Addressing global uncertainty and sensitivity in first-principles based microkinetic models by an adaptive sparse grid approach

Döpking, Sandra; Plaisance, Craig; Strobusch, Daniel; Reuter, Karsten; Scheurer, Christoph; Matera, Sebastian

doi:10.1063/1.5004770

Cited by 32 publications

(34 citation statements)

References 50 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For adsorbates, we assume that the binding energies (and thence heats of formation) are accurate to within ± 0.3 eV. 44,47,51 However, although the global uncertainty in adsorbate heats of formation might be ± 0.3 eV, the relative uncertainties between adsorbates is correlated, which is consistent with the LS concept. 31 As an approximate method for dealing with this correlation, we assume that all adsorbates that bind to the surface through the same element should be perturbed in the same direction.…”

Section: Parametric Uncertainty In Mechanism Generationmentioning

confidence: 67%

“…50 The uncertainty in model parameters should be propagated to the outputs of the model, e.g. conversion, turnover frequency (TOF) 44,45,49,51 and to identify the path with the highest occurring frequencies in a mechanism. [51][52][53] However, given the large uncertainty in model parameters, some pathways or intermediates might have been overlooked because of the very complex landscape of the potential energy surface.…”

mentioning

confidence: 99%

See 1 more Smart Citation

Quantifying the Impact of Parametric Uncertainty on Automatic Mechanism Generation for CO2 Hydrogenation on Ni(111)

Kreitz¹,

Goldsmith²,

West³

et al. 2021

Preprint

View full text Add to dashboard Cite

Automatic mechanism generation is used to determine mechanisms for the CO2 hydrogenation on Ni(111) in a two-stage process, while considering the uncertainty in energetic parameters systematically. In a coarse stage, all the possible chemistry is explored with gas-phase products down to the ppb level, while a refined stage discovers the core methanation submechanism. 5,000 unique mechanisms were generated, which contain minor perturbations in all parameters. Global uncertainty assessment, global sensitivity analysis, and degree of rate control analysis are performed to study the effect 1 of this parametric uncertainty on the microkinetic model predictions. Comparison of the model predictions with experimental data on a Ni/SiO2 catalyst find a feasible set of microkinetic mechanisms that are in quantitative agreement with the measured data, without relying on explicit parameter optimization. Global uncertainty and sensitivity analyses provide tools to determine the pathways and key factors that control the methanation activity within the parameter space. Together, these methods reveal that the degree of rate control approach can be misleading if parametric uncertainty is not considered. The procedure of considering uncertainties in the automated mechanism generation is not unique to CO2 methanation and can be easily extended to other challenging heterogeneously catalyzed reactions<br>

show abstract

Section: Parametric Uncertainty In Mechanism Generationmentioning

confidence: 67%

mentioning

confidence: 99%

Quantifying the Impact of Parametric Uncertainty on Automatic Mechanism Generation for CO2 Hydrogenation on Ni(111)

Kreitz¹,

Goldsmith²,

West³

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…In order to investigate the impact of uncertainty in the free energies arising from uncertainty in the electronic energies computed with DFT, 58 and in the entropic contributions approximated with the rigid-rotor harmonic-oscillator approximation, 38 an ensemble of 100 noisy samples was generated by drawing one random number from a Gaussian distribution with mean zero and standard deviation 2 kcal mol −1 , and adding this to the relative free energy of each intermediate state. Transition states were modified by adding a uniformly distributed random number times this noise term as a scaling-relation-type correlation between the states.…”

Section: Quantification Of First-principles Uncertaintymentioning

confidence: 99%

First-principles-informed energy span and microkinetic analysis of ethanol catalytic conversion to 1,3-butadiene on MgO

Boje¹,

Taifan²,

Ström³

et al. 2021

Preprint

View full text Add to dashboard Cite

Kinetic modeling of single-step catalytic conversion of ethanol to 1,3-butadiene is necessary to inform accurate process design. This paper uses first-principles-informed energy span and microkinetic analysis to explore the reaction free energy landscapes and kinetic limitations of competing reaction pathways on a MgO (100) step-edge. Previous studies suggested mechanisms proceeding via both dehydrogenation and dehydration of ethanol, and highlighted sensitivity to conditions and catalyst composition. Here, we use the energy span concept to characterize the theoretical maximum turnover and degree of turnover frequency control for states in each reaction pathway, finding the dehydration route to be less active for 1,3-butadiene, and suggesting rate-determining states in the dehydrogenation, dehydration, and condensation steps. The influence of temperature on the relative rate contribution of each state is quantified and explained through the varying temperature sensitivity of the free energy landscape. A microkinetic model is developed to explore competition between pathways, interaction with gas-phase species, and surface coverage limitations. This suggests that the turnover may be significantly lower than predicted solely based on energetics. Turnover frequency determining states found to have high surface coverage include adsorbed ethanol and two longer, oxygenated hydrocarbons. The combined energy span and microkinetic analysis permits investigation of a complex system from two perspectives and helps elucidate conflicting observations of rate determining steps and product distribution by considering both energetic and kinetic limitations. The impact of uncertainty in the energy landscape is quantified using a correlated error model. While the range of predictions is large, the average performance and trends are similar.

show abstract

“…Mean-field microkinetic models-developed by combining electronic structure properties with macroscopic reaction parameters, such as reaction temperature and pressure 1,2 -are used to obtain fundamental insights into the reaction kinetics occurring on the solid/gas interfaces. However, the success of such physics based models is critically dependent upon reliable estimate of the adsorption energetics of various elementary reactions [3][4][5][6] . In the last decade, improvement in exchange-correlation functionals made it possible to estimate adsorption energies with high fidelity, which enabled computational catalysis modeling a surrogate scheme to replace time-consuming experimental methods 7,8 .…”

Section: Introductionmentioning

confidence: 99%

A Bayesian framework for adsorption energy prediction on bimetallic alloy catalysts

et al. 2020

View full text Add to dashboard Cite

For high-throughput screening of materials for heterogeneous catalysis, scaling relations provides an efficient scheme to estimate the chemisorption energies of hydrogenated species. However, conditioning on a single descriptor ignores the model uncertainty and leads to suboptimal prediction of the chemisorption energy. In this article, we extend the single descriptor linear scaling relation to a multi-descriptor linear regression models to leverage the correlation between adsorption energy of any two pair of adsorbates. With a large dataset, we use Bayesian Information Criteria (BIC) as the model evidence to select the best linear regression model. Furthermore, Gaussian Process Regression (GPR) based on the meaningful convolution of physical properties of the metal-adsorbate complex can be used to predict the baseline residual of the selected model. This integrated Bayesian model selection and Gaussian process regression, dubbed as residual learning, can achieve performance comparable to standard DFT error (0.1 eV) for most adsorbate system. For sparse and small datasets, we propose an ad hoc Bayesian Model Averaging (BMA) approach to make a robust prediction. With this Bayesian framework, we significantly reduce the model uncertainty and improve the prediction accuracy. The possibilities of the framework for high-throughput catalytic materials exploration in a realistic setting is illustrated using large and small sets of both dense and sparse simulated dataset generated from a public database of bimetallic alloys available in Catalysis-Hub.org.

show abstract

Addressing global uncertainty and sensitivity in first-principles based microkinetic models by an adaptive sparse grid approach

Cited by 32 publications

References 50 publications

Quantifying the Impact of Parametric Uncertainty on Automatic Mechanism Generation for CO2 Hydrogenation on Ni(111)

Quantifying the Impact of Parametric Uncertainty on Automatic Mechanism Generation for CO2 Hydrogenation on Ni(111)

First-principles-informed energy span and microkinetic analysis of ethanol catalytic conversion to 1,3-butadiene on MgO

A Bayesian framework for adsorption energy prediction on bimetallic alloy catalysts

Contact Info

Product

Resources

About