From Quanta to the Continuum: Opportunities for Mesoscale Science

Home , Kate (text editor)

About the Department of Energy’s Basic Energy Sciences Program Basic Energy Sciences (BES) supports fundamental research to understand, predict, and ultimately control matter and energy at the electronic, atomic, and molecular levels. This research provides the foundations for new energy technologies and supports DOE missions in energy, environment, and national security. The BES program also plans, constructs, and operates major scientific user facilities to serve researchers from universities, national laboratories, and private institutions.

About the “Basic Research Needs” Report Series Over the past ten years, the Basic Energy Sciences Advisory Committee (BESAC) and BES have engaged thousands of scientists from academia, national laboratories, and industry from around the world to study the current status, limiting factors, and specific fundamental scientific bottlenecks blocking the widespread implementation of alternate energy technologies. The reports from the foundational Basic Research Needs to Assure a Secure Energy Future workshop, the following ten “Basic Research Needs” workshops, the panel on Grand Challenge science, and the summary report New Science for a Secure and Sustainable Energy Future detail the key basic research needed to create sustainable, low carbon energy technologies of the future. These reports have become standard references in the scientific community and have helped shape the strategic directions of the BES-funded programs (science.energy.gov/bes/news-and-resources/reports/basic-research-needs/). 1 Science for Energy Technology: Strengthening the Link between Basic Research and Industry 2 New Science for a Secure and Sustainable Energy Future 3 Directing Matter and Energy: Five Challenges for Science and the Imagination 4 Basic Research Needs for Materials under Extreme Environments 5 Basic Research Needs: Catalysis for Energy 6 Basic Research Needs for Electrical Energy Storage 7 Basic Research Needs for Geosciences: Facilitating 21st Century Energy Systems 8 Basic Research Needs for Clean and Efficient Combustion of 21st Century Transportation Fuels 9 Basic Research Needs for Advanced Nuclear Energy Systems 10 Basic Research Needs for Solid-State Lighting 11 Basic Research Needs for Superconductivity 12 Basic Research Needs for Solar Energy Utilization 13 Basic Research Needs for the Hydrogen Economy 14 Basic Research Needs To Assure A Secure Energy Future

About this Report Over the course of this study the Mesoscale Science Subcommittee of the Basic Energy Sciences Advisory Committee engaged hundreds of colleagues in town hall meetings, webinars, and web site interactions in order to identify important and timely priority research directions for mesoscale science, as well as the capabilities required to address these challenges. This report outlines the need, the opportunities, the challenges, and the benefits of mastering mesoscale science. Further information can also be found at www.meso2012.com.

On the cover Simulation showing water molecules (purple) and acid groups (yellow and red) in the confined environment of a separations membrane (green) relevant to many energy technologies. To understand and control the selectivity of the separation process and the degradation of membranes and electrolytes in complex electrochemical environments, we need to understand the mesoscale membrane/liquid interface in addition to understanding molecular level details of transport and reactivity. The figure illustrates these different scales. Source: Michel Dupuis and Ram Devanathan, Pacific Northwest National Laboratory. FROM QUANTA TO THE CONTINUUM: OPPORTUNITIES FOR MESOSCALE SCIENCE

A Report from the Basic Energy Sciences Advisory Committee

Chair: John Hemminger University of California, Irvine

U.S. Department of Energy

September 2012

Prepared by the BESAC Subcommittee on Mesoscale Science

Co-chairs: George Crabtree Argonne National Laboratory and University of Illinois at Chicago and John Sarrao Los Alamos National Laboratory New Science for a Secure and Sustainable Energy Future

Basic Energy Sciences Advisory Committee

Chair: John Hemminger (University of California-Irvine)

Simon Bare (UOP LLC) William Barletta (Massachusetts Institute of Technology) Nora Berrah (Western Michigan University) Gordon Brown (Stanford University) Sylvia Ceyer (Massachusetts Institute of Technology) Yet-Ming Chiang (Massachusetts Institute of Technology) George Crabtree (Argonne National Laboratory and University of Illinois at Chicago) Peter Cummings (Vanderbilt University) Beatriz Roldan Cuenya (University of Central Florida) Frank DiSalvo (Cornell University) Roger French (Case Western Reserve University) Bruce Gates (University of California, Davis) Laura Greene (University of Illinois at Urbana-Champaign) Allen Goldman (University of Minnesota) Ernie Hall (GE Global Research) Sharon Hammes-Schiffer (Pennsylvania State University) Bruce Kay (Pacific Northwest National Laboratory) Kate Kirby (American Physical Society) Max Lagally (University of Wisconsin) William McCurdy, Jr. (University of California, Davis and Lawrence Berkeley National Laboratory) Mark Ratner (Northwestern University) John Richards (California Institute of Technology) John Spence (Arizona State University) Douglas Tobias (University of California, Irvine) John Tranquada (Brookhaven National Laboratory)

Designated Federal Officer: Harriet Kung Associate Director of Science for Basic Energy Sciences iv ii New Science for a Secure and Sustainable Energy Future

Subcommittee on Mesoscale Science

Co-chairs: George Crabtree, Argonne National Laboratory (University of Illinois at Chicago) John Sarrao, Los Alamos National Laboratory

Paul Alivisatos (University of California, Berkeley) William Barletta* (Massachusetts Institute of Technology) Frank Bates (University of Minnesota) Gordon Brown* (Stanford University) Roger French* (Case Western Reserve University) Laura Greene* (University of Illinois at Urbana-Champaign) John Hemminger** (University of California, Irvine) Marc Kastner (Massachusetts Institute of Technology) Bruce Kay* (Pacific Northwest National Laboratory) Jennifer Lewis (University of Illinois at Urbana-Champaign) Mark Ratner* (Northwestern University) Anthony Rollett (Carnegie Mellon University) Gary Rubloff (University of Maryland) John Spence* (Arizona State University) Douglas Tobias* (University of California, Irvine) John Tranquada* (Brookhaven National Laboratory)

* BESAC Member ** BESAC Chair

v iii From Quanta to the Continuum: Opportunities for Mesoscale Science From Quanta to the Continuum: Opportunities for Mesoscale Science

Table of Contents

Basic Energy Sciences Advisory Committee ii Subcommittee on Mesoscale Science iii Contents v Executive Summary 1

Introduction 3 Hallmarks of Mesoscale Behavior 3 Mesoscale Architectures 6 Periodic Lattices 6 Chemical Bonds 8 Membranes and Cells 8 The Gap Between Top Down and Bottom Up 9 The Meso Opportunity 11

Priority Research Directions for Mesoscale Science 13 Mastering Defect Mesostructure and its Evolution 14 Regulating Coupled Reactions and Pathway-Dependent Chemical Processes 20 Optimizing Transport and Response Properties by Design and Control of Mesoscale Structure 27 Elucidating Non-Equilibrium and Many-Body Physics of Electrons 31 Harnessing Fluctuations, Dynamics, and Degradation for Control of Metastable Mesoscale Systems 36 Directing Assembly of Hierarchical Functional Materials 43

Capability Gaps for Mesoscale Science 52 Synthesis Capabilities 54 Characterization Capabilities 56 Theory and Simulation Capabilities 59 Workforce Considerations 62

Realizing the Meso Opportunities 65

Conclusion 67

Appendix: Community Input to the BESAC Meso Subcommittee Process 71

v From Quanta to the Continuum: Opportunities for Mesoscale Science From Quanta to the Continuum: Opportunities for Mesoscale Science

Executive Summary

We are at a time of unprecedented challenge and opportunity. Our economy is in need of a jump start, and our supply of clean energy needs to dramatically increase. Innovation through basic research is a key means for addressing both of these challenges. 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, where defects, interfaces, and non-equilibrium structures are the norm. 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 present report explores the opportunity and defines the research agenda for mesoscale

• Imagine the ability to manufacture at the mesoscale: that is, the directed assembly of mesoscale structures that possess unique functionality that yields faster, cheaper, higher performing, and longer lasting products, as well as products that have functionality that we have not yet imagined.

• Imagine the realization of biologically inspired complexity and functionality with inorganic earth-abundant materials to transform energy conversion, transmission, and storage.

• Imagine the transformation from top-down design of materials and systems with macroscopic building blocks to bottom-up design with nanoscale functional units producing next-generation technological innovation.

This is the promise of mesoscale science.

science—discovering, understanding, and controlling interactions among disparate systems and phenomena to reach the full potential of materials complexity and functionality. The ability to predict and control mesoscale phenomena and architectures is essential if atomic and molecular knowledge is to blossom into a next generation of technology opportunities, societal benefits, and scientific advances.

1 From Quanta to the Continuum: Opportunities for Mesoscale Science

Mesoscale science and technology opportunities build on Realizing the mesoscale opportunity requires advances the enormous foundation of nanoscience that the scientific not only in our knowledge but also in our ability to observe, community has created over the last decade and continues characterize, simulate, and ultimately control matter. to create. New features arise naturally in the transition Mastering mesoscale materials and phenomena requires to the mesoscale, including the emergence of collective the seamless integration of theory, modeling, and simu- behavior; the interaction of disparate electronic, mechan- lation with synthesis and characterization. The inherent ical, magnetic, and chemical phenomena; the appearance complexity of mesoscale phenomena, often including of defects, interfaces and statistical variation; and the self- many nanoscale structural or functional units, requires assembly of functional composite systems. The meso- theory and simulation spanning multiple space and time scale represents a discovery laboratory for finding new scales. In mesoscale architectures the positions of indi- science, a self-assembly foundry for creating new func- vidual atoms are often no longer relevant, requiring new tional systems, and a design engine for new technologies. simulation approaches beyond density functional theory and molecular dynamics that are so successful at atomic The last half-century and especially the last decade have scales. New organizing principles that describe emergent witnessed a remarkable drive to ever smaller scales, mesoscale phenomena arising from many coupled and exposing the atomic, molecular, and nanoscale structures competing degrees of freedom wait to be discovered and that anchor the macroscopic materials and phenomena applied. Measurements that are dynamic, in situ, and multi- we deal with every day. Given this knowledge and capa- modal are needed to capture the sequential phenomena bility, we are now starting the climb up from the atomic of composite mesoscale materials. Finally, the ability to and nanoscale to the greater complexity and wider hori- design and realize the complex materials we imagine will zons of the mesoscale. The constructionist path up from require qualitative advances in how we synthesize and atomic and nanoscale to mesoscale holds a different kind fabricate materials and how we manage their metastability of promise than the reductionist path down: it allows us to and degradation over time. We must move from serendipi- re-arrange the nanoscale building blocks into new combi- tous to directed discovery, and we must master the art of nations, exploit the dynamics and kinetics of these new assembling structural and functional nanoscale units into coupled interactions, and create qualitatively different larger architectures that create a higher level of complex mesoscale architectures and phenomena leading to new functional systems. functionality and ultimately new technology. The reductionist journey to smaller length and time scales gave us While the challenge of discovering, controlling, and manip- sophisticated observational tools and intellectual under- ulating complex mesoscale architectures and phenomena standing that we can now apply with great advantage to to realize new functionality is immense, success in the the wide opportunity of mesoscale science following a pursuit of these research directions will have outcomes bottom-up approach. with the potential to transform society. The body of this report outlines the need, the opportunities, the challenges, and the benefits of mastering mesoscale science.

2 From Quanta to the Continuum: Opportunities for Mesoscale Science

Introduction

The immense diversity of materials in the macroscopic world—hard, soft, viscous, conducting, insulating, magnetic, liquid, and gaseous—is made up of only a hundred or so distinct kinds of atoms representing the elements of the periodic table. The differences in the size, complexity, and operating principles of atoms and macroscopic materials are enormous. Atoms are one-tenth of a nanometer in size, have a relatively simple structure composed of negatively charged electrons circulating a positively charged nucleus in well-defined orbits, and obey the “digital” rules of quantum mechanics, where energy, charge, and mass are quantized in indivisible fundamental units. The macroscopic materials we encounter every day are tens of microns or more in size—at least 100,000 times larger than atoms—and display dramatically more complex structures and behavior obeying the “analog” rules of continuous classical mechanics, where energy, charge, and matter are, for all practical purposes, continuous and infinitely divisible. The enormous differences separating atoms and bulk materials appear at first sight to be irreconcilable. They are connected, however, bya sequence of mesoscale architectures and phenomena that form, step by step, a staircase reaching from atoms to bulk materials that can be experimen- tally 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.

Hallmarks of Mesoscale Behavior The emergence of mesoscale behavior from nanoscale origins is marked by distinctive features in the role of atomic granularity of matter, quantization of energy, collective behavior of many identical units, interaction of disparate degrees of freedom, the appearance of defects, fluctuations and statistical variation, and heterogeneity in structure and dynamics. These hallmarks of mesoscale behavior distinguish it from the adjacent realms of quantized nanoscale behavior at smaller scales and continuous macroscopic behavior at larger scales.

At nanoscale dimensions, the size of atoms (0.1 nm) and the typical spacings between them (0.5 nm) are dominant features. The atomic granularity of matter and the exact positions of individual atoms are key determinants of nanoscale structure and dynamics. At mesoscale dimensions of order ten to a hundred times the typical atomic spacing, the presence of individual atoms is less obvious and their impact less significant. Mesoscale behavior is captured by the average density of atoms in a volume containing thousands of atoms, and their associated fluctuations. In these ensembles, the exact positions of individual

3 From Quanta to the Continuum: Opportunities for Mesoscale Science

atoms are often not important and may not even be known. Dependence The Particle in a Box of behavior on the average density of Every beginning student of physics, atoms instead of their exact positions chemistry, and materials science learns is a hallmark of mesoscale behavior. the “particle in a box,” one of the few elementary and exactly solvable problems of quantum mechanics and a simple, Quantization of energy rules the engaging illustration of the transition from n=3 behavior of nanoscale systems, where nano to meso. Quantum mechanics tells n2h2 energies fall into discrete, quantized us that the energy of an electron confined E E = 2 levels separated by large forbidden to a box of width a (of nano- or meso- 8ma gaps not accessible to the system. scale dimensions) is quantized—it can n=2 The large spacing between quantized assume only discrete values, given by the simple formula indicated in the figure (see energy levels prevents the system n=1 caption for details). In practice, only the from responding to stimuli in its envi- lowest energy levels are occupied. ronment that have smaller energies, a such as temperature, light, or elec- The central point is that as the size of the Quantized energy levels of an electron (of tric and magnetic fields. At larger box increases from nano to meso, the spacing between allowed energy levels mass m) confined in a nanoscale or meso- mesoscale dimensions, the spacing shrinks and they become dramatically scale box of width a. The electron can assume between energy levels falls, following denser, giving the electron much greater only discrete, quantized energy levels given by the formula above, where n is an integer the “particle in a box” analogy (see choice in levels to occupy and much more (n=1, 2, 3…) and h is Planck’s constant. For a flexibility in its behavior. The freedom sidebar “The Particle in a Box” at nanoscale box, a ~ 1 nm, the electron has only right), until it becomes smaller than of a larger box with dense energy levels a few widely spaced energy levels available to allows the electron to participate in new other relevant energies. Access to it, restricting its interactions with its surround- phenomena, a primary reason for the many densely spaced energy levels ings. For a mesoscale box, a ~ 100 nm, the greater range and complexity of meso- allowed energy levels are 10,000 times denser, at the mesoscale opens a broad array scale over nanoscale behavior. As the box dramatically increasing the interactions with of potential responses of the system, grows to macroscopic size, the spacing neighboring particles and with temperature, significantly enriching the diversity of between allowed energy levels shrinks to light, and electric and magnetic fields in the environment. interactions with the environment and insignificant values and energy becomes effectively continuous, smoothly joining to with neighboring components. the classical mechanics that governs the ingly rare. The heterogeneity of mesoscale familiar macroscopic world. systems is likewise based on energy argu- Collections with a large number of ments: inserting a boundary between two members such as electrons, atoms, The higher density of allowed energy distinct phases requires less impact on or magnetic moments behave in ways levels at the mesoscale compared to the densely packed energy spectrum of the nanoscale is the root of many hall- mesoscale systems than on the sparsely that individual or small numbers of marks of mesoscale behavior. The collec- packed spectrum of nanoscale systems. members cannot. Flocks of birds, tive behavior of many electrons, such schools of fish, and packs of wolves as screening of point charges or strong The particle in a box illustrates yet another are examples from the animal world correlation due to Coulomb repulsion, connection between two key mesoscale hallmarks. The importance of atomic displaying collective behavior that expresses subtle behavioral features that granularity and the quantization of energy is qualitatively different, more effec- require closely spaced energy levels to convey. The statistical variation of meso- track the size of the box and, therefore, tive and more functional than that of scale structure and dynamics is due to the each other as the size of the box grows. individual members. Similar kinds of many choices of states enabled by closely For boxes 100 times larger than atoms, collective behavior apply to nanoscale spaced energy levels, allowing small vari- for example, the positions and influences components. The electric fields of point ations in configuration with little change of individual atoms are hardly resolvable; simultaneously, the spacing between charges, for example, are screened in the energy; for nanoscale systems the allowed energy levels falls to small values. by rearrangement of surrounding itin- wider spacing between energy levels drives up the energy cost of such varia- Thus, these two key hallmarks of meso- erant charges in metals, semiconduc- tions or defects and they are correspond- scale behavior rise and fall together. tors, and ionic solutions. To achieve full screening, the size of the system

4 From Quanta to the Continuum: Opportunities for Mesoscale Science

must be comparable to the screening Mesoscale systems typically have of freedom, spintronics for control length, which depends primarily on many modes of response to stimuli of charge by magnetic fields (see the density of itinerant charges. For in the environment, including phys- sidebar “Giant Magnetoresistance: semiconductors and ions in solution ical deformation; chemical change; A Triumph of Mesoscale Science” the screening length can be one to transport of electrons, ions, or heat; on page 35) and piezoelectricity for hundreds of nanometers, mesoscale magnetic or electric polarization; and conversion of applied voltage to lengths that modulate and enable the emission or absorption of light. Each mechanical displacement (see figure collective phenomena of screening. of these modes corresponds to an at left). These interacting degrees of Collective behavior appears in every independent degree of freedom, and freedom can be homogeneous, as kind of variable, including electronic, within each degree of freedom the in superconductivity where the elec- ionic, mechanical, electromagnetic, response can be mild or intense. At the trons, phonons, and superconduc- and chemical, each described by its mesoscale these responses become tivity coexist in the same physical own mesoscale response and charac- much richer and more developed, space, or heterogeneous, as in solar teristic collective length. as the energy levels associated with cells where a semiconductor absorbs each degree of freedom the photon, creates an excited elec- become densely packed tron, and passes it to metallic wires and overlap with the for transmission in an external circuit. levels belonging to other degrees of freedom. Imperfection and statistical variation constitute a mesoscale feature in The interaction of dispa- sharp contrast to the relative perfec- rate degrees of freedom tion of nanoscale objects. Nanoscale often produces profound systems are typically more perfect outcomes. In metals, because they have fewer atoms, there phonons associated are fewer ways for them to deviate with mechanical degrees from the perfect structure, and the of freedom normally energy cost of these deviations is scatter electrons and relatively high. Mesoscale systems, increase the resistivity. in contrast, have many more atoms At low enough temper- (a few thousand to a few hundreds of ature, however, the thousands or more) and more closely electron-phonon inter- spaced energy levels, so that the action produces just the energy cost of alternative configura- opposite effect—super- tions is relatively lower. As a result, conductivity, where the defects are more common in meso- resistivity collapses to scale systems, and statistical variation zero. Other examples of about a mean is a universal feature of interacting degrees of mesoscale behavior. freedom include electro- chemistry for storage and The variation of mesoscale systems release of electrons in has profound effects. While all elec- Interacting mesoscale degrees of freedom illustrating chemical bonds, photo- trons and all atoms of a given element giant piezoelectricity, resulting in conversion of voltage chemistry for the conver- are identical, copies of mesoscale to mechanical displacement. (Upper panel) Compound sion of light energy to systems vary in small details, the cantilever of Pt(60 nm)/Pb(Mg Nb )O -PbTiO (270 1/3 2/3 3 3 chemical bond energy, basis for biological mutation and nm)/SrRuO3 (100 nm)/SrTiO3 (13 nm). (Lower panel) Displacement of the cantilever tip vs. applied voltage. Source: photovoltaics for the evolution by natural selection. The S. H. Baek et al., Science 334, 958 (2011). conversion of light energy defects in inanimate mesoscale mate- to electronic degrees rials are just as profound, disrupting

5 From Quanta to the Continuum: Opportunities for Mesoscale Science

the behavior of the otherwise perfect decay of bridges and the performance dynamics. The unifying element in lattice by scattering electrons, of solar cells. Degradation science— these hallmarks is the very meaning of preventing atoms from moving plasti- observing, predicting, and mitigating the Greek root meso: intermediate, in cally in response to external stresses, degradation due to defect accumu- between, middle. Mesoscale behavior and interrupting the transport of heat. lation—is key to mesoscale science lies between the nanoscale world of Mesoscopic and larger lattices never and has high societal impact. atoms, molecules, and small assem- show the mechanical, electrical, and blies with relatively perfect structure thermal properties of a perfect lattice; The final mesoscale hallmark consid- displaying simple behavior, and the instead, their macroscopic properties ered here is heterogeneity. Nanoscale macroscopic world of bulk materials are determined by structural defects systems are typically homogeneous, with imperfect structure and endless and imperfections separated by because introducing a boundary and variation, complexity, subtlety, and mesoscale distances. creating two or more distinct phases functionality. costs energy. Nanoscale crystals tend On the positive side, meso- and to be single grain instead of multi- macroscopic properties can be inten- grain, and nanoscale magnets like- Mesoscale tionally tailored by ntroducing specific wise contain a single magnetic Architectures kinds and densities of defects on domain. Mesoscale systems, in The staircase of mesoscale transi- meso length scales. On the negative contrast, are large enough to be tions from atomic, molecular, and side, the accumulation of defects at heterogeneous (see figure at left) and, nanoscale to the more familiar bulk for example, to spontane- macroscopic behavior is expressed ously break into coexisting through a hierarchy of architec- phases at a first-order tures that organizes the constituent phase transition or retain atoms and electrons into functional an artificially constructed structures. At the atomic scale, two composite structure of distinctly different types of archi- two or more distinct tecture stand out: periodic lattices components created by and covalent chemical bonds. At bottom-up self-assembly the mesoscale, a third architectgure or top-down fabrication. emerges, biological membranes and cells, the building blocks for life. Mesoscale behavior These architectures guide the course establishes itself by of the higher level mesoscale transi- many routes; there is no tions, as illustrated below. single hard and fast rule for defining its onset or Periodic Lattices Heterogeneous mesoscale self-assembly of a block its character. Mesoscale For many materials, the basic orga- copolymer (PS-b-PMMA) on a silicon wafer. The behavior reveals itself, nizing structure is a lattice of atoms, straight lines are patterned by a lithographic template. however, by one or more Source: Bryce Richter, University of Wisconsin, Madison where the loosely bound outer (http://www.news.wisc.edu/newsphotos/epitaxial.html) of the hallmarks illus- atomic electrons are shared with and S. O. Kim et al., Nature 424, 411-414 (2003). trated here: reduced many neighboring atoms to form a influence of atomic granu- periodic array. The lattice of atoms the mesoscale often degrades the larity, diminished spacing between creates a mechanical structure with properties of materials, leading to discrete quantized energy levels, its own mechanical responses, such crack formation and propagation collective behavior of many identical as deformation and vibration, that and, ultimately, to macroscopic frac- components, interaction of disparate are intrinsic to the lattice and depend ture and failure. This degradation can degrees of freedom, the appearance on its crystallographic structure be rapid, as in the collision of two of defects and statistical variation, and composition. The mechanical objects, or slow, as in the gradual and heterogeneity in structure and

6 From Quanta to the Continuum: Opportunities for Mesoscale Science

responses of the lattice are collec- producing an impressively wide range The mechanical and electronic tive mesoscale phenomena with no of behavior spanning metals, insula- degrees of freedom interact by scat- analog in the behavior of individual tors, and semiconductors. Electrical tering to produce resistivity that dissi- atoms. Large enough crystals display conductivities vary over 20 orders pates energy and impedes the motion heterogeneous structures, with grains of magnitude, depending on lattice of electrons through the lattice. At low of different orientation separated by structure and the electronic configu- temperatures, however, the electron- grain boundaries, or isolated defects ration of its constituent atoms. The phonon interaction produces a very such as missing or interstitial atoms, electronic structure of lattices, like its different and striking phenomenon: dislocations, and stacking faults mechanical behavior, is a collective superconductivity, where resistivity separated by mesoscopic distances. response that is intrinsic to the lattice collapses to zero and electrons move These heterogeneous structures and has no analog in the behavior of freely throughout the lattice without and defects dramatically change the its atomic constituents. loss of energy. Superconductivity is macroscopic electrical and thermal rich with new mesoscale behavior, conductivity and the strength and Although the lattice itself has a clear including new mesoscale objects— ductility of the otherwise perfect atomic granularity at nanoscale Cooper pairs and superconducting lattice. dimensions, the mechanical and vortices—whose collective behavior electronic responses associated with is described by new meso length The mechanical structure of the lattice the lattice become continuous as scales, the superconducting coher- together with the loosely bound outer the lattice grows in size, the number ence length and magnetic pene- atomic electrons begets a second of quantized energy levels prolifer- tration depth. Among known architecture: the electronic struc- ates, and the spacing between the superconducting materials, these ture of the lattice, a set of electronic quantized levels becomes arbitrarily superconducting mesoscale lengths degrees of freedom distinct from small. This continuum of mechan- range from a few to hundreds of the mechanical degrees of freedom. ical and electronic energy states for nanometers. The electronic degrees of freedom large lattices allows their properties appear as energy bands arising from to be described by familiar contin- Magnetism takes mesoscale behavior the overlap of the outer electrons uous formulations based on elasticity a step further. Each atom in the lattice of the atoms and are capable of theory for mechanical properties (see may have a local magnetic moment figure at left) and by clas- due to the orbital motion and spin of sical electrodynamics its inner shell electrons. The lattice for electronic properties, structure brings these local moments where the natural vari- close enough together to interact ables are density of atoms through dipolar and exchange inter- or electrons instead of actions, which cause neighboring the positions of individual atomic moments to point in the same atoms or electrons. Thus, direction to form ferromagnets, or in the lattice, the most basic opposite directions to form antiferro- architecture of solids, magnets. The large magnetic polariza- displays atomic granu- tion energy of ferromagnetic domains larity and quantized energy limits their size, however, introducing levels at nanoscopic a mesoscale magnetic length much scales, transitioning to larger than the atomic scale of the continuous matter, energy, lattice structure but smaller than elasticity, and conductivity macroscopic lengths. A periodic lattice of atoms showing the atomic granularity at larger scales, a hallmark of matter in which mechanical responses such as elastic of mesoscale behavior. A host of mesoscale magnetic deformation are collective properties of the lattice structure. phenomena are enabled by the Source: Florian Marquardt, Wikimedia Commons.

7 From Quanta to the Continuum: Opportunities for Mesoscale Science

photon absorption produces electron emission. The pattern continues throughout the mesoscale staircase: collective behavior produces new degrees of freedom and new characteristic length scales, which interact with other degrees of freedom to produce additional new levels of collective behavior and so on. Discovering the organizing principles Chemical bonds between porphyrin complex

at each level and applying them to on the left that donates electrons to the C60 design and control phenomena and complex on the right, which accepts and holds electrons for further chemical reactions. functionality at higher levels are grand Source: Source: Liu et al, J. Am. Chem. Soc. Magnetic domains on the surface of challenges of mesoscale science. 134, 1938 (2012). yttrium iron garnet film. Source: Del Mar Photonics (http://www.dmphotonics.com/ atoms. Like lattices, extended molec- Scanning_Probe_Microscopy/Magnetic%20 Chemical Bonds ular structures have mechanical, structure%20of%20surface%20domains%20 Like the lattice, the chemical bond is thermal, magnetic, and ferroelec- in%20Yttrium%20Iron%20Garnet%20 a fundamental architectural element (YIG)%20film.htm) tric degrees of freedom that interact that underpins a host of macroscopic to produce new architectures and materials and phenomena. Chemical architecture of magnetic moments phenomena. Unlike lattices, chemical bonds fold outer shell electrons into on a lattice, including polarization bonds can connect chemical units of tight configurations between two or domains, polarization switching, very different structure and function, at most a few neighbors to produce hysteresis, and fluctuations (see such as joining porphyrin complexes new molecules. Carbon dioxide and figure above). The behavior of these that donate electrons to C bucky- water are examples, the first being 60 mesoscale domains can be described balls that accept and hold electrons a linear molecule and the second, by continuous formulations of for later use in chemical reactions (see an angular molecule with hydrogen- magnetism, with magnetic moment figure above). Chemically bonded oxygen bonds oriented at the precise density as the natural variable. These structures need not have a regular angle of 105°. The rich diversity continuous formulations describe structure, allowing them wide freedom of chemically bonded structures the macroscopic bulk behavior quite to configure locally to enhance a extends from simple small molecules precisely with little or no need to refer- particular phenomena or function- such as water and carbon dioxide to ence their atomic and lattice origins. ality; in addition, they have chemical one-dimensional polymer chains that degrees of freedom that dynami- can be linked at their ends to form The staircase of higher level archi- cally make or break bonds to change co-polymers and then cross-linked tectures and phenomena arising composition, configuration, and func- to produce two- or three-dimensional from lower level structures is rich tion and to release or store energy rigid or flexible networks. Solutions, and pervasive. Ferroelectric degrees and electrons. The chemical bond is a suspensions, colloids, and foams are of freedom interact with ferromag- primary mechanism for self-assembly additional examples of architectures netic degrees of freedom to produce of complex and functional meso- constructed of various kinds of chem- multiferroics, where magnetic fields scale structures like mesoporous ical interactions. switch electric polarization or vice membranes, composites, photonic versa. There are also magneto-elastic, crystals, or polymer ferroelectrics. The molecules and extended struc- chemomechanical, and photo- The versatility and scope of the chem- tures produced by chemical bonds electric interactions, where strain ical bond in molecules, polymers, and differ from lattices in that they need affects the magnetic polarization or networks make it a pervasive and not be periodic, and their electrons the rate of chemical reactions, and key architectural element for building are usually confined to the bonding functional mesoscale systems.

8 From Quanta to the Continuum: Opportunities for Mesoscale Science

Architectures based on chemical with topologically distinct insides and bonds show the hallmarks of meso- outsides (see figure below). The inside scale behavior: can be filled with proteins and other • polymers exhibit flexible line-like machinery that can be selected, orga- mechanical behavior that tran- nized, maintained and reproduced in scends atomic granularity; a protected environment to accomplish specific functionalities such • particle size directly affects as energy production, locomotion, spacing between quantized or communication with the outside energy levels and the optical by transporting ions and molecules absorption and emission spectra through the cell wall. of suspensions; Micrograph of stem cell biopsy. Source: • electron transfer rearranges The cell along with its internal Y. Chung et al., Cell Stem Cell 2, 113-117 chemical bonds and couples mesoscale machinery is a remarkably (2008). structural, mechanical, electronic, versatile structural and functional and thermal degrees of freedom; module, the smallest unit of life. blocks of multicellular architectures and Single-cell bacteria are pervasive in that perform complex functions that • the role of defects, statistical vari- the natural world from hydrothermal a single cell cannot, such as muscle ation, and heterogeneity is key to vents to radioactive waste to flexing and photosynthesis that molecular systems such as solu- Antarctic ice cores, performing a require the interaction of disparate tions, ionic liquids, and colloids. host of functions such as creating mechanical, chemical, and electronic and recycling nutrients; promoting degrees of freedom. Biology displays Membranes and Cells digestion in humans, cockroaches, many compelling examples of the and termites; and fixing nitrogen from Membranes composed of lipid bilayers mesoscale staircase of architectures the atmosphere (see sidebar “Enlisting are a basic architecture that supports and functionalities (see sidebar “The Bacteria as an Energy Workforce” on an enormous number of biological Mesocale Inspiration of Biology” on page 42). Stem cells (see figure at phenomena and functionalities. The page 10). right) have the remarkable capability lipid bilayer is a derived architecture to develop into many different cell based on polymers with hydrophilic types during growth, such as a muscle The Gap Between Top heads and hydrophobic tails, which cell, a red blood cell, or a brain cell. are themselves composed of chemi- Down and Bottom Up The assembly of mesoscale building cally bonded atoms. The presence Colonies of cells display collec- blocks into higher level architec- of water encourages the polymers to tive behavior by signaling each tures and phenomena is a contin- line up head-to-head and tail-to-tail other through chemical and physical uous process extending well beyond to form flexible membranes that can means, and cells are the building the dimensions that are natural for fold themselves into closed shapes humans to manipulate. Biology has created living architectures as large as the Great Barrier Reef, which is

Lipid bilayer folded into a closed over 2000 km long and dominates cell with topologically distinct its regional environment by playing inside and outside. Biology uses host to an enormous diversity of other the protected inside to contain living species. Geology has stratified mesoscale machinery to produce energy, transport cargo within the Earth’s structure by density into the cell, and reproduce the cell. a viscous mantle supporting floating Source: Wikimedia Commons, tectonic plates that execute an intri- http://commons.wikimedia.org/wiki/ cate dance creating mountains and File:LiposFreome_scheme-en.svg.

9 From Quanta to the Continuum: Opportunities for Mesoscale Science

The Mesocale Inspiration of Biology Biology displays many compelling features of mesoscale science. science. Mesoscale science, however, has the potential to go The building blocks of biology are cells, the smallest units of life, well beyond biology. Biology works primarily with bottom-up consisting of a lipid bilayer membrane deformed into a closed self-assembly of atoms and molecules, developing complex container with a collection of chemicals, polymers, and functional structures and functionalities from smaller and simpler building units on the inside and access to a diverse chemical environment blocks over millions of years by mutation, natural selection, and on the outside. The machinery inside the cell produces energy in evolution. Mesoscale science embraces not only bottom-up self- mitochondria to power the cell, moves cargo along cytoskeletal assembly but also top-down design and fabrication, starting with tracks to build and operate the cell, and transports molecules macroscopic materials that are cut to size as components and and ions like water, sugar, or calcium through channels in the cell then assembled to a pre-conceived rational design. The promise wall to the environment. of mesoscale science is joining bottom-up self-assembly with top-down rational design to produce targeted outcomes and The structures of cells—membrane, mitochondria, and cytoskel- functional systems using simple components, little waste, meso-, eton—define a versatile architecture adaptable to many tasks. At nano- or atomic level interconnects, high efficiency, and long life. the same time, the cell is the building block for the next higher level of architecture: multicellular tissues such as blood vessels, Unified bottom-up self-assembly and top-down rational design bone, or muscle; the combination of disparate tissues into organs, dramatically improve two features of biology: production of such as the heart, the brain, and the stomach; and the combina- targeted, pre-conceived outcomes instead of random variation of organs into independent plants or animals. tions, most of which are ultimately discarded, and shortening of the development time for complex functional structures from the Each architectural level of biology produces a new level of many generations required for biological evolution to a few years complexity and functionality building on the levels below and required for rational design and implementation. contributing to the levels above. At the lowest levels, molecules and polymers obey quantum mechanics, where the granularity of atoms and the quantization of energy are dominant, features that give way to the continuous matter and energy of classical mechanics as the size of the system grows. Biology illustrates other mesoscale transitions as well: • the interaction of disparate degrees of freedom, such as mechanical, electronic, ionic, and fluid, to create new phenomena and functionality; • increasing complexity enabling greater functionality as the number and diversity of interacting units grow; • the onset of statistical variation or “defects” in a collection of complex but nominally identical objects, the basis for mutation, natural selection, and evolution; and • the integration of component parts into coordinated systems with heterogeneous structure and function that work harmo- Cell structure of an onion scale, from an inner layer of the onion. Each cell niously to produce a single composite functionality. is approximately 250 microns wide. Source: http://commons.wikimedia. org/wiki/File:Onion_Cells.jpg. The staircase of increasingly complex architectures and functionalities exemplified by biology is a proof of mesoscale principles, and serves as a model and inspiration for inorganic mesoscale

10 From Quanta to the Continuum: Opportunities for Mesoscale Science

earthquakes as they shift and collide. dizzying number of types of cells and Humans are learning to manipu- architectures for organizing them. late the bottom-up, self-assembled Biological architectures accomplish a world, creating tools that operate host of chemical, physical, electrical, well beyond the scale of our eyes and mechanical, and energy conversion hands. Telescopes extend our vision functionalities, using energy from the to the very large and microscopes to sun extracted by photosynthesis to the very small, while tools with greater power their self-assembled machinery. resolution based on x-rays, electrons, Biology and geology are beacons of neutrons, and scanning probes allow inspiration, proving the principle and diffraction, imaging, and spectros- showing the way to a much richer, copy at the atomic scale. more functional, and more efficient self-assembly paradigm. While we can see the atomic scale with the aid of electron microscopes, Top-down lithography and bottom- we cannot yet build structures at the up self-assembly differ much atomic and nanoscale with the lati- more profoundly than just the size, tude, flexibility, and precision of the complexity, and functionality of the top-down methods we use in the structures they can make (see figure macroscopic world. These top-down at right). They differ in operational methods allow the pre-conceived philosophy as well. Top down starts design of stand-alone functional units with a large block of raw material and that we connect together with macro- continuously whittles it away like a Bottom up versus top down. Bottom-up scopic links such as electrical wires, sculptor, leaving the final structure it self-assembly makes intricate structures with mechanical linkages, and channels wants and casting aside the unwanted atomic precision and little waste, starting with for fluid flow. The smallest objects we detritus that often makes up the atoms and molecules. Upper panel shows the can make with top-down methods majority of the original block. Bottom self assembly of silicon nanoparticles (blue) on carbon black trees (black) forming rigid are ~10 nm using electron beam up, however, uses only the atoms porous microspheres for high energy density lithography. Functional units requiring and materials that it needs, assem- lithium battery electrodes. Lower panel shows several top-down components are bling them into functional units with top-down lithography: a simple structure at proportionately larger, well above the little or no waste. Because bottom 10 nm precision following a pre-determined template with significant waste. Joining scale of bottom-up self-assembly. up works at smaller scales than top bottom-up with top-down fabrication down, it gains the advantage of the techniques is a primary challenge and In contrast to top-down design and relative perfection and reproducibility opportunity for mesoscale science. Source: A. Magasinski et al., Nature Materials 9, fabrication, bottom-up self-assembly of its manufacturing process. In addi- 353358 (2010) (upper panel); A. Han et al., starts with atoms and molecules tion, bottom up connects functional Nano Lett. 12, 1018 (2012) (lower panel). to build architectures that support units together directly at the atomic new phenomena and functional- or mesoscale, without the need for ities at larger scales. We are begin- extended wires, mechanical linkages, The gap between top down and ning to learn to use self-assembly, and channels for transporting elec- bottom up is rich with meso oppor- inspired in part by the examples of trons, force, and hydraulic or chemical tunities. Self-assembly, though origi- biology and geology. The complexity fluids. This direct connectivity simpli- nating in atomic and nano space, and functionality of the structures fies the operational design, eliminates quickly transitions to meso space as that we can self-assemble, however, opportunities for failure, and saves the architectures and phenomena to pale in comparison to those of the the material and assembly cost of the be integrated become more complex. natural world. Biology, in particular, interconnects. Biology and geology are guiding has devised through trial and error a

11 From Quanta to the Continuum: Opportunities for Mesoscale Science

examples; much of their functionality arises from complex of the mesoscale remain relatively unexplored. What laws structures at or above the 10-100 nanometer level. Bridging govern self-assembly, or the stability of dynamic steady the top-down/bottom-up gap is a prime meso chal- states, or the localization of strongly correlated electrons? lenge for developing sophisticated self-assembly and for These uniquely mesoscale phenomena are pervasive controlling metastability in these systems. At present the in the materials world; understanding their occurrence, two approaches are virtually exclusive. We do not design operation, and potential is a powerful step forward for new top down with the intention of incorporating bottom-up science and technology. components, nor do we design bottom up with the intention of connecting bottom-up components with top-down Over the course of this study the Mesoscale Science links. Bridging the gap would provide many advantages, Subcommittee of the Basic Energy Sciences Advisory including the ability to communicate with self-assembled Committee engaged hundreds of colleagues in town hall molecular units using conventional electronics, a task that meetings, webinars, and web site interactions in order now, for the most part, is beyond our practical reach (see to identify important and timely research directions for sidebar “Mesoscale Manufacturing” on page 50). mesoscale science, as well as the capabilities required to address these challenges. Six priority research directions The Meso Opportunity emerged from this study: The meso opportunity embraces and exploits the enor- • Mastering Defect Mesostructure and its Evolution mous foundation of nanoscience understanding and • Regulating Coupled Reactions and Pathway- expertise that the research community has created over Dependent Chemical Processes the last decade and continues to create. Examples include • Optimizing Transport and Response Properties by nanoscale structural and functional units such as quantum Design and Control of Mesoscale Structure dots, core-shell nanoparticles, catalytic clusters, polymer systems, and host-guest chemistry, and nanoscale tech- • Elucidating Non-equilibrium and Many-Body Physics niques such as bottom-up self-assembly and micro-beam of Electrons structure determination and spectroscopy with electrons, • Harnessing Fluctuations, Dynamics and Degradation x-rays, and neutrons. These nanoscale building blocks, for Control of Metastable Mesoscale Systems phenomena, and techniques can be interfaced and inte- • Directing Assembly of Hierarchical Functional Materials grated at the mesoscale to form new levels of architectures of greater complexity and functionality that dramatically In exploring these priority research directions, it became enlarge the horizon of science and technology. The meso clear that a next generation of synthesis, characterization, opportunity for new, cheaper, and more efficient tech- and modeling capabilities was required to enable success. nology and solutions to societal problems is significant These, as well as the essential challenge of integrating and timely. these capabilities, are described herein.

Beyond new technology and solutions to societal problems, In what follows, each of the priority research directions is the mesoscale is fertile ground for new basic science. In the discussed in detail. The remaining chapters discuss both bottom-up mesoscience world, the harmonious interaction the integration challenge and the specific gaps in synthesis, of components and architectures will lead to new orga- characterization, and modeling capabilities. Workforce nizing principles that will require observation, analysis, and needs and recommendations for realizing the mesoscale simulation to discover and understand. We have discov- opportunity are then discussed. Finally, a perspective on ered the organizing principles of quantum mechanics for potential outcomes resulting from exploration at the meso- the atomic, molecular, and nanoscale and of classical scale is offered. physics for the macroscale, but the organizing principles

12 From Quanta to the Continuum: Opportunities for Mesoscale Science

Priority Research Directions for Mesoscale Science

The present study reveals significant mesoscale science opportunities and iden- tifies what this subcommittee has defined as the most promising emerging directions for future research. Through interactions and discussions with the community, we have identified a set of priority research directions (PRDs). In each of the PRDs, the science opportunity is discussed, the mesoscale challenge is emphasized, an approach to addressing these challenges is described, and a vision for success is articulated. These PRDs span the traditional disciplines of condensed matter physics, materials science, and chemistry, and because of their interdisciplinary nature, defy simple characterization within these traditional boundaries. They span the Five Challenges for Science and the Imagination that the Basic Energy Sciences Advisory Committee (BESAC) previously identified, and success in their pursuit will impact the full array of energy technology challenges.

13 From Quanta to the Continuum: Opportunities for Mesoscale Science

Mastering Defect lattices to magnetic domain struc- for evaluating the positive or negative tures. Such defects and interfacial effect of defects. System performance Mesostructure and structures may not be correlated with integrates over numerous interme- its Evolution microstructural properties. Second, diate characteristics and parameters Characterizing and controlling the we seek to control materials func- and usually declines with time as patterns and evolution of mesoscale tionality at the mesoscale not only defect populations grow and agglom- heterogeneity are key to optimizing by minimization of deleterious defect erate. Thus, predicting the spatial and and exploiting a wide range of mate- types but also by controlled introduc- temporal evolution of defect popula- rials performance and functionality. tion of defects that enhance properties tions from their initial value is a key or performance. Without mastering mesoscale challenge. These predic- Opportunity science at the mesoscale, translating tions, however, must be motivated Defects in materials and structures atomic and nanoscale phenomena and validated by experimental obser- comprise a wide range of meso- into macroscopic materials solutions vation of defect evolution, which in scale features, including interfaces, for society and the economy will be real time can take decades. Therefore, dislocations, vacancies and intersi- inefficient and slow. Delivering value accelerated testing modalities are tials. Defects are always present and from the nanoscale requires mastery needed to project long-term behavior. distributed nonuniformly, producing of defects at the mesoscale. Unfortunately, scientific insight at the mesoscale heterogeneity which is mesoscale is often inadequate to part and parcel of materials proper- The profound advances that we move beyond brute force approaches ties and behavior. Defect patterns are witnessing in materials science, to accelerated testing. typically extend to the boundaries particularly a combination of compu- of materials and are described by tational capabilities and character- Thus, the science requires (a) iden- mesoscale dimensions. Thus, a key ization tools, make this area ripe for tifying defects and tracking their element of mesoscale science is exploitation (see figure on page 15). dynamic evolution—a major chal- understanding and controlling the Both computation and experiment lenge—and (b) understanding the formation of individual defects on can now address, for the first time, mechanisms by which defects alter the atomic or nanoscale level and a large range of the defect types system performance through their projecting their collective influence to and populations of interest, whether long-range/long-time behavior. The the macroscale properties and perfor- they are dislocations, pores, inter- impact follows from the fact that, if we mance of materials. faces, grains, domains, or any other could establish the science, the path defect with mesoscale influence. The from fundamental understanding at Defects are discontinuities in material capability is such that not only indi- the nanoscale to higher performing, properties and behavior, which typi- vidual defects but also distributions of longer lasting products will be shorter cally occur at mesoscale separation in defect features or field values can be and more effective. Engineers can materials systems and architectures. computed and measured, including adopt such new knowledge and use While defects are often associated statistical analysis of rare features and it for enhanced technological solu- with microstructural features that are correlations between quantities. Some tions in, for example, advanced classified according to topology (i.e., defects and defect distributions lead manufacturing. point vs. line vs. areal vs. volumetric to fault-tolerant behavior, while others defects), we intend a broader conno- can become sites for damage initia- Meso Challenges tion and ultimately materials failure. tation, in two senses. First, we regard Because defects and interfaces are Understanding and controlling meso- defects as features that constrain by definition mesoscale constructs, scale defects structures will lead to either the extent of collective behavior the meso challenges associated with enhanced macroscale performance. or the utility of mesoscale systems mastering defect structure and evolu- formed as large assemblies of tion are myriad. Here, we highlight It is important to focus on system components, i.e., functional defects, three examples, spanning materials which occur in everything from crystal performance as the figure(s)-of-merit

14 From Quanta to the Continuum: Opportunities for Mesoscale Science

types (mesoporous solids), relevant properties of the constituent mate- a two-sided nature to crack forma- physical properties (strength and rials may be known, the mechanical tion and materials failure. Considered damage), and the consequences of properties of assembled porous a detriment for the built environment, fabrication approaches (top-down structures at the nano-, meso-, and it is a benefit for recovering the vast and bottom-up defect manipulation). microscale—including aerogels, reserves of petroleum trapped in foams, and engineered lattices—are shale. A predictive understanding of Porous and Mesoporous Solids relatively unexplored. Even the funda- microstructure-based heterogeneity The mechanical and transport proper- mental mechanisms determining the evolution and its consequences for ties of materials containing vacancy emergent strength of the bulk form material damage/failure and phase structures and pores with meso- are not well understood. The utility transformation is presently lacking. scopic dimensions are of practical of such materials, especially in crit- This gap prevents us from predicting importance to the development of ical structures, depends on main- material response under severe high-strength, low-weight materials. taining the integrity of the porous loading conditions. In strong, light-weight materials, structure in the presence of extreme enhanced structural and functional stresses, be they radiation-induced, Effect of Pores on Materials performance can increase energy effi- mechanical, or chemical. Geological Behavior. The dimensions of a ciency in, for example, transportation materials are a good example: under- porous structure, which can range systems. standing porous structures and their from the nanoscale to microscale, formation is crucial to improving the affect the effective bulk elastic prop- Integrity and Transport in Porous performance of carbon sequestration erties of the solid. Knowledge of the Materials. While the mechanical and gas extraction from hydraulically mechanical properties of mesopo- fractured or “fracked” rocks. There is rous materials is essential to gaining

15 From Quanta to the Continuum: Opportunities for Mesoscale Science

a basic understanding of the mecha- easily dislocations flow. The mecha- crack initiation is highly variable and nisms responsible for strength. One nisms by which alloying elements microstructure dependent. Improved must be able to distinguish pores control the balance among inter- understanding of how to control the from other defects. What is the struc- face strength, dislocation motion, defects would have many benefits. tural stability of pores? Can pores and ductility remain unknown in evolve into networks of defects? How many cases. Ductility is not simply Engineered Assembly and do networks evolve in the presence a strength balance, however; it also Self-organization of environmental stresses? Dynamic depends on the stress-strain curve Our ability to understand, engineer, damage/failure and phase trans- and the extent to which strain hard- and exploit atomic and nanoscale formation processes are believed ening occurs, once more pointing out phenomena is gated by how well we to occur by a series of nucleation, the lack of a nano-to-meso under- can manage defect mesostructures growth, and coalescence events. The standing of dislocations. comprising defect populations and linkage of these events to the details their correlated patterns. We inten- of the material chemistry and micro- Damage Initiation. Evolution of mate- tionally say to manage defects, not to structure of porous materials is poorly rials microstructure under loading eliminate them, because defects are understood. includes accumulation of damage to always present when atoms, mate- the structure. This damage may be rials, or structures are aggregated at Plastic Flow, Loading Leading beneficial as, for example, disloca- large length scales. Defects may be to Damage tion accumulation provides strain created, transformed, or activated Structural metallic alloys are among hardening that in turn allows load during synthesis of materials and the most important and pervasive redistribution in structures, or it may structures, during operation of the materials classes. Their usefulness be life-limiting when cracks or voids resulting devices, or by subsequent in structures depends on their ability form. Cracks and voids may arise external influences such as radiation to distribute load via plastic deforma- through weakest-link failure such damage and thermal or mechanical tion. Plastic strain occurs mainly via as grain boundary cracking. They stress. The need to manage defects the flow of dislocations, which are line can also arise through fluctuations is critical in two domains—the engi- defects in crystals, and sometimes in dislocation density. Even under neered assembly of smaller compo- via twinning, which also depends on macroscopically uniform deformation nents into mesoscale systems, and the line defects in a complicated fashion. in single crystals, dislocations always exploitation of natural self-assembly. Although individual dislocations are form patterns. This behavior points very well understood as isolated line to the importance at the mesoscale Engineered Assembly. Energy appli- defects at the nanoscale, their collec- of distributions and the properties of cations provide rich examples of the tive, mesoscale behavior remains their tails, where high concentrations need to manage defects in engi- a challenge. By contrast to elastic of defects trigger rare but ultimately neered systems (see figure on next moduli, which are routinely calculated dominant damage events. Critical page). While mesoscale architectures from first principles, there is no avenue events, such as crack formation, are (assemblies of nano-wires, nano- to compute a stress-strain curve, labeled as damage initiation at the tubes, and nanoporous materials) as required by engineers, starting at mesoscale because, while they may promise power and energy benefits the atomistic level because we lack result from a smooth progression at for electrochemical energy storage, anything remotely resembling a statis- the nanoscale, they mark the begin- the increased surface area that tical mechanics of dislocations. ning of a well-defined path to ultimate enables them offers more channels failure at the mesoscale. An example for defect generation; for example, Ductility, Dislocations, and of a property that limits the lifetime as battery nanoelectrodes swell and Interfaces. One mesoscale property and/or performance of many mate- shrink during ion transfer with the that is basic to metals is ductility or rials is resistance to cyclic loading electrolyte, cracks may degrade the ability to flow plastically without or fatigue. In many materials, the essential passivation layers, leading breaking, which depends on how extent of loading that leads to fatigue to dendrite formation that shorts the

16 From Quanta to the Continuum: Opportunities for Mesoscale Science

device and risks fire or explosion. In applications. These emerge sponta- measurements (see sidebar “Imaging nanophotonic devices for photovol- neously rather than being imposed Mesoscale Structure” on page 19). taic solar energy capture, defects may externally. Our ability to identify and Small angle scattering is appropriate trap and de-excite charge carriers or manage these competing interactions for studying the pore sizes: diffraction, destroy the long-range nanostructure and the associated defects or defect for measuring interatomic spacing organization, degrading performance. mesostructures thus defines how and strain, and radiography, for deter- Other examples include mesoscale much value these phenomena can mining macroscopic dimensions. A heterostructures, where interfaces offer in practice. challenge for future instrumentation control properties and may enhance is to establish the simultaneous appli- or damage performance. Research Directions cation of such techniques. Specific directions include: New experimental capabilities must Self-organized Assembly. Pheno- be developed to observe the evolu- • Development of new methods to mena such as superconductivity, tion of three dimensional defect find, observe, and model func- magnetism, and multiferroic behavior formation and evolution in situ with tional defects evolving in three rely on long-range collective interac- micron-scale resolution while the dimensions and in sub-surface tions of atomic-scale components material is exposed to extreme envi- systems. Such information is a in complex materials. Structural or ronmental conditions. New theory and prerequisite to managing and compositional defects can substan- simulation capability must be fostered mastering the progressive evolu- tially degrade or even destroy these at the length scale of these nucleation tion of isolated defects into meso- phenomena and their utility for appli- and growth events in order to best scale defect patterns that limit the cations. On the other hand, the extent extract insight from the experiments range of collective response or of long-range collective behavior is and tangibly test hypotheses. X-ray into extended defect assemblies limited by other, typically unknown, or neutron scattering techniques that locally alter the mechanical mechanisms, which constitute the are appropriate for these kinds of response of the system and, “defects” of importance in these for example, initiate cracks that

Electrochemical Storage. Power and energy density for batteries and supercapacitors can be enhanced by using materials in nanostructured and microstructured formats, enabling fast ion and electron transport pathways. Functional defects may occur at any of these length scales, e.g., randomness at the nanoparticle scale introducing highly tortuous ion-transport paths or discontinuous electron transport paths, or dimensional variability that impedes transport in the fabricated microstructures. Adapted from Ming Tang (Lawrence Livermore National Laboratory).

17 From Quanta to the Continuum: Opportunities for Mesoscale Science

grow and ultimately fracture the Implicitly, we are always looking for between resolution and speed. For system. Defect evolution, i.e., an an aggregate property that controls example, in fatigue crack forma- explicit focus on dynamic defect materials performance. Dislocations tion, much of fatigue life comprises phenomena, must acquire the are a prime example. We understand waiting for crack(s) to form. Thus, importance already attributed individual dislocations very well at a substantial fraction of materials to the static study of structural the nanoscale, even in low symmetry lifetime derives from the underlying defects. It is the dynamic evolu- materials. When millions of disloca- plastic deformation mechanisms, tion of defect structures and tions flow and interact, however, their i.e., dislocation flow and localization. patterns in materials systems that pattern formation is not well under- Ultrasonic cyclic testing is being used directly translate to performance stood and requires supercomputing to perform accelerated testing, but a and reliability limitations. to simulate, even when only a few great deal of theoretical work needs • Development of defect decoration dislocations are present. And yet, to be done to relate these results to techniques, employing selective this pattern formation controls where nanoscale understanding. This under- chemistry or other phenomena, we find slip steps, surface rough- standing and new capability will allow for finding and watching func- ness and cracks, void formation, and us to simulate these statistical events tional defects. cracking of protective surfaces. Thus, under extreme loading conditions of the collective behavior of dislocations strategic importance and to design • Conduct of accelerated testing controls materials performance but new materials for improved perfor- in new modalities, based on in a way that we do not know how to mance. We will also be able to design advanced mesoscale character- quantify or calculate. techniques for accelerated testing ization methods, for the predic- and to establish the validity of such tion and controlled evolution of A key idea is that, although defects are procedures. functional defects. traditionally thought of as detrimental, • Integration of data mining strate- they can be engineered to promote Potential Impact gies, as developed for applica- useful and functional behavior. The Bulk material properties need to be tions from security to marketing, efficient study of defects requires quantitatively connected to atom- with data-rich materials science accelerated testing, remaining istic, fundamental understanding instrumentation (e.g., computed mindful that there is a trade-off in a predictive manner. The gaps in tomography

and orientation mapping), particularly at advanced user facilities, to answer questions such as the following: How do we find features in large three dimensional computed tomographic images? How do we correlate data from disparate sources such as images and acoustic emission? Electromigration Failure. At high current density, metal atoms can diffuse along grain boundaries, leading to void formation, Electromigration Failure. At high current density, metal atoms can diffuse along grain boundaries, leading to void forma- elevated current density and heating around the void, and ultimately breakage of wiring. In a serendipitous event, Francois d’Heurle tion, elevated current density and heating around the void, and ultimately breakage of wiring. In a serendipitous event, (IBM) discovered that dopants (e.g., Cu or Si) introduced into Al wiring segregate at grain boundaries and retard electromigration. MFrancoisesoscale d’Heurlescience o (IBM)ffers p rdiscoveredomise to dra thatmati cdopantsally shor ten(e.g., the Cutim eor be Si)tw eenintroduced such ad vintoanc eAls. wiringGiven thesegregate proposed at co grainmbina boundariestion of large scandale retardsimula tielectromigration.Mesoscaleon and advances in characte rscienceization, toffershis rep promiseresents a to p rdramaticallyoblem that m ishortenght have thebeen time so lvbetweened far m osuchre rap advances.idly. Photo fGivenrom fo rtheum. xcproposedpus.com .combination of largescale simulation and advances in characterization, this represents a problem that might have been solved far more rapidly. Photo from forum.xcpus.com. 18 From Quanta to the Continuum: Opportunities for Mesoscale Science

our current mesoscale knowledge empirical field in which macroscale example, development of a scientific severely limit our ability to optimize behavior is measured and variability basis for accelerated testing would materials performance. accounted for by means of safety be immensely valuable. An old but Advancing the connections between factors, for example, by limiting oper- still relevant example is how Paul D. individual defects at the nanoscale, ational stresses to half the projected Merica’s theoretical description of which are mostly well understood, limit. Improved understanding would phase diagrams for aluminum alloys and defect systems and architec- help to define more-resistant, more- in the early 20th century rapidly accel- tures at the macroscale, which limit reliable ways to assemble materials erated development and utilization of materials performance and are poorly while maintaining beneficial micro- diverse aluminum alloys in industry. understood, will accelerate discovery structure. Quantitative formulations There is a similar potential impact and enable active defect manage- of the links between defect distribu- from mastering defect mesostructure ment. Today, materials performance tions and materials performance will and evolution on almost the whole in relation to mechanical/thermal/ promote innovation of new products range of manufactured devices and electrochemical cycling is a highly and services with enhanced materials machinery from fuel cells to airplanes. reliability (see figure on page 18). For

Imaging Mesoscale Structure and Materials Response Using High Energy X-rays High Energy X-ray Diffraction Microscopy (HEDM) has emerged as a powerful measurement technique that probes mesoscale structure and evolution inside of bulk materials. The ability to see inside of millimeter-sized samples with microscale resolution of polycrystal orientations and strain states promises to allow rigorous tests of computational models of materials response to thermal, mechanical, and chemical stimuli. This new and unique capability can be expected to lead to deeper understanding of structure-property relationships and to predictive models of materials responses. Coupling HEDM to modeling can play an important role in shortening the time scale for development of new materials and better understanding of service lifetimes in a broad spectrum of applications.

HEDM and similar techniques are implemented at third-generation synchrotron facilities such as the Advanced Photon Source at Argonne National Laboratory. These sources yield high brilliance beams at energies >50 keV that can be focused to the microscale and that penetrate millimeter distances into materials across much of the periodic table. State-of-the-art area detectors allow collection of large terabyte data sets that probe complex materials problems in unprecedented detail. The reduction of these data sets to physical quantities relies on the availability of high-performance computing systems. Making comparisons to model calculations or employing models to compare directly to observed diffraction images also requires advanced computational techniques and will require the development of statistical methods of comparison and the use of the mathematics of optimization.

The images at left show two representative data sets illustrating the power of HEDM microstructure mapping, a method that yields the orientation field that specifies the orientation of crystal unit cell axes over three dimensions. Future work will combine these measurements with elastic strain sensitivity. The top figure shows a volumetric map of ~0.3 mm3 of nickel as mapped with HEDM. This same volume of material has been re-measured successively as the sample was annealed five times. The data sets are being used to test traditional models of grain growth, a central problem in microstructural science for at least 60 years. The data allow the evolution of the grain boundary network to be tracked both by statistics and through the tracking of individual boundaries. The lower figure shows a combined HEDM map and tomographic measurement of a copper wire being deformed in tension in the experimental setup. The gray tomographic image (cut away on the left side) shows the shape of the wire after substantial strain. The small red object in the center of the puckered narrow section of the wire is a void that formed during ductile deformation. The colored regions represent crystalline grains. These measurements will allow the correlation of damage (intra- grain defect accumulation and void formation) with specific microstructural features. Very recently, the resolution of HEDM has been dramatically improved by a series of new lensless x-ray imaging tools, such as tomographic ptychography, and diffractive imaging. Source: R.M. Suter, Carnegie Mellon University.

19 From Quanta to the Continuum: Opportunities for Mesoscale Science

Regulating Coupled can be exceedingly complex at the better fundamental understanding of nano- to microscales. Adding to this emergent mesoscale colloidal and Reactions and complexity are the chemical reac- interfacial phenomena. Armed with Pathway-dependent tions occurring along these pathways such knowledge, we might be able Chemical Processes that can irreversibly alter permeability to determine how to trigger colloidal Characterizing and controlling the and porosity as well as the composi- instability under the extreme chemical mechanisms and kinetics of pathway- tions of fluid, gas, and solid phases, and physical conditions typical of the dependent chemical processes on depending on the physicochemical subsurface environment such that the mesoscale are central to a broad conditions. Subsurface mesoscale fractures are sealed by aggregation of spectrum of important technological colloidal and interfacial phenomena nanoparticles. One of the key issues are another area of critical impor- with chemical trapping of CO is the areas such as carbon sequestra- 2 tion, heterogeneous catalysis, energy tance. The current theories of colloidal slow kinetics of mineral carbonation storage systems, functional materials, stability are not adequate for accu- reactions, resulting from low reac- and reactive transport and interfacial rate predictions of the behavior of tive surface areas as well as surface chemical processes in the subsurface colloids under the extreme conditions passivation caused by formation of (see figure on next page). in subsurface environments; in partic- unreactive layers of silica. Solutions ular, we do not know how to trigger to these problems could potentially

their different functionalities in meso- lead to commercial CO2 sequestration Opportunity porous media. operations on a scale large enough to Multiphase physicochemical trans- sequester the billion-plus metric tons port and interfacial processes are An example of the important role of CO2 produced annually from the pervasive in synthetic permeable played by the flow of natural fluids burning of fossil fuels. media for energy conversion and in and gases through solid media is natural systems such as the Earth’s the recovery of oil and shale gas In the synthetic world of materials surface. The photo of Earth from the from subsurface reservoirs through and catalysis, pathway-dependent 1972 Apollo 17 mission shows that the enhanced recovery methods involving mesoscale chemical processes are Earth’s surface is dominated by inter- injection of water/CO2 mixtures under also of enormous importance. An faces involving solids, liquids, and supercritical conditions. Another is excellent example is the polymer gases on a global scale and provides the sequestration of CO2 in subsur- electrolyte membrane (PEM) fuel cell, a hint of the enormous importance of face rocks, through either physical which uses H2 fuel and O2 from air to interfacial chemical reactions among trapping in depleted oil and natural produce electricity by catalytic split- these three phases of matter in Earth’s gas reservoirs or chemical trapping ting of H2 into protons and electrons highly fractured crust. These coupled by reaction of CO2 with appropriate on the anode side of the membrane, reactions control the composition of minerals such as Mg silicates; these then transport of only the protons our environment, including the atmo- processes result in mineral carbon- through the membrane to the cathode sphere, oceans, groundwaters, and ation and permanent CO2 seques- side while the electrons travel along the soils derived from interactions tration. One of the key issues in CO2 an external circuit to the cathode, of atmospheric gases and natural sequestration through physical trap- producing an electrical current. For waters with solid phases. Aqueous ping is leakage through tight frac- practical application, the PEM fuel cell fluids, liquid hydrocarbons, and gases tures and well grout in the subsurface requires a variety of improvements, flow through soils, sediments, and environment. This problem could be including better-performing catalysts permeable rocks along pathways that mitigated significantly if we had a and membranes.

20 From Quanta to the Continuum: Opportunities for Mesoscale Science

Pathway-Dependent Chemical Processes. Understanding such processes on the mesoscale would have many benefits, including accurate prediction of interfacial chemical reactions in natural systems such as the Earth’s surface, mitigation of leakage associated with carbon sequestration in the subsurface, and development of improved membranes for fuel cells.

There are many mesoscale chal- transport in mesoscale architec- and storage, and gas separations. lenges associated with membranes tures. We do not know how to engi- Another example is the emerging in various applications, such as those neer this integration at present. An field of reticular chemistry, which

used for H2 fuel cells and for reverse opportunity in this area is the design includes the design and synthesis of osmosis in the purification of drinking and synthesis of hybrid materials that a new class of mesoporous materials water. For example, integrating nano- combine “hard” inorganic nanocrys- known as metal-organic framework materials into mesoporous architec- tals with “soft” polymeric materials structures, which can be used for tures would allow the benefits of the in one, two and three dimensional gas storage, purification, separation, high surface area of nanomaterials to mesoscale architectures that could catalysis, sequestration, hydrogen

be combined with the efficient mass be used in CO2 capture, H2 generation storage, and other applications.

21 From Quanta to the Continuum: Opportunities for Mesoscale Science

We have little fundamental under- The design of mesoscale catalytic designing new porous materials for standing of the properties of liquids in processes presents an opportunity energy harvesting, energy storage, confined geometries. An example of for major advances in the control of stored-energy actuation, or catalytic an opportunity in this area is exploi- pathway-dependent chemical reac- applications. In many cases, unique tation of the self-organization of tions that could lead to entirely new charge-transfer-induced changes in water around small molecules such catalytic processes for the efficient physical properties are observed as as methane, resulting in clathrate conversion of complex feedstock surface chemistry is manipulated. hydrates, and around protein mole- materials into fuels and chemicals Addressing this important question cules, affecting their binding, recog- based on new understanding of how will lead to predictive capabilities and nition, and enzyme catalysis. Local to control coupled reaction/transport feedback loops for designing new solvent organization must also play processes. Molecular science has mesoporous materials for energy a role in non-enzymatic catalysis, but made great strides in understanding storage and catalysis. we have no systematic understanding how individual active sites promote of how or when this occurs. This lack bond breaking/making processes. This is the right time for addressing of understanding of the structure Control of catalytic conversions, which some of the challenges associated and properties of confined water and involve multiple coupled reaction with pathway-dependent chemical other solvents opens new opportuni- steps, will require an understanding processes at the mesoscale because ties for novel approaches to promote of how to transport intermediates of major advances in high-perfor- and control chemical transformations between active sites with differing mance computation and develop- that take into account local solvent functions and to design active sites ment of new molecular-level and organization and interactions. for specific transformations. Although mesoscale simulation codes. This is progress toward understanding also an opportune time for advancing Another example of how under- catalytic processes at individual our understanding of mesoscale standing pathway-dependent chem- active sites has been significant, we phenomena because of new devel- ical reactions could help advance currently do not know how to control opments in imaging tools, such as energy-related needs is the use of interactions between active sites that coherent diffractive imaging, synchro- hierarchically structured porous elec- are needed to direct the cascade tron-based 3-D tomography, and trodes for energy storage in batteries. of reactions in complex, multi-step x-ray absorption nanospectroscopy Most batteries today consist of catalytic conversions. Biology has that is element and chemical-state porous electrodes produced by mastered such complex conver- specific and provides information simple random packing of nanopar- sions by using a variety of meso- not obtainable with laboratory x-ray ticles and microparticles. The central scale processes to control coupled sources. question that needs to be addressed reaction and transport processes at is how to organize porous channels at the level of enzymes and cells. We Meso Challenges different length scales in battery elec- have the opportunity to learn from In the natural world, pore-scale (~μm trodes to maximize charge transport nature how to design new inorganic to cm) and field-scale (~10 cm and efficiency and minimize chemical and catalytic materials and processes. larger) heterogeneity has long been mechanical degradation. The tailoring Another area ripe with opportunity in recognized as a source of complexity of porous electrodes on the meso- heterogeneous catalysis addresses and uncertainty in subsurface fluid scale is a largely unexplored research the question of how electronic struc- transport. It is clear, however, that direction that, if successful, could ture is correlated with the morphology the largely ignored mesoscale (100 accelerate advancement in energy and mesoscopic structural stability of nm to μm) plays a critical role in storage systems and push the perfor- catalysts. Moreover, understanding multiphase flow of CO and brine in mance and lifetime limits of batteries 2 how mechanical and electronic prop- subsurface rocks. It is also becoming by optimizing “degrees of freedom” at erties evolve as the surface chem- clear that chemical reactions between the mesoscale that have rarely been istry of solids is manipulated is one these gas-fluid mixtures and the explored. of the fundamental questions in

22 From Quanta to the Continuum: Opportunities for Mesoscale Science

mineral-lined pathways though which challenge is to develop a fundamental of nano-meso architectures through they flow can have a major influence on understanding of the microscopic mesoscale simulations and advanced (i) the connectivity of the pathways, processes that occur on the surfaces characterization. (ii) the ability to recover oil and natural of porous materials (or meso-scale gas more efficiently and with less constructions of nanomaterials) under Research Directions detrimental environmental conse- various environmental conditions. Experimental, theoretical, and quences (e.g., serious groundwater In this context, it is crucial to under- modeling approaches are needed to pollution from fracking of shale-gas stand the role of charge transfer on address the challenges discussed reservoirs), and (iii) the safe, long- surface stress, as well as the correla- above and thereby lead to new under- term sequestration of CO in subsur- tion between the change in electronic 2 standing of pathway-dependent face rocks. Understanding reactive structure and morphology or struc- chemical processes and coupled transport at the mesoscale provides tural stability. In the area of energy reactions at the mesoscale. In all of the opportunity to develop a rigorous storage, the mesoscale challenges the areas involving flow of liquids and understanding of the major controls are how to fabricate and characterize gases through mesoporous media, on CO migration at the lower end mesostructures in porous electrodes 2 ultra-high resolution imaging capable of the spatial scales where quantum with flexibility, precision, and effi- of distinguishing between different meets continuum descriptions of the ciency, and how to accurately model elements and different chemical multiphase physics needed to bridge charge transport and degradation states of a given element is needed. between mesoscale and field-scale in electrodes containing multi-scale Conventional x-ray tomography theory and experiments. A number (nano, meso, micro) porous channels. methods using commercial comput- of challenges must be addressed In the area of fluid and gas trans- erized tomography (CT) scanners are if progress is to be made in solving port through mesoporous media, widely used to provide 3-D images of these problems. We must develop the challenges include the following: rock cores and other porous media, better ways of characterizing meso- (1) better observation, characteriza- but spatial resolution is limited, and scale pore structures of complex, tion, and quantification of complex there is no element or chemical-state heterogeneous geological media and and coupled meso- and micro-pore specificity. Recent developments in their evolution under field-relevant scale processes, (2) more effec- synchrotron-based CT scanning at conditions. In addition, we must be tive ways to translate, abstract, and different x-ray energies coupled with able to follow chemical reactions with simplify their collective behavior to diffraction-based microscopy as well elemental specificity at the mesoscale achieve predictive continuum models, as improvements in resolution to the in real time in order to understand (3) new approaches for understanding few-nanometer level through tomo- how reaction intermediates and final gas flow in mesoporous media (the graphic ptychography and scanning products affect porosity and perme- continuum description of fluids is microscopies, coupled with simulta- ability in a dynamic fashion. not adequate in the slip or transi- neous nano-XANES (x-ray absorption tional flow regimes, and molecular near edge structure) spectroscopy, The same mesoscale challenges dynamics simulations are prohibitively are beginning to provide unique infor- discussed above are also true for expensive due to the large number of mation on compositional heterogene- the synthetic world. For example, in molecules that must be considered), ities at the mesoscale and complex mesoscale catalytic processes, the (4) improved knowledge of the opti- 3-D nano-meso architectures that goal is to learn how to control coupled mum architectures and bonding motifs can be optimized for fast mass trans- reactions and transport over 10s to for efficient gas transport through port. These studies will provide the 100s of nanometers, rather than at mesoporous inorganic-organic hybrid necessary feedback for nanoporous the macroscopic levels encountered, materials, (5) better integration of materials synthesis. A similar imaging for example, in scale-up of catalytic nanomaterials into mesoporous approach will provide new insights to processes. New emergent behavior architectures using robust synthesis the design and functioning of cata- is expected from reaction/trans- approaches based on templating and lysts and battery electrodes under in port coupling at the mesoscale. The self-organization, and (6) optimization

23 From Quanta to the Continuum: Opportunities for Mesoscale Science

operando conditions (see sidebar on transformations occurring during the with gas-specific binding modalities “Synchrotron-Based High-Resolution flow of liquids and gases through in nanocrystals, enables targeted 3-D Tomography In Operando porous geological media are needed end-use design not possible today. Conditions” on page 25). Based on to understand how to enhance the New metal-organic framework struc- these early successes, new hard kinetics of mineral carbonation reac- tures have similar potential for enor- x-ray synchrotron beam-line facilities tions required for permanent seques- mous improvements in selective gas

are needed in the U.S. to facilitate in tration of CO2 as well as the long-term storage. situ characterization studies of meso- integrity of shale and clay caprocks In the area of heterogeneous catal- porous materials. In addition, small that prevent the leakage of CO2 angle x-ray and neutron scattering will from deep saline aquifers. There are ysis, detailed understanding of how be useful in characterizing the meso- many fruitful research avenues that to control coupled reaction/transport scale pore structure of membranes as will lead to major advances in our processes will enable the design of well as shales, clay rocks, and sand- understanding of pathway-dependent entirely new catalytic processes for stones under the extreme conditions chemical processes in mesoporous efficient conversion of complex feed- of the subsurface. New functional media of all types. stock materials to fuels and chemi- optical probes, such as nanotubes cals. In addition, new understanding with tunable optical and physical Potential Impact of solvent organization in enzymatic and non-enzymatic catalysis, as well properties, as well as high-speed near There are many potential impacts as intentional encapsulation of cata- infrared microscopy, can be used to of mesoscale research on pathway- lysts, has the potential to enable more visualize transport through mesopo- dependent chemical processes. In general control of reaction selectivi- rous media. the area of geological sequestra- ties and rates. tion of CO , successful commer- In theory and modeling, advances 2 cial-scale physical and/or chemical in high performance computing In the area of energy storage, trapping of CO will help solve one now allow a variety of approaches, 2 fundamental understanding of how of the major environmental prob- including use of the lattice Boltzmann nanoscale surface properties of lems facing humankind (see sidebar method to model transport and inter- battery electrode materials affect “Earth Science: A Meso Frontier” on facial processes in mesoporous macroscopic performance will lead page 26). In a more general sense, permeable media, as well as kinetic to development of a predictive capa- understanding what drives mesopo- Monte Carlo modeling of complex bility and feedback loop for designing rous flow and transport will poten- catalytic processes. At a more funda- new porous electrode materials. tially benefit areas as diverse as drug mental level, density functional theory This understanding will also accel- delivery, ultrafiltration, and nanoflu- (DFT) studies coupled with ab initio erate advancement in energy storage idic energy conversion. In synthetic thermodynamics of solid surfaces in systems and push the performance mesoporous systems, novel polymer- contact with liquids and gases can be and lifetime limit of battery elec- nanocrystal hybrid materials with used to provide critical information on trodes by optimizing “degrees of controlled nanoscale to mesoscale the solid-liquid interface structure and freedom” at the mesoscale that have architectures represent a new form of its properties under reactive condi- rarely been explored. The develop- matter that could potentially enable tions. In addition, controlled synthesis, ment of nanomaterials with integrated unprecedented control over selective assembly, and deposition of catalytic mesoporosity will improve the power gas storage, separations, and trans- scaffolds are critical to model devel- density, efficiency, and throughput of port. The broad range of gas solu- opment. Experimental studies of next-generation energy storage and bilities in polymers, when combined the kinetics of reactions and mineral conversion materials, and thus has game-changing potential.

24 From Quanta to the Continuum: Opportunities for Mesoscale Science

Synchrotron-Based High-Resolution 3-D Tomography under In Operando Conditions

A research team from SLAC National Accelerator Laboratory (Piero Pianetta et al.), Lawrence Berkeley National Laboratory (Jose Cabana et al.), FBK in Povo, Italy (F. Meirer), and Xradia has characterized NiO/Li battery electrodes under in operando conditions in an electrochemical cell using 3-D tomography and elemental mapping at the 40-nm resolution level. Detailed, spatially resolved information on the chemical state of Ni was provided by x-ray absorption near-edge structure (XANES) spectroscopy at a pixel size of 15 nm. The transmission x-ray microscope on beam-line 6-2 at the Stanford Synchrotron Radiation Lightsource (SSRL) was used for this imaging and nanospectroscopy. The 3-D x-ray tomograph and Ni K-edge XANES spectrum of the NiO battery electrode below show that NiO (red) is being reduced to Ni metal (green) during the first charge cycle of the operating battery, and that Ni metal is not fully reoxidized to NiO, which shortens battery lifetime and function. This state-of-the-art 3-D tomographic imaging that is element and chemical-state specific has provided important new insights about the irreversible changes that occur in the battery electrode materials that directly address the important problem of how to organize porous channels at different length scales in battery electrodes to maximize charge transport efficiency and minimize chemical and mechanical degradation.

This same 3-D tomographic approach has provided critical new information on mesoscale changes in Fe2O3-Fe2TiO5 catalysts in a project led by Bert Weckhuysen and Frank de Groot at SSRL, and it is likely to provide important new insights about path-dependent

chemical processes at the mesoscale in all of the other areas discussed in this section, including CO2 capture and sequestration and gas-phase separations. Current laboratory-based x-ray tomographic imaging using commercially available CT scanners is limited to spatial resolutions of ≈ 0.5 µm and is not capable of element- or chemical state-sensitive mapping. Thus, this new synchrotron-based tunable energy, hard x-ray tomographic imaging, represents a major new development that is potentially transformative.

25 From Quanta to the Continuum: Opportunities for Mesoscale Science

Earth Science: A Meso Frontier From groundwater flow to petroleum formation and extraction to carbon sequestration, mesoscale phenomena control the Earth’s dynamics and its human impact.

The pores in rocks and soils play a primary Oil and gas form by chemical reac- role in the formation and dynamics of the tion of organic matter in sedimentary earth’s crust. Sedimentary rocks that rock promoted by the high pressure and cover much of the earth’s crust are formed temperature of buried strata. After forma- by accumulation of mineral and organic tion times of millions to hundreds of material that were created by weath- millions of years, these fossil fuels may ering and erosion and transported to the be released if the rocks are sufficiently place of deposition by water, wind, ice, or permeable and rise through pores until glaciers. The mineral and organic grains stopped by an impermeable cap rock to form strata that are compacted by pres- form reservoirs that can be tapped by sure from above to form rocks of various drilling. Less permeable organic shales porosities, depending on the size, compo- retain their oil and gas, a rich source of sition, and dispersity of the constituent unconventional fossil fuel that can be grains and the pressure and heat applied tapped by hydraulic fracturing to produce to the buried strata. porosity that releases the trapped petroleum to flow to a reservoir or well. The pores in sedimentary rocks are critical to the dynamics of the earth, including Carbon sequestration in the pores of the availability of groundwater, the forma- sedimentary rock provides an opportu- tion and extraction of oil and gas, and the nity to store, for thousands of years, the sequestration of carbon dioxide. Ground- carbon dioxide produced by fossil fuel Pores in sandstone, a sedimentary rock formed water, the largest source of available combustion that now supplies over 80% by accumulation of many sizes and shapes of freshwater, flows through pores in sedi- of the world’s energy. Migration of carbon mineral and organic grains. The pores range in size from nanometers to microns, depending mentary rock to replenish aquifers that dioxide underground, the possibility of on the dispersity of the grains and the pressure provide water for agriculture, drinking, and leakage to the atmosphere where it would and temperature of formation. Source: sanitation. As groundwater flows, it reacts contribute to global warming and climate http://darkwing.uoregon.edu/~drt/ with and dissolves the surrounding porous change, and reaction with the rocks in Classes/201_99/Rice/Seds.html. rocks, providing nutrients for plants and which it is sequestered are critical issues increasing or decreasing pore size and that require mesoscale science to resolve flow properties. The mesoscale dynamics before sequestration can be implemented of groundwater flow are critical to solving on a significant scale. challenges of freshwater supply, aquifer replenishment, and contamination with toxic pollutants.

26 From Quanta to the Continuum: Opportunities for Mesoscale Science

Optimizing Transport but high charge density favors large- Meso Challenges grained material. The constraints of The research challenges are best and Response reproducible cycling and good elec- described in terms of specific exam- Properties by Design trical connectivity provide further ples; in all cases, a common theme and Control of challenges. emerges: the need to discover, design, Mesoscale Structure fabricate, and probe mesoscale- There are other situations in which ordered superstructures and archi- Competing degrees of freedom and the important functional property tectures. We begin with a problem system constraints involving coupled of a material emerges from the self- relevant to batteries and supercapaci- electrons, photons, and atomic organized mesoscale structure in tors. The transport of ions, such as structures determine intermediate response to competing interac- lithium, into and out of the electrodes length scales that can potentially be tions. For example, high-temperature of a battery causes expansion and exploited in functional mesoscale superconductivity appears in mate- shrinking of the electrodes. The repe- systems. rials in which antiferromagnetism tition of these processes as a battery competes with the mobility of the is cyclically charged and discharged Opportunity charge carriers. Competition among tends to cause the electrodes to dete- In the process of optimizing mate- antiferromagnetic and Coulomb inter- riorate, thus reducing the amount of rials for energy-related applications, actions and the tendency of electrons charge that can be stored. To improve a compromise between competing to delocalize frequently lead to elec- the performance and lifetime, the requirements often leads to structural tronic modulations on an intermediate strain effects can be reduced if the architecture on an intermediate length length scale. Somehow, supercon- electrodes are formed from small scale. This observation is especially ductivity emerges with long-range grains of material. At the same time, true when one needs to couple light order. Another example is the case the grains of the mesostructured elec- or electric and magnetic fields, with of colossal magnetoresistance mate- trode require good electrical conduc- the motion of electrons, ions, or rials. There, interactions involving tivity to allow electrons and ions to spins. For example, in designing an magnetic moments, electronic separate or recombine in an efficient efficient solar cell to convert sunlight charges, and lattice distortions result fashion. At present, the design of to electricity, one has to balance the in mesoscale structures that are very electrodes proceeds by trial and error. challenges of using as much of the sensitive to temperature and applied By characterizing and modeling such incident light as possible to generate magnetic field, resulting in a strong systems, we will develop the capability electrons (or electron-hole pairs) while coupling between electrical conduc- to design and control the mesoscale effectively collecting the electrons at tivity and applied magnetic field. structure and composition in order an electrode (and sending the holes to optimize capacity and sustainable to the opposite electrode). The light The importance of mesoscale struc- performance simultaneously. absorption and electron transport ture to the design and optimization functions put different demands on of functional materials and systems Semiconductor quantum dots have the architecture. Another example has become increasingly apparent in shown great promise as the active involves the design of improved recent years. Experimental and theo- material for light-harvesting and light- batteries for storage of electrical retical tools have developed to the emitting devices that are efficient energy (see figure on next page). The point that we have an opportunity to and have low cost. Plasmonic metal coupling of ion transport to ion inser- make substantial progress in deter- nanoparticles can greatly enhance tion and removal from electrode mate- mining and understanding the rules absorption, emission, and transport rial results in competing demands. To behind external and intrinsic meso- of excitations in such devices. Full get ions into and out of the electrode scale structure, so that we can design exploitation of these effects requires material, one would like a large surface and control materials and systems to advanced understanding and control area relative to the volume, pointing improve their performance. over the interactions among disparate to small-grained electrode material, nanoscale ingredients in an extended structure. The challenge is to design,

27 From Quanta to the Continuum: Opportunities for Mesoscale Science

fabricate, and probe mesoscale- object by running an electrical current. realistic for a practical thermoelectric ordered superstructures, arrays, and For this process to be efficient, there device as the device must be quite architectures. must be little heat transport by lattice large compared to the substructure, vibrations. Experiments have shown which would be expensive to create. The transport of heat by electrons in that this condition can be achieved The challenge is to find suitable mate- thermoelectric materials allows one by the suitable nanostructuring of rials with natural mesoscale structure to convert thermal energy to elec- an appropriate material. The artificial or to find processes for effectively trical energy or, conversely, to cool an nanostructuring of materials is not assembling smaller objects into a practical device.

Self-adaptive Electrodes. Conventional battery electrodes with predetermined crystalline structures often undergo phase transitions upon inter- calation of transporting ions. These phase transitions cause swelling of the electrode materials, resulting in local atomic rearrangements, limited diffusion of ions, and ultimately capacity degradation. In recent work, a new bio-inspired evolutionary approach has been demonstrated. Starting from amorphous, low-crystallinity materials, repeated voltage cycling leads to interconnected porous mesoscale electrodes that undergo self- organization into an optimized crystalline phase. This is a promising approach that requires further testing and study. Source: H. Xiong et al., J. Phys. Chem. C 116, 3181-3187 (2011); S. Tepavcevic et al., ACS Nano 6, 530-538 (2011).

28 From Quanta to the Continuum: Opportunities for Mesoscale Science

Resistance-free transport of electrons Characterization

in high-temperature superconductors A range of characterization tech- typically coexists with self-organized niques is needed. One certainly electronic heterogeneity that exists needs to be able to map the spatial on a length scale of ten times the structure with diffraction techniques interatomic spacing. By controlling that employ x rays, neutrons, and the chemical and lattice structure, electrons. One would also like to one should be able to enhance the obtain chemical and electronic varia- current-carrying capacity and oper- tion, as determined by spectroscopic ating temperature of superconducting techniques such as photoemission compounds. There is no theoret- and optical absorption, with the ical consensus on the nature of the same spatial resolution. In some electronic heterogeneity or its effect cases, one needs to map magnetic on superconductivity. Experimental Superconductivity. Correlated-electron order and orientation. It is important characterizations of the electronic materials can exhibit self-organized meso- to be able to measure variations in inhomogeneity are challenging and scale structures. This image shows elec- order or structure when various fields tronic modulations in the copper-oxide high- incomplete (see figure at right). are applied, and to follow the time temperature superconductor Bi Sr CaCu O , obtained by spectroscopic- dependence of such variations. 2 2 2 8+δ Ferromagnets couple their magneti- imaging scanning-tunneling microscopy. Image is 12 nm on a side. The relationship zation to an external magnetic field. In between the modulations and the supercon- a multiferroic material, the magnetic ductivity remains to be understood. order is coupled to a ferroelec- Y. Kohsaka et al., Science 315, 1380 (2007). tric modulation of the lattice, thus allowing the magnetic order to couple to an applied electric field and the Research Directions ferroelectric order to respond to an Synthesis applied magnetic field (see figure at Exploratory synthesis remains key to far right). The response of real mate- discovering new possibilities in materials is complicated by the formation rial behavior and performance. While of domains, with the orientation of the we aspire to gain the theoretical under- magnetization and/or electric polar- standing that will allow rational design ization varying from one domain to of optimal systems, the science of the next. For example, a ferromagnet materials with mesoscale structure is tends to break up into domains with presently dominated by experiment. the magnetization reversing direc- A broad range of synthesis efforts tion from one domain to the next; the is necessary, from growth of single

size of the domains is determined by crystals, to atomic-scale growth of Polarization. A multiferroic material can have a multilayered films, then to assembly the competition between the energy complex domain pattern. The top image shows cost of forming domain walls and the of nanoparticles. The performance the ferroelectric domain pattern at the surface

dipole interaction energy between the of particular materials can often be of a crystal of YMnO3, obtained with an optical domains. The domain structure occurs tested in the laboratory with measure- microscope after selectively etching the surface to distinguish domains of upward and downward on an intermediate length scale that ments of conductivity, photoabsorp- polarization. The domain areas are not equal impacts the way in which the material tion, etc. Characterization of the because the crystal was cooled from a higher couples to applied fields. An improved mesoscale structure often requires temperature in an electric field. The coloring of the lower figure reflects the existence of three inequiv- understanding of such domain struc- unique and specialized equipment alent structural domains, each of which can have tures is necessary to enable control of available at national user facilities. two polarization directions. One would like to be functional properties. able directly to measure both the structural and magnetic characteristics of each domain. Source: S. C. Chae et al., PNAS 107, 21366 (2010).

29 From Quanta to the Continuum: Opportunities for Mesoscale Science

Theoretical Modeling Potential Impact on an intermediate length scale. The Multiscale modeling is essential. The importance of mesoscale struc- ability to design and control meso- Many accurate theoretical techniques ture to the design and optimization scale structure of materials could have been developed for modeling of functional materials and systems lead to a more efficient electrical grid, the properties of materials on the has become increasingly apparent improved energy storage, efficient nanoscale, but it is challenging to scale in recent years. In the process of conversion of waste heat to elec- up these methods to span mesoscale optimizing materials for energy- trical energy, improved photovoltaics structure. Granular models must be related applications, a compromise and light-emitting diodes, and better developed and tested against experi- between competing requirements electrical motors and generators (see ment. The goal is to develop predic- often leads to structural architecture sidebar “Mesophotonics for Energy tive capabilities across scales. Applications” below).

Mesophotonics for Energy Applications

In mesophotonics, we explore artificially created materials (meta- Photonic bandgap crystals (PhCs) are an example of this materials) in which the structure is tailored at length scales smaller approach: they behave like “semiconductors for light”: light of than the wavelength of light of interest. Because the substruc- the frequencies within the bandgap is prohibited from propaga- ture is smaller than the wavelength, light cannot really resolve tion, while the frequencies outside of the bandgap are allowed to the structure: it appears to the light as if it were propagating in propagate. This way, by tailoring the mesostructure, one can tailor a uniform medium, whose physical properties are substantially the photonic density of states almost at will, with good agree- modified. By tailoring the mesostructure, we can tailor the laws ment between simulation and experiment. For example, within the of physics (as far as light is concerned) almost at will. This way, photonic bandgap, the density of states is zero, while at some we can create artificial materials that display physical phenomena other frequencies, it can be dramatically enhanced. PhCs have that do not exist in any naturally existing materials. For example, been explored, for example, to implement some of the lowest meta-materials have been used to demonstrate negative refrac- power-threshold lasers. Moreover, since thermal emission of a tion, as well as to implement invisibility cloaks. hot body depends on the density of states, one can also tailor thermal emission (as well as absorption) of hot bodies almost at will. Source: Y.X. Yeng et al., PNAS 109, 2280-2285 (2012).

30 From Quanta to the Continuum: Opportunities for Mesoscale Science

Elucidating Non- With their light masses, electrons temperature superconductors and the experience the mesoscale transi- manganese perovskites, have phase equilibrium and tion to quantized energy levels at diagrams with many complementary Many-Body Physics much larger length scales than atoms and competing metallic, insulating, of Electrons (see sidebar “The Particle in a Box” and magnetic phases, including novel pseudogap, charge ordered, and Mesoscale architectures that confine on page 4), as large as 10, 100, or nematic electron phases where strong electrons in semiconducting quantum even 1000 nm. Confinement on such correlation plays a central role. These dots, nanocrystals of metals, magnets large length scales is relatively easily competing interactions give rise to and ferroelectrics, and novel materials achieved in semiconducting quantum intrinsic inhomogeneity and local such as carbon nanotubes and litho- dots, small magnets, and nanocrys- structure on the mesoscale. The inter- graphically patterned graphene offer tals. The wide range of phenomena action of these intrinsic mesoscale new horizons for discovery science mediated by electrons and the structures with externally imposed and innovative technology. abundance of confining geometries make mesoscale electronic behavior mesoscale architectures has not been a fertile field for discovering new explored significantly. Confinement Opportunity phenomena, identifying organizing on strong correlation length scales Electrons occupy a special place in principles, and creating functionality is likely to dramatically change the materials phenomena. Electrons are by designing and imposing meso- observed behavior. fast and agile compared to atoms, scale structures. Confinement effects with a mass a thousand times lighter in semiconductor quantum dots have One strong correlation effect that has than the lightest atoms, and atto- applications in transistors, solar cells, been explored in confined geometries second response times a million light-emitting diodes, diode lasers, is the Kondo effect, where a fixed local times faster than the picosecond medical imaging, and quantum moment is partially or fully compen- times typical of atomic motion. Fast computing. Metals, magnets, and sated by a cloud of itinerant electrons moving electrons give metals their superconductors show equally strong magnetically polarized in the opposite high conductivity, excited electrons confinement effects, giving rise to direction. In the last decade theory enable the absorption and emission of surface enhanced electromagnetic and experiments have examined photons, and shared electrons create fields, surface plasmon propagation, the cloud of Kondo spin compensa- chemical bonds between atoms. magnetic vortex dots, and enhanced tion surrounding a single impurity Electrons mediate the interactions superconducting critical fields. moment confined in quantum dots of among atoms through their energy ~100 nm dimensions, a tour de force bands, chemical bonds, electrostatic Among electronic confinement effects, of mesoscale science revealing the and van der Waals forces, magnetic those involving strong electron corre- hidden details of the Kondo effect moments and exchange interac- lation arising from the Coulomb repul- (see figure on next page). Similar tions, and hydrophilic/hydrophobic sion of like charges or the magnetic mesoscale confinement experiments attractions or repulsions. Two of the interaction of electronic moments on other strong electron correlation five Grand Challenges in Directing are rich in consequences and remain phenomena remain to be explored Matter and Energy: Five Challenges largely unexplored. Strong correlation and will be equally revealing. for Science and the Imagination deal is at the very forefront of materials explicitly with electrons: How do we challenges, spanning Kondo compen- Quantum information processing control materials processes at the sation, heavy electron behavior, and raises a host of mesoscale science level of electrons? How do remark- Mott-Hubbard localization leading to challenges (see sidebar “Giant able properties of matter emerge from metal-insulator and superconducting- Magnetoresistance” on p. 35). the complex correlations of atomic or insulator transitions. Strongly corre- Quantum computing relies on the electronic constituents? lated materials, such as the high quantum mechanical entanglement of

31 From Quanta to the Continuum: Opportunities for Mesoscale Science

initial states of quantum bits (“qubits”) and spin compensation, localiza- to produce a final state representing tion, and local moment formation is a calculated result. The calculation is a major challenge and opportunity for defined by the nature of the quantum mesoscale science. interference among qubits; for certain calculations quantum computing The experimental side of confining is much faster than conventional strongly correlated electron materials computing. To be effective, the qubits is equally challenging. Appropriate must interact with each other strongly Quantum dot geometry for measuring the mesoscale boxes must be found to enough to reach the final state but Kondo spin compensation cloud around confine these materials on length interact with the environment only a single magnetic moment. The small scales that vary considerably quantum dot on the left contains about weakly to avoid disturbing the qubit 25 electrons and plays the role of a single depending on charge and spin densi- interactions that define the calcu- magnetic moment. It communicates with the ties. There is also a fundamental chal- lation. The challenge is to control large quantum dot on the right (red), which lenge to communicate the behavior of provides a Kondo compensation cloud. The the interaction of multiple degrees confined electronic phenomena to the leads on the left (blue) allow the Kondo resis- of freedom: strong enough interac- tivity to be measured. Source: R.M. Potok et macroscopic world. Electrical trans- tion among qubits to do the calcula- al., Nature 446, 167-171 (2007). port is a common probe of confined tion, weak enough interaction with systems, attractive because it probes the environment to avoid noise that the quantized energy levels directly, On the theoretical side, the roles of disturbs the calculation, and, after and the transport voltage drop charge correlation arising from the the final state is reached, a moderate grows as the confinement becomes Coulomb interaction and spin correla- interaction to read out the result stronger. Many other probes based tion arising from magnetic interactions without destroying it. on thermodynamic, scattering, are fully intertwined by the coexis- and spectroscopic measurements tence of charge and spin on each The requirement of quantum become weaker and less obvious electron—a spatial charge correlation computing to minimize interaction as the size of the sample shrinks. implies a spatial spin correlation and with external degrees of freedom is Scanning tunneling spectroscopy is a vice versa, and a local magnetic field counter to the usual goal of meso- good example of a revealing transport further splits spin correlations into scale science, which is to maximize probe that can map local electronic up and down components. When all the interactions among multiple behavior with atomic resolution over spins are itinerant, the spin compen- degrees of freedom to promote emer- mesoscale fields of view. sation of the Kondo effect and the gent behavior. Although opposite in charge correlation of the Coulomb outcome, the two objectives rest on In transport measurements electrons interaction reach their maximum a common foundation: experimen- must be injected into and removed complexity and subtlety. Electron tally observing, theoretically under- from the confined sample from meso- density is a key feature of strong corre- standing, and functionally controlling or macroscopic leads or tunneling tips, lation affecting both the charge and the interaction of disparate degrees “boxes” that are much larger than the spin compensation lengths; it differ- of freedom. Whether the goal is to confined object, whose energy levels entiates correlation effects in metals promote or inhibit, a firm observa- are continuous and whose electronic and semiconductors. Mott-Hubbard tional base and predictive under- states are spread homogeneously localization introduces another funda- standing of the interacting degrees of over large dimensions. The essential mental variable: if electronic correla- freedom are essential. elements of the confined state, quan- tion becomes too strong, itinerant tized energy and spatial variation of charges become localized, and they Meso Challenges electron density, are easily lost in the may further acquire a local magnetic transition to the relatively featureless The meso challenges for strongly moment that induces a new Kondo landscape of the leads (see figure correlated electrons fall into theoret- compensation cloud. Sorting out the on next page). Beyond the loss of ical and experimental components. roles and the interactions of charge

32 From Quanta to the Continuum: Opportunities for Mesoscale Science

information on confinement, putting taken by the spin compensation of Convenient systems for studying the macroscopic leads on meso- or nano- the Kondo effect: shrink the size of mesoscale aspects of strong elec- scopic objects is a spatial contradic- the box. Theoretical predictions in the tron correlation are the high tempera- tion: the small size advantage of a late 1990s showed the richness of the ture superconducting copper oxide confined object is lost in the large Kondo effect confined to a quantum perovskites and the structurally leads or tunneling tips required to dot, predicting the size of the similar magnetic manganese oxide communicate with it. This challenge Kondo spin compensation cloud to perovskites. The copper oxides have of communicating confined nanoscale be of mesoscale dimensions a canonical phase diagram full of behavior to the macroscopic world is (~100-1000 nm). This theoretical work strongly correlated behavior (see a fully mesoscale challenge. stimulated a series of experiments figure on next page). The six distinct looking for the Kondo compensation normal and superconducting phases of the magnetic moment of a single are all strongly correlated, extending impurity atom embedded in a semi- from the Mott-Hubbard antiferro- conducting quantum dot. Transport magnetic insulator on the left to the experiments have observed confined Fermi liquid on the right. An addi- Kondo effects in GaAs/AlGaAs tional feature of this phase diagram is quantum dots, silicon metal-oxide- the proposed quantum critical point semiconductor field-effect transis- under the superconducting dome, tors (MOSFETs), carbon nanotubes, which would introduce an entirely and single molecules. The success different kind of mesoscale behavior of these experiments has stimulated as the coherence lengths of the two a host of suggestions for exten- competing phases diverge. Metal nanoparticle (yellow) positioned sions of the confined Kondo effect between two leads (red) illustrating the size disparity between a confined nanoscale by breaking the spin symmetry with Nanoparticle arrays are a second object and the much larger meso or a magnetic field or with polarized and strikingly different approach tfor macroscale electrodes that inject and leads, connecting the dot to a resis- bringing out the mesoscale aspects of remove electrons from the nano-object. The electronic behavior (see second figure size disparity of the “boxes” representing the tive environment, imposing finite-size nanoparticle and the leads creates a dramatic effects through the leads, embed- on next page). These arrays are an artifi- mismatch in the quantized energy level ding the dot in a mesoscopic ring cial mesoscale architecture consisting structure and confined spatial variation of the to perform quantum interferometry, of confined nanoparticles, typically electronic states in the nanoparticle and the continuous or nearly continuous energy level and connecting the dot to supercon- 1-20 nm, arranged in a periodic array structure and extended spatial variation of ducting leads. These approaches in one, two, or three dimensions. The the electronic states in the leads. The quan- are rich with new behavior and ideas internal structure of the nanoparticles tized electronic behavior of the nanoparticle embracing novel theory and experi- can be crystalline (in which case they can be lost as electrons transition to the more continuous electronic structure of the leads. ment. The quantum dot approach can are called “nanocrystals”) or disor- Capturing the confined electronic behavior of be adapted to explore other strong dered. The nanoparticles are typically the nanoparticle and communicating it to the correlation effects such as electro- coated with ligands to keep the parti- larger meso or macroscopic world through cles from agglomerating; the ligands transport measurements is a major meso- static screening, heavy fermions, scale challenge. Image: Cees Dekker group Mott-Hubbard localization, and local also serve as linkers that connect the at TU Delft, The Netherlands. moment formation associated with nanoparticles in the array, determine localization. The long range of the their spacing, and may provide a path Coulomb, dipole, and exchange inter- for transmitting electrons or energy Research Directions actions guarantees that these strong between neighbors. One route to exploring mesoscale correlation phenomena are rich with confinement in strongly correlated unexplored mesoscale features. electron materials follows the path

33 From Quanta to the Continuum: Opportunities for Mesoscale Science

Nanoparticle arrays provide one solu- size is a critical parameter, allowing matter physics and chemistry can tion to the challenge of communicating the energy level spacing between be recast in mesoscale terms via confined behavior to the macroscopic quantized levels within the particle to nanoparticle arrays, opening wide world. Each nanoparticle commu- be adjusted through the size of the horizons of new or emergent behavior. nicates with its neighbors through confining box. The physical separa- tunneling or the linker ligands, main- tion and the degree of communication The synthesis of nanoparticle arrays taining the confined geometry and the between neighboring nanoparticles presents major opportunities for quantized energy level structure on constitute a second knob, tunable mesoscale science that are just begin- each side of the interaction. Contact by the length of the linker ligands ning to be explored. We are learning to with the external world is made by connecting them. The composition macroscopic leads at the edges of of the nanoparticles introduces enor- the array and covers many nanopar- mous flexibility, embracing metals ticles. The confined behavior is lost with strong or weak electronic correla- in those nanoparticles contacting the tion and wide or narrow bands, semi- leads, but survives in the many more conductors with low carrier density, nanoparticles away from the contact magnets, ferroelectrics, and multifer- area. The confined transport behavior roics. Composite nanoparticles, such in the pristine areas of the array domi- as core-shell or Janus nanoparticles, nates the transport properties and add functionality to the individual signals the confinement phenomena elements of the array. Nanocrystals to the outside world. that display strong electronic correlation, such as the high temperature The architecture of nanoparticle arrays superconductors, allow exploring the Array of nanometer-size gold nanocrystals offers many knobs to probe and tune effects of mesoscale confinement on, connected by linker ligands. Source: R.P. mesoscale behavior. Nanoparticle for example, the doped Mott-Hubbard Andres et al., J. Vac. Sci. Technol. A 14, 1178 insulator, the pseudogap, the (1996). strange metal, or the superconductor. The shape of the nano- fabricate arrays by a variety of means, crystals introduces anisotropy, including slow evaporation of the and arrays with two or more solvent from a colloidal suspension distinct kinds of nanocrystals of nanocrystals (“drop casting”), or (such as magnetic and metallic) directed assembly by DNA or polymer in ordered or disordered configu- templates. Binary arrays with two rations allow the interaction of distinct nanocrystal species and sizes distinct degrees of freedom. All have been made, as have arrays with of the hallmarks of mesoscale quasicrystalline structure. The field behavior—granularity, quantized is in its infancy. We are now learning energy level spacing, collective to control the structural parameters Phase diagram for copper oxide high temperature behavior, interacting degrees of of arrays including degree of order, superconductors (shown as a function of hole doping freedom, defects and statistical number of nanoparticle species, and or oxygen content). There are at least six strongly correlated phases, one of which is superconducting. Corre- variation, and heterogeneity— number of dimensions. As the fabri- lation is strongest on the left, dominated by the antifer- can be rationally designed and cation challenges are conquered and romagnetic Mott-Hubbard insulator, and decreases to systematically tuned in nanopar- synthesis of more complex architec- the right. The mesoscale aspects of strong correlation ticle arrays. The range of meso- tures is mastered, the challenge of are different in each phase and remain largely unexplored. The proposed quantum critical point (QCP) scale behavior that can be fabricating targeted functionalities will on the T = 0 line induces an entirely different kind of explored is virtually unlimited. come to the fore. Solar cells sensitive mesoscale behavior, as the coherence lengths asso- In principle, every phenomenon to specific photon energies, memory ciated with the competing phases diverge. Source: of ordinary atomic condensed bits as small as 1 nm, supercapacitors, C.M. Varma, Nature 468, 184-185 (2010).

34 From Quanta to the Continuum: Opportunities for Mesoscale Science

semiconductor lasers, advanced thermoelectrics, and biological and Giant Magnetoresistance: A Triumph of mechanical sensors are some of the Mesoscale Science potential functional applications. Discovered in 1988 and generally considered to mark the birth of the Potential Impact field of spintronics, giant magnetoresistance exploits the mesoscale hall- The confinement of electronic marks of interacting degrees of freedom and heterogeneous structures behavior in mesoscale architecture to create new mesoscale phenomena and functionality and has perma- (such as semiconducting quantum nently changed the landscape of computer memory. dots; metallic, magnetic, and ferro- Giant magnetoresistance controls the electric nanocrystals; carbon nano- motion of the charge on an electron tubes; and lithographically patterned by manipulating its spin, exploiting the graphene sheets) results in a host of coupling between these disparate degrees new behavior that is just beginning of freedom. A typical configuration is to be explored. Confined electronic shown in the figure. Ferromagnetic CoFe behavior described by quantized layers surround a non-magnetic MgO layer. As an electron crosses the sandwich, its energy levels differs qualitatively spin magnetic moment is oriented by the and often dramatically from bulk first ferromagnetic layer and carried across macroscopic behavior governed by the intermediate layer. The second ferro- continuous energy distributions. The magnetic layer blocks or passes the electron if its spin is opposite or aligned with Mesoscale trilayer sandwich consisting of promise of mesoscale science is the ferromagnetic layers of CoFe enclosing a the polarization of the second layer. The opportunity to tune the degree of nonmagnetic layer of MgO. Switching the magnetic polarization of the second layer confinement to any arbitrary level, magnetic polarization in the ferromagnetic is thus a switch for the electric current, and to connect multiple confined CoFe layers from parallel to antiparallel capable of altering it by up to 200%. switches the electrical resistance by up electronic boxes expressing different For memory applications, the size of the to 200%. The interacting magnetic and charge, spin, and mechanical degrees current reads whether the two ferromag- electronic degrees of freedom create new of freedom to produce new behavior netic layers are set parallel or opposite. mesoscale phenomena and functionality in a heterogeneous structure. Source: S. P. Parkin et and ultimately new technology that Giant magnetoresistance embodies al., Nature Materials 3, 862-867 (2004). performs faster, is cheaper, and boasts several hallmarks of mesoscale science. greater efficiency and durability. A It exploits the coupling between charge the era of high-density information storage host of applications to polymer-based and spin, two distinct degrees of freedom, that is now universally employed in digital nanocomposites, energy storage and to create a new phenomena, switching electronics. The road from discovery to electric currents with a magnetic field. conversion, environmental monitoring market was remarkably fast, only nine Its heterogeneous structure combines years before its introduction to commer- and remediation, biological diagnosis different functionalites in different compo- cial devices. In 2007, Peter Grunburg and and therapy, quantum computing, and nents, magnetic layers for setting and Albert Fert were awarded the Nobel Prize electronic and optoelectronic devices filtering the electronic spin, and a non- for Physics for discovery of giant magneto- can be explored. The opportunities magnetic layer for transmitting the electric resistance. Giant magnetoresistance illus- for innovation in mesoscale discovery current. It is also explicitly dynamic, trans- trates many of the benefits of mesoscale porting and processing charge and spin science, identification of new orga- science: discovery of new phenomena, across the composite system. creation of new technology, and manu- nizing principles, and creation of facturing at the mesoscale using minimal new technology are just beginning to Giant magnetoresistance spawned material and atomic level interconnects emerge. succeeding generations of concepts while producing high efficiency, low cost, and devices, including spin valves and long life, and high economic and societal magnetic tunnel junctions, that enabled impact.

35 From Quanta to the Continuum: Opportunities for Mesoscale Science

Harnessing inspired cargo delivery platforms, and self-repairing (through decreased and environmentally responsive, self- metastability), but at the same time Fluctuations, assembly/disassembly processes) all also responsive to their local envi- Dynamics, and require successful control of length ronments and amenable to external Degradation for and time scale evolution. In drug control (through increased meta- delivery platforms, for example, the stability). The detailed manipulation Control of Metastable targeted delivery time and location of this fluctuation-enabled control Mesoscale Systems must be controlled by adapting inher- would, for example, allow us to engi- The inherent metastability of complex ently metastable features of the self- neer the release time of cargo delivery behaviors in mesoscale systems, assembled drug vectors to induce platforms, control the assembly- including biological and human- them to disassemble and release their disassembly kinetics of self-assem- made systems, is revealed not only drugs on an external cue from outside bled macromolecular aggregates, and on multiple length scales but also on the body. Controlling this kind of modify the lifecycles and extend the multiple time scales. This metasta- metastability and degradation occu- overall lifetimes of photovoltaic tech- bility engenders near-equilibrium fluc- pies the center stage of many areas nology to improve its sustainability tuations and longer time dynamics. of mesoscale science. and reduce costs. Ultimately, these conditions can be exploited to introduce real-time Harnessing the metastability of Meso Challenges responses to environmental cues, mesoscale systems holds the Metastable mesoscale systems mitigate materials degradation, and promise of developing systems and exhibit fluctuations, correlations, and extend useful life. processes that are self-organizing dynamics that lead to interesting

Opportunity At the mesoscale, control of complex systems whose behavior reflects properties across a wide range of length and time scales requires knowledge not only of the averages of pertinent observables as one usually assumes, but also statistical variances and time-dependent fluctuations around the time averages (see figure at right). Only with this knowledge can we reach the goal of controlling short-time and mesoscale dynamics with externally driven energy inputs or externally imposed structural modi- fications, with the ultimate aim of countering completely the degradation and failure that occurs over long time scales.

Human-made complex mesoscale inorganic- or bio-systems (such as Performance of various metastable mesoscale systems. These systems exhibit fluctuations, those employed in long-lived photo- dynamics, and degradation processes that span many length and time scales. Source: Roger voltaic technologies, biologically French, Case Western Reserve University.

36 From Quanta to the Continuum: Opportunities for Mesoscale Science

physical phenomena whose proper Lifetime and Degradation of a stress-and-response framework understanding requires fundamental Energy Technologies that extends from initial formation to physical insights on a wide range of In photovoltaic power applica- complete functional degradation of length and time scales. For example, tions, modules and their associated the materials (see figure below). electromagnetic field fluctuations in processes must have long lifetimes, dielectric media produce universal 25 years or more, in order to make Bio-inspired Rational Design van der Waals–London dispersion solar power a commercially viable of Cargo-Delivery Platforms interactions, electronic correlations energy technology. Current qualifica- Based on Plant Viruses in materials give rise to new elec- tion standards are based on safety A central goal of nanotechnology is the tron and other quasi-particle states and initial performance; they do not fabrication of hybrid or cargo-loaded leading to complex optical properties, serve as lifetime qualification tests, systems that programmably self- and atomistic correlations and fluc- nor has the necessary lifetime and assemble at the atomic level. Biology tuations are essential to nucleation, degradation science been developed has achieved this capability through phase transitions, and growth in inho- to provide quantitative and mecha- evolution. We are now developing bio- mogeneous media. nistic understanding. Advancing inspired delivery platforms and hybrid this reliability physics, or physics of systems that use plant-derived viral The temporal behavior of mesoscale failure, requires new metrology and nanoparticles engineered to encap- systems bridges a vast range of life- metrics at the mesoscale for identi- sulate cargos that are mineralized, times from initial formation dynamics fying and quantifying the degrada- metallized, or chemically engineered to time-dependent behavior of the tion mechanisms and rates by using with ligands that enable site-specific pristine system, to individual particle degradation and defect generation, and all the way to aggregation of defects into more complex, larger units leading to eventual degradation and final disintegration of the system. The relevant evolution embraces length and time scales associated with formation and growth as well as degradation and disintegration. Current understanding of fluctuations and dynamics in driven systems, either close to or far away from equilibrium, is lacking in detail and depth due to the high degree of complexity (>>107 atoms) and the long span of length (1012) and time (>1024) scales. However, recently developed tools and insights from the computational, biological, physical, chemical, and materials sciences could pave the way for breakthroughs in harnessing the dynamics of these systems for Power loss as a function of time for a photovoltaic (PV) module or power plant. Mitigating key degradation modes and lifetime penalties can increase a PV module’s lifetime performance. time-responsive functionalities and The key degradation modes and rates are essential components of a practical reliability for longer life without the need for physics model and must account for the degradation of a module in real-world environmental unnecessary oversimplifications. conditions. L. S. Bruckman, R. H. French, unpublished.

37 From Quanta to the Continuum: Opportunities for Mesoscale Science

delivery and deposition (see figure at thermodynamic modeling and chemistry, materials science, theoret- bottom of page). Viral nanoparticles measurements of viral nanoparticle ical biophysics, and practical bioengi- are dynamic structures that can be stability and assembly energetics neering would introduce a continuous disassembled and reassembled into related to cargo loading and release. feedback loop extending over multiple various configurations, depending These different approaches range space and time scales, where the on control signals from the environ- over many orders of magnitude in optimal parameters determined from ment. The challenge is to precisely time, starting at the high frequency the modeling framework would be control self-assembly and disas- of optical properties of materials out provided directly to the experimental sembly of components that encapsu- to the time scales that define equi- setup/pharmaceutical formulation, late cargos and protect them during librium conditions. The new para- and would direct modification of the transport to the deposition site. Such digm for a rational design framework modeling scheme (see figure immedi- a system could be used to assemble based on a synergy of quantum ately below). a semiconductor device. This can be accomplished only if the self-assembled viral nanoparticle states are sufficiently metastable that soft control of their environment precipitates the disassembled state that delivers their cargo to the targeted location at the targeted time.

Environmentally Responsive Long-range Interaction Driven Self-assembly/Disassembly A rational design framework to control fluctuations on different time scales would enhance the design of nanotechnology processes. For example, for viral nanoparticles the rational design framework would be based on optimized assembly/disassembly parameters obtained from atomic- structure-based quantum chemical calculations, concurrent measure- Mesoscale science feedback loop of measurement, learning, modeling, engineering, and application that informs the design, production, and lifecycle performance of viral nano- ments of mesoscale properties of particles loaded with cargos. Source: Rudolph Podgornik, University of Ljubljana, Slovenia. its components, and biophysical

A mesoscale case study: lifecycle of a viral nanoparticle, from formation of amino acids and coated proteins to assembly of viral nanoparticle, then its transport and delivery and disassembly, followed by deposition of the cargo. Source: Nicole F. Steinmetz, Case Western Reserve University.

38 From Quanta to the Continuum: Opportunities for Mesoscale Science

Research Directions At equilibrium, complex mesoscale engineer specific dynamical response Metastable mesoscale systems systems are controlled by short- in various environmental conditions, encompass multiple space and and long-range interactions coupled as in targeted cargo delivery or in time scales, including fluctuations, to quantum mechanical and clas- controlled self-assembly. dynamics, degradation, and evolu- sical thermal fluctuations. In equi- tion in complex energy landscapes in librium with the environment, such A second class of metastable systems response to varying external inputs. systems can exhibit pronounced and and/or materials are those that are The research needed to control these controllable structural and functional not in equilibrium, such as kineti- complex interacting/evolving systems changes in response to changes in cally frozen or quenched systems (for that span wide length and time environmental conditions (see figure example, atomic or molecular glasses) scales can be subdivided into distinct below). The proteins and viruses that are forced to reside in states that regimes: equilibrium thermodynamics, in complex biological systems are are often very far from equilibrium. close to equilibrium dynamics, far examples of this kind of equilib- Adding energy to the system from the from equilibrium kinetics, and actively rium response. This characteristic environment can trigger an “annealing driven dynamics. can be further harnessed in order to step” toward equilibrium, providing a mechanism to steer the system away from metastability toward a new quenched state with different structure, durability, or functionality.

A third metastable class is driven systems, such as biomolecular motors involved in viral DNA pack- aging and release, in which the addition of an energy source to the system (for example, from adenosine triphos- phate) drives it away from equilibrium and toward metastability, as seen in mutant variants of partially filled or overfilled bacteriophages.

Across all of these heterogeneous equilibrium and nonequilibrium systems, fluctuations, dynamics, and degradation play a critical role. Stabilization/destabilization of mesoscopic systems via thermal fluctuations of the van der Exploiting fluctuations in various order Waals–London dispersion (vdW-Ld) interactions. The thermal fluctuations give rise to the parameters coupled via quantum and Matsubara frequency terms, which are the classical contribution to the vdW-Ld interactions, classical degrees of freedom creates while the zero frequency term is the quantum mechanical contribution. Source: V. Adrian opportunities for universal organizing Parsegian, University of Massachusetts, Amherst. pathways, avenues for environmental

39 From Quanta to the Continuum: Opportunities for Mesoscale Science

responsiveness, and scenarios for discovering improved lifetime and sustainability. As these metastable systems evolve in time, they exhibit dynamics over short time scales that can be used to perform designed

tasks, and over longer time scales that control the life of the system. The metastable structure can be intentionally changed by response to environmental cues, leading to evolution through many time and length scales (see figure at right). These kinds of designed structural changes can lead to several salutary outcomes: 1. Reversible assembly/disassembly of soft matter and bio-systems whose structure is formed as a consequence of long-range interactions (as compared to the

irreversible formation of covalent bonds), and that can be respon- Biological systems display metastable mesoscale behavior, including self-organization across scales. Upper image: Length scale for proteins self-organizing into living materials. sive to changes in environmental Lower image: Self-organization of treadmilling polymers that utilize chemical energy to do conditions and variables or moved mechanical work. Source: Margaret Gardel, University of Chicago. to new environmental conditions, as in cargo delivery systems. and resulting work is performed self-assembles and later disassem- 2. Systems near or far away from as an outcome of energy degra- bles or disintegrates with program- equilibrium that evolve driven by dation. Once the energy source mability, a property biology has an instantaneous energy shock in is removed, they revert to their naturally achieved through evolu- the direction of a preset form and quiescent equilibrium structure. tionary processes. This requires structure, far removed from that simultaneous control of the proper- dictated by equilibrium thermo- Potential Impact ties of the system on multiple length dynamics alone. Examples of this and time scales. The mesoscale The potential impact of harnessing condition are quenched disorder component is crucial here because fluctuations, dynamics, and degra- of many types in kinetically frozen our control of these systems is neither dation in order to control metastable systems and their evolution over wholly microscopic, though we are mesoscale systems is significant, short or very long time scales. dealing with nanostructures, nor is both for bio-inspired and inorganic it wholly macroscopic, because we 3. Mesoscale metastable systems systems (see sidebar “Opportunities need not steer a macroscopic number which are explicitly driven by a to Enlist Bacteria as an Energy of cargo-carrying virus-like nanopar- continuous energy input, such as Workforce” on page 42). biological motors and contractile ticles. Frequently it suffices to control tissues with an abundant energy In the bio-inspired rational design properly only a small number of effec- source (see figure above). These of cargo delivery platforms based tively targeted molecular vehicles. By systems are far from equilibrium on plant viruses, the goal is to fabri- harnessing inherent fluctuations and while energy input continues, cate a delivery platform that first metastability of the self-assembled

40 From Quanta to the Continuum: Opportunities for Mesoscale Science

molecular carriers, researchers will degradation in performance to a For energy systems and medical be able to drive them by design down level without utility. The lifetime and devices and technologies, lifetime well-defined dynamical paths that degradation of complex mesoscale requirements of greater than 20 or eventually lead to the place and the systems, such as photovoltaic cells, 40 years correspond to dynamical time where they are needed. are intrinsically metastable and domi- processes spanning the range from nated by large numbers of defects, 10-18 or 10-15 seconds to 109 seconds. Mesoscale system performance interactions, and competing degrees This suggests that a fundamental over a lifetime is a significant chal- of freedom. For example, the degra- challenge is studying the time evolu- lenge. Typically, the focus has been dation of a photovoltaic module over tion of metastable systems over 27 only on initial performance and the its lifetime presents a significant orders of magnitude. Typically, we initiation of degradation, a far easier challenge to forecasting the cost of have been unable to span 6 or 9 task. Lifetime correlates closely with solar technologies and their level- orders of magnitude in time without the nature of change from a robust ized cost of electricity and net present being overwhelmed by the limita- initial structure to a degraded condi- value. Understanding the degrada- tions of our assumptions, approximation through damage accumulation. tion modes, mechanisms, and rates tions, and methodologies. Advances Mesoscale properties and behavior of decay of photovoltaic modules in in mesoscale theory and computa- deviate from those set by initial a vast array of real-world conditions tion have the potential to significantly values due to real-world interactions would give insights into ensuring high expand this range with corresponding and exposures, leading to either performance of future generations of impact. catastrophic failure or continuous solar technologies.

41 From Quanta to the Continuum: Opportunities for Mesoscale Science

Enlisting Bacteria as an Energy Workforce Cell signaling enables large colonies of cells to coordinate their functions and work in concert, presenting mesoscale opportunities to disarm pathogens and reprogram function to produce energy.

Biological systems involve vast numbers of living cells whose interconnected networks define function. Protein synthesis, transloca- tion, and modification, for example, present a complex network at the nanoscale that is guided by a diversity of cues. The interactions between cells that determine phenotype are also guided by signaling events, wherein cells respond to cues of various forms— from structurally distinguishing small molecules and charge-specific ions produced by nearby cells to mechanical forces imposed by neighboring tissues. Then, populations of cells, including consortia of bacteria or tissue composed of differentiated cell types, guide complex physiological systems such as the gastrointestinal tract. Identifying and re-programming the machinery that enables bacterial colonies to communicate present a major opportunity for them to, for example, generate energy. These opportunities include: • Cell signaling. Unraveling the language Cell Signaling. (a) Small signaling molecules are made and sensed by cells, enabling coordi- of cell signaling is both a scientific grand nated behavior among bacterial colonies and stimuli/responsive behavior between different challenge and a major mesoscale fron- cell and tissue types. (b) Signaling molecules are manufactured by metabolic pathways tier—involving large numbers of cells, within cells, e.g., the autoinducer-2 family involved in bacterial attack. (c) In vitro microfluidic multiple cell types, length scales from platforms present one approach to diagnosing and modulating these cell signaling processes. Sources: K.K. Carter et al., Metabolic Eng. 14 (2012) 281-288 (2012); X. L. Luo et al., Lab on a small molecules on the nanoscale to Chip 8, 420-430 (2008); Y. Cheng et al., Adv. Func. Mater. 22, 519-528 (2012). larger cells and tissues on the microscale, and time frames ranging from small molecule binding, enzyme reactions and genic bacteria is what enables their • Bacteria as an energy workforce. A diffusion for signaling entities to evolu- mischief, but research to understand, parallel opportunity exists at the meso- tion of cell groups and tissues among detect, and re-engineer their communi- scale in understanding and directing species. The implications for society are cation and behavior is in its infancy. ubiquitous bacterial colonies to generate profound if research can unravel nature’s energy, either as biofuels (e.g., from • Re-programming bacterial behavior. complex network of nano- and meso- cellulose) or as electricity (e.g., from Identifying and re-programming the scale cell-signaling processes. microbial fuel cells). The efficiency of machinery that enables bacterial colo- these sources depends on how well • Bacteria in human health. Bacterial nies to communicate presents a major we can understand and re-engineer cells substantially outnumber human opportunity with implications from bacterial behavior from biomolecular cells in our bodies, particularly in the biomedicine to energy. In medicine, processes to signaling between cells and digestive tract, and they are essential traditional antibiotics to control patho- longer range interactions with various to health. The importance of bacteria genic bacteria work by killing them and tissue types. Microbial fuel cells also and their contribution to our meta- all nearby healthy bacteria, forcing the require efficient interfacing of cell popu- bolic well-being is becoming increas- remaining cells to mutate, evolve, and lations to electrical wiring that maintains ingly apparent. In the last decade, the do more damage. Accordingly physi- connections as well as cell health. Thus, “microbiome” has received significant cians, concerned about their increasing enlisting bacteria for energy produc- attention, particularly as a target of next- inefficacy, are more reluctant to admin- tion will likely require that we modify the generation sequencing technologies ister antibiotics. New mechanisms to biochemical machinery at the subcellular racing to identify commensal species redirect/re-engineer bacterial activity level, assemble robust architectures for as well as the emergence of patho- offer promise, e.g., by interfering with evolving bacterial colonies, and control gens. Dangerous bacterial attacks have the networks by which signaling mole- the signaling cues which determine this become commonplace and routinely cules are made, relayed, and trans- evolution—challenges centered at the newsworthy—tainted spinach, methi- duced, and thereby altering the collec- mesoscale. cillin resistant Staphylococcus aureus tive behavior of extended bacterial and anticipated “superbugs”, infections populations. This is an inherently meso- initiated by catheters in hospitals, and scale challenge. others. Signaling between these patho-

42 From Quanta to the Continuum: Opportunities for Mesoscale Science

Directing Assembly Meso Challenges in more extensive, creative fashions to achieve the goals of mesoscale mate- Synthesis and Self-assembly of of Hierarchical rials science. Beyond this, molecular Programmable Building Blocks Functional Materials architectures will become more intri- The integration of disparate materials Molecular-Based Building Blocks. cate and complex. Through the inter- classes by “top-down” and “bottom- The rational design and synthesis of mediary of block copolymers, with up” approaches, now quite distinct structure-directing, molecular building increasing numbers, chemical types, and representing different operating blocks (surfactants, foldamers, and various architectural connectivi- principles and spatial scales, is an amphiphilic peptides, charged elec- ties of blocks, the science of synthetic underpinning focus of directed meso- trostatic complex formers, multi- macromolecules will progress toward scale assembly. block polymers, etc.) are required to the realm of protein folding and self- guide the hierarchical self-assembly assembly. At the same time, through of desired architectures over large Opportunity the tools of synthetic biology, geneti- distances (see figure below). Toward cally programmed synthesis of amino A grand challenge in materials this objective, a fundamental under- acid polymers, comprising natural and science is to create multiscale func- standing and ability to precisely unnatural amino acids, polypeptides tional structures that encode infor- manipulate the intra- and inter-molec- and their analogs (e.g., polypeptoids) mation; adaptively respond to their ular forces that enable and guide may increasingly develop new mate- environment; and capture, trans- self-assembly are needed. The full rials based on their mesoscale struc- port, and utilize energy. The ability to range of non-covalent forces (van der ture. Exploiting the self-assembly design and realize the complex and Waals, π-stacking, metal-ion coor- possibilities with polymers containing composite materials that we envisage dination, hydrogen-bonding, elec- charged blocks that can complex with will require significant advances in trostatic, hydro- and solvo-phobic, oppositely charged segments makes how we synthesize materials, how steric, and packing) will be deployed the analogy to proteins even stronger. we process and assemble them, and how we control their composition and architecture. The parameter space available to us is simply enormous, thereby necessitating a concerted experimental and theoretical approach, particularly considering that both equilibrium and nonequilibrium states of matter are of interest. The emphasis in mesoscale synthesis and assembly is on integration of disparate materials classes by “top- down” and “bottom-up” approaches. Bridging the gap and allowing these two powerful approaches to influence, complement, and reinforce each other are a major challenge and

opportunity of mesoscale science. Modeling challenges in the synthesis and self-assembly of molecular-based building blocks. Source: Greg Exarhos, Pacific Northwest National Laboratory.

43 From Quanta to the Continuum: Opportunities for Mesoscale Science

Particle-Based Building Blocks. aspects due to building block shape some of the character of atoms and Toward this objective, a fundamental and rigidity. The size and shape of molecules that arise from the symme- understanding is needed of the often the self-assembled clusters are highly tries of atomic and molecular elec- subtle and competing consequences sensitive to both interparticle interac- tronic orbitals. Translation of these of anisotropic repulsive and attractive tions (e.g., charge, hydrophobicity) nanoscale particle symmetries into forces (typically strong and spatially and “Janus balance.” Intentional vari- materials structures at the mesoscale short ranged) and coupled transla- ation of this latter parameter, defined is a current experimental challenge. tional-orientational Brownian dynamics by α, can be achieved by scalable over a wide range of length and time synthetic methods. Dynamic Synthesis via Biomimetic scales, problems that have no coun- Architectures terpart in traditional materials science To date, only a limited range of While the synthesis of functional based on atoms or homogeneous programmable building blocks has materials as static or even respon- spherical particles. As a result, supra- been realized, yet by tailoring both sive entities remains a challenging colloidal assemblies in both isotropic particle shape and chemical hetero- frontier area of research, a higher and anisotropic configurations can be geneity, programmable surface level of sophistication—the dynamic produced that possess widely varying forces acting in concert with the synthesis of functional materials— connectivity, porosity, and properties nonadditive consequences of steric remains virtually unexplored. Dynamic (see figure below at left). packing constraints will enable new synthesis would enable on-demand assemblies and growth, re-configurability, and resto- dynamical behavior ration, characteristics that are mostly to be realized. unachievable in today’s materials or Computationally, rudimentary in concept. To meet the more insight into requirements of mass and energy structural possi- transport for dynamic materials bilities has been synthesis, as well as the regulatory gained from simula- capabilities needed for control, new tion of dense particle mesoscale materials and associ- systems, where ated science and technology are shape or “patchi- needed in the area of artificial circu- Advanced synthesis routes for self-assembly with ness” (beyond Janus) latory systems. This understanding particle-based building blocks. Source: T. Yong Hang, Lawrence Livermore National Laboratory. confers on particles will enable the design of vascularized

The power of this concept is shown in the illustrative example (see figure at right) in which amphiphilic Janus microspheres self-assemble into a variety of motifs, including stable Janus clusters, whose composition and shape can be uniquely specified (as opposed to polydisperse collections) as well as long chiral semiflexible chains. The chemical heterogeneity that drives cluster formation is akin to the polar/ nonpolar balance in molecular and polymeric surfactants, but with qualitatively new geometric Assembly pathway between Janus microsphere clusters and chain-like objects (top). Graphical representation of “Janus balance,” as defined by α (bottom left). Epifluorescence images of tetrahedral Janus clusters (middle) and chains (right). Source: Q. Chen et al., Science 331, 199 (2011). 44 From Quanta to the Continuum: Opportunities for Mesoscale Science

solids and functional fluids flowing Directed Assembly of Hierarchical functional structures by fluid printing through the channels. The challenges Functional Architectures strategies. In the solid state, methods of dynamic materials synthesis will Directed assembly is a powerful that control fracture at soft material require, for example, fluids capable approach for integrating top-down interfaces enable a different form of of delivering reagents, catalysts, and and bottom-up assembly strate- printing, with complementary capabil- synthetic precursors with spatial and gies. The ability to deterministically ities for producing hierarchical assem- temporal gradients in concentration assemble functional objects into blies of materials and devices in a appropriate to achieve the desired planar and three-dimensional config- highly deterministic mode (see figure synthetic target. In one embodiment, urations would enable significant below at right). These fundamental we envision the guided delivery of advances in the design and perfor- suspended carrier particles in flowing mance of materials for light capture vascular streams. The size, interfacial and utilization, battery electrodes, and chemistry, and mechanical character- lightweight structures. Achieving this istics of the particles will govern their mastery requires a deep integration partitioning between channel wall of expertise in the design, synthesis, and fluid stream, thereby enabling assembly, and modeling of a broad complex transport mechanisms, such class of materials in myriad shapes, as rolling, as a possible means to facil- including nanoparticles, microwires, itate surface-guided particle delivery monocrystalline semiconductor plate- (see figure below). Interfacial interac- lets, and so on. tions between particles will also allow particles to communicate, thereby Massively parallel assembly of micro/ controlling the stoichiometry and nanostructured materials represents concentration of synthetic precursors a capability that is critically enabling delivered to the active zone. Finally, for a wide array of energy-relevant we imagine the need for particles to devices. Starting from the fluid state, be equipped with triggering mecha- particles and nanowires suspended nisms for stimuli-initiated deployment as individual elements or assembled of their payloads. clusters can be used as inks to form

AAnn ep epitaxialitaxial multilayermultilaye approachr approach to tbulko bu quantilk - ties of semiconductor building blocks. (a) Sche- matic illustration of a multilayer of GaAs/AlAs Surface-guided particle delivery. Source: J. Moore et al., University of Illinois at before and after release by etching of AlAs. (b) Urbana-Champaign. Corresponding profile. (c) Cross-sectional image after partial etching of AlAs. (d) A large collection of GaAs plates derived from complete etching. Source: J. Yoon et al., Nature 465, 329-333 (2010).

45 From Quanta to the Continuum: Opportunities for Mesoscale Science

advances lead to compelling options piezoelectric platelets. The power of form of single-phase and composite in engineering due to their ability to this approach lies in the sophistica- nanostructures. capture certain characteristics that tion in both device architecture and are prominent in biological systems, materials heterogeneity realized when An exciting opportunity for mesoscale including (1) the combined use of top-down and bottom-up approaches science is to advance these mate- hard and soft components in hybrid are truly integrated. rials beyond the realm of electric-field structures for unusual mechanical control of magnetism to include a and geometrical forms (e.g., flexible Functional Complex Oxide Hetero- new property—temperature (or heat). “skin-like” architectures); (2) heterostructures. Complex oxides, such as The coupling between electric field, geneously integrated collections of multiferroic and magnetoelectric mate- ferroelectricity, and magnetism in dissimilar materials in wide ranging rials, that simultaneously possess two these materials generates a collec- sizes and shapes for increased levels or more order parameters—ferroelec- tive mesoscale effect where multiple of functionality; and (3) dynamic tricity, ferromagnetism, or ferroelas- degrees of freedom work together tunability in structure and confor- ticity—have garnered considerable to enhance response by generating mation for responsive designs. The attention. Magnetoelectricity refers to the maximum field-induced entropy specific systems of interest are impor- a linear magnetoelectric effect or the changes possible via so-called tant, because they offer a combina- induction of magnetization by an elec- magneto-electro-caloric and pyro- tion of unique features and levels of tric field or polarization by a magnetic electric-magnetic effects (see figure performance that would be difficult or field. Researchers have been driven below). These systems could poten- impossible to achieve by other routes. by the promise of coupling magnetic tially be used for waste heat energy and electronic order parameters and conversion (complementary to ther- Flexible “Skin-Like” Architectures the potential to manipulate one with moelectrics); low-power, refrigerant- for Light Capture and Utilization. the other. Such materials come in the free solid-state cooling; advanced It is expensive to grow large, high- quality wafers of III-V semiconductors and intimately integrate them on silicon or amorphous substrates such as glass or plastic, thereby restricting their use. Solid “inks” composed of thick, multilayer epitaxial assemblies can be created in a single deposition sequence on a growth wafer, then distributed over large areas by transfer printing, in a step-and-repeat mode. This innovative approach has been used for creating flexible “skin- like” architectures for high-speed electronics, efficient photovoltaic modules, and inorganic light-emitting diode arrays for solid-state lighting. This approach has been adopted by Semprius to make/manufacture concentrator photovoltaic cells (Top) Schematic of various multicomponent multiferroics: single layer, single with world-record efficiency. One phase multiferroics, and two different composite multiferroics (layered and can readily imagine extending these nanostructured). (a) Overlap of functional materials giving rise to multiferroics. (b) Diagram showing the cross-coupling of electric field (E), magnetic field (H), concepts to other energy-harvesting and stress (σ). We now add temperature as a new point on the tetrahedron. devices, such as those composed of Source: L.W. Martin, University of Illinois.

46 From Quanta to the Continuum: Opportunities for Mesoscale Science

electron emitters; and thermal to multi-variable and multi-scale with nickel metal followed by template imaging. problems in order to decouple various removal. structural and functional properties. To meet this challenging objective, The fabrication challenge is to build New methods for scalable directed a multi-faceted approach is required these architectures, which often are assembly are needed to reliably that combines directed assembly of multi-dimensional, multi-scale, and produce three-dimensional multiscale heterogeneous materials; advanced multi-material. Toward this grand architectures that are composed of multi-modal experiments that bring challenge, advanced microstructural multiple materials for both struc- together electric field, magnetic field, design, using flexure and screw theory tural (e.g., strength, stiffness, thermal and temperature (as well as the corre- as well as topology optimization, expansion, and Poisson ratio) and sponding material susceptibilities); could be combined with new directed functional (e.g., electronic, optical, and multi-scale models from atom- assembly techniques to create novel magnetic, and energy) applications. istic first-principle studies of new material systems and architectures materials and effects to large-scale, with previously unachievable prop- Another recent example is the meso- device-level models of efficiency and erty combinations. Much of this structuring of electrodes for electro- loss. enhanced performance can be real- chemical energy storage (batteries), ized by tailoring material architecture which offers potential to concurrently Three-Dimensional Hierarchical at the nano-to-millimeter scale. One optimize electrical and ionic connec- Architectures. Recent efforts have powerful example of this approach is tivity, and to maximize the active- focused considerable attention on the the ultralight metallic lattice (see figure material volume fraction. Minimization design and directed assembly of below), which is created by coating a of strain-induced damage—in partic- three-dimensional hierarchical archi- three-dimensional polymer template ular, in emerging high-energy density tectures composed of diverse classes of functional materials. There are many potential applications—lightweight structural materials, battery electrodes, metamaterials, and well beyond—that would benefit from the unparalleled properties and performance realizable through advances in mesoscale engineering.

Material properties are governed by the chemical composition and spatial arrangement of constituent elements at multiple length scales. This condition fundamentally limits material properties with respect to each other, creating trade-offs when selecting materials for a specific application. Decoupling these property relationships could result in previ- Design, assembly, and cellular architecture of ultralight metallic lattices. (A) 3D ously unobtainable performance. For polymer template created by ultraviolet patterning, (B) a conformal nickel-based example, new lightweight structural thin film formed by electroless plating followed by etch removal of the template, materials are needed that exhibit (C) image of the lightest microlattice produced by this approach (0.9 mg/cm3), low density, yet high stiffness. The and (D and E) images of two as-fabricated microlattices along with a breakdown of relevant architectural features. Source: T. A. Schaedler et al., Science 334, 962 design challenge is to find solutions (2011).

47 From Quanta to the Continuum: Opportunities for Mesoscale Science

materials—is a unique opportunity for battery electrodes promise to lead to The central premise of bottom-up mesostructured systems. The chal- the discovery of new phenomena and multi-scale modeling approaches is to lenge is to identify the architectures functionalities from both current and use electronic structure methods and and materials, which might be quite emerging energy storage materials state-of-the-art molecular dynamics different from those in use today, with high performance and low cost and Monte Carlo sampling techniques that take full advantage of meso- that will transform the energy storage to calculate the fundamental inter- scale phenomena and functionalities, landscape. actions that arise between distinct and to develop scalable methods to components of a material. These self- or deterministically manufacture Templated self-assembly of block interactions are subsequently fed into these systems. copolymers is a demonstrated means mesoscale-level simulations, which of spanning the nanoscale (domain seek to predict the emergent interac- Electrode mesostructuring can come structure in ordered block copolymers) tions or contributions to the free energy in many forms, including colloidal to mesoscale (functional patterns for that arise in pure and composite self-assembly, direct ink writing, and applications ranging from integrated systems as a result of enthalpic holographic methods. Understanding circuit manufacture to biomaterials and entropic forces. A particularly building block interactions, coupled development). The extension of this important challenge for mesoscale with synthesis of building blocks with directed self-assembly capability modeling efforts is to describe the unique structures and chemistries, will into three dimensions is just being structure and dynamics of complex enable realization of advanced elec- explored. materials over multiple length and trode designs. Sacrificial templates time scales, thereby enabling predic- that are later removed enable struc- Research Directions tion of hitherto unknown nonequilib- turing of broad classes of materials. All elements of directed assembly rium states of matter. As alluded to As one illustrative example of this described thus far, from the building throughout the previous sections, approach, colloidal crystal templating blocks to the strategies of process the assembly processes envisaged in has been exploited to create a new assembly, rely on guidance provided emerging mesoscale-based technol- class of bicontinuous cathodes by multiscale and mesoscale models ogies are inherently nonequilibrium, composed of three interpenetrating to design new materials, to devise requiring the use of flow, voltage, networks that provide highly effi- new processing strategies, and to or surface fields to attain desirable cient electron and ion transport path- assist in the interpretation of experi- metastable states of matter and trap ways due to their high surface area mental data. the proposed materials in such states. and short diffusion paths (see figure below). Future efforts on mesoscale In the context of particle assembly, new models are now capable of describing structure at a molecular level, while treating fluctuating hydro- dynamic interactions in multipar- ticle systems and in the presence of surfaces and interfaces. Such models permit prediction of the assembly that occurs as a function of particle

Illustration of bicontinuous cathode composed of 3D interpenetrating networks of

nickel, LiMnO2, and porosity engineered to exhibit high power and energy density. Source: H. Zhang et al., Nature Nanotech. 6, 277 (2011).

48 From Quanta to the Continuum: Opportunities for Mesoscale Science

structure and applied fields and are Potential Impact makeup, shape, and/or softness. essential, for example, in the context The overarching goal of directed Importantly, achievement of this goal of dynamic synthesis with biomi- assembly is to establish the scien- requires a deep integration of exper- metic architectures. In the context of tific basis and assembly pathways tise in assembly, characterization, and nanostructured polymeric materials, required to enable the formation computational/theoretical modeling recent approaches have been shown of “designer” materials and archi- of building blocks, clusters, and to be capable of describing entangle- tectures. Of specific interest is the mixtures thereof. With this knowledge, ments, microphase and macrophase establishment of the fundamental one can create novel forms of matter separation, and confinement or free knowledge required for assembly and in a rational manner via directed interfaces in composite systems in a disassembly of information-encoded assembly and disassembly—whose concurrent manner. Such approaches building blocks with spatial and spatial organization and connectivity enable prediction of the dynamics of temporal control. This effort hinges are dynamically tunable over multiple morphology formation in hierarchical on the ability to model, synthesize, length scales ranging from those over materials, and provide the means to and assemble building blocks whose which intermolecular and/or interpar- design out-of-equilibrium processes exotic motifs embed information from ticle forces act to progressively longer that lead to controlled, specific the nanometer to the microscale scales associated with the individual outcomes. via anisotropies in their chemical building blocks, clusters, and supra- cluster assemblies.

49 From Quanta to the Continuum: Opportunities for Mesoscale Science

Mesoscale Manufacturing Manufacturing entirely at the mesoscale using bottom-up self-assembly of functional components connected directly at the mesoscale without macroscopic wires, mechanical linkages, or fluid channels enables faster, cheaper, more efficient, and longer lasting products.

Manufacturing at the mesoscale is a naturally occur on the mesoscale; manu- dramatically new and compelling outcome facturing and assembling them exclusively of mesoscale science. Its building blocks on the mesoscale without recourse to the will be mesoscale units displaying the macroscale techniques and interconnects hallmarks of mesoscale behavior such as that we now routinely employ is a major collective response, interacting degrees operational and strategic opportunity. of freedom, heterogeneity, and densely spaced energy levels (see Introduction: The benefits of manufacturing on the “The Hallmarks of Mesoscale Behavior” mesoscale are dramatic and legion. on page 3). These building blocks will be Employing bottom-up self-assembly assembled at the mesoscale with nano- or incorporates all the starting materials

mesoscale interconnects. into the final device, with little or none remaining as waste. Connections between Mesoscale arrays can be self-assembled from Mesoscale manufacturing differs funda- mesoscale functional units will be made nanocrystals that have pure, alloyed, core- mentally from the miniaturization tech- at the mesoscale through interfaces shell or cluster-in-cluster internal structures, niques of manufacturing that have been designed and implemented at the atomic and arranged in a variety of one, two and three dimensional lattice configurations. The proper- so successful in advancing digital elec- and molecular level. Because these direct ties of the arrays are tunable by a host of tech- tronics. Moore’s Law reflects the long “atom to atom” interconnects avoid extra- niques (see text) and are fundamentally different success of semiconductor electronics: the neous wires, mechanical linkages, and from those of the individual nanocrystals or bulk number of transistors on a chip doubles flow channels, efficiency will rise and the materials of the same composition. Source: approximately every 18 months. Moore’s risk of failure will fall. The simpler design J. Henzie et al., Nature Materials 11, 131 (2012). Law is robust because inventors have and construction of mesoscale devices followed a top-down process, finding mean longer lifetime and lower cost. sity of programmable behavior in these ways to make macroscopic components arrays is enormous, and the opportunity Arrays of single crystals of order 1-15 smaller, faster, and cheaper by miniaturiza- for applications in biological diagnosis nm in diameter are a promising route to tion. Mesoscale manufacturing starts from and therapy, polymer-based nanocom- mesoscale manufacturing. Nanocrystal a different premise: that the innovations posites, energy storage and conversion, arrays display a wide range of meso- of the future will exploit new phenomena environmental monitoring and reme- scale behavior that is different from both firmly rooted in mesoscale behavior, occu- diation, and electronic and optoeletronic bulk materials of the same composition pying the transition region between nano devices is virtually limitless. The capability and individual isolated nanocrystals. The and macro. One can imagine sequential of synthesis and characterization of nano- behavior of arrays can be tuned by many catalytic reactions for producing alterna- crystal arrays is in its infancy, with far more techniques: the size, shape, composi- tive biofuels, or self-assembly on massive phenomena remaining to be discovered tion, and structure of the nanocrystals; scales of membranes that purify water, than have yet been found. Incorporating the chemical linkers binding them; and or mesoscale structuring of battery elec- these arrays into mesoscale devices their physical proximity. The arrays them- trodes that coordinate the electronic, without using macroscopic techniques selves can be mono- or polydisperse, ionic, and mechanical degrees of freedom and interconnects is a primary challenge and be assembled in one-, two-, or three- needed to convert electricity into chemical for mesoscale science. bonds and vice versa. These processes dimensional configurations. The diver-

50 From Quanta to the Continuum: Opportunities for Mesoscale Science

Tuning Bottom-up Self-Assembled Energy Converters with Mesoscience Mesoscale features dramatically alter the photovoltaic performance of nanocrystal assemblies.

The energy conversion performance of composite mesostructures, such as nanocrystal arrays linked with organic molecules, can be tuned over wide ranges by adjusting their interfacial behavior. The behavior of electrons excited by light or electrical potential in nanocrystal arrays is dictated by the organic linking ligands, and organic molecules are an infinitely diverse toolbox by which we can tune the interfacial chemistry of the individual nanoparticles. In turn, assembling nanocrystal arrays into multilayer “sandwich” structures yields materials with exceptionally high flux of charge and energy across the layer interfaces, and a set of “knobs” to chemically control these fluxes. The result is a collection of high-efficiency mesoscale energy converters. The power of this approach is illustrated with the self- assembly and photovoltaic performance of nanocrystal arrays.

Lead sulfide nanocrystals (~10 nm size) connected with oleate linkers (~2 nm separation) are self-assembled into a nanocrystal array. The size and shape of the nanocrystals and their separation are governed by the organic linkers connecting them. The mesoscale features of these arrays are intermediate between atomic and macroscopic dimensions; they govern the macroscopic photovoltaic performance of the array.

Like the lead sulfide nanocrystal arrays at the left, the photoelectron

Photovoltaic properties of lead sulfide monolayer films. Nanocrystals separated by oleate (length ~2 nm, bottom transfer rate of cadmium selenide nanocrystals arrays depends critically on the length and composition of the organic ligands linking the The photovoltaic properties of lead sulfide nanocrystal arrays depend nanocrystals. Cadmium selenide nanocrystal arrays are prepared with critically on the organic linkers connecting the nanocrystals. Nanocrystals mercapto-alkylcarboxylate linkers whose length varies with the number of separated by oleate (length ~2 nm) have negligible dark (light off) and alkyl units from n=1 to n=15 (top left). Layers of the redox-active polymer photocurrent (light on) densities (bottom left), both less than one pico- poly(viologen) (middle left) are alternated with the nanocrystal arrays to ampere per square centimeter. Nanocrystal arrays in which the oleate form a multilayer “sandwich” (top right). Photoexcitation of the stack ligands have been exchanged for butylamine (length ~0.3 nm) have large transfers electrons from the cadmium selenide nanocrystal array to the dark and photocurrent densities approximately 10 nano-amperes per poly(viologen). The rate of electron transfer, however, depends critically square centimeter (top left). The organic linkers in nanocrystal arrays are on the number of alkyl groups (and thus the length) of the organic ligands a powerful knob to adjust the performance of next generation photocells. linking the nanocrystals. The upper curves in the lower figure show the dramatic change, where the longer linker ligands (CdSe-C15) maintain lower values for the normalized absorption for much longer times than do the short linker ligands (CdSe-C1). This strong dependence of electron transfer rate on nanocrystal array configuration provides a control “knob” for tuning phenomena and creating new functionality in solar cells for producing electricity and for driving photochemical reactions. Source: M. Tagliazucchi et al., ACS Nano 5, 9907-9917 (2011).

51 From Quanta to the Continuum: Opportunities for Mesoscale Science

52 From Quanta to the Continuum: Opportunities for Mesoscale Science

Capability Gaps for Mesoscale Science

Realizing the meso opportunity requires advances not only in our knowledge but also in our ability to observe, characterize, simulate, and ultimately control matter. The areas of knowledge that are poised for breakthroughs have been discussed in the preceding priority research directions. In this section gaps in our capabilities to make, measure, and model materials on the mesoscale are highlighted. Of at least equal importance as the research infrastructure is the human capital required to explore the frontiers of mesoscale science. This section also discusses workforce considerations for mesoscale science.

First and foremost among the capability gaps for realizing the meso opportunity are advances in our ability to integrate theory, modeling, and simulation seam- lessly with synthesis and characterization. The inherent complexity of mesoscale phenomena, often including several nanoscale structural or functional units, requires theory and simulation spanning multiple space and time scales. New organizing principles that describe emergent mesoscale phenomena arising from many coupled and competing degrees of freedom wait to be discovered and exploited at the mesoscale. An integrated and coupled approach in which measurement directly influences theory-guided synthesis will accelerate the discovery of new mesoscale phenomena and architectures.

Measurements that are dynamic, in situ, and multimodal must replace static characterization. The mesoscale is an inherently dynamic regime, where energy and information captured at the nanoscale are processed and transformed to create new outcomes. Given the complexity and dynamism of mesoscale phenomena, multiple measurements of transient phenomena are key. Spatially diverse measurements in three dimensions capturing coupled transformations in different parts of the mesoscale assembly are critical. Complex mesoscale ensembles are not identical and inevitably display a range of behaviors; effective measurements must capture the range of possible outcomes and not just characterize a “representative” sample.

Finally, the ability to design and realize the complex and composite materials we imagine will require qualitative advances in how we synthesize materials, both in moving from serendipitous to directed discovery and in advancing our 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.

53 From Quanta to the Continuum: Opportunities for Mesoscale Science

Synthesis Capabilities dimensions of material layers (e.g., through conventional lithographic thicknesses) must achieve explicit approaches, which are more diffi- Creating the materials, structures, and architectures that access the benefits relationships to optical wavelengths, cult and expensive in the submi- of mesoscale phenomena requires which are themselves defined at cron regime. Soft lithography (e.g., advances in at least three broad cate- the mesoscale. Because the inter- molding and transfer printing) offers gories of synthesis capabilities: play between collective behavior other options. Spatially selective • Directed synthesis to create and defects is critical to mesoscale material deposition, employing either complex materials, controlled phenomena, capability for highly selectivity of a chemical process or interfaces, and patterned controlled synthesis conditions offers focused beam deposition, is a more structures; the possibility to intentionally intro- challenging method. duce defects (e.g., impurities) from • Assembly processes and patterning strategies for fabricating which correlations can be developed Self-assembled Structures. New heterogeneous, spatially defined between defect populations and methods for directed synthesis are structures; and mesoscale phenomena. One must needed to create structures on the also remain open to serendipitous mesoscale. Examples range from • Computational tools that simulate synthesis of mesoscale structures these synthesis and assembly discovery of new phenomena and steps to facilitate formation of the new architectures through explor- such as carbon microtubes (100- desired structures and estimation atory synthesis. Characterization 1000 nm in diameter) to electrocat- of their resulting properties and techniques to identify defects and alytic features formed by galvanic behavior. their properties and behavior, ideally displacement reactions. Surface in situ during synthesis, are of course area considerations make meso- and More generally, significantly enhanced required. nanoscale structures both credible capability and capacity for materials candidates for energy applications, synthesis, fabrication, and processing Interface Control. Control of inter- yet much less attention has been are needed to fully explore known faces accompanies this require- devoted to synthesis of mesoscale and discover unknown mesoscale ment, because surface and interface features. phenomena and architectures in properties can strongly influence pursuit of controlled functionality. the materials synthesized on them. Assembly Processes Interface control plays an impor- Multistep Synthesis of Hetero- tant role from complex oxides to Directed Synthesis geneous Structures. Assembly of Dimensional Control. For new soft materials, influencing properties mesoscale systems often requires phenomena on the mesoscale to such as the quality of epitaxial align- heterogeneous structures, which emerge and be exploited, mate- ment in the former and soft material are formed through a sequence rials synthesis must be highly properties (e.g., polymerization prod- of synthesis and patterning steps. controlled—in composition, micro- ucts and kinetics) in the latter. Various synthesis processes and structure, characteristic length scale, corresponding equipment are already and reproducibility. The challenge of Patterning. Directed synthesis may known and available. They are based understanding collective behavior require or benefit from patterning on physical, chemical, and electro- often arises from multilayer hetero- of material configurations to meso- chemical phenomena in both vapor structures whose attributes depend scale dimensions in order to under- and liquid phase. However, directed on synchronization of material layers stand collective phenomena (e.g., synthesis to reproducibly elicit meso- at mesoscale dimensions or from multiferroic behavior) arising at scale phenomena imposes new the matching of length scales arising the mesoscale. This patterning or requirements, such as in-situ thick- from distinct and competing degrees templating can arise in three dimen- ness measurement and control, or of freedom. Optical and plas- sions, including in the direction of self-limiting process mechanisms monic structures are an example: growth. It may be accomplished

54 From Quanta to the Continuum: Opportunities for Mesoscale Science

(e.g., atomic layer deposition). for mesoscale research is to develop New Design Modalities. A particu- Surface and interface control may efficient synthesis and experimenta- larly exciting notion is the possi- require in-situ processing equipment tion protocols as combinatorial strat- bility that mesoscale designs could that enables successive steps without egies for varying nanostructure array decouple structure and function, exposure to air and other chemical configurations to reveal the parame- based on additive manufacturing of contaminants. ters and mechanisms controlling new micron to millimeter features. While mesoscale phenomena. use of materials configurations to Nanostructure Assemblies. Impor- accomplish particular behavior is tant mesoscale phenomena also Computational constrained by structure-function arise from nanostructure assemblies Synthesis/Assembly coupling, assembling different struc- covering mesoscale dimensions. tures from mesoscale components Materials and Process Modeling. As an example, nanowire battery offers a pathway to decoupling. The Experimental synthesis is expensive structures packed close together structures of micro-electro-mechan- in time and cost; major advances are with high aspect ratio are promising ical systems (MEMS) offer examples needed in modeling and simulation of pathways to future storage tech- of the effective separation of structure synthetic approaches to guide experi- nology. However, their spatial prox- and function, albeit targeted more at ment. A rich portfolio of approaches imity, driven by high aggregation at applications in which new phenomena is currently available to predict the mesoscale, raises important new do not emerge at the mesoscale. outcomes of processes, sequences, questions of transport limitations and and consequences for material and coupled electrochemical phenomena structure properties. It includes infor- Availability of Synthesis that would limit performance. Thus, mation on surface chemistry during Infrastructure tools are needed not simply to make deposition/etching, 3-D profile evolu- Distributed Synthesis Facilities. the nanostructures, but to synthesize tion, materials properties derived from Experimental facilities needed for high-aspect-ratio nanostructures in ab initio density functional theory, synthesis span a broad spectrum well-controlled high-density arrays and continuum mechanics and of equipment types and associated extending to the mesoscale. These multiphysics behavior of structures, infrastructure (chemicals, materials, can then be patterned to form func- all enhanced by high-performance etc.). Multistep fabrication of multi- tional connections (e.g., contacts) to computing infrastructure. component mesoscale structures the arrays. requires even more individualized Mesoscale Context. Unique facilities. This situation places value Combinatorial Strategies for phenomena at the mesoscale present on synthesis resources widely distrib- Mesoscale Phenomena. Synthesis new opportunities for computational uted and widely available to serve of functional nanostructure arrays and theoretical research. Collective the specific needs of meso research begins with self-assembly of aperi- behavior underscores the need to groups, rather than a reliance on a few odic scaffolds (e.g., porous gels) develop and implement into modeling major national facilities. To be sure, or ordered templates (e.g., anodic frameworks and codes a coupling some synthesis capabilities at national alumina and colloidal particle arrays), between statistical mechanics of facilities, following the successful followed by wet or dry deposition mesoscale assemblies and mate- example of the Nanoscience Research techniques that emphasize thickness rial properties that leads to emergent Centers, as well as at end stations control and uniformity/conformality phenomena. Multistep, multicom- at existing characterization (elec- over formation of complex 3-D nano- ponent synthesis involving multiple tron, neutron, and photon) facilities structure shapes. Such synthesis can length and time scales demands new designed to foster in situ measure- be facilitated by self-assembly, self- means for bridging synthesis steps ment, are very desirable for research limiting processes, and self-aligned and scales (e.g., system reduction, at the mesoscale, but they cannot structures. However, the challenge sampling, filtering, and projection replace the need for widely distrib- techniques). uted capabilities.

55 From Quanta to the Continuum: Opportunities for Mesoscale Science

Cultivation of Synthesis Expertise. challenges. Nevertheless, gaps exist An even more compelling case for in our characterization capabilities for distributed synthesis facilities may be fully exploring mesoscale opportuni- the alignment of mesoscale research ties. Needs include capabilities for with broad-based technologies and • Time-dependent spatial and advanced manufacturing. In fields functional characterization of from energy research to new mate- materials and structures with rials to biomedical devices, it is the mesoscale resolution in four ability to synthesize, fabricate, and dimensions (space and time); assemble materials and structures at • In situ, operando measurement the mesoscale that translates into an to explore the dynamics of meso- economic and social benefit through scale phenomena in the environ- technology-based products. The ment in which they are active, competitive posture of the U.S. work- including synthesis environments; place can be the direct beneficiary of Graphic depiction of an ultrashort pulse of tera- and educating scientists and engineers hertz light (yellow arrow) distorting a manga- nite crystal lattice. The induced rearrangement • Sustained measurements over the at all levels in the skills represented of magnetic moments on Mn atoms (red and in mesoscale synthesis and related blue arrows) and atomic displacement patterns lifetime of a component, including research areas. The relevance and (suggested by the red and blue diamonds) are long-duration measurement tools quality of their experience are notably probed with short x-ray laser pulses. Changes and the means to artificially age in the ordering are probed with a time resolu- enhanced by the availability of state- tion of 250 fs. Experiment performed at the materials through accelerated of-art equipment and computational LCLS. Source: M. Först et al., Phys. Rev. B 84, testing. infrastructure throughout their daily 241104(R) (2011). routine. Implicit in these characterization capabilities is the challenge of very large data sets and the need to sepa- Characterization rate on the fly the signal from the Capabilities noise or risk a massive data storage The Office of Basic Energy Sciences challenge. Further, significant needs (BES) is the principal steward of exist for additional end stations and large-scale characterization facilities instruments, including those capable in the U.S., and recent investments of producing relevant sample environ- have significantly increased the suite ments, as well as advanced detectors of available tools. Currently available in order to take maximal advantage of and emerging capabilities at large- the newly available sources. scale facilities such as the Spallation Neutron Source (SNS) at Oak Ridge Current State National Laboratory (ORNL), Linac A 3D reconstruction of a 20-µm lithium-ion Ideally, one would like to have a Coherent Light Source (LCLS) at the battery electrode obtained with a transmission universal tool, something like a SLAC National Accelerator Laboratory hard x-ray microscope at the NSLS, with resolu- “scientific Swiss army knife,” with the tion of 25 nm. Measurement of 1,441 2D images (see figure at top right), and National following capabilities: Synchrotron Light Source (NSLS) II required 4 hours, with a speed increase of 1,000 expected when applied at NSLS II. Source: • Map structure in three spatial at Brookhaven National Laboratory J. Wang et al., Appl. Phys. Lett. 100, 143107 dimensions and as a function of (BNL) (see figure at bottom right), as (2012). well as advances in laboratory-scale time, including for nonperiodic measurement tools, position the molecules. community well to address mesoscale • Cover length scales from 10-10 to 10-3 m and time scales from 10-15 to 109 s. 56 From Quanta to the Continuum: Opportunities for Mesoscale Science

• Have sensitivity to chemical BNL, and the Stanford Synchrotron for people (humans) with CT scans. elements and to magnetic Radiation Lightsource (SSRL) at A new approach, which can provide moments and electric moments SLAC, provide high intensity beams fine spatial resolution, involves “lens- simultaneously with structural of photons from infrared light to less” imaging. Rather than making the resolution. x-rays. The scattering of infrared light beam spot smaller and imaging one • Detect vibrational, electronic, from a sample can provide sensi- point at a time, this approach uses a and magnetic excitations with tivity to chemical bonds (through their large spot size and collects coherent energies from 10-9 to 101 eV and vibrational energies) and the ability to diffraction measurements with small their associated excited state carry electrical currents (through the displacements of the spot position dynamics. electronic conductivity). Ultraviolet on the sample. The spatial image of light is commonly used to probe the sample is reconstructed from the • Measure as a function of applied the electronic structure of materials data via computer. The time required electric field, magnetic field, pres- through photoemission spectroscopy. for measurements can vary widely. sure, or chemical environment as X-rays can probe crystal structure via An individual “snapshot” might take well as extremes of temperature. diffraction or absorption spectros- 10-3 s, while high-resolution 3D copy, and new imaging modes have imaging can take minutes to hours. While such a universal tool may be recently been developed to provide beyond our reach, we do possess both in-situ zone-plate imaging under Inelastic x-ray scattering provides several different toolboxes, each controlled environments and lens- information on how atoms and elec- containing a range of sophisticated less imaging for hard x-rays at high trons fluctuate within materials. To instruments that provide complemen- resolution. One can obtain elemental obtain the necessary energy resolu- tary capabilities. These toolboxes are and chemical sensitivity by tuning the tion, one must throw away most of exemplified by the BES user facili- x-ray energy to an atomic transition. the beam intensity, resulting in rela- ties that have been constructed with With some atomic transitions, it is tively low rates of data collection. As past investments. They can be distin- also possible to obtain information on a consequence, inelastic instruments guished by the types of particles that the local magnetization. have not yet focused on spatial reso- are used to probe materials: photons, lution or time-dependent studies. neutrons, and electrons. Each has its With all of these techniques the beam own advantages and limitations, and can be rastered across the sample The LCLS at SLAC is an x-ray free all are needed to provide the desired (or the sample through the beam) electron laser that provides very range of capabilities and approach the to spatially map characteristics. In short pulses of light (10-14 s) with vision of a universal tool. Importantly, this mode, the spatial resolution is remarkable flux. Beam lines provide these large-scale tools are comple- limited by the size of the beam spot capabilities for soft and hard x-rays. mented by a widely distributed array focused on the sample, which can be A typical application is to excite a of characterization capabilities. Where in the range of 10-7 to 10-6 m. Depth sample with an impulse, usually from possible, combining these probes sensitivity varies. A technique such an optical laser, and then follow the with each other simultaneously to as photoemission is surface sensi- evolution of the sample with time enable multi-modal measurements is tive (an advantage for some applica- using x-ray pulses. Delay times a mesoscale opportunity. tions) but cannot sample the bulk of a between the initial impulse and the specimen. Hard x-rays can penetrate probe can be resolved on the scale of Synchrotron light sources, including 10-6 to 10-3 m, but with the tradeoff 10-13 s. The time between x-ray pulses the Advanced Light Source (ALS) that the more penetrating x-rays do is 10-2 s, so the pump and probe at Lawrence Berkeley National not provide elemental sensitivity. measurements must be repeated Laboratory (LBNL), the Advanced One application of x-rays is tomog- many times to map the time evolu- Photon Source (APS) at Argonne raphy, in which one can map the tion of the sample. One challenge National Laboratory (ANL), the density of materials by assembling is that the energy of individual x-ray NSLS (and soon to arrive NSLS-II) at a series of images just as is done pulses can be sufficient to damage

57 From Quanta to the Continuum: Opportunities for Mesoscale Science

the sample; clever schemes are used Variation of the density of CO in to move samples in the beam so that 2 pores of coal relative to bulk liquid each pulse probes pristine material density as a function of pore size. (information travels at the speed of Data obtained from small-angle light, and damage at the speed of neutron scattering measurements sound, enabling use of “probe before performed at ORNL High Flux Isotope Reactor and the Center for destroy” measurement methodolo- Neutron Research of the National gies). It is clear that the full potential Institute of Standards and Tech- of x-ray free electron lasers has yet to nology. Results are relevant to the challenge of carbon sequestration. be realized, and various concepts are Source: A. P. Radlinski et al., Lang- being developed to tap this potential, muir 25, 2385 (2008). including at the mesoscale.

Neutron sources include the SNS and the High Flux Isotope Reactor (HFIR) at ORNL and the Lujan Neutron Scattering Center at Los Alamos National Laboratory (LANL). Whereas x-rays and electrons scatter from the electron density in a sample, neutrons scatter from atomic nuclei and magnetic moments. As a result, neutrons can probe the entire volume of samples with thicknesses greater than 10-2 m. Real-space imaging is done with spatial resolution of 10-4 m, with particular sensitivity to hydrogen- The location of a layer of green fluorescent protein molecules within a polymer film determined with resolution of 10-9 m using neutron reflectivity measurements at the SNS. containing compounds. Variations Source: V. Kozlovskaya et al., Soft Matter 7, 11453 (2011). in crystal structure can be mapped in 3D with a resolution of 10-9 m3 scattering instruments having energy available through BES’s Nanoscience (1 mm x 1 mm x 1 mm). Frequently, one resolutions ranging from 10-6 to 1 eV. Research Centers (and various is interested in the statistical distribu- In applications to chemical spectros- university-based nanocenters). Trans- tion of particle, pore, or domain sizes, copy, one can fingerprint molecular mission electron microscopy (TEM) and this can be determined with small- bonds by their vibrational frequen- allows atomic-scale imaging of thin angle neutron scattering (see figure at cies. Neutron capabilities are not as samples. Beyond determining posi- top right). The structures of polymer ubiquitous as photon sources, and tions of atoms, the types of atoms films or magnetic multilayers can be measurements can require more time can be resolved with electron-energy- determined by neutron reflectivity and greater expert knowledge by the loss spectroscopy. TEM is ideal for with spatial resolution of 10-9 m (see user. Nevertheless, neutron capa- characterizing grain boundaries and figure at bottom right). With diffrac- bilities are particularly well suited interfaces between distinct materials. tion and more-generalized scattering to a number of mesoscale science A particular opportunity for electronic instruments, it is straightforward to challenges. scattering is to extend their exquisite measure spatial correlations among spatial resolution into the time domain atomic magnetic moments, including Electron beam microcharacterization (i.e., dynamic electron microscopy) the excitations of such systems. centers are located at ANL, LBNL, and into a greater suite of environ- Atomic vibrations and molecular and ORNL, with further capabilities mental conditions; initial efforts are motions are probed with inelastic

58 From Quanta to the Continuum: Opportunities for Mesoscale Science

underway in these regards. A strength techniques, employing selective be an obstacle to effective mesoscale of the Nanoscience Research Centers chemistry or other phenomena research, where it is often desirable is their physical co-location with to enable the real-time tracking to use multiple techniques to probe photon and neutron sources. This of active sites, can be crucial in multiple length and time scales (simul- proximity is yielding positive results in finding and watching functional taneously, if possible). This barrier can the coupling of synthesis and theory defects while they perform in their be reduced by suitable investment in with characterization as well as the local mesoscale architecture and instrument staff and software devel- acquisition of a suite of multi-modal, at heterogeneous reactive sites opment at facilities to allow efficient in situ measurements of relevance to during chemical processes. collaboration and data processing mesoscale science. • Long-duration measurements and on the research projects of outside accelerated testing in new modal- users. Similarly, before a user takes a Characterization Needs ities, based on advanced meso- sample to a facility, the quality of the sample needs to be characterized in As indicated in the above section, an scale characterization methods, the laboratory in which the sample important step toward realizing the promise major advances in was synthesized. Much needed is an meso opportunity is to continue to prediction and controlled evolu- appropriate scale of support for such explore and stretch the limits of the tion of functional defects and also mid-scale characterization instrumen- capabilities of newly available large- enable the harnessing of fluctua- tation as a complement to the large- scale facilities. While the sources tions and the inhibition of degra- scale facilities. of probe beams are relatively well in dation for intrinsically metastable hand, significant advances in probe systems. optics, detectors, sample environ- • Data mining strategies, as devel- Theory and Simulation ments, and data handling/processing oped for applications from secu- Capabilities are needed. Particular challenges rity to marketing, must become an An essential element of the meso- identified in the priority research essential component of data-rich scale challenge is to discover the directions include: materials science instrumentation underlying theoretical principles that • New methods to find and watch (e.g., computed tomography and give rise to emergent phenomena as structural and functional defects orientation mapping) now avail- the system size increases from atom- evolve in 3D and sub-surface able, particularly at advanced istic to macroscopic. Theoretical tools systems are prerequisite to user facilities, to answer ques- and approaches do not exist today to bridge these processes across scales mastering defect mesostructure tions such as: How do we find in space, time, and energy. Currently, and evolution. Defect evolution, features in large 3D CT images? multiscale phenomena are treated in i.e., an explicit focus on dynamic How do we correlate data from isolation or coupled only through ad disparate sources such as images phenomena, must acquire the hoc scale separation and parameter importance already attributed to and acoustic emission? This is passing. A major theoretical challenge static structural defects, since particularly acute when the data is to develop a seamless and unified defects in materials systems challenge is at least five dimen- theoretical framework that describes evolve in time and directly trans- sions: three spatial dimensions, physicochemical phenomena across late to performance and reliability temporal resolution, and the multiple spatiotemporal scales, espe- in physical systems. To effectively environmental state in which the cially in the presence of mesoscale direct assembly of mesoscale measurement is taking place. defects and interfaces that cause the architectures, one must observe mesoscale to be a unique scale in the these processes during the To efficiently take advantage of evolution from atomistic to continuum. Further, there are unknown unknowns assembly events. new characterization capabilities, that arise at the mesoscale and give rise a researcher often requires a high • In situ, in operando measure- to unanticipated phenomena through level of experimental expertise in a ments including defect decoration coupled and competing interac- single technique. This demand can tions. The complementary theoretical

59 From Quanta to the Continuum: Opportunities for Mesoscale Science

challenge is to develop reliable many- body quantum models to incorporate emergent meso phenomena to yield first principles design of materials with targeted functionality. Finally, these models must be operational at a level that can guide characterization methods in extracting relevant observables and direct synthesis in tailoring mesoscale architectures to yield the controlled functionality we seek. Thus, the third element of the challenge is to realize these advanced multi-scale and As represented by the trajectory (red) of a system between two stable states (blue many-body models in a broadly avail- circles), evolution of complex systems involves rapid local fluctuations and rare, able and well-documented form (e.g., frustrated, large-amplitude transitions between basins of stability/metastability. widely distributed community codes) Source: G.K. Schenter, Pacific Northwest National Laboratory. in order to empower and motivate a broad suite of synthesis and character- domains of local equilibria and their find the optimal and most economical ization efforts. associated fluctuations. In metastable connection between scales to avoid systems such domains are frequently unnecessary simulation or the need Recent advances in computational connected through a complex to track dynamics of irrelevant fluc- power and software development network. New tools are needed to tuations or flows. have enabled partial success in this identify the large-amplitude, rare, frus- pursuit, including development of trated, and fluctuation-driven motions Across scales, the simulation tools that reliable, quantum-based models of that underlie this connectivity. need to be employed and enhanced atomic-scale structure and reactivity, can be partitioned into four classes: and large-scale continuum simula- At the core of this scientific investiga- (1) models of molecular interaction, (2) tions (e.g., adaptive grid methods to tion is an effort to develop “mesoscale sampling techniques, (3) exploration solve partial differential equations) of principles” that couple simulation tech- of essential collective variables and complex structures and transport in niques and theoretical frameworks at order parameters, and (4) projection materials in three dimensions with low different length and time scales (see and coarse graining methods. symmetry. The advent of “exascale figure on next page). At each scale, computation” holds further promise techniques exist to describe trans- Models of molecular interac- for mesoscale science breakthroughs. port, flow, diffusion, transformation, tion include fully correlated elec- equilibrium, fluctuations, and tran- tronic structure calculations, which Owing to their small particle size and sient response to external stimuli. The allow characterization of excited large surface areas most nano- and fact that the appropriate order param- states. Reduced (mean field) levels mesoscale structures are inherently eters, coordinates, or field variables of electronic structure include far from equilibrium and intrinsically that are used to characterize the state density functional theory and semi- metastable (see figure at right). We of a system are not universal is at the empirical methods either based on need to develop theoretical/simula- heart of the scientific challenge in approximations to Hartree-Fock or tion tools to elucidate how a system realizing a unified mesoscale model. methods based on the tight-binding evolves to achieve a state of persis- Finding the formal relation between Hamiltonian. The use of efficient tent metastability. This requires under- scales is required for the multiscale and accurate electronic structure standing how systems organize by simulation to be accurate and predic- methods will be imperative to ascer- characterizing the nature of the atten- tive. Furthermore, it is essential to the tain reduced levels of description, dant collective motions, elucidating efficiency of simulation techniques to where the relevant short-range/fast the energy landscape, and identifying degrees of freedom can be accurately

60 From Quanta to the Continuum: Opportunities for Mesoscale Science

folded into a mean field description at large scales that will contain the relevant aspects of important correlation effects. The rigorous determination of this reduction will enable one to define the important principles to extract from one scale to another.

The importance of sampling the relevant fluctuations, especially tempo- rally, is significant at every scale. The sampling methods are, in principle, independent of the complexity of the interaction potentials. Statistical sampling methods will be all important in defining collective variables or order parameters to provide input into a reduced description of the relevant interactions. Important examples of methods that allow the elucidation Schematic of the mesoscale principles that need to be developed at each scale to of complex collective variables are predict the association of complex nanostructures. The lower left corner depicts meta-dynamics and its extensions, the use of quantum mechanics at the angstrom scale that will, in turn, determine as well as parallel tempering. Even the appropriate description of aqueous ion solvation. The principles of solvation govern the interaction of ions with surfaces. These surfaces then obtain a surface for reduced forms of the interaction charge distribution, which dictates the effective interactions, ultimately leading to potential, these methods will require the formation of complex structures at the micron scale. Source: M.D. Baer and modern computational resources to C.J. Mundy, Pacific Northwest National Laboratory. ensure their effectiveness. Additional methods such as transition path become dominant or important. properties such as stability. Novel sampling and other path sampling Many coarse-grained simulations use theoretical methods of reduction must techniques will become important in interaction potentials derived from be developed that go well beyond identifying basins of stability/meta- potentials of mean force. When ions the mean field approaches, such as stability. In principle, one could use that are contained in a solvent or Poisson-Boltzmann, to derive effec- a reduced form of the interaction electrolyte interact significantly with tive dielectric continuum theories that potential to provide a rough estimate nanoparticles, other approaches must contain correlations imported from of the free energy landscape by using be formulated to obtain a reduced molecular simulation. the aforementioned methods. The description of the interaction. Such free energetics and the concomitant electrostatic effects can play a domi- In the extension of theoretical repre- kinetics could be refined with higher nant role and are currently underrep- sentations to larger scales, explicit levels of theory to obtain more accu- resented in simulations of colloidal molecular-level detail describing rate estimates of the rates and ther- suspensions and continuum simula- the ensemble must be replaced by modynamics, which could then be tions of porous media. Many simula- an appropriate set of reduced field passed to another scale for further tions of suspended particles leave variables that propagate in space analysis. out interactions between charged and time. Traditionally, these fields colloidal particles with dissolved consist of conserved quantities such Challenges at the molecular scale counter ions that, in turn, determine as number, momentum, and energy are borne out when electrostatics

61 From Quanta to the Continuum: Opportunities for Mesoscale Science

density. The focus on the larger description of flow and transport in materials science and the engineering scale will be to relate the results of bulk fluid to include multiple surface disciplines, but also specialized areas detailed molecular simulations to species, 2D transport along surfaces, within disciplines such as organic and field descriptions of transport and nonuniform properties on the surface, inorganic chemistry, atomic, molecular and optical physics, and biomolecular material properties. Partitioning the and appropriate couplings between materials. Traditional disciplines reflect system into relevant volumes that surface fields and fields defined in the history of reductionist science that can be described by using molecular- the bulk fluid. Additional complexity divides phenomena into component arises if the surfaces are moving or level detail or the appropriate field parts, each with a rich and deep foun- description of fluid dynamics, dielec- deformable, as is the case in fluid- dation to be elaborated. Nanoscience tric behavior, elastic properties, fixed fluid interfaces and surface depo- shows us that the boundaries between impenetrable volume, etc., will be sition/dissolution. In addition to traditional disciplines become fuzzy the challenge. The rigorous coupling correctly formulating these equations, and tend to disappear at small length of molecular detail described by substantial computational challenges scales. Mesoscience takes this a quantum mechanics or Newton’s are associated with solving them step further, requiring active cross- equations of motion to fluid descrip- numerically for realistic geometries. disciplinary research to explore new tions of the system described by Understanding the utility and exten- architectures and phenomena arising Navier-Stokes equations or elastic sions of smooth particle hydrody- from interactions among nanoscale degrees of freedom and mesoscale solid descriptions of the system namics, dissipative particle dynamics, collective responses, defect configura- described by linear or nonlinear solid the diffusion equation, the Langevin tions, fluctuations and heterogeneity. A mechanics will require new math- equation, and kinetic Monte Carlo new emphasis on creating an interdis- ematical formulations of numeri- and lattice Boltzmann descriptions of ciplinary workforce led by established cally efficient projection techniques. motion will be required. A significant scientists at the research frontier and As mentioned before, electrostatics component will be the solution of extending to the training of students, needs to be efficiently introduced by the relevant partial differential equa- postdocs and early career scientists extending dielectric continuum theo- tions using unstructured 2D and 3D is needed to embrace the potential of ries, where a more accurate mean- meshes. The coupling of 2D and 3D mesoscience. field ionic approximation is achieved fields via complex boundary condi- through inclusion of a correlation from tions presents unique challenges for Certain areas of interdisciplinary the molecular scale. developing the mathematical and science stand out as especially impor- simulation foundation necessary for tant for mesoscience. Self-assembly Ultimately, the appropriate mesoscale elucidating organizational principles is a ripe opportunity for the new cross-specialization area of computa- principles can be used to develop at the mesoscale. tional materials synthesis. Predicting improved field-based representa- complex self-assembled architec- tions of reaction, flow, and transport Workforce tures on the basis of starting compo- at scales ranging from nanometers nents and assembly conditions is a to microns. Such representations are Considerations rich and mostly unexplored oppor- Research in materials science, relevant for understanding how many tunity. The current state of the art of condensed matter physics and chem- systems behave in practical settings, self-assembly targets controlling the istry, especially at the nanoscale, both in the natural environment and in formation of complex structures; the provides a strong base for the devel- industrial applications. coming high-payoff opportunity is opment of mesoscience, including targeting not only structure but also the capabilities discussed above. These new field-based descrip- previously unachievable functionality Understanding and controlling emer- tions can still be quite complicated. and performance. Developing a work- gent phenomena at the mesoscale force proficient in next-generation self- Explicitly incorporating informa- requires new modalities of research assembly requires fluency in materials tion from microscopic properties will and training that cross boundaries chemistry, condensed matter, layer- require extending the usual continuum separating not only traditional disci- by-layer growth, and the new area of plines such as chemistry, physics, computational synthesis.

62 From Quanta to the Continuum: Opportunities for Mesoscale Science

Understanding mesoscale pheno- be devised. The workforce that easily New modalities of training are needed mena requires an emphasis on relating crosses the theory, modeling, charac- to implement the constructionist prin- collective properties and dynamics terization and synthesis worlds does ciples of mesoscience. The traditional to the generation and evolution of not yet exist. Building such a workforce single mentor/advisor mode of training lower-dimensional defect structures. requires bringing these specialized graduate students can be extended to An enhanced focus on defects inten- communities together in innovative include dual or triple mentorship, with tionally introduced into controlled, ways that remain to be designed and close communication and advising by low-defect synthesis processes may implemented. two or more mentors from different be an important enabler. Experimental fields or specializations. Two advisors, efforts can be profoundly assisted by The need for interdisciplinary and cross- for example, can implement training in combinatorial approaches for experi- specialization cooperation creates a computational synthesis, one expert mental design and by development second challenge. The breadth of knowl- in each field, mentoring one or more of a portfolio of modeling and simula- edge required for mesoscience is much students in a cooperative effort. Every tion tools, especially when coupled to wider than for traditional reductionist boundary that needs to be bridged, in situ characterization. A further need science. Highly cooperative research such as multi-modal characterization from computational infrastructure is teams strategically chosen to develop spanning spectroscopy, structure and for means to couple the statistical specific opportunities are needed. Such dynamics by x-rays, neutrons and mechanics of mesoscale assemblies teams are emerging in centers such as electrons, can be approached in the to materials properties and hierarchical the Energy Frontier Research Centers same way. Incentives for dual or triple modeling frameworks that bridge and the Energy Innovation Hubs. These mentoring and advising can be imple- both synthesis processes and length/ models should be embraced and mented, such as research funding, time scales. Such advances present exploited, identifying best practices fellowships and student or faculty the possibility of powerful mesoscale and lessons learned and extending their awards that require cross-boundary designs that enhance functionality by reach to the specific disciplines and training. decoupling structure and function. specializations needed to achieve the potential of mesoscience. The organiza- The Office of Science’s Early Career Theory, modeling, characterization tional principles, operating modes, and Research Program is an exciting and synthesis are traditional areas of reward systems for interdisciplinary and opportunity for fostering the meso- specialization within materials and cross-specialized research teams must scale workforce. Compelling opportu- chemical sciences that must be much be developed and implemented. nities exist for students and post-docs more intimately merged to pursue the to acquire skills in both experiment opportunities of mesoscience. The To realize the mesoscience oppor- and theory, as well as to gain hands- complexity of mesoscience, and the tunity in a timely manner, the meso- on synthesis experience through joint sheer number of components, param- scale workforce of the future must mentoring arrangements. The need is eters, phenomena and functionalities be nurtured aggressively. The next particularly acute for instrument scien- it comprises cannot be embraced generation of mesoscientists must tists to exploit newly available charac- without strong mutual interaction. be trained in cross-disciplinary and terization sources to impact mesoscale Computation must rise to the challenge cross-specialization science. They science through multi-modal measure- of describing and ultimately predicting are best trained by hands-on experiments in collaboration with users. more complex materials behavior; and ence working directly in the integrated Importantly, these researchers will not characterization must include compu- frontier research teams. They must only define the future of mesoscale tation and modeling to integrate and see and experience the new inclusive science but also represent an oppor- organize the wealth of information it paradigm during their formative years tunity for developing a diverse tech- will produce by the new generation to fully embrace, deploy and extend nology and manufacturing workforce of multi-modal, in situ, space- and its power. Their first hand familiarity well suited to the industries and prod- time-resolved coordinated measure- with the full range of cross-boundary ucts arising from materials and nano- ments. Data management and mining research including its intellectual, tech- technology development. strategies must be developed; new nical and human dimensions is the theoretical formulations that describe fastest and most direct approach to mesoscale emergent behavior must realizing its benefits.

63 From Quanta to the Continuum: Opportunities for Mesoscale Science

64 From Quanta to the Continuum: Opportunities for Mesoscale Science

Realizing the Meso Opportunity

The promise of mesoscale science is discovering and managing the connections among phenomena, degrees of freedom, and functional units to create materials with new behavior, higher performance, longer life, and lower cost, advancing the science of rational design of materials with targeted functionality. The mesoscience approach to materials is qualitatively different from conventional macroscale approaches that rely on serendipitous discovery followed by top-down reductionist analysis of the new macroscopic phenomena to identify their nanoscale origins. Instead, the constructionist meso approach emphasizes bringing disparate nanoscale phenomena and degrees of freedom together to discover or direct new emergent behavior at the mesoscale. Bottom-up synthesis streamlines the innovation process, eliminating the distraction of pre-existing and fully formed macroscopic behavior that competes with alternate innovative or emergent outcomes, and simplifying the building blocks to their most basic atomic, molecular, or nanoscale level. The promise and vision of mesoscience are to make mesoscale emergence a predictive theoretical enterprise steeped in numerical modeling and implemented through mesoscale synthesis and characterization.

Realizing the meso opportunity requires developing new approaches to observation, characterization, synthesis, fundamental theory, phenomenology, and numerical modeling. Beyond the technical means of realizing mesoscience, there is a strong cultural component that must be fostered. The next generation of mesoscientists must not only be fully comfortable with synthesis, characterization, and analytical theory/numerical modeling, but also actively pursue all three of these basic activities in her/his mesoscience research. This requires establishing early training with multiple mentors and team research as the mesoscience standard to cover the broad range of required expertise.

The prior section outlined needed technical directions in synthesis, characterization, and theory/modeling required for effective mesoscience, and the cultural

65 From Quanta to the Continuum: Opportunities for Mesoscale Science

shifts in approach, integration, and team-oriented research that will bring the mesoscience opportunity to fruition. The Mesoscale Science Subcommittee has distilled these six specific actions that will significantly advance mesoscale science:

1. Increase investment in small- and intermediate-scale lab instrumentation for mesoscale science; 2. Develop detectors, sample environments, instruments, and end stations for mesoscale research that fully capitalize on the large-scale sources available at national user facilities; 3. Grow the cohort of beam line and facility scientists defining the frontiers of increasingly sophisticated mesoscale measurements at user facilities; 4. Attract and retain the best and brightest graduate students and postdoctoral researchers to be the mesoscale energy scientists of the future; and establish dual or triple mentorship/advisor paradigms to train them in cross-boundary mesoscience; and 5. Stimulate the formation of multi-disciplinary research groups that include theorists and experimentalists and span from discovery science to use-inspired research.

66 From Quanta to the Continuum: Opportunities for Mesoscale Science

Conclusion

The mesoscale opens a new frontier of scientific opportunity. Mesoscale science builds on the remarkable nanoscience insights that the research community is now producing. The reductionist drive to understand bulk materials and phenomena from the top down has revealed the central role of atomic, molecular, and nanoscale structure in shaping bulk behavior. We are now ready to explore the opposite direction, a bottom-up journey from atomic, molecular, and nano structures and phenomena to the broader horizons and more complex behavior of the mesoscale. The mesoscale brings profound changes, replacing the atomic granularity of matter and the quantization of energy with continuous matter and energy, and enabling the onset of collective behavior of ensembles of particles, the interaction of coupled and competing degrees of freedom, the appearance of defects and fluctuations that profoundly alter the behavior of perfect structures, and the formation of heterogeneous composite systems whose parts work cooperatively to transform energy, charge, and mass. Building on knowledge from the nanoscale, these emergent mesoscale features can be manifested in innovative architectures to create targeted new materials phenomena and functionalities, and, ultimately, disruptive technologies with the potential to outperform existing technologies and dominate the competitive market. Seizing the opportunity to develop mesoscale science and to dominate the innovation opportunities it spawns brings an enduring competitive advantage for future economic growth.

The promise of mesoscale science is rich. Manipulating mesoscale architectures assembled from nanoscale building blocks enables virtually endless possibilities to discover, design, and enhance complex phenomena and functionalities. Directed assembly offers unprecedented opportunity to create complex structures and functionalities, inspired in part by biology. Light can be manipulated in photonic crystals, metamaterials, and surface plasmon polaritons to promote chemical reactions, harvest energy from sunlight, and enhance the performance of light-emitting diodes. New generations of electrodes for batteries and fuel cells can be designed to promote the coordinated motion of electrons, ions, and gases and to maximize efficiency and energy density. Mesoporous membranes with defined charge and chemical profiles lining the pores can be designed to separate carbon dioxide, purify water, and catalyze chemical reactions.

There are new organizing principles for mesoscale behavior to be discovered. We understand quantum mechanics for atomic, molecular, and nanoscale phenomena, and classical mechanics for macroscale behavior, but we do not know the organizing principles governing the mesoscale. What principles govern

67 From Quanta to the Continuum: Opportunities for Mesoscale Science

self-assembly, for example, or the appearance of new behavior from the interaction of competing degrees of freedom? New phenomena will emerge as mesoscale architectures are tuned to promote collective behavior or the interaction of targeted degrees of freedom. As mesoscale systems grow in size, defects, interfaces, and fluctuations appear and can be manipulated to program the thermal, electronic, and mechanical response of the bulk. Mesoscale science thus represents a discovery laboratory for finding new phenomena, a self-assembly foundry for creating new functional systems, and a design engine for new technologies.

This report describes the richness and scope of the mesoscale frontier in six priority research directions: • 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

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.

68 From Quanta to the Continuum: Opportunities for Mesoscale Science

While the challenge of controlling mesoscale phenomena is enormous, the potential benefits are significant and far- reaching. Success will enable transformational outcomes for society. Mesoscale science has the potential to re-energize innovation, deliver high-performance, low-cost technologies, and provide a competitive foundation for enduring economic growth. Transformational mesoscale outcomes include • The ability to manufacture at the mesoscale: that is, the directed assembly of mesoscale structures that possess unique functionality that yields faster, cheaper, higher performing, and longer lasting products, as well as products that have functionality that we have not yet imagined. • The realization of biologically inspired complexity and functionality with inorganic earth-abundant materials to transform energy conversion, transmission, and storage. • The transformation from top-down design of materials and systems with macroscopic building blocks to bottom-up design with nanoscale functional units producing next-generation technological innovation.

It is both the magnitude of the challenge in bridging quanta to the continuum and the potential dividend in controlling the mesoscale that have energized the research community and motivated this report.

69 From Quanta to the Continuum: Opportunities for Mesoscale Science From Quanta to the Continuum: Opportunities for Mesoscale Science

Appendix: Community Input to the BESAC Meso Subcommittee Process

To foster engagement with the broader scientific community and to collect input on the frontiers of mesoscale science, the Meso Subcommittee organized town hall meetings at the American Physical Society meeting in Boston (February 2012), the American Chemical Society meeting in San Diego (March 2012), the Materials Research Society in San Francisco (April 2012), and in Chicago in May 2012. It also hosted an American Chemical Society webinar in April 2012. More than 1000 researchers participated in one or more of these outreach activities. These efforts led to more than 100 contributed quad charts that are listed by title and contributor below. They can also be found at www.meso2012.com.

Mesoscale Science of Damage Accumulation in Long-Lived Systems, R. French (Case Western Reserve University)

Role of Microscopic Quenched Disorder in Macroscopic Stability, R. Podgornik (University of Massachusetts)

Failure Mechanism in Materials and Meso-Scale Structures, W. Ching (University of Missouri, Kansas City)

Time- and Length-Scales of Radiation Damage Processes, S. Kucheyev (Lawrence Livermore National Laboratory)

Self-organization by Irradiation and Plastic Deformation, Pascal Bellon (University of Illinois at Urbana-Champaign)

Plasticity of Crystal Interfaces, S. Mesarovic (Washington State University)

Mesoscale Structural Changes from Atomic Scale Radiation Damage, D. K. Shuh (Lawrence Berkeley National Laboratory)

Optimizing Flow Stress of a BCC Metal at T/Tm<0.1, Anthony Bryhan

Damage Accumulation in, and Lifetimes of Materials, Tony Rollet (Carnegie Mellon University)

Microstructure Based Heterogeneity Evolution Leading to Phase Transformation and Damage/Failure Events, C. A. Bronkhorst (Los Alamos National Laboratory)

Harnessing Non-linear Effects for Mesoscale Devices, D. Lopez (Argonne National Laboratory)

Fluctuation De/stabilization of Mesoscale Structures, V. A. Parsegian (University of Massachusetts)

Magnetization Dynamics: The Inherently Mesoscale Problem, A. Hoffman (Argonne National Laboratory)

71 From Quanta to the Continuum: Opportunities for Mesoscale Science

Lithium Batteries: Composite Electrode Materials by Mesoscale Functionality from Interface-Dominated Oxides, Design, J. Flake (Louisiana State University) Brandon Chung (Lawrence Livermore National Laboratory)

Wetting and Fluid Transport Phenomena under Mesoscale Rational Design of Electrode/Electrolyte/Membrane Confinement, A. Checco (Brookhaven National Laboratory) Interfaces for Energy Storage, Michel Dupuis (Pacific Northwest National Laboratory) Making Better Foams, R. Chau (Lawrence Livermore National Laboratory) Ionomer and Solvent Structure Interactions in Electrodes, R. Borup (Lawrence Berkeley National Laboratory) The Role of Defects in Phase Transitions and Kinetics, M. Akin (Lawrence Livermore National Laboratory) Mesoscopic Perspectives in Separations Science, Ross Ellis (Argonne National Laboratory) Chemistry at Extremes, M. Akin (Lawrence Livermore National Laboratory) Interfacial Durability of Meso-scale Systems, Debbie Myers (Argonne National Laboratory) Thermal Conductivity of Thin Films at High Temperature (> 4000 K) and Pressure (> 100 GPa), D. Fratanduono Thin Boundary Layer Heat Transfer, Charles Agosta (Clark (Lawrence Livermore National Laboratory) University)

Mesoscale Phases: Control of Global Structural and Functional Mesoscale Systems, Gary Rubloff (University of Physical Properties, R. Jin (Louisiana State University) Maryland)

Nanoscale Materials Engineering for Mesoscopic Meso Challenge: Multi-Valent Interactions in Poly- Functionality, W. Plummer (Louisiana State University) electrolytes, Matt Tirrell (University of Chicago)

Electrocatalysts: Active Sites and Interfaces of Clusters, Meso-photonics for Energy Applications, Marin Soljacic J. Flake (Louisiana State University) (Massachusetts Institute of Technology)

Dynamic Tomography: Polymers, Batteries, and H2 Storage, Controlling Optical Properties of Functional Mesoscale Les Butler (Louisiana State University) Systems via Multi-Scale Interfacial Interactions, Stephen K. Doorn (Los Alamos National Laboratory) 3D Mesostructured Battery Electrodes, Paul Braun (University of Illinois, Urbana-Champaign) “Upscaling” Nanoscale Thermoelectrics: The Mesoscale Design Challenge for Real-World Energy Needs, Jennifer Functional Metal-Organics for Gas Storage, J. Spence Hollingsworth and Andrei Piryatinski (Los Alamos National (Arizona State University) Laboratory)

Self-adaptive Electrodes for Energy Applications, T. Rajh Design and Control of Interfaces in Soft Materials, Sergei (Argonne National Laboratory) Tretiak (Los Alamos National Laboratory)

Multifunctional Metal-Organic Frameworks, V. Zapf (Los Heteroatom-Doped Graphene-Sheet Structures in High- Alamos National Laboratory) Performance Li-Ion Battery Anode, Gang Wu and Piotr Multistep Multiscale Control and Delivery, Mark Jensen Zelenay (Los Alamos National Laboratory) (Argonne National Laboratory)

72 From Quanta to the Continuum: Opportunities for Mesoscale Science

Role of Fluctuations in Formulating Organizing Principles Hierarchically Structured Porous Electrodes for Energy in Meso-scale Systems, Roger French (Case Western Storage, M. Tang (Lawrence Livermore National Laboratory) Reserve University) Integrating Mesoporosity into Nanoporous Materials, Controlling Coupled Ferroic Domains at Meso-scopic J. Biener (Lawrence Livermore National Laboratory) Length Scale, Stephen K. Doorn (Los Alamos National Elastic Properties for Mesoporous Structures under Laboratory) Extreme Conditions, N. Butch (Lawrence Livermore Emergent States in Oxide Interfaces, Alexander Balatsky National Laboratory) (Los Alamos National Laboratory) Fluid, Mass, and Energy Transport in Porous and Quasi- Cooperative Phenomena in Excitonic-Plasmonic Systems: porous Media, A. Tompson (Lawrence Livermore National Plasmon-Enhanced Quantum-Dot Devices, Victor I. Klimov Laboratory) and Jeffrey M. Pietryga (Los Alamos National Laboratory) Emergent Transport Phenomena in Natural Systems, Mei Emerging Mesoscale Conduction Mechanisms: Charge Ding (Los Alamos National Laboratory) Transport in Nanoscale Assemblies and across Complex Gas Transport in Mesoporous Inorganic-Organic Hybrids, Interfaces, Victor I. Klimov and Istvan Robel (Los Alamos Jeff Urban (Lawrence Berkeley National Laboratory) National Laboratory) Reactive Transport through Mesoporous Media, Gordon Shuffling Materials Chemistry: Controlling Emergent Brown (Stanford University) Phenomena in Layered Materials, Michael Janicke and Mark McCleskey (Los Alamos National Laboratory) Multi-physicochemical Transport/Interfacial Processes in Permeable Media, Qinjun Kang and Piotr Zelenay (Los Meso Magnets: Understanding the Quantum to Classical Alamos National Laboratory) Transition, I. K. Schuller and L. J. Sham (University of California, San Diego) Visualization and Control of Transport and Flow through Mesoporous and Crowded Environments, Juan G. Duque, Quantum Design of Materials with New Functionalities, Jared J. Crochet, and Stephen K. Doorn (Los Alamos Valerii Vinokur (Argonne National Laboratory) National Laboratory) Alternate Explosives Signatures for Detection, David S. Gas Flow in Mesoporous Media, Qinjun Kang (Los Alamos Moore (Los Alamos National Laboratory) National Laboratory) Functional Materials in Extreme Subsurface Conditions, Confinement in Meso Pores-Macromolecular Crystals and Amr Abdel-Fattah (Argonne National Laboratory) Sensors, I. K. Schuller (University of California, San Diego) Next-Generation Theory and Modeling Approach for Role of Solvent Meso-structures in Catalysis, R. Williams, Multiphase Flow in Porous Media, H. H. Liu (Lawrence J. Gordon, and D. Thorn (Los Alamos National Laboratory) Berkeley National Laboratory) Self-assembled Heterostructures for Energy, S. Darling How Is Electronic Structure Correlated with Morphology (Argonne National Laboratory) and Mesoscopic Structural Stability? J. Biener (Lawrence Livermore National Laboratory)

73 From Quanta to the Continuum: Opportunities for Mesoscale Science

Optically Directed Assembly of Mesoscale Hybrid Ultrasonically Guided Assembly of 3D Periodic Structures, Structures, S. Sankaranarayanan (Argonne National Dipen Sinha (Los Alamos National Laboratory) Laboratory) Artificial Leaves: Integrated Soft, Biological & Composite Self-assembly and Dis-assembly of VNPs with Cargos, Mesoscale Systems for Enhanced Bioenergy Production, N. Steinmetz (Case Western Reserve University) Amr Abdel-Fattah (Los Alamos National Laboratory)

Directed Assembly of Pre-programmed Building Blocks Nanomaterial-Based Complex Architectures for Structural at the Mesoscale, T. Han (Lawrence Livermore National and Functional Heterogeneity, H. Zbib (Washington State Laboratory) University)

Proteins Self-organize into Living Materials, M. Gardel Driving Structural Phase Transitions in Periodic (University of Chicago) Mesostructures, J. Crowhurst (Lawrence Livermore National Laboratory) Self-assembled Peptide Amphiphiles to Control Cellular Response, G. Schatz (Northwestern University) Taming Complexity from Atoms to the Mesoscale: NP/ Organic Hybrids, P. Duxbury (Michigan State University) Understanding and Controlling Self-assembly of Inorganic Nanoparticles through Mass-Selected Ion Deposition, Julia Hierarchically Structured Porous Electrodes for Energy Laskin (Pacific Northwest National Laboratory) Storage, M. Tang (Lawrence Livermore National Laboratory)

Designing Solution Self-assembly Routes to Meso-scale Microcatalyst Engineering, C. Chang (National Renewable Materials, Greg Exarhos (Pacific Northwest National Energy Laboratory) Laboratory) Interfacial Control of Polymer Properties, Frances Houle Design of Mesoscale Catalytic Processes, Roger Rousseau (Lawrence Berkeley National Laboratory) (Pacific Northwest National Laboratory) A Novel Substrate-Free Synthesis of Carbon Microtubes, Self-organization of Materials under Irradiation, Igor Hoon T. Chung and Piotr Zelenay (Los Alamos National Veryovkin (Argonne National Laboratory) Laboratory)

Hierarchical Assembly of Confined Nanomaterials into Advanced Characterization of Mesoscopic Materials, Multifunctional Mesoscale Arrays, Igor I. Slowing (Iowa C. Fadley (Lawrence Berkeley National Laboratory) State University) Rapid Evolution of Mesoscale Material Structure by DTEM, Self-assembly of Inorganic Nanoparticles, George Crabtree B. Reed (Lawrence Livermore National Laboratory) (Argonne National Laboratory) Connecting Mesoscale Structure to Macroscopic Material Self and Guided Assembly in Biology, Douglas Tobias Performance, J. Bernier (Lawrence Livermore National (University of California, Irvine) Laboratory)

Programmable Polymer Membrane Composite Meso- High Energy Diffraction Microscopy: Enabling Higher- materials, Gabriel A. Montano (Los Alamos National Order Characterization of Polycrystalline Microstructures, Laboratory) J. Bernier (Lawrence Livermore National Laboratory)

74 From Quanta to the Continuum: Opportunities for Mesoscale Science

Supra-grain Correlations and Materials DNA, M. Kumar Integrating Mesoporosity into Nanoporous Materials, (Lawrence Livermore National Laboratory) J. Biener (Lawrence Livermore National Laboratory)

Microcatalyst Engineering, C. Chang (National Renewable Computational Design and Additive Manufacturing of Energy Laboratory) Mesoscale Architectures for High Performance Materials, C. Spadaccini (Lawrence Livermore National Laboratory) Ionic Liquid at Interfaces and under Confinement, B. Ocko (Brookhaven National Laboratory) Predicting Mesoscale Material Kinetics with Atomistically- based Computation, R. Rudd (Lawrence Livermore Strongly Nonlinear Waves at the Mesoscale: Design of National Laboratory) Metamaterials, E. Herbold (Lawrence Livermore National Laboratory) Meso-scale Investigation of Particle Transport in Porous Media, Y. Kanarska (Lawrence Livermore National Isotopic Probes of Ionic Diffusion and Solid Surfaces, Laboratory) D. J. DePaolo (Lawrence Berkeley National Laboratory) Meso-scale Investigation of Mechanically and Thermally Multi-dimensional Soft X-ray Microscopy of Mesoscale Induced Reactions in Reactive Solid Mixtures and Behavior, Peter Fischer (Center for X-Ray Optics/Lawrence Structures, I. Lomov (Lawrence Livermore National Berkeley National Laboratory) Laboratory) 4D Imaging of Mesoscale Structures, Wah-Keat Lee Application of Mesoscale Simulations toward the (Brookhaven National Laboratory) Development of Predictive Models for Deformation and Dynamic 3D Imaging, Dula Parkinson (Lawrence Berkeley Failure of Heterogeneous Granular Materials, T. Antoun National Laboratory) (Lawrence Livermore National Laboratory)

Mesoscale Plasticity by Synchrotron X-Laue Micro- Theory & Computation for Self-assembly, Chris Mundy diffraction and Reverse Modeling Simulation, Nobumichi (Pacific Northwest National Laboratory) Tamura (Lawrence Berkeley National Laboratory) Computational Design of Electronic Mesoassemblies, X-Ray Studies of Meso Materials, Dennis Mills (Argonne S. Whitelam (Lawrence Berkeley National Laboratory) National Laboratory) Mesoscale Heterogeneous Reactivity: Rare Events and The Potential of High Energy X-ray Diffraction and Surfaces, Andrew Stack (Oak Ridge National Laboratory) Microtomography for the Study of UO under Processing 2 Modeling of Electronic Dynamics and Transport at Multiple and Operating Conditions, Don Brown and Mark Bourke Time- and Length-Scales, Sergei Tretiak (Los Alamos (Los Alamos National Laboratory) National Laboratory)

3D Imaging of Mesoscale Architectures with Nanoscale Atomistic to Mesoscale Modeling of Material Defects Resolution, John Spence (Arizona State University) and Interfaces, Irene J. Beyerlein (Los Alamos National In-situ Monitoring of Dynamic Phenomena during Phase Laboratory) Transformations, Amy Clarke and Jason Cooley (Los Alamos National Laboratory)

DISCLAIMER This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States government nor any agency thereof, nor any of their employees, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise, does not necessarily constitute or imply its endorsement, recommendation, or favor- ing by the United States Government or any agency thereof. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof.

The report is available online at http://science.energy.gov/bes/news-and-resources/reports/basic-research-needs/.

Publishing support services provided by Argonne National Laboratory