As the world becomes increasingly interconnected, modernized, and populated, the demand for energy across the globe is growing at an unprecedented rate. This growth in energy demand has an undeniable impact on increasingly pressing social issues including, climate change, energy security, energy economy, atmospheric chemistry, and air quality. Finding a way to address these issues on a rapid timescale is more important than ever. A common thread running through all of these challenges is that they can be partially or fully addressed with the development of new chemical energy conversion technologies which, in turn, rely on a comprehensive understanding of gas phase kinetics.
Examples of promising technologies include renewable fuels (i.e. methanol and hydrogen) and/or reliable, efficient, and clean engines that can accommodate renewable fuels. The development of such technology would enable the use of renewable fuels, thereby reducing emissions and cutting down on harmful byproducts released into the atmosphere. Computational simulations have become a powerful approach for developing and advancing energy technology in a safe, efficient, and effective manner. These computational approaches model reacting flows and are generally known as computational fluid dynamics (CFD). However, in order for these CFD simulations to work effectively and make meaningful predictions, the sub-models used to describe the underlying chemistry (gas phase kinetics) must be accurate; information about underlying chemistry is provided to computational simulations via a chemical kinetic model/mechanism, which describes the chemical reactions that drive the fuel oxidation within the system being simulated. Regarding combustion specifically, the reliability of predictive simulations depends on the availability of accurate data and models not only for chemical kinetics, but also thermochemistry and transport.
Further complicating the problem, combustion and chemical kinetics provide a unique challenge in regard to obtaining accurate predictive models; underlying chemical kinetics mechanisms may require unprecedented accuracy to obtain truly predictive combustion modeling. For example, it has been shown in computational simulations that uncertainties in any of several kinetic parameters can yield uncertainties large enough in the physical system being modeled to cause system failure, thereby reducing the effectiveness of computational design approaches that could accelerate technology development. Hence, a strong need exists to develop a method that significantly reduces uncertainties in chemical kinetics parameters to meet the accuracy demands of advanced computational design tools. To this end, it is useful to draw on inspiration from existing methods in the field of combustion and chemical kinetics as well as tangential fields; the most compelling inspiration can be found in the field of thermochemistry in the form of the Active Thermochemical Tables (ATcT).
This work presents a novel, analogous approach for chemical kinetics called MultiScale Informatics, or MSI for short. The MSI approach identifies optimized values and quantified uncertainties for a set of molecular parameters (within theoretical kinetics calculations), rate parameters, and physical model parameters (within simulations of experimental observables) as informed by data from various sources and scales. The overarching objectives of this work are to demonstrate how the MSI approach can be used to determine physically meaningful optimized kinetics parameters and quantified uncertainties, unravel webs of interconnected rate constants in complex reaction systems, resolve discrepancies among data sets, and touch on key elements of MSI’s implementation.
To demonstrate how these objectives are met, the MSI approach is used to explore the kinetics of three reaction sub-systems. The studies of these sub-systems will demonstrate some key elements of this approach including: the importance of raw data for quantifying the information content of experimental data, the utility of theoretical kinetics calculations for constraining experimental interpretations and providing an independent data source, and the subtleties of target data selection for avoiding unphysical parameter adjustments to match data affected by structural uncertainties.
For the first sub-system explored (CH₃ + HO₂), the MSI approach is applied to carefully selected (mostly raw) experimental data and yields an opposite temperature dependence for the channel-specific CH3 + HO2 rate constants as compared to a previous rate-parameter optimization. While both optimization studies use the same theoretical calculations to constrain model parameters, only the present optimization, which incorporates theory directly into the model structure, yields results that are consistent with theoretical calculations.
For the second sub-system explored (HO₂ + HO₂), the MSI approach is applied to carefully selected experimental data, leveraging the hydrogen reaction system from the first study with the addition of high level theory calculations for the reaction of HO₂ + HO₂. Recent high-level theoretical calculations predict a mild temperature dependence for HO₂ + HO₂, which is inconsistent with state-of-the-art experimental determinations that upheld the stronger temperature dependence observed in early experiments. Via MSI analysis of the theoretical and experimental data, alternative interpretations of the raw experimental data that uses HO₂ + HO₂ rate constants nearly identical to theoretical predictions are identified – implying that the theoretical and experimental data are actually consistent, at least when considering the raw data from experiments. Similar analyses of typical signals from low-temperature experiments indicate that an HOOOOH intermediate – identified by recent theory but absent from earlier interpretations – yields modest effects that are smaller than, but may have contributed to, the scatter in data among different experiments. More generally, the findings demonstrate that modern chemical theories and experiments have progressed to a point where meaningful comparison requires joint consideration of their data simultaneously.
The third sub-system explored builds a larger web of interconnected reaction systems in an attempt to achieve data redundancy and demonstrate how interpreting coupled reaction systems is necessary to accurately determine many key rate constants. The ability of the MSI method to interpret raw experimental data and untangle rate constant reaction systems is demonstrated. The study also reinforces how implementing theory into the model structure is imperative to yield results that are consistent with experimental data as well as theoretical calculations and achieve physically realistic branching ratios.
Finally, this work will present how results from all the studied reaction systems culminate into a complex hydrogen/syngas combustion model validated against data from various combustion experiments.
| Identifer | oai:union.ndltd.org:columbia.edu/oai:academiccommons.columbia.edu:10.7916/h5qc-fd63 |
| Date | January 2023 |
| Creators | LaGrotta, Carly Elisa |
| Source Sets | Columbia University |
| Language | English |
| Detected Language | English |
| Type | Theses |
Page generated in 0.0021 seconds