The Ray Phaneuf Research Group

Among the most important and exciting directions in materials research today is toward understanding and controlling materials at the nanometer scale. Much of the driving force for this comes from the continued decrease in individual device size, and increase in device densities called for in the semiconductor road map for technology, but there are numerous other applications for materials whose properties are dominated by the confinement of electrons, phonons and other particles to dimensions comparable to their wavelength, i.e. quantum effects. My research is centered on techniques for fabricating and characterizing nanometer scale structures, in directing their rapid self-assembly and in using nanometer scale structures to enhance the efficiency of devices which involve their interaction with light (click here for a list of publications).

One approach to making and characterizing such structures involves using the scanning tunneling microscope to manipulate or deposit small numbers of atoms. We’ve used the technique of evaporating a thin film of gold or aluminum onto our STM tip, and then applying a voltage pulse between sample and tip to produce nanometer-scale dots of metal on a silicon surface. We then used the same tip both to image the resulting dots, and to measure the local density of electronic states (click here for a short display of our results).

Although the size- and position-control afforded by this approach are superb, this sort of “direct-write” approach is intrinsically slow, and unlikely to meet the demand of macroscopic arrays of nanometer scale structures called for in practical applications. On the other hand lithographic approaches suffer from insufficient resolution to produce the nanometer scale structures we are interested in. Technology is thus turning to “self-assembly to combine small size and large numbers of structures.

Self-organization in nature often leads to pattern formation at characteristic length scales, a phenomenon which has long fascinated scientists. It also presents a clue as to how bottom-up fabrication of densely packed arrays of nanometer scale devices might be achieved practically. The general idea is to use a template to direct the assembly that occurs during subsequent deposition or growth. Fabrication of the template typically involves either spontaneous pattern formation arising from a balance between thermodynamic driving forces and kinetic processes, or lithography followed by etching to produce artificially patterned substrates. The first approach is relatively simple and clean to carry out, however the characteristic length scales of the patterns which form are determined mainly by material properties, resulting in a limited range of variation. The second approach is labor intensive, particularly if the detailed structure of individual cells within the pattern is to be controlled.

Research in Directed Self Assembly – “Teaching Nature How to Build”

Alternatively, a third, hybrid approach can be used, with an initial lithographically determined pattern setting conditions for the physical driving forces which control evolution during a subsequent sample processing step, i.e. directed self-organization, in which a template is fabricated to guide nature in deciding how to assemble very large numbers of structures quickly. To make this practical, it is crucial to understand the interactions between the underlying physical mechanisms involved, and in particular how these interactions vary with lateral length scale. Some of the recent directions of the research carried out by my group has been toward gaining such an understanding, by probing the effect of the characteristic template length scale in directed self-organization during growth and annealing of semiconductor surfaces.

We start with a template, which we define in a number of different ways-lithography followed by etching, and “nanoscraping” are the two approaches we are using at present. The first relies on depositing a “resist”, usually a polymeric film onto a surface, and then selectively changing the properties of the regions to be removed or left in place. This can be done by irradiating it through a mask with UV light, or irradiating selected regions top a scanned electron beam-both change the resistance of the polymer to being washed away by a chemical “developer”. Nanoscraping uses an atomic force microscope to selectively scrape away resist; and allows for extremely small patterns to be made. In both approaches the surface, parts of which are covered with resist, others bare, is exposed to a plasma which etches away the top layers. Finally the resist is removed chemically, leaving a clean but topographically patterned surface, a template. Our templates typically made up of arrays of cells, with the pattern lateral size and spacing varied systematically from cell to cell. In this way we are able to measure the lateral size dependence quickly, and under otherwise uniform conditions during such processes as epitaxial growth, annealing and reactive ion etching.

Fig. 1. Patterned GaAs(001) surface containing arrays of cylindrical pits, 50 nm deep, with varying diameter and spacing.

The beauty of this approach is that many patterns can be explored simultaneously, allowing the effect of the pattern length scale to be explored. An example is shown in Figure 1, where we have defined pits of varying diameters and spacings onto a gallium arsenide surface. We now grow more GaAs onto this template, and observe the effect of the length scale on how new structures self-assemble during growth. We find that large period structures amplify during growth: the pits effectively grow deeper, while those whose period is below a certain characteristic size relax during growth. This characteristic size moves to larger values as we grow thicker films, and so eventually even large period structures relax, but the surface shows a transient instability. We’ve explored the temperature dependence, and find that the nature of the instability changes beneath ~540 C; rings of material form around pits during growth beneath this. We explain this change based on competing kinetic effects: one is associated with a barrier that atoms feel on difusing across a step from above, which is important at low temperatures. The second is a faster collection of atoms by larger terraces, important at high temperatures. In upcoming work, we’ll extend our patterns downward to nanometer dimensions.

Fig. 2. (Left) Atomic Force Microscopy Height profiles across patterned GaAs(001) for 1 um pits spaced at 2 um, vs. thickness of grown film, (Right) with varying diameter and spacing. (right) Measured peak-valley height of patterns vs. initial diameter for different grown thicknesses. Pattern lengths larger than the peak value amplify, those beneath it decay.

In related experiments, we have patterned stepped silicon surfaces, and measured how the length scale of our pattern affects the self assembly of bunches of steps during heating in vacuum. Beneath a characteristic length scale bunches of straight steps form, while above this the bunches form near sinusoidal shapes, with the waviness of the bunches decaying at the same rate as the height. This length scale is set both by the stiffness of the steps and their interactions. Using a simple model in which steps move across the surface in response to these effects and sublimation we’re able to reproduce the pattern size dependence of the self-organization of bunches of steps during annealing. (Click here for a (347 MB !) movie showing the relaxation for the smallest period structure)

Fig.3. Self assembly of straight step bunches during annealing of short period patterned stepped Si(111), but wavy step bunches for long period patterns. The decay times for step bunch waviness (IP) and height (OOP) are different for small periods, but converge for large pattern structures.

Research in Light – Plasmon Interactions: “Using Nanoparticles to Enhance Optical Phenomena”

Another of the directions of the research that my group and I are carrying out is using arrays of noble metal nanoparticles to enhance optical processes due to interactions with particle plasmons. It’s interesting to note that Nanotechnology has been around in some sense for hundreds of years. Some of the brilliant colors of stained glass windows in cathedrals across Europe derive from the formation of nanoclusters of certain metals when added to glass. Gold clusters yield red glass, for example.

A much more contemporary application of the interaction of light and nanoparticles is in fabrication of biosensor arrays or biochips, which are often based upon fluorescence. The appeal of this is the ease with which biochemists can add a fluorescent tag to molecules which in turn bind to certain biomolecules of interest. This in fact is the basis of the so-called sandwich array. There has been a good deal of work done investigating a related technique, called surface enhanced Raman scattering, or SERS, which allows a different type of molecular fingerprinting. There huge enhancements, up to 10^14 have been reported. Much less work has gone into optimizing and understanding the enhancement which comes from the interaction of light from fluorescent molecules and metallic nanoparticles. Fluorescence of course involves both absorption and radiative decay, at two different wavelengths, due to the Franck-Condon Effect. Our results show evidence that the observed nanoparticles enhancement of fluorescence involves both processes.

We use electron beam lithography to creat arrays of silver nanoparticles on a substrate. This allows us control over the shape, size and spacing of the particles-all of which effect the plasmon resonance frequency. To avoid quenching of the fluorescence we need a finite spacing between the fluorescent molecule and the nanoparticles. We use a combination of proteins to space a variety of fluorescent tags several nm from the surface of the surface of the Ag particles. We measure the fluorescence using a scanning laser microscope and a filter which cuts out the incident ligh, but allows a band of wavelengths through to the detector. It allows us to quickly determine the optimum combination of size an spacing as seen in the figure to the right.. The upward shift in the optimum size in going from green light to red points to an enhanced absorption of the incident light, since the plasmon resonance is known to shift to lower frequency for larger particles. On the other hand, we find that the maximum in intensity coincides with a minimum in the fluorescence life time showing that the emission is also enhanced. Our present work is aimed toward a quantitative, and predictive understanding of nanoparticle-enhanced fluorescence, and related phenomena.

Other Research Activities

Since many of the applications for nanometer scale structures are for devices, we have also been investigating other electron microscopies to determine what their capabilities are in measuring local potential, dopant concentration and charge density (click here for a short description of some recent STM results). Among the most powerful of these techniques is low energy electron microscopy (LEEM) (click here to see a an example LEEM result from our group or here to see an overview article on LEEM published in Physics Today). My group (click here to see group members at work) and I have designed and built a LEEM, adding it to the “arsenal” of techniques we are using in working toward directed self assembly of nanoscale structures. Another capability we have is molecular beam epitaxy, which gives us extreme control over the composition of structures built up by growth of ultrathin films of semiconductors.Photoemission electron microscopy (PEEM), particularly spectroscopically resolved is particularly powerful in probing these quantities. Using a scanning x-ray photoemission microscope at the Elettra Synchrotron Light Source in Trieste, Italy we found it is possible to “image” the bending of electron energy levels across the depletion layer of a Si pn-junction for the first time (click here to see a short description of these results).

Presently we are carrying out a NSF-NIRT funded collaboaration with Professor Gottlieb Oehrlein at the UM, Professor David Graves at UC-Berkeley, Professor Grant Willson at UT-Austin and Dr. Azar Alizehda at General Electric, concerning the stability of the surface of model polymer resists during plasma etching. Line-edge roughening in this system is of crucial concern, as it threatens to limit the minimum feature size obtainable in ULSI and beyond. My group is again adopting a patterning approach to perturb polymer films over a range of spatial frequencies, and investigating the response during subsequent etching.

Research Group

Our group, past and present consists of research scientists, postdocs, graduate students and undergraduate students with a varied technical and cultural background. It includes physicists, materials scientists and electrical engineers from the US, Taiwan, Iran and Korea, working on research which varies from lithographic patterning, to molecular beam epitaxy, to nanoparticle enhanced fluorescence, to numerical simulations based on finite difference methods, and direct integration of differential equations.

Shy-Hauh (Sherman) Guo is a graduate student in the department of Electrical and Computer Engineering. Sherman uses electron beam lithography to create arrays of silver and gold nanoparticle whose size shape and spacing are varied in a combinatorial manner to tune the particle plasmon resonance. He has been studying the underlying mechanisms for nanoparticle-enhanced fluorescence from nearby fluorescently-tagged molecules, which he measures with scanned fluorescence microscopy. He has also developed numerical simulations of the electric field around arrays of nanoparticles, based upon the finite difference time domain method (FDTD). Sherman has published papers in Applied Physics Letters and the Journal of Fluorescence, is lead author on a recently submitted paper on the effect of an active substrate on fluorescence enhancement. And has given talks at the APS March Meeting and the Physical Electronics Conference.

Dominic Britti is a graduate student in the Department of Materials Science and Engineering, and a graduate of the Department of Physics. Dominic is using Scanning Tunneling Luminescence (STL) as a means of probing the electronic energy levels of structures with nm-scale resolution. Dominic has participated in collaborative research with the Group of Professor Ross Rinaldi, at the National Nanotechnology Laboratory in Lecce, Italy, probing transport through nanometer scale structures, and the variation in fluorescence intensity with the separation between a fluorescent molecule and a nearby nanometer-sized silver sphere; this work also involved collaboration with De-Hao Tsai, a graduate student in the research group of Professor Michael Zachariah of the Department of Mechanical Engineering at the UM.

Shu-Ju (Phoebe) Tsai is a graduate student in the Department of Materials Science and Engineering. Phoebe’s research involves the use of size selected noble metal nanoparticles to enhance the interaction of molecules and light. She has been working with Sherman Guo and De-Hao Tsai on experiments on nanoparticle-enhanced fluorescence, in this case employing size selected particles fabricated by spray pyrolysis and differential mobility analysis, and deposited onto a silicon substrate. Phoebe has also developed numerical simulations based upon the dynamic dipole approximation which show that the substrate plays an active role in the enhancement, reshaping the distribution of intense electric field during illumination with light. Phoebe gave a talk at the 2007 APS March Meeting in Denver, and is coauthor on a recently submitted paper on the effect of an active substrate on fluorescence enhancement.

Tsung-Chen (T. C.) Lin is a graduate student in the Department of Materials Science and Engineering, who recently joined our group. T. C.’s research involves studying the stability of prototype resists against roughening during reactive ion etching, as part of the National Science Foundation Nanoscience Interdisciplinary Reseach Team (NIRT) project which we are carrying out in collaboration with Professor Gottlieb Oehrlein’s group at the UM, Professor David Graves’ group at UC-Berkeley, Professor Grant Willsons’ group at UT-Austin, and Azar Alizehda at General Electric. T. C. is adopting a Nanopatterning approach toward probing the stability of the plasma-resist interface.

Dr. Tim Corrigan is a Postdoctoral Research Scientist with our group. His research concerns the interaction of light with nanometer-scale noble metal nanoparticles, including enhanced fluorescence, and nano-optronics/metamaterials. Tim is an expert in electron beam lithography, and device processing techniques. He has collaborated with the group of Professor Ross Rinaldi at the National Nanotechnology Laboratory in Lecce, and Professor Michael Zachariah at the UM, and was a Visiting Professor at Shandong University in China. Tim has published papers in the Journal of Fluorescence, Applied Physics Letters, Nanotechnology, Journal of Applied Physics, Journal of Vacuum Science and Technology the Proceedings of the SPIE

Dr. Hung-Chih Kan is a former Assistant Research Scientist in my group, and has recently joined the Physics Department at National Chung Cheng University in Taiwan as an Associate Professor. His research with my group included: experimental investigation on enhancement of fluorescence from metallic nano-particles and numerical simulation for surface plasmonic resonance from nano-particles (see figure below) to probe the physical process that causes the enhancement of the fluorescence. He also worked on experimental studies of evolutions of patterned semiconductor surfaces under processing conditions, such as homo-epitaxial growth or high temperature annealing, as potential recipe for directed self-organization for nano-scale fabrication, and performed simulation to provide explanation for the experimental observations. He continues to collaborate with our group as a Visiting Professor

Electric field distribution for a silver nano-particle with diameter of 200nm in the presence of a light wave propagating in the +x direction, with the E field initially polarized in the +y direction. The wave length for the incident light is 200 nm for (a)-(c), and 400nm for (e)-(f). According to the simulation the maximum extinction (resonance) occurs at 400nm wavelength. Panel (a) and (d) show the X component of the E-field, (b) and (e) the Y component, and (c) and (f) the E². Color levels for (a),(b),(d) and (e): red for positive, blue for negative, and black indicated zero level. For (c) and (f), black indicates 0 (V/cm) ², and bright yellow for 10 (V/cm) ².

Dr. Tabassom Tadayyon-Eslami recently received her PhD in our group, carrying out investigations of the length-scale dependence of unstable growth on patterned GaAs(001). In this work, she employed photolithography and reactive ion etching to create arrays of cylindrical pits on the surface, whose size and spacing are varied in a combinatorial manner. She found that there is a transient instability during molecular beam epitaxial (MBE) growth at standard conditions, and that the nature of this instability changes qualitatively upon lowering the temperature through a value which approximately coincides with that of thermodynamic preroughening. She established that this was coincidental, however, and identified atom-scale mechanisms which explain the instability and its temperature dependence. She published papers in Physical Review Letters, Physical Review B and Applied Physics Letters based upon this work, and gave talks on it at the APS March Meeting and the Physical Electronics Conference. Tabassom is presently working at the US patent office in Arlington, VA

Dr. Taesoon Kwon is a recent PhD graduate from my group. Taesoon’s research involved patterning for self organization during annealing and reactive ion etching of materials of great technological interest for the semiconductor industry: silicon, nanoporous silica and model resists. Her research also included a study the stability of prototype resists against roughening during reactive ion etching, as part of the National Science Foundation Nanoscience Interdisciplinary Reseach Team (NIRT) project which we are carrying out in collaboration with Professor Gottlieb Oehrlein’s group at the UM, Professor David Graves’ group at UC-Berkeley, Professor Grant Willsons’ group at UT-Austin, and Azar Alizehda at General Electric. Taesoon has published papers in Applied Physics Letters, Nanotechnology and Journal of Vacuum Science and Technology B, and gave talks at the Ameican Physical Society March Meeting, the American Vacuum Society Symposium and The Physical Electronics Conference. She is presently a plasma process engineer at Micron, in Boise, ID.

Dr. Jeong-Young Park is a former Postdoctoral researcher with my group. His work with us involved the use of the scanning tunneling microscope in probing the time response of working device structures and in creating and modifying nanostructures at surfaces. Jeong-Young has published papers in Science, Applied Physics Letters, Journal of Vacuum Science and Technology and Surface Science. He is now a Research Scientist at Lawrence Berkeley Laboratories carrying out studies on Hot electron generation from catalytic reaction in catalytic nanodiode, Atomic or molecular scale properties of surface and interface, Nanotribological properties of quasicrystal surface and their relation with the surface atomic structure, and Charge transport properties and mechanical properties in organic molecules or bio-materials with conductive atomic force microscopy

Teaching

One of the courses I teach at the University of Maryland is one that I’ve developed, on techniques used to characterize materials which have been structured at the nanometer scale, ENMA698T. This reviews important topics such as quantum confinement and tunneling as well as intermolecular forces, and also covers scanning tunneling microscopy, atomic force microscopy, electrostatic and magnetic force microscopies, near-field scanning optical microscopy, scattering techniques and transport. I teach it in part from the recent literature, and ask the students review a recent paper in class. The final project involves each student writing a research proposal based upon one of the techniques we discuss. In one lecture I talk about my own research, and I also invite other Faculty to discuss their work characterizing nanometer scale structures in other lectures. At the end of these research lectures we visit labs, letting the students see and ask questions about working experiments.

A second course I’ve been teaching is Diffusion, Kinetics and Phase Transformations. Topics include diffusion in substitutional solid solutions, interstitial diffusion, nucleation and growth theories, solidification, diffusional transformations and growth of crystalline solids. I supplement traditional lectures with results from the literature, stressing the application of course material to problems of interest in current research being carried out with a battery of techniques, new and old, including LEEM, FIM, STM, TEM and Scattering.

I’ve also taught the Senior Capstone Design Course, Introduction to the Materials, an Introductory Course in Nanotechnology for Freshman, and Introdution to Engineering Design. In all case I stress active learning, and awareness of how the course material relates to problems of interest in research today.