The Delplot analysis for reaction network elucidation involves the use of measured species and their yields as a function of reactant species. Highly reactive intermediates, such as free radical and other mechanistic intermediates, however, will often have large reaction rate constants relative to other species. This can cause the product classification by the Delplot technique, which uses only observable species, to be confounded. The present work addressed this issue, where the limitation of the use of the Delplot technique to classify the fleeting intermediate species was explored. To achieve this aim, the series reaction mechanism given by A ->(k1) B ->(k2) C was used to synthesize the kinetic data at various values of reaction rate constant ratios lambda = k(2)/k(1). At low lambda, it was found that kinetic data can be collected through the reaction time and the product ranks are as indicated by the mechanism. At high lambda, the Delplot analysis will disguise the reaction mechanism if the kinetic data are not collected at ultralow conversion. The present results also suggest that, during parameter estimation for the determination of kinetic parameters, no more than 2 orders of magnitude should separate the initial guesses for the reaction rate constants if all of the species in the network are observable.

This chapter describes the reformulation of both the description of the Monte Carlo reaction system and its states in order to model composition-dependent transition parameters and address chemical kinetic coupling. It utilizes a simple prototype kinetically coupled reaction model to verify the accuracy of a series of different computational algorithms aimed at handling molecular interactions through the redefinition of system state and state space. The limitations of the full-memory approach motivated the development of an iterative self-consistent concentration approach, somewhat akin to the self-consistent field methods in quantum mechanical computations. Development of a general mathematical framework for describing both molecule–molecule and molecule–catalyst kinetic interactions would therefore be quite useful in the computational design of catalytic hydrocarbon conversion systems. The Monte Carlo simulation reaction modeling approach chronicles the discrete transformations of an N-component system throughout time. The composition profile for the diaromatics exhibits one of the fastest initial rates. The rate, however, quickly attains an equilibrium-like limit.

A molecular-level kinetic model was developed for triglyceride hydroprocessing. Triglyceride molecules were defined on the basis of the 8-22 carbon fatty acids commonly present in renewable diesel feeds. A reaction network detailing the hydrodeoxygenation, decarboxylation, and decarbonylation parallel pathways of the triglyceride chemistry was constructed. The final network contained 476 species and 1709 reactions. The network was used to build a kinetic model based on experimental data for coconut and soybean oil hydroprocessing at various temperatures, pressures, and catalyst contact times. Parameter optimization for the kinetic parameters was performed for two different catalysts. The final kinetic model provided good agreement with experimental results. Diesel cetane number and cloud point property models were also constructed and optimized on the basis of experimental data. These property models were used to study the product diesel cetane number versus cloud point trade-off to determine the end-use properties of the product fuel.

Strategies to reduce the computer time to access the information in molecular-level kinetic models (MLKMs) were evaluated. A triglyceride hydroprocessing MLKM was used to generate data sets for small ranges of input parameters simulating three output parameters. The data sets were used to generate multilinear regression, polynomial regression, decision tree regression, gradient boosting regression, and artificial neural network data-driven model (DDM) representations of the MLKM. All of the DDMs were able to predict results very quickly (<< 1 s). The predictive accuracy for the DDMs was compared to the polynomial regression, gradient boosting regression, and artificial neural network models, providing the best models over the entire range of the input parameters selected. However, in narrow input parameter ranges, multiple multilinear models and decision tree models also provide good accuracy, with the added benefit of easily understood parameters and faster solution times. Additionally, multilinear regression models had much lower data requirements than the decision tree regression and artificial neural network models. The major downside to all of the DDMs was shown to be the great loss in accuracy once the input parameters exceed the range of the input parameters in the data sets used to optimize the DDMs. This suggests that the extrapolation capability of the DDMs is very low, and as such, new data should be generated from the MLKM every time predictions are required outside the range of the underlying DDM data.

The use of the Delplot analysis to interpret the experimental results from the parallel-series reaction network A→k1B→k3C; A→k2D under the influence of reaction reversibility was examined. A reactor model was used to test two experimental feeds, one with zero product species concentrations and the other with non-zero product species concentrations. The equilibrium constants K1, K2, and K3 were varied between 0.2 and 1.5 for each reaction in the network. When CA0 = 2.0 M and CB0 = CC0 = CD0 = 0 M, it was found that reversibility did not change the analysis provided by the first-rank Delplots; i.e., species “B” and “D” are primary, while species “C” is non-primary. When CA0 = 2.0 M and CB0 = CC0 = CD0 = 0.5 M, it was found that the effect of the reversibility altered the information provided by Delplot analysis. For instance, the first-rank Delplot classified species “B” as a primary product when K1 ≥ 0.5 but non-primary when K1 = 0.2. A similar conclusion was also obtained for species D; i.e., species “D” ...

The reaction network for the oxidative dehydrogenation of n-butane to butadiene was examined. Delplots for experiments with n-butane, 1-butene, and 2-butene feeds were constructed. These analyses revealed that butadiene formed from both 1-butene and 2-butene. The experimental ratio of 2-butene and 1-butene was far from the equilibrium value, suggesting that the 2-butene to 1-butene equilibrium was not established, supporting the existence of the direct reaction of 2-butene to butadiene.

The kinetics of the hydrolysis of heavy hydrocarbons in supercritical water were probed using density function theory (DFT) and molecular dynamics (MD) simulation of the probe molecule dibenzyl eth...

The upgrading of light naphtha (C-5-C-6 stream) to gasoline blending components has been the subject of intensive research at both academic and industrial laboratories. The combination of high volatility and low-octane number has made this stream surplus at many refineries worldwide. This review presents the latest developments in selected catalytic upgrading processes and a brief discussion on the reaction mechanism and reactor models. A majority of the review falls within the development of catalysts for n-hexane isomerization to hydrocarbon isomers with a high octane number. There are three types of isomerization catalysts that include Pt/Al2O3-Cl, Pt/SO4-ZrO2, and Pt/zeolite. Efforts are ongoing to improve the catalyst performance for higher selectivity and catalyst lifetime. Very little work has been published on the conversion of n-pentane mainly as a result of its low activity and the limited options available for its transformation to gasoline blending components. Other approaches discussed in the review include dimerization and oligomerization of C-5-C-6 alkenes and methylative homologation. The review covers literature published during the period of 2000-2018.

Accurate forecast of the hourly spot price of electricity plays a vital role in energy trading decisions. However, due to the complex nature of the power system, coupled with the involvement of multi-variable, the spot prices are volatile and often difficult to forecast. Traditional statistical models have limitations in improving forecasting accuracies and reliably quantifying the spot electricity price under uncertain market conditions. This paper presents a hybrid model that combines the results from multiple linear regression (MLR) model with an auto-regressive integrated moving average (ARIMA) and Holt-Winters models for better forecasts. The proposed method is tested for the Iberian electricity market data set by forecasting the hourly day-ahead spot price with dataset duration of 7, 14, 30, 90, and 180 days. The results indicate that the hybrid model outperforms the benchmark models and offers promising results under most of the testing scenarios.

A molecular-level kinetic model for the hydroprocessing of methyl laurate was constructed. The reaction network was deduced using experimental observations in the context of the delplot method for the discernment of product rank. The resulting 45 species and 83 reactions were used to construct the set of material balances in the kinetic model. Kinetic parameters of the model were determined by minimizing the difference between model outputs and experimental data for methyl laurate hydroprocessing. Differences in reactivity as a result of catalyst metal composition were modeled via the catalyst family concept. The model results show good agreement with the experimental results for a range of process conditions.

The Delplot technique for the classification of experimentally derived reaction products as primary, secondary, tertiary, etc. was extended to address experiments in which reaction products are present in non-zero concentrations in the feed. This allows use of the Delplot method for the complex feedstocks common in energy applications, such as crude oil, where lumped models containing boiling-point-defined pseudocomponents will frequently have non-zero product amounts in the experimental feedstocks. We found that the existence of the intermediate products in the feed changes the behavior of the Delplots. For example, in the A →k1 B →k3 C; A →k2 D sequence with B0 ≠ 0, the Delplot can suggest B to be primary, secondary, or neither depending upon the ratios k1/k2 and A0/B0. These findings suggest key additional experiments covering different limits to be able to identify the proper classification of the intermediate products in the reaction network.

ADVERTISEMENT RETURN TO ISSUEEditorialNEXTACS Virtual Issue on Multicomponent Systems: Absorption, Adsorption, and DiffusionJ. Ilja Siepmann*J. Ilja SiepmannMore by J. Ilja Siepmannhttp://orcid.org/0000-0003-2534-4507, Joan F. BrenneckeJoan F. BrenneckeMore by Joan F. Brenneckehttp://orcid.org/0000-0002-7935-2134, David T. AllenDavid T. AllenMore by David T. Allenhttp://orcid.org/0000-0001-6646-8755, Michael T. KleinMichael T. KleinMore by Michael T. Kleinhttp://orcid.org/0000-0001-5444-1512, Phillip E. SavagePhillip E. SavageMore by Phillip E. Savagehttp://orcid.org/0000-0002-7902-3744, George C. SchatzGeorge C. SchatzMore by George C. Schatzhttp://orcid.org/0000-0001-5837-4740, and Françoise M. WinnikFrançoise M. WinnikMore by Françoise M. Winnikhttp://orcid.org/0000-0001-5844-6687Cite this: J. Chem. Eng. Data 2018, 63, 10, 3651Publication Date (Web):October 11, 2018Publication History Published online11 October 2018Published inissue 11 October 2018https://pubs.acs.org/doi/10.1021/acs.jced.8b00842https://doi.org/10.1021/acs.jced.8b00842editorialACS PublicationsCopyright © 2018 American Chemical Society. This publication is available under these Terms of Use. Request reuse permissions This publication is free to access through this site. Learn MoreArticle Views880Altmetric-Citations9LEARN ABOUT THESE METRICSArticle Views are the COUNTER-compliant sum of full text article downloads since November 2008 (both PDF and HTML) across all institutions and individuals. These metrics are regularly updated to reflect usage leading up to the last few days.Citations are the number of other articles citing this article, calculated by Crossref and updated daily. Find more information about Crossref citation counts.The Altmetric Attention Score is a quantitative measure of the attention that a research article has received online. Clicking on the donut icon will load a page at altmetric.com with additional details about the score and the social media presence for the given article. Find more information on the Altmetric Attention Score and how the score is calculated. Share Add toView InAdd Full Text with ReferenceAdd Description ExportRISCitationCitation and abstractCitation and referencesMore Options Share onFacebookTwitterWechatLinked InRedditEmail PDF (348 KB) Get e-AlertscloseSUBJECTS:Absorption,Adsorption,Diffusion,Mixtures,Separation science Get e-Alerts

A molecular-level kinetic model was constructed for the hydroisomerization/hydrocracking of a hydrotreated deasphalted oil feed. A kinetic network was developed for the lube base oil production containing 1105 molecular species and 15 991 reactions grouped into nine reaction families. The molecular composition of the feedstock was reconstructed by minimizing the difference between experimental data and simulated mixture properties. More specifically, hydrocarbon types, carbon number distribution, and sulfur level were matched very accurately. To model the kinetics and reduce the computational load, the Langmuir-Hinshelwood-Hougen-Watson rate law parameters were constrained using the linear free-energy relationship principles. Using the experimental process and product data of hydroisomerization over a commercial catalyst, the reaction kinetics were optimized on the basis of 141 data points. Excellent agreement was found between experimental and simulated properties of the lube base product of hydroprocessing at three different temperatures.

The demand for improved gasoline product quality has helped make molecular-level models become more preferred for the modern refinery. Building the molecular compositional model is an essential first step for this quantitative molecular management of gasoline streams. Gas chromatography equipped with flame ion detection (GC-FID) is commonly used in the gasoline detailed hydrocarbon analysis (DHA). The combination of GC-FID analysis and molecular-level modeling is thus very attractive. In the present study, we developed a gasoline compositional model based solely on GC-FID as input. To suppress the negative influence of peak coelution, we developed a statistics-based peak tuning algorithm to obtain individual compound resolution at higher carbon number range. Using the tuned result as input, the molecular-level gasoline compositional model was built by estimating the quantitative percentages of the species in a predefined molecular library (573 molecules). The molecular-level compositional model has good extensibility and can link to the molecule-based physical properties prediction model. The model has been verified via applications on various gasoline samples. The prediction of research octane number for large-scale gasoline samples was also revealed.

A detailed kinetic model for a continuous catalytic reforming (CCR) process was developed. The model included 447 naphtha molecules (C1-C12) that underwent 1469 reactions. Paraffin and naphthenic isomers up to C9 components were fully depicted, whereas aromatic isomers were fully described up to C10. Coking kinetics and the corresponding deactivation of the catalyst were integrated into the model. The steady state kinetic parameters were tuned using pilot plant data for a widely used industrial catalyst. To enable the use of commercial plant data, the energy balance and catalyst moving mechanism of typical CCR reactors were also formulated. The model was then used to simulate an industrial unit loaded with the same catalyst after deactivation calibration by adjusting a few deactivation parameters. The results showed that calculated PONA fractions, individual aromatic species, and the temperature drops of each reactor were in good accord with industrial data.

The use of Delplots to deduce the key features of reaction networks using nonisothermal kinetics data was examined. Using Delplots, a product's network rank, i.e., the number of reaction steps required for its formation from a specified reactant "A," is generally obtained by extrapolating plots of y(i)/x(A)(r) vs x(A) to x(A) = 0, where at isothermal conditions, contact time was varied to provide the range of conversion supporting the extrapolation. The presently described work addressed the common experimentalists' technique of using temperature, rather than contact time, to provide the range of conversion. To assess any uncertainties thus introduced, the effect of changing the temperature of kinetic measurements has been addressed for the parallel-series reaction network A (k1)-> B (k3)-> C; A (k2)-> D with B-0 = O. The relative activation energies of the key reactions were varied by 6 kcal/mol with respect to that for k(1), and temperature was varied between 200 and 1000 K. The resulting Delplot information can appear to suggest different reaction networks if the activation energy difference is too large and the temperature range too wide. The Delplot method classifies species B to be a primary product at low temperature when E-2 > E-1, while it appears to be a secondary product when E-2 < E-1. We suggest, as rough guidelines, that varying temperature to provide variations in conversion in the kinetic study is reasonable for E-1 similar to 50 kcal/mol if the activation energy difference E-21 is in between 3 and -3 kcal/mol.

The thermochemistry and kinetics of hydrolysis in supercritical water were probed using density function theory (DFT). Four molecules (propane, dimethyl ether, 1,3-diphenylpropane, and dibenzyl ether) were selected for this study to compare the reactivity of molecules with and without a heteroatom on a saturated carbon. We found that the activation energy for compounds with a heteroatom attached to saturated carbon was lower than that for fully hydrocarbon systems. The fastest reaction among the four molecules was that for dibenzyl ether with water. The activation energy and pre-exponential factor of the dibenzyl ether reaction with water is rationalized in the context of experimental values.

The catalytic cracking of light paraffinic crude oil with an API gravity of 51 degrees was compared using two laboratory testing techniques, a fixed-bed microactivity test (MAT) unit and a fixed fluidized-bed advanced cracking evaluation (ACE) unit. Both units were operated using equilibrated FCC catalyst (E-Cat), MFI zeolite (ZSM-5), and E-Cat/MFI (equal mixture with MFI) at two temperatures (550 and 600 degrees C) and a constant catalyst-to-oil ratio of 4.0. Despite the different hydrodynamics in MAT and ACE reactors, both units gave similar catalyst ranking based on the conversion of 221+ degrees C feed fraction at 550 and 600 degrees C in the order of E-Cat > E-Cat/MFI > MFI, which is attributed to diffusion limitation of MFI catalyst. While both testing techniques showed variation in product yield structure (dry gas, LPG, naphtha, and unconverted 221+ degrees C) over the three catalysts, the ACE unit gave significantly higher coke yield compared with MAT. The highest yield of light olefins was obtained over E-Cat/MFI (29 wt %) at 600 degrees C in MAT compared with MFI (23 wt %) and E-Cat (21 wt %). The effect of high temperature (650 degrees C) on crude oil cracking in ACE showed an increase in conversion and light olefins yield for all catalysts as well as in thermal cracking case (no catalyst) associated with a decrease in naphtha yield. The highest yields of light olefins (35 wt %) was obtained at a naphtha yield of 37 wt % over E-Cat/MFI compared with 30 and 41 wt %, respectively, for no catalyst. However, the operation at high temperature introduced the adverse effects of thermal cracking resulting in high yields of dry gas (14 wt % for E-Cat and 17 wt % for no-catalyst), which reflects a significant contribution of pyrolytic cracking reactions.

The computer-aided reconstruction of gasoline composition is an active area of petroleum and petrochemical research as a result of the demand for molecular-level management of the petroleum feed streams. To that end, in this work, a molecular compositional model based on a predefined representative molecular set was built that allows for the conversion of conventional bulk property data to an approximate molecular composition. The selection of representative molecules was based on their presence in gasoline molecular compositional measurement and their potential contribution to the key physical properties. Around 170 hydrocarbons and heteroatom species were chosen as predefined identities of molecules that can exist in a gasoline sample. The physical property data of all of the representative molecules were collected, and suitable mixing rules for the gasoline range stream were applied for the accurate prediction of bulk properties. The approximate concentration of representative molecules was obtained through fitting the predicted bulk property to the measured data. The methodology was verified through intensive tests on various gasoline samples, including straight-run naphtha, catalytic cracking gasoline, coking gasoline, and reformates. The modeling was also accomplished in a sequential order using basic to advanced measurements to find the optimum number of measurements required for detailed composition evaluation on various feedstocks. The propagation of error in the experimental measurement and prediction method on composition has been evaluated.