Selection and performance comparison of jet fuel surrogates for autothermal reforming

Three fuel mixtures were investigated as possible surrogates for low-sulfur JP-8. The selected fuel mixtures were chosen based on a desire to match hydrocarbon chemical composition classes found in real jet fuels. The surrogate fuels selected consisted of single, binary and tertiary-component mixtures of n-dodecane, decalin and toluene in liquid volume ratios of 10:0:0, 9:1:0 and 7:1:2. The hydrocarbon components selected represented the largest chemical classes within JP-8 of normal paraffin, cyclo-paraffin and aromatic. The surrogate fuels and individual surrogate fuel components were reacted in an atmospheric pressure autothermal reformer with noble metal catalysts under conditions of steam-to-carbon ratio of 2.0, fuel equivalency energy flow of 3.3 kW thermal, space velocities of 21,000-28,000 h⁻¹ and variable oxygen-to-carbon ratios of 0.8-1.2. For all fuels investigated fuel conversion of greater than 96% could be achieved. The single component n-dodecane proved to be the least reactive resulting in lower hydrogen yields, lower reforming efficiency and increased olefin products in the reformate. The binary mixture of n-dodecane and decalin resulted in a closer match with JP-8, but did not correlate well in terms of fuel conversion and hydrogen yield. Aliphatic mixtures also exhibited greater olefin production. The three-component mixture of n-dodecane/decalin/toluene provided the best correlation to JP-8 and appears to be a good three-component surrogate fuel, particularly over the operating range of oxygen to carbon ratio of 0.95-1.10.     

The development and experimental validation of a reduced ternary kinetic mechanism for the auto-ignition at HCCI conditions,proposing a global reaction path for ternary gasoline surrogates

To acquire a high amount of information of the behaviour of the Homogeneous Charge Compression Ignition (HCCI) auto-ignition process, a reduced surrogate mechanism has been composed out of reduced n-heptane, iso-octane and toluene mechanisms, containing 62 reactions and 49 species. This mechanism has been validated numerically in a 0D HCCI engine code against more detailed mechanisms (inlet temperature varying from 290 to 500 K, the equivalence ratio from 0.2 to 0.7 and the compression ratio from 8 to 18) and experimentally against experimental shock tube and rapid compression machine data from the literature at pressures between 9 and 55 bar and temperatures between 700 and 1400 K for several fuels: the pure compounds n-heptane, iso-octane and toluene as well as binary and ternary mixtures of these compounds. For this validation, stoichiometric mixtures and mixtures with an equivalence ratio of 0.5 are used. The experimental validation is extended by comparing the surrogate mechanism to experimental data from an HCCI engine. A global reaction pathway is proposed for the auto-ignition of a surrogate gasoline, using the surrogate mechanism, in order to show the interactions that the three compounds can have with one another during the auto-ignition of a ternary mixture.     

Development of a detailed kinetic model for gasoline surrogate fuels

A detailed chemical kinetic model to describe the autoignition of gasoline surrogate fuels is presented consisting of the fuels iso-octane, n-heptane, toluene, diisobutylene and ethanol. Model predictions have been compared with shock tube ignition delay time data for surrogates of gasoline over practical ranges of temperature and pressure, and the model has been found to be sensitive to both changes in temperature and pressure. Moreover, the model can qualitatively predict the observed synergistic and antagonistic non-linear blending behaviour in motor octane number (MON) for different combinations of primary reference fuels (PRFs) and non-PRFs by correlating calculated autoignition delay times from peak pressures and temperatures in the MON test to experimental MON values. The reasons for the blending behaviour are interpreted in terms autoignition chemistry.     

Experimental determination of laminar burning velocity for butanol and ethanol iso-octane blends

The potential of butanol as an additive in iso-octane used as gasoline fuel was characterized with respect to laminar combustion, and compared with ethanol. New sets of data of laminar burning velocity are provided by using the spherical expanding flame methodology, in a constant volume vessel. This paper presents the first results obtained for pure fuels (iso-octane, ethanol and butanol) at an initial pressure of 0.1 MPa and a temperature of 400 K, and for an equivalence range from 0.8 to 1.4. New data of laminar burning velocity for three fuel blends containing up to 75% alcohol by liquid volume are also provided. From these new experimental data, a correlation to estimate the laminar burning velocity of any butanol or ethanol blend iso-octane-air mixture is proposed.     

Structure of evaporating single- and multicomponent fuel sprays for 2nd generation gasoline direct injection

In the present paper the effect of fuel properties on spray formation and evaporation was investigated for a hollow-cone spray of a piezoelectric injector for Direct Injection Spark Ignition (DISI) engines. Late injection timing in a high-pressure atmosphere (1.5 MPa, 200 °C) was simulated in an injection chamber. Liquid and vapor phase structure of the hollow-cone spray were studied with 2D-Mie scattering, laser-induced fluorescence (LIF) as well as phase-Doppler anemometry (PDA). The spray structure was investigated for several alkanes with high and low volatility (n-hexane, n-heptane, iso-octane, n-decane) and a three-component mixture of the n-alkanes with similar fuel properties like a multicomponent gasoline fuel. It is found that the rapid evaporation of high volatility fuels can lead to spray destabilization, whereas low volatility single-component fuels overestimate radial spray propagation and vortex formation. For iso-octane the droplet size distribution is shifted to smaller droplets and the spray appeared to be less dense compared to n-heptane despite almost identical boiling behavior. However, the much higher viscosity of iso-octane determines the internal nozzle flow which results in a reduced injected fuel mass and changed atomization. A well defined three-component fuel models the global spray characteristics as well as the droplet size, droplet momentum distribution and evaporation behavior of the used multicomponent gasoline fuel very precisely. Small amounts of low volatility fractions delay the droplet evaporation and support the overall spray stability also for multicomponent mixtures. This leads to an increased spray width as well as larger droplet sizes and momenta. The evaporation characteristic of multicomponent fuels at increased ambient pressure is complex. At the studied injection conditions it is situated between the two limiting cases of distillation-like behavior and coevaporation of the components. Moreover, the results in comparison with theoretical estimations indicate a demixing of light and heavy boiling fractions in the three-component and multicomponent fuel under conditions which are typical for DISI strategies with late injection.     

Optimization of oxidative desulfurization of thiophene using Cu/titanium silicate-1 by box-behnken design

The oxidative desulfurization of thiophene in a synthetic mixture of thiophene and iso-octane was investigated with copper loaded titanium silicate-1 (TS-1) catalyst in presence of hydrogen peroxide as oxidising agent and the conversion was enhanced by 22% at 240 min on addition of 1.05 wt.% copper in TS-1. The optimal design of experiments using box-behnken method was employed to evaluate the effects of individual process variables such as, reaction temperature, amount of catalyst and moles of hydrogen peroxide per mole of thiophene and their optimum values were found to be 70 °C, 0.45 g (22.5 mol/L of iso-octane) and 19.9 mol (in 20 ml of iso-octane), respectively, to achieve a conversion of 93%. The influence of mass transfer effects on the desulfurization reaction was minimized by selecting proper degree of agitation and catalyst size. An empirical kinetic model was used to interpret the rate data. The apparent activation energy was found to be 28.67 kJ/mol.     

Investigations on surrogate fuels for high-octane oxygenated gasolines

Gasoline is a complex mixture that possesses a quasi-continuous spectrum of hydrocarbon constituents. Surrogate fuels that decrease the chemical and/or physical complexity of gasoline are used to enhance the understanding of fundamental processes involved in internal combustion engines (ICEs). Computational tools are largely used in ICE development and in performance optimization; however, it is not possible to model full gasoline in kinetic studies because the interactions among the chemical constituents are not fully understood and the kinetics of all gasoline components are not known. Modeling full gasoline with computer simulations is also cost prohibitive. Thus, surrogate mixtures are studied to produce improved models that represent fuel combustion in practical devices such as homogeneous charge compression ignition (HCCI) and spark ignition (SI) engines. Simplified mixtures that represent gasoline performance in commercial engines can be used in investigations on the behavior of fuel components, as well as in fuel development studies. In this study, experimental design was used to investigate surrogate fuels. To this end, SI engine dynamometer tests were conducted, and the performance of a high-octane, oxygenated gasoline was reproduced. This study revealed that mixtures of iso-octane, toluene, n-heptane and ethanol could be used as surrogate fuels for oxygenated gasolines. These mixtures can be used to investigate the effect of individual components on fuel properties and commercial engines performance.     

A kinetic modeling study of self-ignition of low alkylbenzenes at engine-relevant conditions

The self-ignition of low alkylbenzenes at engine-relevant conditions has been studied with kinetic modeling. A previously developed chemical kinetic model for gasoline surrogate fuels [J.C.G. Andrae, R.A. Head, Combust Flame 156 (2009) 842-51] was extended with chemistry for ethylbenzene and m-xylene resulting in an overall model consisting of 150 species and 759 reactions. In model validation, comparisons were made between model predictions and experimental data of ignition delay times measured behind reflected shock waves, laminar burning velocities collected at elevated temperature and pressure and species profiles in a high-pressure single pulse shock tube. Generally good agreement was found and the model is sensitive to changes in mixture strength, pressure and temperature. Shock tube ignition delay modeling results for ethylbenzene and m-xylene also compare well to the ones for toluene. The rate controlling step for the ignition of ethylbenzene in the current mechanism is the reaction with ethylphenyl radical and oxygen. Ignition delay time for m-xylene was found to be very sensitive to reactions involving hydrogen atom abstraction from fuel by hydroxyl and oxygen and to branching reactions where methylbenzyl reacts with oxygen and hydroperoxide. The validated mechanism was used to study fuel chemistry effects when blending ethylbenzene with the paraffinic fuels iso-octane and n-heptane. A sensitivity- and flow path analysis showed that a higher consumption of hydroperoxide by ethylphenyl than expected from the contribution of neat ethylbenzene in a fuel mixture with iso-octane inhibits both iso-octane and ethylbenzene ignition. This can explain the observed increase in ignition delay time and octane number for fuel mixtures compared to neat fuels.     

Temperature and pressure dependence of molecular adsorption on single wall carbon nanotubes and the existence of an “adsorption/desorption pressure gap”

The interaction of acetone with single wall carbon nanotubes (SWCNTs) was studied by temperature programmed desorption with mass spectrometry (TPD-MS), after reflux, sonication, or exposure to 7.6 Torr of acetone vapors at room temperature. Acetone molecules adsorb strongly on SWCNTs desorbing at ∼400-900 K, corresponding to desorption energies of ∼100-225 kJ/mol, as intact molecules. Exchange of intact adsorbed molecules with gas phase species was observed in successive dosing of hydrogenated and deuterated acetone molecules. The desorption energies reported here are in stark contrast to the desorption energies (∼75 kJ/mol) reported earlier for SWCNTs interacting with acetone under high vacuum at cryogenic temperatures. This result suggests activated adsorption/desorption, and is also observed for adsorption of ethanol, methane, n-butane and 1,3-butadiene on SWCNTs and on carbon black. Quantum chemical calculations suggest that adsorption in interstitial channels of bundles formed of large-diameter SWCNTs is possible and can account for high desorption barriers, a result of strong dispersion interactions between neighboring SWCNTs.     

Laminar burning velocities of n-heptane, iso-octane, ethanol and their binary and tertiary mixtures

Measurements of the adiabatic laminar burning velocities of n-heptane, iso-octane, ethanol and their binary and tertiary mixtures are reported. Non-stretched flames were stabilized on a perforated plate burner at 1 atm. The Heat Flux method was used to determine burning velocities under conditions when the net heat loss from the flame to the burner is zero. Initial temperatures of the gas mixtures with air were 298 and 338 K. Uncertainties of the measurements were analyzed and assessed experimentally. The overall accuracy of the burning velocities was estimated to be better than ±1 cm/s. These new measurements were compared with the literature data when available. Experimental results in lean ethanol + air mixtures are systematically higher than previous measurements under similar conditions. Good agreement for n-heptane + air flames and for iso-octane + air flames was found with the experiments performed in counter-flow twin flames with linear extrapolation to zero stretch.     

Fractionation of pitches by molecular weight using continuous and semibatch dense-gas extraction

A countercurrent, multistage, dense-gas extraction technique with reflux was investigated for the fractionation of carbonaceous pitches. Two modes of operation were investigated: continuous-stripping and semibatch operation. For example, continuous stripping with dense-gas toluene in the supercritical state, a positive column temperature gradient from 330 to 380 °C, and a pressure of 49 bar was used to strip the monomer and dimer species from an A-240 petroleum pitch feed, yielding a high molecular weight (mol wt) bottoms product rich in trimer and higher oligomers. Afterwards, semibatch operation was used with supercritical, dense-gas toluene, a temperature gradient of 330 to 380 °C, and pressures from 84 to 111 bar to fractionate the above bottoms product, yielding a trimer-rich overhead (average mol wt (M_w) = 800) and a tetramer and higher residue with M_w ∼ 1000. Considering the two operations as a unit, a combined selectivity factor of ∼350 was obtained. Not only is this at least an order of magnitude better than what can in principle be accomplished by conventional, single-stage solvent extraction, but such extraction is inapplicable to our system because of the insolubility of the pitch fractions of interest in typical liquid solvents. Matrix-assisted, laser desorption/ionization time-of-flight mass spectrometry (MALDI) was used to verify that separation was indeed occurring by mol wt and to study the relationship between the M_w, softening point, and C/H ratio of the fractions produced.     

Chemical composition of fuels and emissions from a coal + tire combustion experiment in a power station

Medium-sulfur bituminous coal and a mixture of 95 wt.% coal plus 5 wt.% tire-derived fuel (TDF) in the form of shredded automotive tires were combusted in a stoker boiler under the same conditions. This paper presents quantitative chemical compositions of the fuels and of the gaseous and particulate emissions. The coal + TDF mixture is considerably richer in Zn than the pure coal as a result of the high Zn content of the shredded tires (∼1 wt.% Zn). Atmospheric emissions of Zn increased from 15 g/h to nearly 2.4 kg/h when coal + TDF was combusted. Similarly, emissions of most other metals and metalloids, as well as those of HCl increased when TDF (∼3000 ppm Cl) was added. The enhanced metal emissions might be due to formation of gaseous metal chloride species in the stack gases. On the other hand, emissions of CO decreased slightly, whereas those of NO_x, SO₂, and total particulate matter remained virtually unchanged. These results help in assessing the environmental impact of energy recovery from scrap tires in stoker boilers.     

Effect of composite aftertreatment catalyst on alkane, alkene and monocyclic aromatic emissions from an HCCI/SI gasoline engine

Designing automotive catalysts for effective control of NO_x, HC and CO emissions under both lean and stoichiometric engine operation is a challenging task. The present work assesses the performance efficiency of a three-zone prototype catalytic convertor in reducing exhaust emissions from a gasoline engine, operating in Homogeneous Charge Compression Ignition (HCCI) and Spark Ignition (SI) mode under lean and stoichiometric conditions. The performance of the convertor for HC oxidation follows the order: lean HCCI > stoichiometric SI > stoichiometric HCCI. The study mainly focused on the quantitative analysis of C1-C7 hydrocarbon compounds before and after the catalytic convertor. The results show that monocyclic aromatic hydrocarbons such as toluene are present at higher concentrations in the exhaust under HCCI operation than in the SI case. On the other hand, benzene concentrations are higher in the SI exhaust. The most common exhaust products of the two engine operating modes are methane, ethylene, propylene, benzene, and toluene. The prototype catalytic convertor eliminates most of the hydrocarbon species in the exhaust under both combustion modes, especially with a lean mixture. Conversion efficiencies for the different hydrocarbon species over the catalyst were in the order of alkenes > alkanes > aromatics. Hydrogen was added upstream of the catalyst primarily to assess its ability to promote NO_x reduction, however it was also found to influence the oxidation characteristics of the catalyst. During H₂ addition, the methane concentration was higher downstream of the catalyst.     

Thermodynamic modeling of dealcoholization of beverages using supercritical CO2: Application to wine samples

The supercritical removal of ethanol from alcoholic beverages (brandy, wine, and cider) was studied using the GC-EoS model to represent the phase equilibria behavior of the CO₂ + beverage mixture. Each alcoholic drink was represented as the ethanol + water mixture with the corresponding ethanol concentration (35 wt% for brandy, 9-12 wt% for different wines and 6 wt% for cider). The thermodynamic modeling was based on an accurate representation of the CO₂ + ethanol and CO₂ + water binary mixtures, and the CO₂ + ethanol + water ternary mixture.The GC-EoS model was employed to simulate the countercurrent supercritical CO₂ dealcoholization of the referred beverages; the results obtained compared good with experimental data from the literature. Thus, the model was used to estimate process conditions to achieve an ethanol content reduction from ca. 10 wt% to values lower than 1 wt%. The model results were tested by carrying out several extraction assays using wine, in a 3 m height packed column at 308 K, pressures in the range of 9-18 MPa and solvent to wine ratio between 9 and 30 kg/kg.     

Adsorption of thiophene and toluene on NaY zeolites exchanged with Ag(I), Ni(II) and Zn(II)

The present study evaluates the adsorption capacity of thiophene and toluene and their competitive behaviour on zeolite NaY exchanged with transition metals (5 wt% Ni, Zn and Ag). The headspace chromatography technique was used to obtain monocomponent apparent adsorption isotherms of thiophene and toluene with NaY, NiY, ZnY and AgY using isooctane as an inert solvent at 30 and 60 °C. Selectivity between toluene and thiophene at saturation capacities were also measured at 30 °C. The adsorption capacity for thiophene increased for the studied adsorbents as follows: NaY < ZnY < NiY < AgY at 30 °C and NaY < NiY < ZnY < AgY at 60 °C. Toluene is less adsorbed, but within the same order of magnitude as thiophene and following the same sorbent order. All adsorbents were moderately selectivity for toluene. Nevertheless, the sulfur content was successfully reduced in the presence of aromatics and olefins in immersion tests with a model fuel mixture. These results show the importance of inserting transitions metals in the zeolitic structure to enhance the adsorption of both aromatic and sulfur containing compounds in organic liquid mixtures, which shows promise to meet environmental standards in transportation fuels.     

Supercritical fluid extraction from dried banana peel (Musa spp., genomic group AAB): Extraction yield, mathematical modeling, economical analysis and phase equilibria

Supercritical fluid extraction from dried banana peel (Musa spp., subgroup Prata, genomic group AAB, popularly known in Brazil as Enxerto) was studied. The aspects investigated were: overall extraction curve (OEC), mass transfer modeling of the yield curves, economical analysis of the process and phase equilibrium data for the pseudo-ternary system of banana peel extract, carbon dioxide and ethanol. The extraction operating conditions evaluated were: pressure ranging from 100 bar to 300 bar, temperature from 40 to 50 °C and constant solvent flow rate of 5.0 gCO₂/min. Experimental extraction data were correlated using three kinetic models based on mass transfer equations (logistic, diffusion and Esquível models). Phase equilibrium measurements were performed using pressure from 64.9 bar to 239.9 bar and mass fraction of supercritical extract from 0.52 to 3.55 wt%. Yield results ranged from 0.6 to 6.9% d.b. (dry basis). The lowest deviation between experimental and correlated data was obtained by the Logistic model, except for the curve at 300 bar and 40 °C which was best represented by the Esquível model. The economical analysis identified the possibility to apply the supercritical fluids to obtain extracts from banana peel in an industrial scale. Phase equilibrium for the supercritical extract from banana peel with carbon dioxide modified by ethanol exhibited liquid-liquid, vapor-liquid (bubble point) and vapor-liquid-liquid phase transitions. A crossover phenomenon for the systems evaluated was observed for pressures between 200 bar and 240 bar, for both groups of assays, i.e., supercritical extraction and phase equilibrium.     

A comparative experimental study of the autoignition characteristics of alternative and conventional jet fuel/oxidizer mixtures

Autoignition characteristics of an alternative (non-petroleum) and two conventional jet fuels are investigated and compared using a heated rapid compression machine. The alternative jet fuel studied is known as “S-8”, which is a hydrocarbon mixture rich in C₇-C₁₈ linear and branched alkanes and is produced by Syntroleum via the Fischer-Tropsch process using synthesis gas derived from natural gas. Specifically, ignition delay times for S-8/oxidizer mixtures are measured at compressed charge pressures corresponding to 7, 15, and 30 bar, in the low-to-intermediate temperature region ranging from 615 to 933 K, and for equivalence ratios varying from 0.43 to 2.29. For the conditions investigated for S-8, two-stage ignition response is observed. The negative temperature coefficient (NTC) behavior of the ignition delay time, typical of higher order hydrocarbons, is also noted. Further, the dependences of both the first-stage and the overall ignition delays on parameters such as pressure, temperature, and mixture composition are reported. A comparison between the autoignition responses obtained using S-8 and two petroleum-derived jet fuels, Jet-A and JP-8, is also conducted to establish an understanding of the relative reactivity of the three jet fuels. It is found that under the same operating conditions, while the three jet fuels share the common features of two-stage ignition characteristics and a NTC trend for ignition delays over a similar temperature range, S-8 has the shortest overall ignition delay times, followed by Jet-A and JP-8. The difference in ignition propensity signifies the effect of fuel composition and structure on autoignition characteristics.     

Contributions of heterogeneous and homogeneous chemistry in the catalytic partial oxidation of octane isomers and mixtures on rhodium coated foams

Catalytic partial oxidation experiments with n-octane, 2,2,4-trimethylpentane (i-octane), and an n-octane:i-octane (1:1) mixture were performed on 80 and 45 ppi Rh-coated α-alumina foam supports at 2, 4, and 6 SLPM total flow rate in order to explore the effects of chemical structure for single components and binary mixtures on fuel reactivity and product distribution. When reacted as single components, the conversion of i-octane is greater than n-octane at C/O>1.1 (both fuel conversions are 100% for C/O<1.1). However, when reacted in an equimolar mixture, the conversion of n-octane is greater than i-octane. All three fuels give high selectivity to syngas (H₂ and CO) on 80 ppi supports for C/O<1. For C/O>1, n-octane produces high selectivity to ethylene while i-octane makes i-butylene and almost no ethylene. The fuel mixture produces these species proportional to the mole fractions of n-octane and i-octane within the reacting mixture. Increasing the support pore diameter decreases the selectivity to syngas and increases H₂O and olefin selectivity.The reforming of all three fuels is modeled using detailed chemistry by decoupling the heterogeneous and homogeneous chemistry in a two-zone plug flow model. Detailed homogeneous reaction mechanisms with several thousand elementary reactions steps and several hundred species are used to simulate experimentally observed olefin selectivities for all three fuels on 80 and 45 ppi monoliths at 2, 4, and 6 SLPM quite well. These results support the hypothesis that a majority of the observed olefins are made through gas-phase chemistry.     

A soot formation embedded reduced reaction mechanism for diesel surrogate fuel

A reduced diesel surrogate fuel chemical reaction mechanism of n-heptane/toluene was developed, the reduced mechanism (referred as the “THU mechanism”) includes 60 species and 145 reactions, and it contains soot formation reactions. The THU mechanism was developed from the existing n-heptane/toluene mechanism (70 species and 313 reactions) of Chalmers University of Technology (referred as the “CTH mechanism”). SENKIN and XSENKPLOT were used to analyze the important reactions and species during n-heptane, toluene oxidation and soot formation processes to formulate the reduced mechanism. Ignition delays of n-heptane and toluene predicted by the THU mechanism match well with the CTH mechanism and shock-tube test data under different conditions. The THU and CTH mechanisms also show similar soot concentration prediction. The global reaction of diesel fuel decomposed into n-heptane and toluene with mole fraction 7:3 was built to accelerate the decomposition and advance ignition timing. Kinetic constants of soot oxidation reactions were adjusted to reduce the soot oxidation rate. The THU mechanism was coupled with the KIVA-3V Release 2 code to model diesel combustion processes in the constant-volume combustion vessel and optical diesel engine of Sandia. The predicted ignition delay, in-cylinder pressure and heat release rate match the experimental results well. The predicted spatial and temporal soot concentration distributions have similar trend with the experiments.