Super-Seed 1: Nanoscale Additive Manufacturing with Programmed Interfaces
Faculty (* = coordinators): G. Engel, G. Galli, P. Guyot-Sionnest, J. Park*, S. Sibener, D. Talapin*
Scientific and manufacturing breakthroughs make revolutionary societal and technological impacts as evidenced by the development of integrated circuits based on silicon. The ever- increasing size of single crystalline silicon wafers and the ability to deposit, pattern, and process various thin films with ever-improving precision became the foundation for the electronic and digital revolution in the 20th century. This is an example of additive manufacturing: it is based on robust materials platform whereby materials are “added” through a variety of means in a “patterned” fashion to control the information or functions. Realizing additive manufacturing at the nanoscale—nanoscale additive manufacturing—, however, is a goal yet to be realized. In order to address this challenge, our seed team aims to develop the universal nanomanufacturing platforms that combine directed assembly, placement and interfacial coupling of functional nanoscale building blocks. With this goal in mind, we explore vertically-programmed nanoscale interfaces that are generated by adding atomically-engineered nanoscale building blocks with precision and scale.
SuperSeed2: Harnessing training and memory to evolve material functionality
(UofC: Aaron Dinner, Margaret Gardel, *Arvind Murugan, Sidney Nagel, Stephanie Palmer; NWU: MichelleDriscoll
Aging as a tool: While it is a well-known and indisputable fact that materials age and deform over time, this common behavior has not been developed into an effective tool to create novel desired functionality in matter. When a material retains a memory of how it was previously manipulated, this can act as a training protocol for enhancing desired behaviors not easily found otherwise. We will present a few examples of how “directed aging” – in which Nature uses a “greedy algorithm” to evolve in response to prior stress history–leadstospecialfunctionalities. Weproposetoexploittheretentionofsuchadvantageousbehavior as a general training paradigm for creating novel material function.
Evolution from biology to materials – illuminating the rules of life: The broad scope of this proposal has its roots in both material science and in lessons learned from the evolution of biological function. Natural selection is a powerful update algorithm for living systems and some of these ideas have made their way into computer science in the form of evolutionary algorithms. Here we take another page from this book and use ideas about how biological material adapts by feeding back memories of its previous environment to determine its future evolution. As a result, our approach borrows ideas from biology while also illuminating the rules of life by providing an alternative platform to study similar conceptual questions.
Genotype-to-phenotype map for materials: Our approach relates materials design to a metaphor at the heart of biology: the genotype-to-phenotype map. As in biology, this map for materials will be complex andrugged,stronglyaffectingtheabilityofasystemtolearn. Forexample,theruggednessofthegenotype- to-phenotype map dictates whether a training method can be completely greedy or whether it needs to allow forsomedeleteriouschangesinhopeofsubsequentlargergains. We ask the central questions: How general is this framework and what material-specific functions can be learned?
Nature of trained systems: How are trained materials different from designed ones? The analogous question about evolved systems in biology has revealed a diverse array of answers that shed light on the limitations and strengths of our training paradigm. For example, biophysical systems are replete with “evolutionary spandrels”, features with no direct benefit but reflect their evolutionary development. On the other hand, proteins and neural networks evolved in changing environments are adaptable and can switch between functions in a way that designed systems rarely can. A fundamental goal of this Seed is to understand the scope and limits of our training paradigm for materials.
Seed 3: Transport Properties of Molecularly Doped Semiconducting Polymers for Thermoelectrics (Shrayesh Patel & Giulia Galli)
A tremendous amount of focus has been placed on the development of optoelectronic polymers for organic electronic devices such as solar cells, field-effect transistors, and light emitting diodes. This focus has led to processable and stable n- and p- type conjugated semiconducting polymers that exhibit charge carrier mobilities approaching or exceeding the value of amorphous silicon.
Along with improved processability and morphology control, the high carrier mobility is due to resiliency to structural and electronic disorder inherent in polymers. Leveraging these high performance semiconducting polymers towards new applications while still advancing the fundamental understanding and molecular design of (opto)electronic polymers is paramount. One particular emerging application relates to organic thermoelectrics – devices that interconvert heat and electricity. Many efficient thermoelectric materials contain elements with poor abundance, e.g. Bi2Te3. In contrast, polymeric semiconductors have potential transformative advantages for cooling applications and thermal energy harvesting of low-grade heat (< 200 °C). First, their thermal conductivity is low relative to inorganic materials (factor 10 lower) while maintaining high electronic conductivity (e.g. ~10 to 1000 S/cm). Importantly, they can be coated over very large areas, or molded into bulk structures readily, and are relatively inexpensive and lightweight.
Overall, the investigation in to the thermoelectric properties of semiconducting polymers is in its early stages and there is little known about how processing and electrical doping influences the relationship between electrical conductivity and Seebeck coefficient (thermopower). In most inorganic semiconductors, the Seebeck coefficient (a) typically decreases with increasing electronic conductivity (s). However, for polymeric materials, the carrier mobility and, in turn, the electronic conductivity can be a complex function of the morphology, thus making it difficult to form such simple connections. To fully realize the potential of organic thermoelectrics, there is a need of advancing the fundamental understanding of structure-property- processing relationships of doped semiconducting polymers and developing detailed charge transport models. Thus, this work leverages an important integrated theoretical/experimental approach to address these challenges.