Search for: All records

Hydrothermal liquefaction (HTL) has remarkable potential for efficient conversion of abundant, decentralized organic wastes into renewable fuels. Because waste is a highly distributed resource with context-dependent economic viability, selection of optimal deployment sites is slowed by the need to develop detailed techno-economic analyses (TEA) for the thousands of potential deployment locations, each with their own unique combinates of scale, proximity to infrastructure/markets, and feedstock properties. An economic modeling framework that requires only easily obtainable inputs for assessing economic performance would therefore allow multiplexed analysis of many thousands of cases, whereas traditional TEA would not be possible for more than a handful of cases. Within such a context, the present study uses machine learning to guide development of a TEA and modeling framework which provides accurate cost predictions using three key inputs – feedstock cost, biocrude yield, and process scale – to estimate the minimum fuel selling price (MFSP) that an HTL process can achieve. The structure of the proposed framework is informed and based on empirical observations of cost projections made by a detailed TEA over a wide range of feedstock costs, biocrude yields, and process scales. A machine learning guided process was used to identify, train, and test a series of models using auto-generated data for training and independently reported data for testing. The most accurate model consists of three terms and requires 6 adjustable parameters to predict independently published values of MFSP (N = 28) to within an average value of ± 20.4%. It is demonstrated that the reduced-order model’s predictions fall within 40% of the corresponding published values 95% of the time, and in the worst case, the associated discrepancy is 45.9%, suggesting that the accuracy of the machine learned model is indeed comparable to the TEAs that were used to build it. Moreover, the terms in the model are physically interpretable, conferring greater reliability to the use of its predictions. The model can be used to predict the dependence of MSFP on biocrude yield, scale, and feedstock cost; interestingly, MFSP is insensitive to biocrude yield and/or scale under many situations of interest and identifying the critical value for a given application is crucial to optimizing economic performance. The proposed model can be also extended to evaluate economic performance of newly developed HTL-based processes, including catalytic HTL, and the methodological framework used in this study is deemed appropriate for the development of machine learned TEA models in cases of other similar waste-to-energy technologies. 

Current solutions to global challenges place tension between global benefits and local impacts. The result is increasing opposition to implementation of beneficial climate policies. Prioritizing investment in projects with tangible local benefits that also contribute to global climate change can resolve this tension and make local communities’ partners instead of antagonists to change; the approach advocated is a new take on “thinking globally, acting locally”. This approach is a departure from the usual strategy of focusing resources on solutions perceived to have the largest potential global impact, without regards to local concerns. Reclamation of polluted mine sites by using fast growing bamboo to remove heavy metals provides a case study to show what is possible. Effective implementation of thinking globally while acting locally will require increased coordination between different types of researchers, new educational models, and greater stakeholder participation in problem identification and solution development. 

Less than 5% of polystyrene is recycled, motivating a search for energy efficient and economical methods for polystyrene recycling that can be deployed at scale. One option is chemical recycling, consisting of thermal depolymerization and purification to produce monomer-grade styrene (>99%) and other co-products. Thermal depolymerization and distillation are readily scalable, well-established technologies; however, to be considered practical, they must be thermodynamically efficient, economically feasible, and environmentally responsible. Accordingly, mass and energy balances of a pyrolysis reactor for thermal depolymerization and two distillation columns to separate styrene from α-methyl styrene, styrene dimer, toluene, and ethyl benzene co-products, were simulated using ASPEN to evaluate thermodynamic and economic feasibility. These simulations indicate that monomer-grade styrene can be recovered with energy inputs <10MJ/kg, comparable to the energy content of pyrolysis co-products. Thermodynamic sensitivity analysis indicates the scope to reduce these values and enhance the robustness of the predictions. A probabilistic economic analysis of multiple scenarios combined with detailed sensitivity analysis indicates that the cost for recycled styrene is approximately twice the historical market value of fossil-derived styrene when styrene costs are fixed at 15% of the total product cost or less than the historical value when feedstock costs are assumed to be zero. A Monte Carlo and Net Present Value-based economic performance analysis indicates that chemical recycling is economically viable for scenarios assuming realistic feedstock costs. Furthermore, the CO2 abatement cost is roughly $1.5 per ton of averted CO2, relative to a pyrolysis process system to produce fuels. As much as 60% of all polystyrene used today could be replaced by chemically recycled styrene, thus quantifying the potential benefits of this readily scalable approach. 

Abstract MF-LOGP, a new method for determining a single component octanol–water partition coefficients ($$LogP$$ $\mathrm{LogP}$) is presented which uses molecular formula as the only input. Octanol–water partition coefficients are useful in many applications, ranging from environmental fate and drug delivery. Currently, partition coefficients are either experimentally measured or predicted as a function of structural fragments, topological descriptors, or thermodynamic properties known or calculated from precise molecular structures. The MF-LOGP method presented here differs from classical methods as it does not require any structural information and uses molecular formula as the sole model input. MF-LOGP is therefore useful for situations in which the structure is unknown or where the use of a low dimensional, easily automatable, and computationally inexpensive calculations is required. MF-LOGP is a random forest algorithm that is trained and tested on 15,377 data points, using 10 features derived from the molecular formula to make$$LogP$$ $\mathrm{LogP}$predictions. Using an independent validation set of 2713 data points, MF-LOGP was found to have an average$$RMSE$$ $\mathrm{RMSE}$= 0.77 ± 0.007,$$MAE$$ $\mathrm{MAE}$= 0.52 ± 0.003, and$${R}^{2}$$ $R^2$= 0.83 ± 0.003. This performance fell within the spectrum of performances reported in the published literature for conventional higher dimensional models ($$RMSE$$ $\mathrm{RMSE}$= 0.42–1.54,$$MAE$$ $\mathrm{MAE}$= 0.09–1.07, and$${R}^{2}$$ $R^2$= 0.32–0.95). Compared with existing models, MF-LOGP requires a maximum of ten features and no structural information, thereby providing a practical and yet predictive tool. The development of MF-LOGP provides the groundwork for development of more physical prediction models leveraging big data analytical methods or complex multicomponent mixtures. Graphical Abstract 

In this study, four geopolymer sorbents GP0, GP10, GP30 and GP50 were synthesized using volcanic ash (VA) and metakaolin (MK) blends as precursors with 0, 10, 30 and 50% MK content by mass, respectively. The materials were characterized by X-ray fuorescence (XRF), X-ray difraction (XRD), Raman spectroscopy, and Brunauer–Emmett–Teller (BET) surface area analyses, revealing successful geopolymerization of the precursors and increasing surface area with increasing MK content. The sorption performance of the VA, MK and VA-MK geopolymers was then evaluated for the removal of cationic methylene blue (MB) dye from aqueous media. Sorption capacity was independent of composition, providing fexibility in sorbent synthesis. Sorption rate, on the other hand, was 3–8 times greater for the VA-MK geopolymers than the precursor materials. The equilibrium adsorption data were suitably explained by the Freundlich model, denoting multilayer adsorption onto a heterogeneous adsorption surface with higher Freundlich afnity constant (KF) for geopolymers than VA. The adsorption kinetics obeyed the pseudo-second-order (PSO) kinetic law with an average of 98% removal efciency in 30 min. MB uptake was pH-dependent and driven by electrostatic chemisorption interactions. These results motivate further studies on the use of locally sourced geopolymers for water purifcation applications. 

Regression ensembles consisting of a collection of base regression models are often used to improve the estimation/prediction performance of a single regression model. It has been shown that the individual accuracy of the base models and the ensemble diversity are the two key factors affecting the performance of an ensemble. In this paper, we derive a theory for regression ensembles that illustrates the subtle trade-off between individual accuracy and ensemble diversity from the perspective of statistical correlations. Then, inspired by our derived theory, we further propose a novel loss function and a training algorithm for deep learning regression ensembles. We then demonstrate the advantage of our training approach over standard regression ensemble methods including random forest and gradient boosting regressors with both benchmark regression problems and chemical sensor problems involving analysis of Raman spectroscopy. Our key contribution is that our loss function and training algorithm is able to manage diversity explicitly in an ensemble, rather than merely allowing diversity to occur by happenstance. 

Mechanical decrystallization and water-promoted recrystallization of cellulose were studied to understand the effects of cellulose crystallinity on reaction engineering models of its acid-catalyzed hydrolysis. Microcrystalline cellulose was ball-milled for different periods of time, which decreased its crystallinity and increased the glucose yield obtained from acid hydrolysis treatment. Crystallinity increased after acid hydrolysis treatment, which has previously been explained in terms of rapid hydrolysis of amorphous cellulose, despite conflicting evidence of solvent promoted recrystallization. To elucidate the mechanism, decrystallized samples were subjected to various non-hydrolyzing treatments involving water exposure. Interestingly, all non-hydrolyzing hydrothermal treatments resulted in recovery of crystallinity, including a treatment consisting of heat-up and quenching that was selected as a way to estimate the crystallinity at the onset of hydrolysis. Therefore, the proposed mechanism involving rapid hydrolysis of amorphous cellulose must be incomplete, since the recrystallization rate of amorphous cellulose is greater than the hydrolysis rate. Several techniques (solid-state nuclear magnetic resonance, X-ray diffraction, and Raman spectroscopy) were used to establish that water contact promotes conversion of amorphous cellulose to a mixture of crystalline cellulose I and cellulose II. Crystallite size may also be reduced by the decrystallization-recrystallization treatment. Ethanolysis was used to confirm that the reactivity of the cellulose I/cellulose II mixture is distinct from that of truly amorphous cellulose. These results strongly point to a revised, more realistic model of hydrolysis of mechanically decrystallized cellulose, involving recrystallization and hydrolysis of the cellulose I/cellulose II mixture.