A Novel Group Contribution Method for the Prediction of the Derived Cetane Number of Oxygenated Hydrocarbons

Abstract

The assessment of the ignition quality of a wide range of oxygenated hydrocarbons is one key challenge in the identification of novel molecular entities qualifying as biofuels or biofuel blend components derived from oxygen-rich lignocellulosic feedstocks. The present contribution summarizes the results from a comprehensive experimental screening campaign targeting a diverse set of pure-component oxygenated hydrocarbon fuels and their ignition characteristics in an ASTM D6890 Ignition Quality Tester (IQT).This constant-volume combustion chamber experiment has been chosen because of its rapid screening potential. The unique collection of data is utilized for group contribution modeling with the aim of unraveling relationships between the ignition delay observed in IQT experiments and the fuel's molecular structure. We propose a simple, yet predictive estimator for the derived cetane number (DCN) of pure oxygenated hydrocarbons covering acyclic and cyclic, branched and straight, saturated and unsaturated hydrocarbons as well as alcohols, ethers, esters, ketones, aldehydes, and aromatic and polyfunctional compounds. The calculation procedure does neither require involved computational tools nor any information that could not be derived from the two-dimensional molecular graph. The model prediction yields a first estimate of the auto-ignition capability of a fuel candidate, if a sufficiently large and pure sample is not available.

Hydrogen‐Economic Synthesis of Gasoline‐like Hydrocarbons by Catalytic Hydrodecarboxylation of the Biomass‐derived Angelica Lactone Dimer

“…As noted above, the measured octane rating of compound 10 is 79 . This agrees well with the value of 78 calculated by using the group contribution model of Dahmen and Marquardt . The same method yields predicted RON values of 70 for 13 and 78 for 14 , which renders them useful blendstocks for motor gasoline.…”

“…[14] This agreesw ell with the value of 78 calculated by using the group contribution model of Dahmen and Marquardt. [18] The same method yields predicted RON values of 70 for 13 and 78 for 14,w hich renders them useful blendstocks for motor gasoline. If the Table 1, entry 7m ixture werec ombined with 10 % v/v ethanol (blending RON = 129) [19] as an oxygenate,t he weighted RON value would be 84.…”

Application of a Rectified Linear Unit (ReLU) Based Artificial Neural Network to Cetane Number Predictions

“…Data for cetane numbers used in the core data set was obtained from sets present in the NREL Compendium of Experimental Cetane Number Data [11] and other sources [6,12,13]. The core data set contains 423 molecules, with their cetane numbers derived from blend measurements, CFR, IQT or other ignition delay methods.…”

“…We have chosen not to exclude compounds that are common outliers across ANN's, as our goal is to provide a generalizable model for a wide variety of compound groups including new unknown data. Instead, common outliers were targeted for further research and their cetane numbers were updated using more recent sources [13].…”

Early‐stage evaluation of biorefinery processing pathways using process network flux analysis

“…2 However in the design of future biorefineries, many design degrees of freedom need to be addressed systematically like the choice of products varying from bulk chemicals to high-value added chemicals [3][4][5] complemented by the questions of lignin utilization in lignocellulosic biorefineries, 6,7 supply chain, [8][9][10] and product design. 11 Even the seasonality and availability of raw materials 12 or the problematic nature of supply and demand 13 are subject to discussion.…”