In the OSTI Collections: Mesoscale Science

Back to In the OSTI Collections Listing for 2015

Article Acknowledgement:
Dr. William N. Watson, Physicist
DOE Office of Scientific and Technical Information

 

The mesoscale's significance 

Understanding deformation and flow at the mesoscale

Experiments and Tools

Mesoscale science for defense

Apparent requirements for progress

References

Research Organizations

Reports available through OSTI's SciTech Connect

Patent available through OSTI's DOepatents

Conferences

Journals

Additional References

 

 

The mesoscale’s significance

 

Well into the 20th century, people deduced that if matter were not infinitely divisible, but instead made of some kind of smallest basic units, it should behave in certain ways, based on what had been learned about the laws of nature from experiments with larger objects.  As it turned out, the deductions were at least partly wrong; in actual experiments, matter behaved in some of the expected ways, but not in other ways.  With this kind of ambiguous evidence, people at first disagreed about whether matter did come in basic units:  if the deductions were wrong, at least one of the assumptions they were based on had to be wrong, and some people figured that the wrong assumption was that matter has basic units. 

 

As further experiments eventually made clear, that wasn’t the error.  The problem was that the known laws of nature, well established by experiment as they were, had been learned mainly from experiments involving large aggregates of atoms, and thus learned incompletely.  It turns out that large aggregates of atoms don’t exhibit certain behaviors as readily as smaller ones do, so that certain features of the laws of physics had remained obscure.  In general, the larger the chunks of matter one deals with, the less apparent certain atomic and subatomic features become.  Because of this, one can often predict macroscopic entities’ behavior accurately just by using 19th-century physical theory, which, as was realized later, is a simplification of a more complete theory of physics.  And since the earlier, classical theory is accurate enough in such cases, people still use it to make quicker and easier calculations of many macroscopic phenomena. 

 

How big does a physical system have to be before its “nonclassical” (or quantum-physical[Wikipedia]) features become too insignificant to take into account?  It depends on the features.  For many features, objects at least a few micrometers in size—about the size of many living cells—are large enough that their peculiarly quantum-physical features aren’t very evident.  But more of these features are extremely evident when dealing with objects about a thousandth of that size—a nanometer—or less.  One nanometer is just a few times larger than the width of one atom within a molecule or larger piece of matter. 

 

Objects whose size is well between these rough limits aren’t large enough for classical theory to describe all of their behavior very accurately, but they are large enough that some of their quantum-physical features are significantly less pronounced than those of individual atoms or molecules.  The last few decades’ advances in the science and technology of nanometer-size entities has clarified the importance of understanding matter at this intermediate mesoscale—a word which in this context means smaller than a few micrometers and bigger than a few nanometers.[Wikipedia]  As a report from the Department of Energy’s Mescoscale Science Subcommittee puts it,

 

The great scientific advances of the last decade and more, especially at the nanoscale, are ripe for exploitation. Seizing this key opportunity requires mastering the mesoscale, where classical, quantum, and nanoscale science meet. It has become clear that in many important areas the functionality that is critical to macroscopic behavior begins to manifest itself not at the atomic or nanoscale but at the mesoscale ….  Mesoscale behavior lies between the nanoscale world of atoms, molecules, and small assemblies with relatively perfect structure displaying simple behavior, and the macroscopic world of bulk materials with imperfect structure and endless variation, complexity, subtlety, and functionality.

 

With our recently acquired knowledge of the rules of nature that govern the atomic and nanoscales, we are well positioned to unravel and control the complexity that determines functionality at the mesoscale. The reward for breakthroughs in our understanding at the mesoscale is the emergence of previously unrealized functionality.  …  The enormous differences separating atoms and bulk materials appear at first sight to be irreconcilable. They are connected, however, by a sequence of mesoscale architectures and phenomena that form, step by step, a staircase reaching from atoms to bulk materials that can be experimentally observed, theoretically understood, and ultimately physically controlled. Mesoscale science entails the observation, understanding, and control of these intermediate-scale architectures and phenomena. It will ultimately lead to next-generation materials and technology that provide innovative solutions to pervasive societal problems including energy security, environmental sustainability, climate change, and enduring economic growth.

 

—From Quanta to the Continuum: Opportunities for Mesoscale Science[DoE], pp. 1, 6; 1, 3. 

 

The report just quoted describes six research priorities that the subcommittee identified in consultation with hundreds of colleagues in town hall meetings, webinars, and other web interactions.  These priorities are listed and elaborated in a slide presentation[SciTech Connect] by the subcommittee’s co-chairs that was prepared for a seminar at the Institute of Basic Science in Seoul, Korea and is now available from Los Alamos National Laboratory. 

 

  • Mastering Defect Mesostructure and its Evolution
  • Regulating Coupled Reactions and Pathway-Dependent Chemical Processes
  • Optimizing Transport and Response Properties by Design and Control of Mesoscale Structure
  • Elucidating Non-equilibrium and Many-Body Physics of Electrons
  • Harnessing Fluctuations, Dynamics and Degradation for Control of Metastable Mesoscale Systems
  • Directing Assembly of Hierarchical Functional Materials

 

“Defect mesostructure” refers to the structural imperfections in materials’ atomic arrangements mentioned in the subcommittee’s report.  Such imperfections are rarely evident in a typical nanometer-sized portion of matter, but are common in mesoscale portions.  The slide presentation mentions that new instruments enable imaging of the initiation and development of material damage from these defects.  Chemical processes are controlled by interfaces between materials, which are generally mesoscale distances apart.  Materials’ responses to various influences, including how electric charge and various forms of energy are transported through them, can be optimized if their mesoscale structure can be controlled.  The mesoscale is also the size at which electrons exhibit some significant nonequilibrium phenomena and many-electron interactions.  Control of materials that stay a relatively long time in a metastable state outside their most stable state may result in materials that repair themselves, so that devices made of them will last longer.  And directing mesoscale assembly of disparate materials by top-down and bottom-up approaches can lead to new materials that perform new functions. 

 

Several of the numerous mesoscale explorations of matter made in recent years are described in reports available through OSTI’s SciTech Connect and DOepatents. 

   Top

 

Understanding deformation and flow at the mesoscale

 

The ways that atoms tend to bond together into a crystalline solid generally lead to the atoms arranging themselves into a lattice, with every atom other than the ones at the crystal’s surface having nearest neighbors at the same distances in the same directions.  The smallest crystals only a few nanometers wide are most likely to have perfectly regular atomic arrangements, but mesoscale crystals have more room for irregularities of various sorts—atoms missing from some lattice points, extra atoms in between lattice points, planes of atoms extending only part way across the crystal instead of all the way across, or other dislocations. 

 

Atoms at or near such dislocations get rearranged when the solid is put under a large-enough stress, so that the site of the original irregularity becomes regular while the atomic arrangement at a nearby site becomes irregular, with newly dislocated atoms.  One can describe this kind of rearrangement in terms of moving atoms, as we’ve just seen, or in terms of the dislocations themselves, treating the dislocations as physical entities that are moved and possibly reshaped by the stress on the solid.  Equations that relate stress to dislocations in this second way, without reference to the numerous individual atoms and their exact positions, in effect describe the material as if it were a continuum whose features varied gradually from point to point instead of a “lumpy” aggregation of atoms.  Using such equations significantly simplifies efforts to calculate how a mesoscale crystal will deform under stress. 

 

Calculating how single crystals of metals deform by solving continuum equations was the aim of work done at Florida State University and described in the report “Statistical Mechanics Modeling of Mesoscale Deformation in Metals”[SciTech Connect].  According to the report’s author, the calculational framework developed by the Florida State group “is likely to form the basis for a predictive mesoscale plasticity theory capable of capturing the hardening behavior and dislocation patterning [in stressed metals]”.  The group also worked with the x-ray spectroscopy group at Oak Ridge National Laboratory to directly compare mathematical models of dislocation dynamics with three-dimensional experimental data for the first time, and contributed to the development of a computer program for mesoscale simulation that has become popular for modeling plasticity, the irreversible shape changes in materials caused by applied forces.  Problems still to be solved when the report was written include continuum-model representation of how particular types of dislocation move and interact, and of multiple dislocations in an arbitrarily stressed material. 

 

 

Figure 1.  Results of simulating the indentation of a copper block to a depth of ~13.0 nanometers, using a model that treats dislocations as basic physical entities instead of as repositionings of atoms.  Top:  Dislocation structure.  Bottom:  Norm (“size”) of a tensor[Wikipedia] that represents the density of dislocations, which varies from point to point in the block.  (From “Statistical Mechanics Modeling of Mesoscale Deformation in Metals”[SciTech Connect], p. 11.) 

 

Metal matrix composites, or MMCs, are a type of material quite different from single metal crystals.  Whereas the single crystals consist of the same material throughout, MMCs are heterogeneous interminglings of a metal or metallic-alloy matrix with small volumes of another (often nonmetallic) reinforcing material.  The properties of MMCs can be tailored to suit a wide range of applications, in some of which the heterogeneous composites are likely to undergo rapid deformation.  This is explained in a report[SciTech Connect] from Sandia National Laboratories on a “Laboratory Directed Research and Development” project[USC2791, 2791a, 2791b, 2792, 2793] to model MMCs in a manner similar to the Florida State group’s analysis of single metal crystals. 

 

MMCs perform well when they’re rapidly deformed, but our understanding of exactly what happens in them during these deformations has been limited.  Experiments in which an MMC is only subjected to static or quasistatic loads tell us little about its rapid deformation.  And even rapid-deformation experiments give us limited information, as long as we lack a detailed understanding of the composite’s internal processes.  “The most common and versatile experimental techniques” for these conditions, according to the Sandia report, involve stressing materials along a single dimension, but the resulting deformations, while approximately one-dimensional from a large-scale point of view, are actually three-dimensional in the material’s microstructure.  Researchers find that how an MMC deforms is affected the most by its heterogeneities when the deformations are about the same size as the largest pure volumes of the MMC’s component materials. 

 

Whereas the Florida State group’s mesoscale approximation took the effects of atomic-lattice dislocations into account by treating them as basic physical entities in a continuum, the Sandia group’s project involved mesoscopically modeling the effects of other smaller-scale phenomena:  the inhomogeneous spatial distribution of reinforcing-material pieces in the MMC’s metal or alloy matrix, the pieces’ irregular shapes, and the texture, grain size, and shape of the matrix material itself.  The project also involved calculating and measuring what happens when an inhomogeneous material undergoes one type of rapid deformation:  sound-wave vibrations generated by impact.  Inhomogeneity in the material makes a particular contribution to how the sound waves’ speed varies with their frequency. 

 

 

Figure 2.  Effect of compressive pulse, lasting about 200 nanoseconds, on a composite bar of aluminum matrix containing 5% of silicon carbide particulates, as simulated with a mathematical model that explicitly approximates the bar’s microstructure.   Top left, top right, bottom left, and bottom right pictures illustrate pulse speeds at various points in the bar at, respectively, 80, 180, 280, and 380 nanoseconds after the pulse started.  (From “LDRD final report : mesoscale modeling of dynamic loading of heterogeneous materials”[SciTech Connect], p. 16.) 

 

It’s not only solid matter that can be usefully approximated as continuous instead of atomic.  As another report from Lawrence Livermore National Laboratory points out, flows of fluids that contain solid particles “are common in nature and industry”.  The approach reported in “Mesoscale simulations of particulate flows with parallel distributed Lagrange multiplier technique” to the 2010 International Conference on Multiphase Flow[SciTech Connect] and published in the journal Computers and Fluids [SciTech Connect] involved, first, solving equations for the particle-carrying fluid flow without accounting for the particles’ solidity, and then modifying that solution by treating the solids as rigid objects that can exert frictional forces on each other whenever they collide.  Although real solids always deform in response to forces, if the particle deformation is small the no-deformation approximation can usefully simplify the calculation.  The mathematical model was tested for cases of fluids flowing through a fixed bed of solid particles, a solid sphere falling in a fluid, and a set of particles flowing in a fluid through a container opening.  Fluid behaviors of the first type are “of crucial importance in industry and nature”, with the determination of pressure drop as a function of flow rate, geometrical constraints of the solid-particle bed, and physical properties of the bed material being “very critical … in hydraulic and pneumatic devices”.  In the third case, that of particles flowing in a fluid through an opening, particles interact with each other.  The researchers compared their simulation of this case with an actual experiment performed at the University of California, Berkeley. 

 

Other things being equal, the computer took somewhat more steps to find solutions of the fluid-motion equations for fluids that had less viscosity.  This, according to the report, could be explained by deficiencies in the way the model represents fluid flow near rigid walls, particularly where solid particles in the fluid are relatively dense; the model’s inadequate accounting for details in the structure of turbulent flow could also be involved.  Still, the tests indicate to the report’s authors that the model’s overall performance “is very good and promises to be a valuable tool” for two kinds of flow problems:  simulating fluids that contain up to a few hundred thousand particles, and calibrating other models of fluid-particle interaction that involve so many more particles (e.g., billions) that the only practical way to analyze their motion is to not account for the particles individually, but treat them as a second fluid continuum. 

 

Figure 3.  Interacting particles flowing in a fluid through an opening:  comparison of actual experiment with a simulation that treats the particles as nondeformable, rigid objects that exert frictional forces on each other when they collide.  (From “Mesoscale simulations of particulate flows with parallel distributed Lagrange multiplier technique”[SciTech Connect, SciTech Connect], p. 43 of 47 (slide presentation)/p. 13 (paper).) 

 

Useful mesoscale models can be constructed for other processes in addition to those involving the mechanical distortion of solids or the flow of molecular fluids.  One such mathematical model was made at the Pennsylvania State University under a subcontract to Lawrence Livermore National Laboratory in a project to explore phenomena involving the transport of electric charge in batteries.  In a battery with two electrodes, each electrode is in contact with an electrolyte and reacts with it.  The reactions result in one electrode (the anode) gaining electrons while the other (the cathode) loses electrons, so that electrons will flow from the anode to the cathode if they’re connected by a conducting path (typically one that runs through a device to be powered by the battery).  The rate of electron transfer at the electrode-electrolyte interfaces is determined by the electrochemical reactions occurring there, and is affected by the way the electrodes are distorted when electrons flow into or out from them.  

 

The development of mesoscale computational models for these processes is described in the report “Mesoscale Modeling Framework Design: Subcontract Report”[SciTech Connect].  Like some other mesoscale models, these treat the materials that make up a battery’s electrodes and electrolyte as continuous instead of accounting for the locations of their individual atoms.  Preliminary analyses of electrodes made of heterogeneous materials show how strain energy is distributed within them and how different orientations of boundaries between the different materials would increase or lessen the strain energy.  The report also notes the importance of accounting for the distribution of the electric potential[Wikipedia] when modeling charge transport in a battery.  A model that accounts for electric potential, charge flow, and the way electrodes distort under stress “will enable the computer simulations … to determine the optimum condition or architecture to maximize the performance of batteries.” 

   Top

 

Experiments and tools

 

One Lawrence Livermore group has combined a mesoscale modeling effort with experiments of their own to see how much could be learned about various mechanical properties of elastic polymers just from measuring their molecular weight distributions with a new technique.  The polymer they studied, polydimethylsiloxane (PDMS)[Wikipedia], is the most widely used silicon-based organic polymer.  PDMS molecules consist of a flexible “backbone” of alternating silicon and oxygen atoms, with each of the silicon atoms bound to two methyl groups (except for the silicons at the ends of the molecule, both of which are bound to three methyls), each methyl group being one carbon atom bound to three hydrogen atoms.  Polydimethylsiloxane gets its name from this structure:  the “poly” refers to the indefinitely many backbone units that make up molecules of this type of material, the “dimethyl” to the two methyl groups attached to each backbone unit, and the “siloxane” to the silicon and oxygen in each unit.  Different-length PDMS molecules can combine by reacting with each other in a variety of ways to produce very different cross-linked network structures with different material properties. 

 

As the Livermore group’s report notes, producing networks with specific properties

 

… can be an Edisonian trial and error task, and there has been much effort spent to determine the structural motifs that control and tune resultant properties.  One key observation from much of this work is that the combination of the complex nature of the siloxane elastomeric composites with frequently complex analytical and modeling techniques often presents significant challenges in the attempt to correlate various data with specific material constituents or network structural motifs.  ...  In response to the above complication, there is a recognized need to investigate the physical chemical behavior of simple, idealized materials in an effort to identify the analytical “origins of response.”

 

The researchers contributed to meeting this need by analyzing simple PDMS-based materials, some being monomodal networks of nominally identical molecules and the rest bimodal networks that combined shorter and longer molecules.  The researchers measured how much each material stretched under given forces, and compared the results with a mesoscale computer model of the materials’ mechanical properties.  The model required input about the distribution of molecular masses in each material, which the researchers determined from additional, indirect measurements using a long-known technique[Wikipedia] that involves magnetism in the particles that make up the molecules. 

 

Many atomic nuclei are magnetic, with different species of nucleus having different magnetization strengths.  Each atom’s electrons are also magnets, having a south-pole-to-north-pole axis as all magnets do.  All these magnets will tend to align their axes more or less parallel or antiparallel to each other, not just in any old directions but in one of a finite number of possible arrangements.  And the entire set of magnetic nuclei and electrons in each molecule will align in one of a finite number of ways with any external magnetic field from a source outside the material.  The alignment directions of these electrons and magnetic nuclei can oscillate, with the oscillations tending to occur at any of a finite set of natural frequencies that depend on how strongly the individual electrons or nuclei interact, which in turn depends on what species of nuclei are in the molecules and on how the nuclei are arranged among the electrons.  When the alignments of all these particles and their magnetic fields do oscillate, the magnetic fields’ oscillations produce oscillating electric fields, which in turn produce oscillating magnetic fields, and so on, resulting in electromagnetic waves that spread out from the oscillating particles into the surrounding space, whose frequencies match the frequencies of the particles that generate them—frequencies in the radio-wave range of the electromagnetic spectrum. 

 

The particles can be driven into producing such waves if the external magnetic field they’re exposed to is itself that of a radio wave that oscillates at those frequencies.  Thus, if a material is exposed to an electromagnetic wave that oscillates with particular frequencies, and magnetic nuclei in the material are found to resonate and produce new electromagnetic waves of those same frequencies, one can determine from the resonance frequencies what kinds of magnetic nuclei are present, in what amounts, and some information about how they’re arranged.  This examination of nuclear magnetic resonance in PDMS-based molecules is how the Livermore researchers determined the distribution of molecular mass in the materials they examined.  The distribution data, used in the researchers’ mesoscale model, let them “reasonably predict a practical mechanical property” of the materials they analyzed.  They give a detailed description of their work in their paper “Linking Network Microstructure to Macroscopic Properties of Siloxane Elastomers Using Combined Nuclear Magnetic Resonance and Mesoscale Computational Modeling”[SciTech Connect], which was published in the journal Macromolecules.  They concluded that

 

from a practical perspective, the results from the present model bimodal networks can serve as a training set that will enable us to further test this methodology on more complex networks such as bimodal networks with extremely short chain lengths, silica-reinforced filled networks, and engineering materials that have been investigated by this group in detail previously.  We anticipate that this potentially non-destructive, non-invasive methodology can also be applied in a straightforward fashion to, for example, in situ monitoring of subtle aging mechanisms or the effects of material usage in harsh environments. 

 

While this project involved mesoscale mathematical models, researchers at Argonne National Laboratory investigated a new mesoscale fabrication process.  Their paper in Physical Review Letters, entitled “Optically Directed Assembly of Continuous Mesoscale Filaments”[SciTech Connect], describes how they synthesized nanoscale particles of gold and carbon into mesowidth filaments using a laser trap[Wikipedia]. 

 

Since laser beams are beams of electromagnetic waves, they can exert forces on small objects that respond to electromagnetic forces.  One force comes from a beam’s tendency to push objects downstream along its path.  But if a laser beam’s intensity increases toward its central axis, objects that interact appropriately will also be forced toward that axis.  Furthermore, a focused beam will be more intense where the beam is narrowest than it is either upstream or downstream from that point, so objects that are themselves upstream or downstream from there will likewise be forced toward where the beam is narrowest.  There is thus a point in the beam, along its axis and just downstream of where it is most narrow, at which an appropriately interacting object will get a downstream push that exactly balances the upstream push along the beam’s intensity gradient, and the same object in the immediate neighborhood of that point will be pushed toward it, thus trapping the object.  The object can be moved around by changing where the beam is most narrowly focused. 

 

In the Argonne investigation, though, the laser beam didn’t just trap or move small particles.  A laser trap was focused onto droplets in which gold and carbon particles a few nanometers in diameter were suspended without being dissolved[Wikipedia].  The laser beam, which was focused within about 30 micrometers of the droplets’ edges, did more than hold the gold and carbon nanoparticles in place:  it would fuse the particles together there.  If the beam remained stationary, a long filament would grow out of the top of the focal region as fused nanoparticles moved upward to be replaced by new nanoparticles, which would in turn be fused to the bottom of the rising filament.  If the beam’s focal region were moved around within the droplet instead, nanoparticles along the path of the focal region would fuse to the existing filament wherever the focal point moved. 

 

Careful observation and computational analysis enabled the Argonne group to narrow down the range of possible filament-forming mechanisms, and to note the following.  First, calculations indicated that the laser beam doesn’t just push the suspended particles around and heat them up, but also heats up the liquid in the droplet so that the suspension is driven into a convective circulation.  This would lead to the newly fused portions of the filament rising from the laser’s focal region and new suspended particles being drawn into the focal region by the convection current.  Second, because suspensions containing only carbon or only gold nanoparticles don’t form filaments, some kind of cooperative effect between the two particle types seems to be involved.  Third, filament images taken with a transmission electron microscope[Wikipedia] showed that the filaments’ gold nanoparticles are encapsulated by an opaque material, “presumably carbon”.  In view of this, the researchers hypothesized that the carbon nanoparticles, at the 400-450 K temperatures[Wikipedia] that they reach when heated by the laser trap, act like a liquid that wets the gold nanoparticles, thus binding them together. 

 

As one way to test this hypothesis against known physical laws, a computer model based on laws of fluid flow and crystal lattice behavior was used to simulate the filament formation.  The speeds at which the model indicated filaments would form for various gold/carbon nanoparticle ratios roughly matched the trends found in the actual experiments: 

 

Given the simplicity of the model, the agreement between experimental and calculated filament deposition rates is remarkably good ... .  One sees that the growth kinetics increases significantly as gold fraction is varied from 90% to 10%.  The convection frequencies are dictated by the location of the particle ... and are higher near the optical trap where particle velocities are highest ... . Hence, the probability of nucleation of the filament is most likely at the optical trap. For given colloidal composition, increased laser power results in increased convection velocities and higher deposition rates. For given laser power, as the gold fraction decreases, the gold-carbon and carbon-carbon collision frequencies become significant. This leads to increased gold-carbon wetting and higher filament growth rates at lower gold fractions. The net growth rate is dictated by the relative frequencies of coalescence and convection. Tuning these interactions at the nanoscale is key to controlling the dynamics of formation of structures from the nanoscale to the mesoscale.

 

Figure 4.  Optically directed assembly of mesoscale structures.  (a) Optical image of a gold-carbon filament obtained with a stationary optical trap[Wikipedia]. The thick black feature is a 7-micrometer diameter carbon fiber.  (b) Hook-shaped gold-carbon structure form by slow translation of the trap.  (c) Lobe-shaped filament form by translating the trap with varying speeds. Slower speeds cause larger filament diameters and vice versa.  (d) Transmission electron microscope image[Wikipedia] of a filament grown on a grid. The sharp tip is caused when the filament stops growing due to local colloid depletion.  (e) Close-up of (d) showing a dense ‘‘neck.’’ This illustrates how filament radial dimensions are influenced by the radial component of the trap potential.  (f ) Schematic model of optically directed assembly: nanoparticles drawn into the optical trap from below by convection irreversibly bind at the focus.  (After “Optically Directed Assembly of Continuous Mesoscale Filaments”[SciTech Connect], p. 095501-1.) 

 


 Figure 5.  Gold-carbon nanoparticle interactions.  (a) Transmission electron microscope image[Wikipedia] of the extreme tip of the gold-carbon filament in Figure 4 (d) above.  (b) Transmission electron microscope image of encapsulated gold nanoparticles within the tip.  (c) Initial gold-carbon nanoparticle configuration for a molecular dynamics simulation.  (d) molecular dynamics simulation result after 10 nanoseconds showing wetting of a gold nanoparticle by carbon atoms in the 450 K range. These results indicate the possibility of encapsulation of gold nanoparticles by carbon.  (e) Initial configuration of gold-gold interaction for the molecular dynamics simulation.  (f) Snapshot showing result after 10 nanoseconds at 500 K.  (After “Optically Directed Assembly of Continuous Mesoscale Filaments”[SciTech Connect], p. 095501-3.) 

The researchers note that structures of filaments assembled by this method could be “excellent candidates for various applications” such as a technique for detecting and distinguishing very small amounts of different materials[Wikipedia] adsorbed onto the filaments.  They also note that “many fundamental questions remain” about how the method of filament assembly works, such as the exact nature of the mechanism that binds the nanoparticles into filaments, and the assembly of other kinds of nanoparticle combinations besides the aforementioned gold-carbon ones, or the silver-carbon combinations that they’ve also demonstrated. 

 

Certifying that mesoscale objects are made to precise specifications requires the ability to reliably measure the objects.  According to a patent issued to Sandia Corporation (“Mesoscale hybrid calibration artifact”[DOepatents]), typical vision-based commercial measuring equipment, despite its ability to resolve mesoscale detail, only has traceable accuracy at the micrometer level.  The patent notes a need to improve on this, and how this can be achieved with an invention it describes:  3-d calibration artifacts that can be measured with a high-accuracy measurement tool to provide both traceable mesoscale accuracy and the sharp edges needed for vision-based probing.  The patent also describes an etching method for making these artifacts. 

 

 

Figure 6.  After the patent “Mesoscale hybrid calibration artifact”[DOepatents], in which this is Figure 1, which is described in the section “Detailed Description of the Invention”:  A wafer (10) into which several linear arrays (12) of rectangular pits and rectangular arrays (14) of rectangular pits have been etched.  An entire wafer may be used as a calibration artifact, or the wafer may be divided to provide several calibration artifacts, where an artifact is defined as the substrate and the etched features that have been formed on it.   While the wafer and its etched features may be macroscopic (on the order of tenths of a meter and of millimeters, respectively), the etched features’ edges can be measured with high-accuracy tools to provide traceable accuracy better than a tenth of a micrometer. 

   Top

 

Mesoscale science for defense

 

The slide presentation to Korea’s Institute for Basic Science mentioned at the beginning of this article gave a general view of opportunities for mesoscale science.  One of that presentation’s authors prepared a more specific presentation on applying mesoscale research to national defense for a workshop in New York the next year, which focused on the first experiments to be done at the significantly upgraded National Synchrotron Light Source (NSLS-II) at Brookhaven National Laboratory.  NSLS-II is a source of intense beams of x-rays, ultraviolet light, and infrared light that’s capable of resolving mesoscale and nanoscale details in matter.  The new slide presentation, which was entitled “Co-Design at the Mesoscale: Opportunities for NSLS-II”[SciTech Connect], shows that the NSLS-II’s capabilities would allow observation at the mesoscale of material synthesis, composition, structure, and defect development.  It also discusses MaRIE (for “Matter-Radiation Interactions in Extremes”), a proposed upgrade at the Los Alamos Neutron Science Center (LANSCE).  MaRIE is designed to inform the development and manufacture of materials that will perform predictably in extreme environments.  This is of particular interest for predicting how explosives’ dynamics will be affected by heterogeneities in their own composition and what will happen over time with the materials in the remaining nuclear-weapons stockpile. 

 

MaRIE is briefly described by different Los Alamos authors in a brochure entitled “MaRIE: A facility for time-dependent materials science at the mesoscale”[SciTech Connect].  The brochure mentions that MaRIE will be able to produce a movie over timescales of nanoseconds to microseconds through thick samples of materials as they undergo a dynamic event.  MaRIE will also include facilities for synthesizing and characterizing materials with design and data visualization tools that focus on the mesoscale.  Much more detail is given by a larger set of authors in a slide presentation entitled “MaRIE 1.0: The Matter-Radiation Interactions in Extremes Project, and the Challenge of Dynamic Mesoscale Imaging”[SciTech Connect], which was prepared for the 11th LANSCE School on Neutron Scattering. 

 

To more quickly qualify, certify, and assess materials in the nuclear-weapons stockpile, the way the materials’ mesoscale structure affects their time-dependent properties needs to be determined.  MaRIE is to do this with an x-ray free-electron laser, simultaneous optical, proton, and electron probes, close linkages of synthesis, fabrication, and characterization, and full integration with advanced theory, mathematical modeling, and computing.  “After time-dependent control of process, structure, and properties during synthesis of materials at the mesoscale,” says the presentation, “MaRIE will subject them to extreme environments” so their dynamic evolution can be observed at the mesoscale—below the size of the material granules—and connected to their “macroscale” behavior.  The presentation describes this in much additional detail, especially the optical techniques to be used and how these determine design requirements for the x-ray source. 

 


 Figure 7.  Slide 7 from the presentation “MaRIE 1.0: The Matter-Radiation Interactions in Extremes Project, and the Challenge of Dynamic Mesoscale Imaging”[SciTech Connect]. 

  Top

 

Apparent requirements for progress

 

The several reports we’ve just discussed describe a sample of recent work along some of the lines that were recommended in the Mescoscale Science Subcommittee report that was quoted at the beginning of this article (From Quanta to the Continuum: Opportunities for Mesoscale Science).  The Subcommittee also deemed other things necessary for immediate and long-term progress in mesoscale science and technology, which they summarized on page 68 in their report’s conclusion.  

 

Realizing mesoscale science opportunities requires advances not only in our knowledge but also in our ability to observe, characterize, simulate, and ultimately control matter. A key need is the seamless integration of theory, modeling, and simulation with synthesis and characterization. Treating the complexity of mesoscale phenomena, often including several nanoscale degrees of freedom and structural or functional units, requires theory and simulation spanning space and time scales over many orders of magnitude. Analytical theory is needed to find and formulate new organizing principles that describe emergent mesoscale phenomena arising from many coupled and competing degrees of freedom. Computational advances are required to extend nanoscale simulations to the mesoscale and beyond. Theory and simulation have an unprecedented opportunity to guide the complex multistep synthesis and directed assembly of mesoscale architectures.

 

Mesoscale measurements that are dynamic, in situ, and multi-modal must replace the static single-property characterization now common in macro and nano regimes. The mesoscale is an inherently dynamic regime, where energy and information captured at the nanoscale are processed and transformed to create new macroscale outcomes. Given the complexity and dynamism of mesoscale phenomena, multiple, simultaneous measurements of transient phenomena are essential. Spatially diverse measurements capturing coupled transformations in different parts of the mesoscale assembly are critical. Because complex mesoscale ensembles are large enough to have defects, fluctuations and statistical variation, they inevitably display a range of behaviors that requires many measurements to capture the distribution of possible outcomes and not just a single measurement of a “representative” sample.

 

Finally, the ability to design and realize the complex and composite materials of the mesoscale will require qualitative advances in how we synthesize materials, both in moving from serendipitous to directed discovery and in gaining the ability to assemble and control materials systems and architectures from smaller structural and functional units. The integration of top-down and bottom-up approaches, now quite distinct and representing different operating principles and spatial scales, is key to mesoscale synthesis. Bridging the gap and allowing these two powerful approaches to influence, complement, and reinforce each other represent major synthesis challenges and opportunities of mesoscale science.

   Top

 

References

 

Wikipedia

(There is an additional common use of “mesoscale” for a different intermediate scale in meteorology, between one kilometer (1,000 meters) and one megameter (1,000,000 meters).) 

 

Research organizations

  Top

 

Reports available through OSTI’s SciTech Connect

  • “Quanta to the Continuum: Opportunities for Mesoscale Science” [Metadata] (slides)
  • “Statistical Mechanics Modeling of Mesoscale Deformation in Metals” [Metadata]
  • “LDRD final report : mesoscale modeling of dynamic loading of heterogeneous materials” [Metadata]
  • “Mesoscale Modeling Framework Design: Subcontract Report” [Metadata]
  • “Mesoscale simulations of particulate flows with parallel distributed Lagrange multiplier technique” [Metadata for conference paper, journal manuscript]
  • “Linking Network Microstructure to Macroscopic Properties of Siloxane Elastomers Using Combined Nuclear Magnetic Resonance and Mesoscale Computational Modeling” [Metadata]
  • “Optically Directed Assembly of Continuous Mesoscale Filaments” [Metadata]
  • “Co-Design at the Mesoscale: Opportunities for NSLS-II” [Metadata]
  • “MaRIE 1.0: The Matter-Radiation Interactions in Extremes Project, and the Challenge of Dynamic Mesoscale Imaging” [Metadata] (slides)
  • “MaRIE: A facility for time-dependent materials science at the mesoscale” [Metadata] (brochure)

 

Patent available through OSTI’s DOepatents

  • “Mesoscale hybrid calibration artifact” [Metadata]

 

Conferences

  Top

 

Journals

  • Computers and Fluids.   “Mesoscale simulations of particulate flows with parallel distributed Lagrange multiplier technique” appears in vol. 48, no. 1 (March 22, 2011), pp. 16-29. 
  • Physical Review Letters.  “Optically Directed Assembly of Continuous Mesoscale Filaments” appears in vol. 106, issue 9 (March 4, 2011), 095501. 
  • Macromolecules.  “Linking Network Microstructure to Macroscopic Properties of Siloxane Elastomers Using Combined Nuclear Magnetic Resonance and Mesoscale Computational Modeling” appears in vol. 44, no. 20 (September 29, 2011), pp. 8106-8115. 

 

Additional references

  Top

 

View Past "In the OSTI Collections" Articles.

 

 

Last updated on Wednesday 27 July 2016