposition of potassium in ultra high vacuum. At non-zero carrier density, charged impurity scattering produces the ubiquitously observed1, 14-18 linear sigma_(n) with the theoretically-predicted magnitude. The predicted asymmetry11 for attractive vs. repulsive scattering of Dirac fermions is observed. Sigma_min occurs not at the carrier density which neutralizes nimp, but rather the carrier density at which the average impurity potential is zero10. Sigma_min decreases initially with nimp, reaching a minimum near 4e2/h at non-zero nimp, indicating that Sigma_min in present experimental samples does not probe Dirac-point physics14, 15, 19, 20 but rather carrier density inhomogeneity due to the impurity potential3, 9, 10.

cond-mat.other↗

arXiv2007

Persistence and survival in equilibrium step fluctuations

Results of analytic and numerical investigations of first-passage properties of equilibrium fluctuations of monatomic steps on a vicinal surface are reviewed. Both temporal and spatial persistence and survival probabilities, as well as the probability of persistent large deviations are considered. Results of experiments in which dynamical scanning tunneling microscopy is used to evaluate these first-passage properties for steps with different microscopic mechanisms of mass transport are also presented and interpreted in terms of theoretical predictions for appropriate models. Effects of discrete sampling, finite system size and finite observation time, which are important in understanding the results of experiments and simulations, are discussed.

cond-mat.stat-mech↗

arXiv2006

Transmission Line Impedance of Carbon Nanotube Thin Films for Chemical Sensing

We measure the resistance and frequency-dependent gate capacitance of carbon nanotube (CNT) thin films in ambient, vacuum, and under low-pressure (10E-6 torr) analyte environments. We model the CNT film as an RC transmission line and show that changes in the measured capacitance as a function of gate bias and analyte pressure are consistent with changes in the transmission line impedance due to changes in the CNT film resistivity alone; the electrostatic gate capacitance of the CNT film does not depend on gate voltage or chemical analyte adsorption. However, the CNT film resistance is enormously sensitive to low pressure analyte exposure.

cond-mat.mtrl-sci↗

arXiv2006

The Effect of Transfer Printing on Pentacene Thin-Film Crystal Structure

The thermal deposition and transfer Printing method had been used to produce pentacene thin-films on SiO2/Si and plastic substrates (PMMA and PVP), respectively. X-ray diffraction patterns of pentacene thin films showed reflections associated with highly ordered polycrystalline films and a coexistence of two polymorph phases classified by their d-spacing, d(001): 14.4 and 15.4 A.The dependence of the c-axis correlation length and the phase fraction on the film thickness and printing temperature were measured. A transition from the 15.4 A phase towards 14.4 A phase was also observed with increasing film thickness. An increase in the c-axis correlation length of approximately 12% ~16% was observed for Pn films transfer printed onto a PMMA coated PET substrate at 100~120 C as compared to as-grown Pn films on SiO2/Si substrates. The transfer printing method is shown to be an attractive for the fabrication of pentacene thin-film transistors on flexible substrates partly because of the resulting improvement in the quality of the pentacene film.

cond-mat.mtrl-sci↗

arXiv2006

Hooge's Constant of Carbon Nanotube Field Effect Transistors

The 1/f noise in individual semiconducting carbon nanotubes (s-CNT) in a field effect transistor configuration has been measured in ultra-high vacuum and following exposure to air. The amplitude of the normalized current spectral noise density is independent of source-drain current, indicating the noise is due to mobility rather than number fluctuations. Hooge's constant for s-CNT is found to be 9.3 plus minus 0.4x10^-3. The magnitude of the 1/f noise is substantially degreased by exposing the devices to air.

cond-mat.mes-hall↗