October 7, 1999

8:15-8:30 Opening Remarks: K. P. Chong, NSF

The National Science Foundation (NSF) has supported basic research for a half century and as a consequence the U.S. is likely to continue to dominate vital markets because diligent funding of basic research does confer a preferential economic advantage. Over this past half century, technologies have been the major drivers of the U. S. economy, and as well, NSF has been a major supporter of these technological developments. According to the NSF Director for Engineering, Eugene Wong, there are three transcendental technologies:
* Microelectronics - Moore's Law: doubling the capabilities every two years for the last 30 years; unlimited scalability; nanotechnology is essential to continue the miniaturization process.
* Information Technology - NSF and DARPA started the Internet revolution about three decades ago; confluence of computing and
communications.
* Biotechnology - molecular secrets of life with advanced computational tools as well as advances in biological engineering, biology, chemistry, physics, engineering including nano and micromechanics and materials.
By promoting research and development at critical points where these technological areas intersect, NSF can foster major developments in engineering and sciences. The mechanics and materials communities will be well served if some specific linkages or alignments are made toward these technologies. It will be of vital interest to find out what are the key roles nano and micromechanics of solids play in these and other emerging technologies and sciences.
 

Session I: Atomistic-scale, nano-structural and collective-defect-scale modeling

8:30-8:35 Session Chair: K. Cho, Stanford University
 

8:35-9:25 Atomistic-scale modeling and molecular dynamics for nano- and micro-structures (J. A. Harrison; U.S. Naval Academy, R. Phillips; Brown)

Molecular Dynamics Investigations of the Tribology of Hydrocarbon Monolayers Judith Harrison, U.S. Naval Academy

Technological applications, such as microelectromechanical systems and magnetic storage devices, have piqued interest in the potential use of organic films as boundary layer lubricants. In these applications, the thickness of the lubricating film is approaching the monolayer regime and adhesion of the molecules to the substrate is critical. As a result, the molecular structure, mechanical properties, and tribological properties of self-assembled monolayer (SAM) materials, such as alkanethiols and alkylsilianes, have been studied a great deal using scanning probe microscopies. These experimental studies of SAM systems have revealed several interesting phenomena. The aim of our recent work has been to elucidate the atomistic mechanisms responsible for these phenomena using molecular dynamics simulations. Extension of this work to larger systems will also be discussed.

Hierarchical Modeling of Plasticity: The Physics of Entanglement Rob Phillips, Brown University

The goal of understanding plastic deformation from the dislocation perspective poses challenges at a number of different scales. This talk will describe the use of ideas both from continuum theory and atomistic analysis in order to examine the entanglement of dislocations as a result of junction formation. The structure and energetics of junctions will be considered and then these ideas will be reexamined from a statistical perspective with the aim being an understanding of the motion of dislocations encumbered by their interactions with obstacles.
 

9:25-10:15 Defects on surfaces and interfaces (D. Srolovitz; Princeton, D. Kouris; Arizona State)

Defects on Surfaces and in Thin Films David Srolovitz, Princeton University

Defects on surfaces and interfaces interact with one another through forces of elastic, electric, magnetic and/or entropic origins. We will present a general approach to describing these interactions and examine elastic interactions in detail. This approach is based on multipole expansions. While this approach tells us the form of the interactions, atomic scale calculations or experiments are necessary for determining their magnitude. We use this approach to predict the orientation dependence of the surface energy (i.e., the Wulff plot) and equilibrium crystal shapes. We also examine the corrections to these results due to entropic factors. Finally, we generalize this approach for describing interactions between defects in thin films which are mediated by the substrate.

Elastic interaction and growth Demitris Kouris, Arizona State University

Many thin-film/substrate systems initially grow in a two-dimensional (2D) or atomically flat mode but eventually undergo a roughening transition, yielding three-dimensional (3D) islands on top of the 2D layers. It is now understood that 3D islanding can occur owing to elastic interaction of surface defects (e.g. adatoms, islands and steps) during growth of crystal surfaces. These interactions influence the kinetics of growth and under some conditions (defined by growth flux and temperature) control the growth mode morphology. We will discuss discrete models of adatoms and steps and their interaction. We will also examine how the elastic field can define pathways for growth. Examples include superstructures and interfacial dislocation arrays leading to templates for quantum-dot formations.
 

10:45-11:35 Nano and micro mechanics of polymers (M. Boyce; MIT, E. M. Arruda; Michigan, Ann Arbor)

Micromechanics, Meso-Scale Modelling and Molecular Dynamics Simulations, of Polymers Mary C. Boyce, MIT and Ellen M. Arruda, University of Michigan, Ann Arbor

The deformation characteristics of amorphous polymeric chains and the entangled and crosslinked networks they form are investigated for a number of applications using modeling scales from molecular level to continuum. Mesoscopic modeling of bimodal elastomeric networks and molecular level Monte Carlo methods are used to understand and predict experimental stress-stretch and birefringence-stretch responses of these systems. Micromechanical modeling of thermoplastic vulcanizates captures the deformation and recovery of these important specialty polymer blends. Molecular dynamic simulations of rubbery polymers by coarse graining of the elastomeric chains is shown to provide molecular justification of continuum models of rubber elasticity. Glassy polymeric systems are similarly investigated via molecular level Monte Carlo simulations which validate the internal state variable approach.
 

11:35-12:25 Material models for damage and failure (H. Gao; Stanford, W. Yang; Tschinghua)

Mechanics Modeling at Micro- and Nano-Scales Huajian Gao, Stanford University

Here we discuss two classes of mechanics problems that arise at micro- and nano-scales. The first class of problems involves strongly coupled diffusion and deformation problems in thin film structures. Such coupling often gives rise to unique morphological defects with crack-like singular stress concentrations. We discuss the formation of surface cusps in nano-scale heteroepitaxial thin films and grain boundary diffusion wedges in micro-scale thin metal films, and show that these defects exhibit crack-like behaviors and act as nucleation sources for dislocations. The second class of problems involves the development of continuum constitutive theories for micro-scale applications. One of such developments is the mechanism-based strain gradient plasticity (MSG) which links the concept of geometrically necessary dislocations with continuum plasticity via Taylor's dislocation hardening model. We discuss the basic idea of the MSG theory and relevant experimental observations from micro-indentation, micro-bending and micro-torsion tests.
 

Microscopic and Nanoscopic Deformation at a Crack Tip Wei YANG, Yongming XING and Fei FANG, Department of Engineering Mechanics, Tsinghua University, Beijing 100084, CHINA

Two issues will be addressed in this presentation concerning the nanoscopic and microscopic deformation at a crack tip. First we consider the case of dislocation emission from a crack tip. A brief introduction of nano-moire method will be given which involves the interferometry of amplified high resolution atomistic image with a reference optical moire grating. The specimens are made by single crystal silicon films, and pre-loaded by the deformation of supporting copper net. Dislocation emission from the crack tip is observed. It is of the Peierls type with a very long tail. Several premature dislocations of half Burgers vector are also observed. These observations, though preliminary, give support (and also raise questions) to the existing Peierls framework of dislocation emission from a crack tip.
We next move to the microscopic deformation at the crack tip of ferroelectric ceramics by domain switching. Our experiment for polycrystal ferroelectrics after out-of-plane poling does ndicates the crack tip domain switch by the application of a lateral electric field. The theory of crack tip domain switch will be presented, which explains both the electric fracture and fatigue crack growth under alternated electric field. The analysis is developed for both in-plane and out-of-plane poling. Implication on how to raise the fracture toughness of ferroelectrics will be addressed.
 

Session II: Mechanics of quantum devices, thin films and micro-structural evolution

1:30-1:35 Session Chair: D. B. Bogy, University of California, Berkeley
 

1:35-2:25 Strain effects on the functions of quantum devices and manufacturing processes (L. B. Freund; Brown, H. Johnson; Boston)

Continuum Modeling of Strain Effects on Functional Characteristics of Quantum Devices L. B. Freund, Brown University

It has been observed that functional characteristics of nano-scale electronic devices vary systematically with device size, even for nominally one-dimensional configurations. The origin of this effect seems to be nonuniform mechanical strain which is induced during device fabrication. A continuum method has been developed to estimate the magnitude of the effect. The elastic strain field in the device is determined by means of the finite element method. This strain field, in turn, is used to specify a deformation potential which modifies the Schrodinger equation, the solutions of which describe confinement characteristics of charge carriers in the device. Results are reported for a resonant tunneling diode and other configurations.

Atomic scale modeling of strain effects on electronic properties of quantum devices H. Johnson, Boston University

Nanometer scale applications of electronic materials may be strongly affected by coupling of electronic and mechanical properties, particularly when lattice or thermal mismatch strain is induced in fabrication. In this talk, atomistic based approaches for modeling electronic and mechanical coupling will be discussed. The use of tight-binding as part of a mixed atomistic/continuum approach will be described; the goal is to calculate local electronic density of states as the electronic property of interest, while simultaneously determining the spatially varying deformation fields in the structure. Approaches will be described both for deformed regions that are uniform at the atomic scale, and for regions that are nonuniform at the atomic scale, such as defects and interfaces.
 

2:25-3:15 Micro-structural evolution (Z. Suo; Princeton, W. C. Carter; MIT)

Configurational forces in nanostructures  Z. Suo, Princeton University

Modem electronic and photonic devices are solid structures of small feature sizes. During fabrication and use, diffusive processes can relocate matter, so that the structures evolve over time. A film may break into droplets, and a conducting line may grow cavities. Stress and electric current have long been understood as forces that drive the changes. Evidence has accumulated that, while important, these forces are insufficient to account for diverse experimental phenomena, suggesting forces of other physical origins also operate. In a structure, collective actions of atoms, electrons, and photons contribute to the free energy. When the structure changes its configuration, the free energy also changes. The free energy change defines a thermodynamic force which, in its turn, drives the configurational change of the structure. This talk illustrates the concept with specific phenomena. Emphasis is placed on physical descriptions of forces of diverse origins. The effects of some of these forces are particularly significant in structures of small feature sizes, say, between a few to hundreds of nanometers.

Incorporating Crystallography in Phase Field Models of Microstructural Evolution  Ryo Kobayashi, Hokaido University, James Warren, National Institute of Technology
W. Craig Carter*, Massachusetts Institute of Technology

We introduce a method of incorporating crystallographic information into diffuse interface methods. This allows proper computation of interface anisotropies during solidification and grain boundary formation. Included in the method is a two dimensional phase field model of grain boundary dynamics. One dimensional analytical solutions for a stable grain boundary in a bicrystal are obtained, and equilibrium energies are computed. By comparison with microscopic models of dislocation walls, insights into the physical accuracy of this model can be obtained. Indeed, for a particular choice of functional dependencies in the model, the grain boundary energy takes the same analytic form as the microscopic (dislocation) model of Read and Shockley.
 

3:45-4:35 Mechanics of thin-films (A. F. Bower; Brown, G. Genin; Washington)

Mechanics of Nano-Electronic Structures  A.F. Bower, Brown University and G. Genin, Washington University

Nano-mechanics has played an important role in the development of thin film structures for microelectronic applications. The field continues to pose further challenges, however, of both a technological and scientific nature. Further miniaturization of circuits and devices will require significant improvements in manufacturing processes, as well as new procedures to control stress in thin film structures. The objective of nano-mechanics is to enable progress in the area by providing the ability to model quantitatively the effects of thermal, mechanical and electrical loading in thin films, as well as accounting for physical phenomena such as: surface forces; surface stress; crystallographic defects; and mass transport; which all play an important role in determining the performance of sub-micron structures. Our presentation will focus on one particular application of nano-mechanics to thin film device technology, namely, efforts to manufacture quantum dot arrays for solid state laser applications. A quantum dot is a nanometer sized cluster of between 10A3 and 10A5 atoms, which is capable of confining charge at discrete energy levels. For laser applications, it is desirable to manufacture periodic arrays of closely spaced, identical dots. The required dot density and size exceeds the limits of conventional lithographic techniques, so alternative manufacturing processes are sought. One promising approach is to exploit the properties of strained semiconductor heterostructures to grow self-organized island arrays. We will review experimental observations of island formation during growth and annealing of strained films, and will illustrate applications of mechanics which have aided in understanding the processes involved. We will conclude by showing some spectacular three dimensional simulations of island formation during deposition, and by outlining issues which must be resolved in order to make further progress in the area.
 

4:35-5:25 Mechanics of micro-sensors and actuators (R. M. McMeeking, UCSB & C. S. Lynch; Georgia Tech)

Micromechanics Applied to Constitutive Behavior and Reliability of Ferroelectric Ceramic Materials  R.M. McMeeking, University of California, Santa Barbara and
C.S. Lynch, Georgia Institute of Technology

Ferroelectric materials display field coupled constitutive behavior that is both nonlinear and hysteretic. The source of the non-linearity and hysteresis is a reorientation of the polar crystal structure in response to applied stress and electric field. The speakers will present models that smear the atomistic reorientations and sub-crystal domain structures into an average single crystal behavior. The single crystal model is used to develop micromechanics simulations of the behavior of ceramic materials. Measured uniaxial and mutiaxial constitutive behavior and R-curve behavior for crack growth will be presented. The experimental results will be
discussed in terms of the micromechanics model, and theories for fracture mechanics and crack growth in ferroelectrics will be presented.
 

October 8, 1999

Session III: Materials testing in small dimension, nano-tribology and nanobiomechanics

8:30-8:35 Session Chair: W. W. Gerberich, University of Minnesota
8:35-9:25 Nano-indentation for materials testing (G. M. Pharr; The University of Tennessee & Oak Ridge NL, W. D. Nix; Stanford)

Nanoindentation for Materials Testing  G.M. Pharr, The University of Tennessee and Oak Ridge National Laboratory; and W.D. Nix, Stanford University

Nanoindentation is one of a growing number of methods for characterizing the mechanical behavior and properties of materials at the micron and sub-micron levels. In nanoindentation methods, properties are derived from analyses of indentation load-displacement data, most of which are based on continuum descriptions of elastic contact. Elastic contact solutions work curiously well in many circumstances, even when the contact is dominated by plasticity, but in some instances, they are clearly deficient. Experimental results combined with finite element simulations of elastic-plastic contact are presented which show when and why the elastic contact solutions fail to provide accurate results. Other experimental observations are used to illustrate how the principles of continuum plasticity can be inadequate, for example, through the discrete nucleation of dislocations in the vicinity of the contact. The observations emphasize the need for new and better models of elastic-plastic contact at small scales. Time allowing, other important methods for small-scale mechanical testing will also be briefly discussed.

9:25-10:15 Nano-tribology AFM-testing (R. W. Carpick; Sandia NL, K.-S. Kim; Brown)

Nanotribology: Results and Prospects with Scanning Force Microscopy R. W. Carpick, Sandia National Laboratories

Instrumental, theoretical and computational techniques for studying tribology have now progressed to the atomic scale. These parallel advances are enabling new understanding of the origins of friction, adhesion, wear, and lubrication. I will highlight recent experimental results obtained with scanning force microscopy which demonstrate these new insights. The most notable examples pertain to single asperity friction measurements, controlled atomic-scale wear, and correlation between molecular lubricant structure and performance. Although these results are significant, instrumental limitations threaten to impede further progress. A more long-term challenge involves how to connect the results to practical systems in the macroscopic world. I will provide suggestions on how to approach these challenges, and discuss opportunities that lie ahead.

Scale Bridging between Nano and Micromechanics of Solid Surfaces Kyung-Suk Kim, Brown University

Transition from nano to micro mechanisms of interfacial slip in single-asperity adhesive-contact friction will be discussed in conjunction with Dr. Carpick's talk. Nonlocality of nano-mechanical response will be addressed as a key concept of the transition model. A nano-mechanical testing of the response nonlocality, a field projection method with high resolution transmission electron microscopy, will be presented. Then, another scale-transition problem associated with surface roughness will be discussed. In addition to the scale transition in tribological response, a scale-dependent transition in chemical-reaction response in chemical etching will be presented. A newly developed surface-roughness evolution spectroscopy (SRES) will be introduced for the study of chemical-reaction response transition. As a byproduct of this basic research, it is found that the SRES can be used as a high-spatial-resolution stress gauge. With these examples some challenging problems in nano and micro mechanics of solid surfaces will be discussed.
 

10:45-11:35 Material properties at small-length scale (E. Arzt; Max Planck Institute, D. R. Clarke; UCSB)

Material Properties at Small Length Scale Eduard Arzt, Max-Planck-Institut für Metallforschung and University of Stuttgart, Stuttgart, Germany
Discussion Leader: David R. Clarke, University of California, Santa Barbara

Deformation in small dimensions is of both academic and practical interest: dislocation plasticity and diffusional creep processes - classical bulk deformation mechanisms - are strongly affected by the microdimensionality in ways which are only partially understood; and as means of stress relaxation these mechanisms can greatly influence the operation and reliability of micro-systems. In this talk an overview will be given of recent systematic experiments using several test techniques, ranging from thermal straining (wafer curvature) to the application of external loads in an X-ray diffractometer (microtensile tester), on similar Cu thin films. The success of different modeling and simulation strategies is critically evaluated. The focus is then placed on the mechanical behavior of Cu and Cu-Al films at high temperatures, where diffusional creep and creep damage processes have for the first time been observed. The experimental results are compared with a recent model by Gao et al. for diffusional flow constrained by a hard substrate. The results show encouraging, but not complete, agreement. Subtle interface effects not present in bulk materials (e.g. the presence or absence of natural oxide layers) seem to control the damage processes, whose investigation therefore requires controlled, extremely clean (UHV) preparation and handling of the films.
 

11:35-12:25 Mechanical testing of biological cells and molecules (G. Bao; Georgia Tech, C. Zhu; Georgia Tech)

Deform DNA and Protein Molecules: May the Force Be with Us Gang Bao, Georgia Tech/Emory University

Recent studies confirm that mechanical forces, including that due to gravity, tension, compression, pressure, and shear can influence the growth, differentiation, secretion, movement, signal transduction, and gene expression in living cells, which in turn can have significant implications to human health. Yet, little is known about how living cells sense the mechanical forces or deformations, and convert these mechanical signals into biological or biochemical responses. This represents a significant challenge, and a wonderful opportunity to researchers in mechanics.
To answer the questions of why and how cells sense and respond to mechanical forces and deformations, it is necessary to establish the connection between mechanics and biochemistry at the molecular level as related to cell behavior and function. One key aspect of this connection concerns proteins whose 3-dimensional conformations determine function. Another important aspect is the mechanics of molecular machines such as motor molecules involved in DNA condensation, replication and transcription. Despite of its importance, to date, protein deformation under mechanical forces has not been well studied due to the small sizes of protein molecules (about 10-100 nanometers), and the small forces involved (about 10-100 piconewtons).
This talk focuses on the mechanics issues in this area and points out the complexities and challenges. Specific research topics are summarized.

Coupling of mechanics and chemistry in biology  Cheng Zhu, Georgia Institute of Technology

Binding via adhesive receptors is essential to many biological processes. The kinetic rates are critical to such interactions since they determine how likely and how rapidly cells bind and how long they remain bound. In contrast to soluble molecules, the association and dissociation of bonds cross-bridging two opposing surfaces usually take place in the presence of forces, as one of the biological functions of the adhesion molecules is to provide mechanical linkage between cells. Physical forces can influence binding interactions of adhesive bonds. As such, the chemical kinetics of receptor-ligand binding is tightly coupled to the mechanics of stretching and breaking these bonds at the molecular level. This talk will discuss 1) how the chemical reaction of receptor-ligand binding is regulated by mechanical forces and 2) how the kinetics of such reaction can be assayed by mechanical means.
 

Session LV: Mechanics of MEMS, reliability and mathematical modeling

1:30-1:35 Session Chair: W. N. Shame, Jr, The Johns Hopkins University
 

1:35-2:25 Mechanics of MEMS (R. Ballarini; Case Western, W.G. Knauss; Caltech)

Measurements of Fracture Toughness of Polycrystalline Silicon Using Micron Size Specimens  R. Ballarini, Case Western Reserve University
Discussion Leader: Wolfgang G. Knauss, California Institute of Technology

Numerous MEMS (microelectromechanical systems) devices have been developed which use polysilicon as the major structural material. However, the relevant material properties required to predict component and system structural reliability (fracture toughness, fatigue crack growth rates, and environmentally assisted crack growth rates) are not well characterized at these size scales, or for polysilicon which has been subjected to MEMS fabrication techniques. We have developed a series of polysilicon fracture mechanics specimens that could be used to develop a data base of these properties, and to determine whether there are differences in failure characteristic between micron size polysilicon structures and bulk structures. The fracture mechanics specimens have been fabricated using standard MEMS (microelectromechanical systems) processing techniques, and so have characteristic dimensions comparable to typical MEMS devices. These specimens are fully integrated with simultaneously fabricated electrostatic actuators which are capable of providing sufficient force to ensure catastrophic crack propagation. Thus the entire fracture experiment takes place on-chip, eliminating the difficulties associated with attaching the specimen to an external loading source. The electrostatic actuator can also be resonated at very high frequencies, mimicking actual MEMS operation and enabling fatigue measurements in reasonable times. Specimens containing micromachined notches and atomically sharp cracks have been tested under monotonic and cyclic loadings. Preliminary results are presented for the statistics of experimentally measured fracture toughness. A description of a stochastic fracture mechanics model that accounts for elastic anisotropy and heterogeneity that is associated with the polycrytalline films is also presented.
 

2:25-3:15 Mechanical reliability of micro-structures (R. H. Dauskardt; Stanford, J. C. Bravman; Stanford)

Mechanical reliability of thin film structures  Reiner H. Dauskardt, Stanford University
Discussion Leader: John C. Bravman, Stanford University

A critical step in the drive towards increasing density and improved performance of ULSI technologies involves the implementation of new metallization and low-K dielectric materials. The reliability of these materials in multi-layer thin-film structures are profoundly influenced by their adhesion with adjacent layers and substrates. Adhesion and integrity of the constituent interfaces are closely related to chemistry and processing conditions. These determine key interfacial parameters, such as interfacial chemistry and morphology, residual stress levels and adjoining microstructures. Similarly, long-term reliability is likely to be controlled by subcritical debonding of interfaces and/or cracking in the layers themselves. These issues are addressed in a range of materials including oxides and polymers. Models describing the kinetics of subcritical debonding including the effects of environment and temperature are reviewed.
 

3:45-4:35 Mathematical modeling of micro-structures (R. V. Kohn; Courant Institute, G. W. Milton; Univ. Utah)

Stress-driven pattern formation  Robert V. Kohn, Courant Institute

Many problems involve a competition between elastic and surface energy. Surface energy dominates at the smallest scales, elastic energy dominates at the largest scales, and stress-driven pattern formation takes place at intermediate scales where the two effects interact. I will discuss two specific problems of this type, namely (a) the implications of faceted surface energy for stress-driven instability of thin films (work with Alma Stancu); and (b) the geometry of misfit dislocations at the interface where a stressed film meets its substrate (work with Antti Pihlaja).

Composites with novel properties  Graeme Milton, University of Utah

Composites can sometimes exhibit properties which are unlike those of the constituent phases. In particular one can combine stiff and compliant materials with positive Poisson's ratios to obtain a composite with negative Poisson's ratio. More generally one can obtain materials with elasticity tensor matching any prescribed positive definite tensor (work with Andrei Cherkaev). I will also discuss the problem of designing a composite which is efficient at guiding stress, or more precisely which has an extreme value for its average stress for a given average strain (work with Alexander Movchan and Sergey Serkov.)

 

POSTER TITLES

Abstract