An Experimental and Modeling-Based Study Into the Ignition Delay Characteristics of Diesel Surrogate Binary Blend Fuels

Abstract

This study examines the combustion characteristics of a binary mixture surrogate for possible future diesel fuels using both a single-cylinder research engine and a homogeneous reactor model using detailed chemical reaction kinetics. Binary mixtures of a normal straight-chain alkane (pure n-hexadecane, also known as n-cetane, C16H34) and an alkyl aromatic (toluene, C7H8) were tested in a single-cylinder research engine. Pure n-hexadecane was tested as a baseline reference, followed by 50%, 70%, and 80% toluene in hexadecane blends. Testing was conducted at fixed engine speed and constant indicated load. As references, two conventional petroleum-based fuels (commercial diesel and U.S. Navy JP-5 jet fuel) and five synthetic Fischer-Tropsch-based fuels were also tested. The ignition delay of the binary mixture surrogate increased with increasing toluene fraction and ranged from approximately 1.3 ms (pure hexadecane) to 3.0 ms (80% toluene in hexadecane). While ignition delay changed substantially, the location of 50% mass fraction burned did not change as significantly due to a simultaneous change in the premixed combustion fraction. Detailed chemical reaction rate modeling using a constant pressure, adiabatic, homogeneous reactor model predicts a chemical ignition delay with a similar trend to the experimental results but shorter overall magnitude. The difference between this predicted homogeneous chemical ignition delay and the experimentally observed ignition delay is defined as the physical ignition delay due to processes such as spray formation, entrainment, mixing, and vaporization. On a relative basis, the addition of 70% toluene to hexadecane causes a nearly identical relative increase in both physical and chemical ignition delay of approximately 50%. The chemical kinetic model predicts that, even though the addition of toluene delays the global onset of ignition, the initial production of reactive precursors such as HO2 and H2O2 may be faster with toluene due to the weakly bound methyl group. However, this initial production is insufficient to lead to wide-scale chain branching and ignition. The model predicts that the straight-chain alkane component (hexadecane) ignites first, causing the aromatic component to be consumed shortly thereafter. Greater ignition delay observed with the high toluene fraction blends is due to consumption of OH radicals by toluene. Overall, the detailed kinetic model captures the experimentally observed trends well and may be able to provide insight as to the relationship between bulk properties and physical ignition delay.