However, the various uses of water are managed through separate processes, and the impact of management objectives for one can result in sub-optimal practices for the other, and will be exacerbated with predictions of greater year-to-year climate variability. Without a coordinated analysis capability, the ability to predict the effectiveness of climate mitigation, adaptation measures, or setting the value of water and energy is severely limited. In this LDRD, we will develop a computation tool and analysis framework for linked climate-water-energy co-simulation. The LDRD’s resulting research will lay the foundation for an overall regional-scale integrated assessment capability. We will develop analysis tools and software to estimate the cost of consuming water to produce energy, and the cost of consuming energy to produce water at regional spatial scales, and decade and multi-decade temporal scales, develop analytical tools to specify the performance requirements of climate models for the aforementioned water-energy capability, develop uncertainty analysis algorithms to map the trade space between model unknowns , and demonstrate the resulting tools and software by analyzing the effects of climate uncertainty on water-energy management for the American River basin and Sacramento urban region of California. Direct chemical imaging of elemental content and impurities with extreme spatial and depth resolution and specificity is required to understand,stacking pots predict and minimize processes that adversely affect the macro-scale properties of solar and other energy systems. A fundamental lack of key analytical techniques capable of providing this information leaves a pressing need for the development of next-generation nanoscale chemical imaging tools.
The objective of this project is to develop a novel ultrafast laser spectroscopy technique based on a two near-field nanoprobe scheme which will overcome current limitations and meet the requirements of a versatile chemical imaging system for detecting and chemically mapping defects in solar energy systems and other energy materials. This project aims to develop a sensitive femto second laser chemical imaging system in which both material excitation and signal detection occurs in the optical near-field vicinity. This chemical imaging system will enable a fundamental understanding of the properties and functionality of new solar material systems at spatio-temporal scales that were previously unattainable. In the second year of the project, both ultraviolet and visible femtosecond laser pulses were coupled to the near-field excitation probe to obtain chemical signatures of different material systems including nanoparticles, crystalline, and amorphous materials. We demonstrated near field visible-range fluorescence originating from ultraviolet femtosecond laser excitation in the optical near-field. Second order diffraction was also observed in the same spectral range, enabling simultaneous femtosecond Rayleigh and femtosecond laser-induced fluorescence signal detection in the near field vicinity with the dual probe near-field system. We further optimized the near-field excitation and detection processes as a way to improve sensitivity and resolution, and compared the signals from near-field excitation/far-field detection to near-field excitation/near-field detection signals from the same material system . Significant improvements in the signal-to-noise ratio were observed in the near-field/near-field configuration, despite the significantly smaller size of excited surface area. Finally, the potential of generating surface plasmon polaritons from a “femtosecond-laser point source” was explored in the near-field/near-field configuration at a Au/glass interface, and the signal intensity was studied as a function of inter-probe distance using visible femtosecond laser irradiation.
These results underline the importance of detecting near-field signals in the near-field vicinity as a way to achieve high sensitivity, high resolution chemical imaging at small spatio-temporal scales. The purpose of this research is to build and apply to test problems a computational platform for the design, retrofit and operation of urban energy grids that include electrical systems, district heating and cooling systems, and centralized and distributed energy storage. The need for this research arises because an integration of renewable energy beyond 30% poses dynamic challenges on the generation, storage and transmission of energy that are not well understood. Such a platform is also needed to assess economic benefits for the integration of co-generation plants that generate combined heating, cooling and power at the district level in order to decrease the carbon footprint of energy generation. To address this need, this project will create a flexible computational R&D platform that allows expanding energy and policy analysis from buildings to district energy systems. Questions that this platform enables to address include where to place energy generation and storage, how to set the price structure, how to trade-off incentives for energy-efficiency versus incentives to add generation or storage capacity at buildings, how to integrate waste heat utilization to reduce the carbon footprint of district energy systems and how to upgrade the electricity grid to integrate an increasing fraction of renewable energy while ensuring grid reliability and power quality. Significant accomplishments have been made in the development of multi-physics models that describe the interaction between buildings and the electrical grid. Regarding multi-physics modeling, we completed the development of more than fifty models for analyzing buildings-to-electrical grid integration. The models are now part of the Modelica Buildings library, an equation-based object-oriented library for modeling of dynamic building energy systems.
The models can represent DC and AC systems under different assumptions such as quasi-stationary or dynamic-phasorial representation. The electrical models can be connected to thermal models of buildings in order to evaluate the impact of electrical and thermal storages, as well as of building controls, on the distribution grid. The models have been validated against standard IEEE procedures defined for testing the correctness of electrical network simulation software. The models, the results of the validation and few examples showing the ability to perform building-to-grid simulation studies were presented at the 2014 BauSIM conference in Aachen . The paper, titled “A Modelica package for building-to-electrical grid integration” won the best paper award.It allows to graphically connect components of cyber-physical systems that advance in time based on continuous time dynamics, discrete time dynamics, or event-driven dynamics, in order to study building-to-grid integration. These languages also allow accessing the mathematical structure of the entire model. Such information has been used for co-simulation and for solving optimal control problems. For example, we demonstrated how simulation models can be reused to solve optimal control problems by means of computer algebra and numerical methods. The problem investigated was to determine the optimal charge profile of a battery in a small district with multiple buildings and photovoltaic systems that minimizes energy subject to voltage constraints. The increasing availability of complete genomic sequences and whole-genome analysis tools has moved the construction of industrial hosts towards rational design by metabolic engineering and systems biology. The current genetic manipulation tool kits available for industrial hosts, however,grow lights are desperately sparse and unpolished in comparison to the array of tools available for E. coli. The goal of this project is to develop a high throughput genome editing tool to facilitate the engineering of novel applications not only in E. coli, but in under exploited industrial producers such as Streptomyces coelicolor and Corynebacterium glutamicum. The original goal of this proposal was to create a secure industrial bacterium by converting all 484 TGA termination codons to TAA in the C. glutamicum genome and then reassigning TGA to encode an unnatural amino acid. In our phase I work, we discovered that the recombineering approach alone could not achieve the frequency of allelic replacement needed to complete codon depletion in a reasonable time frame. We concluded that a more efficient genome editing tool would be needed for this project. Recent work on the Clustered Regularly Inter spaced Short Palindromic Repeat adaptive immune system of prokaryotes has led to the identification of a DNA endonuclease called Cas9 whose target sequence specificity is programmed by small spacer RNAs in the CRISPR loci. By editing spacer sequences we can direct Cas9 to cut endogenous DNA targets, thereby forcing cells to repair themselves in a predictably mutagenic manner. Such Cas9 mediated cleavage in vivo is more efficient, effective, and potentially multi-plexable than any other tools available for genomic engineering. Our most significant accomplishment has been to develop a reproducible and efficient protocol for engineering E. coli DNA in vivo. Our method uses the Streptococcus pyogenes CRISPR-Cas9 system in combination with λRed recombineering proteins in E. coli. We have created a mobile plasmid with both Cas9 and λRed activities and used it successfully in performing genome editing in all E. coli strains in hand. This protocol has been successfully used to modify gene loci in living E. coli cells within a 3 weeks time frame. The developed Cas9 toolkit and protocol have already been used in several bio-energy research projects.
We have also received requests and started disseminating the toolkit and protocol to general scientific community. We have also succeeded in developing informatics tools to aid in the design of CRISPR spacer constructs given a targeted range of genomic sequences. This tool would be handy in the design of Cas9 genome editing at scale. As we had predicted, our approach provides a significantly faster turnaround time to modify genetic codes than any available tools. We are hopeful that this method will be generally applicable to non-E. coli hosts, which will greatly aid our future goal of modifying genetic codes of industrial microbes. The purpose of this project is to develop sensitive and selective biosensors for a diverse set of target chemicals as a way to provide a high-throughput functional screening method for molecule production in microbial cells. Advances in DNA synthesis and combinatorial DNA assembly allow for the construction of thousands of pathway variants by varying both the gene content as well as the expression levels of the pathway components, a technique commonly referred to as pathway refactoring. However, a lack of sufficiently sensitive, selective, and scalable technologies to measure chemical production presents a major bottleneck that limits our ability to fully exploit large-scale synthesis efforts. We will develop and deploy novel biosensors systems based on both protein and RNA molecules that have been previously shown to respond to the presence of small molecule ligands. In the case of protein-based sensors, we will use synthetic biology approaches to modify the ligand specificity of a known transcription factor . We will screen for ligand-dependent TF function by placing TF binding sites in front of GFP, such that GFP activation should only be observed in the presence of a ligand. We will test the affinity and response of the TF mutant library to a variety of relevant ligands by using several rounds of selection using fluorescence activated cell-sorting . Samples collected after each round of selection will be sequenced using next-generation sequencing methods and we will seek to understand the relationship between TF ligand affinity and sequence evolution, as this will facilitate more rational engineering approaches. In the case of the nucleotide sensor, we will develop a system in which cell survival is linked to ligand production by coupling the switch to a chemical selection system used during cell growth. We will then deploy this system to screen a library of 20K pathway variants to select and further characterize high molecule producing E.coli strains. Selected strains will be sequenced and we will use modeling approaches to identify the key variables and bottlenecks associated to molecule production. Over the course of this LDRD funding, we have successfully developed proof of principle for an end-to-end system to screen for gene regulatory sequences in an unbiased manner. This work has been published in Nature Methods, and an additional small project resulting from this work has been reported in Biology Open. Briefly, we have shown that we can clone hundreds to thousands of random sequences into a precise location in the mouse genome that is linked to a reporter gene, which is activated when sequences are behaving as enhancers. The targeted cells can be flow sorted to isolate those cells that are actively expressing the reporter gene, and the sequences responsible for this reporter expression can be identified through DNA sequencing. To date, we have used this method to test the embryonic stem cell enhancer activity of more than 0.5Megabases of mouse or human genomic sequence in 1kilobase increments. To apply this method to a broader range of cell types, a major aim of this proposal, we have coupled the ES cell reporter assays we developed with in vitro differentiation and showed that we can accurately identify enhancers active in cardiac and neuronal cell populations.