Elsevier

Combustion and Flame

Volume 212, February 2020, Pages 388-402
Combustion and Flame

An experimental study of n-dodecane and the development of an improved kinetic model

https://doi.org/10.1016/j.combustflame.2019.11.014Get rights and content

Abstract

This work presents new fundamental combustion experimental datasets and propose an improved kinetic model for n-dodecane, a widely used surrogate component for jet fuels and diesel. In the experiments, ignition delay times for n-dodecane, were measured over low-to-high temperature range by combining a heated rapid compression machine (RCM) and a heated shock tube (ST). The measurements cover a wide range of temperatures (621–1320 K), pressures (8, 15 bar) and equivalence ratios (0.5–1.5). Flow reactor oxidation was also carried out over temperature range of 600–1100 K and at equivalence ratios of 0.5, 1.0 and 1.5 under atmospheric pressure to provide species evolution profiles of O2, CO, CO2 CH4, C2H4 etc. These datasets were used to develop a mechanism for n-dodecane. New rate coefficients were adopted from recent literature to improve the performance of the model, leading to a model composed of 737 species and 3629 reactions. The present experimental data, as well as a wide range of datasets in literature, were used to evaluate the performance of the model. An overall good agreement of the predictions with the experimental results was observed. This model is anticipated to facilitate the development of kinetic models for jet fuel surrogate.

Introduction

It is well established that accurate simulation using kinetic mechanism of surrogate fuels contributes to the understanding of the interaction between chemical reaction pathway and flow [1], [2], [3] for complex real fuels, and is helpful for identifying and controlling the pollutant formation, e.g., soot [4], [5], [6], [7]. Obviously, the maturity of kinetic mechanisms of pure surrogate components determine the performance of the surrogate mechanism. N-dodecane is a long-chain linear alkane widely existing in real transport fuels. Due to the appropriate carbon number and relevance to jet fuels, it is often used as the representative n-alkane in multi-component jet fuel surrogates and sometimes as a single-component surrogate for jet fuels [8], [9], [10], [11], [12], [13]. It also serves as a component in some proposed diesel surrogates [14,15]. Therefore, a good understanding of the combustion chemistry of n-dodecane is important to the design of modern engines and the optimization of combustion strategies.

The combustion chemistry of n-dodecane has been a focus of interest in the past decades, and there have been extensive researches focusing on experimental characterization and kinetic modeling of this fuel. Previous experimental studies for n-dodecane have been performed in shock tubes (ST), flow reactor, jet-stirred reactor (JSR) and flames, including the measurements of ignition delay time (IDT), speciation of pyrolysis and oxidation, and laminar flame speed. The previous experimental studies are illustrated in Table 1. The kinetic targets provided in these fundamental combustion experiments are indispensable for the development of kinetic mechanisms. Note that the IDT of n-dodecane, a crucial global kinetic parameter for the validation of kinetic model, was seldom studied over low-to-intermediate temperature range using RCM. Although conventional aero-engines usually work at high temperatures, the knowledge about the low temperature chemistry of n-dodecane is significant in the context of using jet fuels in diesel or HCCI type engines. Furthermore, the available IDT measurements for n-dodecane were all preformed in air. However, the promising low temperature combustion (LTC) technology relies on very lean or extremely diluted combustion [16], it is of great significance to characterize the autoignition of fuels under diluted conditions.

Parts of the above fundamental combustion experiment databases have been used as kinetic targets and helped to pave the way for the development, validation, refinement and reduction of n-dodecane combustion kinetic models. In addition, some kinetic studies also contributed to kinetic mechanism development for n-dodecane. For instance, Wang et al. [33] investigated the initiation mechanism and product distributions of pyrolysis and combustion of n-dodecane; Zhao et al. conducted a computational study on the unimolecular decomposition of n-dodecane [24]. Based on these previous works, several detailed mechanisms for n-dodecane have been published, including the models proposed by Ranzi et al. (called Ranzi's model) [34], Biet et al.(called Biet's model) [35], Wang et al. (called JetSurF model series) [26,36,37], Westbrook et al. (called LLNL model) [38,39], Mzé-Ahmed's et al. (called Mzé-Ahmed's model) [28], as well as a very recently developed model by Zeng et al. (called Zeng's model) [25]. In addition, several reduced or optimized models [7,14,[40], [41], [42] were derived from these detailed models. Detailed mechanism with thousands of species and elementary reactions is very useful in many circumstances and is a prerequisite to derive reduced models, and thus determines the performance of the derived reduced mechanism. Therefore, of particular interest in this work is detailed models or optimized models, while less attentions are paid to reduced models. The status of the development of detailed mechanism is briefly reviewed as follows. Ranzi et al. developed a semi-empirical and lumped mechanism for heavy linear alkanes including n-dodecane for both pyrolysis and oxidation spanning low to high temperature ranges, and later updated and improved it in the following studies [43,44]. However, these models failed to simulate the experimental data at intermediate and high temperatures, [43] which demonstrated the need for further improvement . Biet's model was developed using computer-aided generation software EXGAS [35]. This model was validated against the flow reactor oxidation experiment and the results showed that the mechanism needed further improvement. The n-dodecane sub-mechanism in JetSurF model was taken from the n-dodecane mechanism proposed by You et al. [45]. Banerjee et al. optimized the sub-mechanism of n-dodecane in JetSurF 1.0 model based on their experimental datasets, leading to the release of the latest version of this model, i.e., Optimized JetSurF 1.0 model [26]. To our knowledge, this model mainly focused on the high temperature chemistry and only very limited reactions were used to describe the low temperature reaction pathway. Hence, its performance in predicting low temperature oxidation needs to be further evaluated. Westbrook et al. [38] proposed a detailed model for n-alkanes from C8 to C16, and later Sarathy et al. [39] updated and extended this model to include the reaction schemes of 2-methylalkanes from C7-C20, leading to the latest LLNL model. This model was used to simulate the experimental results in many studies [[20], [21], [22], [23],[26], [27], [28],46], and the observations implied the necessity to revisiting the kinetics to improve its predictions. Starting from the LLNL model, Narayanaswamy et al. obtained a skeletal model by using mechanism reduction and then revisited the rate constants for some reaction classes, leading to the release of Narayanaswamy's model [42]. This model was extensively validated against literature data, but the observed discrepancy suggests that further improvements are required to improve its performance. Another improved n-dodecane model derived from LLNL model was derived by Cai et al. [41], who updated the model with the optimized rate rules. Mzé-Ahmed's et al. [28] proposed a detailed mechanism to describe combustion of n-dodecane over the entire temperature range. The model was compared with their JSR oxidation speciation profiles and literature data, and was found to overestimate the IDTs at low-to-intermediate temperatures. Very recently, Zeng et al. [25] developed a detailed model for n-dodecane and compared it with their pyrolysis data, as well as extensive datasets in literature. Although good agreements were observed between simulations and experimental data, the low temperature reaction scheme needs further evaluation.

In summary, the experimental studies on the autoignition characteristics of n-dodecane are inadequate, leaving a gap under low-to-intermediate temperature range and diluted conditions. On the other hand, discrepancies were observed when using the existing kinetic models to simulate the experimental databases, implying further improvements are needed for these models. Furthermore, some of these experimental data were available after the publication of the above-mentioned kinetic models, and thereby those models were not validated against all existing data. Therefore, there are still some experimental databases that can be utilized for further model evaluation and improvement. These research insufficiencies motivate the interest of the present study. This study aims to improve the knowledge of n-dodecane oxidation chemistry by (1) providing a reliable IDT data set over low-to-high temperature ranges by combing a heated RCM and a heated ST; (2) providing speciation information of oxidation in a flow reactor and (3) proposing an improved kinetic model of good performance over wide conditions.

Section snippets

Ignition delay time measurements

The test fuel used in this study is n-dodecane with a purity of 99% provided by MACKLIN. The IDT measurements were conducted in the heated RCM and ST at Shanghai Jiao Tong University (SJTU). The details of the two facilities can be found in our previous studies [47], [48], [49], [50], [51], [52], and will not be repeated here. Since n-dodecane is a heavy hydrocarbon with high boiling point, a preheating temperature of 115 °C was used to ensure the complete vaporization of the fuel. The

The need for an improved model

IDTs are one of the kinetic targets that models should be able to predict. The present experimental data was firstly used to evaluate the existing models, including Ranzi's model [44], optimized JetSurF 1.0 model [26], LLNL model [39], Narayanaswamy's model [42], Cai's model [41] and Zeng's model [25]. Figure 3 shows a comparison between the simulations using the above models and experimental data under several conditions. It is found that there is great discrepancy between the experimental

Autoignition characteristics of n-dodecane under various conditions and model simulation

The autoignition delay times of n-dodecane/O2/N2 mixture were studied over wide conditions to cover various pressures, equivalence ratios, and dilution ratios. Figures 5–8 shows the effects of different parameters on the IDTs of n-dodecane. In order to facilitate the evaluation of the present model, the Const. simulations using constant volume constraint and RCM simulations for RCM data were displayed as well.

Figure 5 shows the influence of pressure on reactivity of n-dodecane by comparing IDTs

Validation of the present model against literature data

To comprehensively evaluate the performance of the model, we use the model to simulate the available experimental data in literature. Here, the databases adopted for evaluation mainly focus on oxidation environment while the pyrolysis data was not considered.

Reaction pathway analysis

The good performance of the model in reproducing the experimental data motivates the interest to perform a detailed kinetic analysis using the mechanism for a better understanding of the combustion chemistry of n-dodecane. In this section, reaction pathway analysis was performed for 15 bar at three temperatures of 650, 800 and 1300 K to cover low, intermediate and high temperatures for stoichiometric temperatures. For the other mixtures and pressures conditions, the branching ratios for each

Conclusion

The autoignition of n-dodecane, which was seldom investigated at low temperatures using RCM previously, was studied over low-to-high temperature ranges by combining a heated RCM and ST in the present study. The experimental conditions cover a wide range of temperature (621–1320 K), pressure (8 and 15 bar), equivalence ratio (0.5,1.0 and 1.5) and dilution ratio (79%, 89.5% and 93%). The effect of these operating parameters on autoignition characteristics of n-dodecane, were comprehensively

Acknowledgment

The authors gratefully acknowledge the financial support from the Key Project of National Natural Science Foundation of China (Nos. 91641202 and 51425602) and Program of Shanghai Subject Chief Scientist (No. 19XD1401800).

References (67)

  • Y.-.X. Liu et al.

    Combustion study of a surrogate jet fuel

    Combust. Flame

    (2019)
  • J. Liu et al.

    A new surrogate fuel for emulating the physical and chemical properties of RP-3 kerosene

    Fuel

    (2020)
  • S.S. Goldsborough et al.

    Advances in rapid compression machine studies of low- and intermediate-temperature autoignition phenomena

    Prog. Energy Combust. Sci.

    (2017)
  • S.S. Vasu et al.

    n-Dodecane oxidation at high-pressures: measurements of ignition delay times and OH concentration time-histories

    Proc. Combust. Inst.

    (2009)
  • D.F. Davidson et al.

    Development of an aerosol shock tube for kinetic studies of low-vapor-pressure fuels

    Combust. Flame

    (2008)
  • J. Shao et al.

    A shock tube study of n-heptane, iso-octane, n-dodecane and iso-octane/n-dodecane blends oxidation at elevated pressures and intermediate temperatures

    Fuel

    (2019)
  • D.F. Davidson et al.

    Multi-species time-history measurements during n-dodecane oxidation behind reflected shock waves

    Proc. Combust. Inst.

    (2011)
  • A. Tekawade et al.

    Time-resolved carbon monoxide measurements during the low- to intermediate-temperature oxidation of n-heptane, n-decane, and n-dodecane

    Combust. Flame

    (2016)
  • T. Malewicki et al.

    Experimental and modeling study on the pyrolysis and oxidation of n-decane and n-dodecane

    Proc. Combust. Inst.

    (2013)
  • M. Zeng et al.

    Investigation of n-dodecane pyrolysis at various pressures and the development of a comprehensive combustion model

    Energy

    (2018)
  • S. Banerjee et al.

    An experimental and kinetic modeling study of n-dodecane pyrolysis and oxidation

    Combust. Flame

    (2016)
  • M.S. Kurman et al.

    Speciation of the reaction intermediates from n-dodecane oxidation in the low temperature regime

    Proc. Combust. Inst.

    (2011)
  • O. Herbinet et al.

    Thermal decomposition of n-dodecane: experiments and kinetic modeling

    J. Anal. Appl. Pyrolysis

    (2007)
  • K. Kumar et al.

    Laminar flame speeds and extinction limits of preheated n-decane/O2/N2 and n-dodecane/O2/N2 mixtures

    Combust. Flame

    (2007)
  • X. Hui et al.

    Laminar flame speeds of transportation-relevant hydrocarbons and jet fuels at elevated temperatures and pressures

    Fuel

    (2013)
  • C. Ji et al.

    Propagation and extinction of premixed C5–C12 n-alkane flames

    Combust. Flame

    (2010)
  • Q.-.D. Wang et al.

    Reactive molecular dynamics simulation and chemical kinetic modeling of pyrolysis and combustion of n-dodecane

    Combust. Flame

    (2011)
  • C.K. Westbrook et al.

    A comprehensive detailed chemical kinetic reaction mechanism for combustion of n-alkane hydrocarbons from n-octane to n-hexadecane

    Combust. Flame

    (2009)
  • S.M. Sarathy et al.

    Comprehensive chemical kinetic modeling of the oxidation of 2-methylalkanes from C7 to C20

    Combust. Flame

    (2011)
  • T. Yao et al.

    A compact skeletal mechanism for n-dodecane with optimized semi-global low-temperature chemistry for diesel engine simulations

    Fuel

    (2017)
  • L. Cai et al.

    Optimized reaction mechanism rate rules for ignition of normal alkanes

    Combust. Flame

    (2016)
  • K. Narayanaswamy et al.

    A chemical mechanism for low to high temperature oxidation of n-dodecane as a component of transportation fuel surrogates

    Combust. Flame

    (2014)
  • X. You et al.

    Detailed and simplified kinetic models of n-dodecane oxidation: the role of fuel cracking in aliphatic hydrocarbon combustion

    Proc. Combust. Inst.

    (2009)
  • Cited by (29)

    View all citing articles on Scopus
    View full text