Search

Experimentally measured values of the laminar flame speed (SL) are reported for the primary reference fuels over a range of unburned-gas temperatures (Tu) spanning from room temperature to above 1,000 K, providing the highest-temperature SL measurements ever reported for gasoline-relevant fuels. Measurements were performed using expanding flames ignited within a shock tube and recorded using side-wall schlieren imaging. The recently introduced area-averaged linear curvature (AA-LC) model is used to extrapolate stretch-free flame speeds from the aspherical flames. High-temperature SL measurements are compared to values simulated using different kinetic mechanisms and are used to assess three functional forms of empirical SL–Tu relationships: the ubiquitous power-law model, an exponential relation, and a non-Arrhenius form. This work demonstrates the significantly enhanced capability of the shock-tube flame speed method to provide engine-relevant SL measurements with the potential to meaningfully improve accuracy and reduce uncertainty of kinetic mechanisms when used to predict global combustion behaviors most relevant to practical engine applications.

Homogeneous and inhomogeneous ignition modes of n-heptane were studied using high-speed imaging in a high-pressure shock tube (HPST). n-Heptane, a fuel with strong negative temperature coefficient (NTC) behavior, was mixed with 4%-21% oxygen in argon or nitrogen and ignited over a wide temperature range (700–1250 K) and at elevated pressures (> 10 atm). Ultraviolet (UV) images of OH* emission were captured through a sapphire shock-tube end wall using a high-speed camera and a UV intensifier. The current study demonstrates the capability to study auto-ignition modes using high-speed imaging in a high-pressure shock tube. Both homogeneous and inhomogeneous auto-ignition events were observed with the latter generally confined to intermediate temperatures and reactive n-heptane mixtures. We also observed that conventional sidewall diagnostic signals are, in many cases, sufficient to identify inhomogeneous ignitions that are not accurately modeled under the assumption of spatially uniform chemistry.

The initiation, propagation, and transition of the autoignition assisted spherical cool flame and double flame are studied numerically and experimentally using n-heptane/air/He mixtures under shock-tube experimental conditions over a wide range of temperatures. The primary goal of the current study is to understand the effects of the ignition Damkohler number, ignition energy, flame curvature, and autoignition-induced flow compression on the propagation of spherical flames to ensure the proper interpretation of shock-tube flame speed measurements at engine-relevant conditions. The results show that at high ignition Damkohler number, there are three different flame regimes, cool flame, double flame, and hot flame. The cool flame speed accelerates dramatically with the increase of ignition Damkohler number. In addition, it is found that the change of flame regime, low-temperature autoignition, flame stretch, and autoignition-induced flow compression result in a complicated non-linear dependence of flame speed on stretch. The results also reveal that the spherical cool flame has much lower Markstein length compared to the hot flame at T > 600 K. Moreover, it is found that both the autoignition assisted cool flame and the trailing hot flame front in the double flame can propagate much faster that the hot flame alone at the same mixture conditions, leading to a nonlinear dependence of flame speed on the mixture initial temperature. The simulated flame trajectories and the flame speed dependence on temperature agree qualitatively well with the shock-tube experiments. A quantitative criterion to ensure the accurate speed measurement of the cool and hot flame is proposed. The present study provides important physical insight and guidance for the flame speed measurement using a shock-tube at engine relevant conditions.

Recently, high-temperature flame speed measurement up to 900 K has been successfully realized at Stanford University using a conventional shock tube, assisted by laser ignition and chemiluminescence image systems. For a near stoichiometric iso-octane/O2/N2/He mixture at atmospheric pressure, their results show that while flame speed increases with increasing unburnt temperature in the low and high-temperature range, it decreases with increasing unburnt temperature in the intermediate temperature regime (750–800 K). The measurement time for the flame initiation and propagation is orders of magnitude shorter than the first-stage ignition delay time, implying that the unburnt mixture upstream of the flame is largely chemically frozen. Therefore, such non-monotonic behavior in the flame speed is unlikely to be induced by autoignition chemistry. To understand this interesting phenomenon, 1-D transient simulation of spherical flame initiation and propagation with different ignition energy has been performed and compared with the experiment data. It is shown that for the relatively low ignition energy 7.5 mJ, an over-driven flame regime characterized by flame speeds and stretch rates well above the steady-state values occurs in the intermediate temperature range. The chemical structure of an over-driven flame involves substantial fuel thermal decomposition in the preheat zone, followed by C1-C4 oxidation in the reaction zone. It is found that the non-monotonicity in the observed flame speed is strongly correlated with the occurrence of over-driven flames in simulations. These results demonstrate a new flame structure which can potentially explain the observed non-monotonic flame speed for iso-octane at high temperatures. Higher ignition energy or smaller ignition kernel is suggested in future experiments to mitigate such over-driven effects.

The average burning velocities S¯b of premixed, spherically expanding iso-octane flames are reported for flame radii from 0.75 to 1.75 cm between 400 and 900 K, near 1 atm. Measurements are perform...

High-temperature methane and propane laminar flame speed measurements were conducted behind reflected shock waves in a shock tube. A high-power Nd:YAG laser was used to spark-ignite the shock-heated gas mixtures and initiate laminar flame propagation. High-speed, OH* endwall imaging was used to record the propagation of the spherically expanding flames in time, and a non-linear stretch correlation was applied and used to determine the unburned, unstretched laminar flame speed. “Low-temperature” ( 750 K) flame speed results are presented for a propane/21% O2-47% N2-32% He mixture (ϕ = 0.8) at initial unburned gas conditions of 764–832 K, 1 atm. The high-temperature measurements fall between kinetic model predictions, but the kinetic model results show significant disagreement, highlighting the need for high-temperature flame speed validation data of this kind. We believe that these results represent the first laminar flame speed measurements conducted in a shock tube, and that the high-temperature results are the highest-temperature, 1-atm flame speed measurements available in the literature.

The constrained reaction volume (CRV) method for shock-tube experiments makes it possible to conduct chemical kinetics studies at nearly constant pressure while inhibiting remote ignition. The application of end-wall imaging revealed, however, that CRV experiments are susceptible to vertical stratification at the interface of the test and buffer gases. This work identifies gravity currents as the mechanism leading to the test gas stratification, providing both a theoretical development and experimental investigation of their behavior in a shock tube. Parametric studies are conducted with both the gate valve and stage-filled CRV methods using a novel beam-split laser absorption diagnostic. The speed of gravity-current-induced mixing in both gate valve and stage-filled CRV experiments is shown to depend on the molecular weight matching of the gases and the fill pressure. Mixing velocity in gate valve experiments is shown not to depend on the gate valve speed. In stage-filled experiments, mixing time is seen to be a strong function of the test gas length. Recommended practices for avoiding gravity-current-induced mixing and stratification are described, including gas density matching, utilizing double diaphragms, increasing test gas length in stage-filled experiments, and implementing a gas interface diagnostic.

A novel, multi-species, combined laser absorption/gas chromatography (GC) speciation diagnostic was used to quantify intermediate species present in the low-temperature oxidation of 1.0% n-heptane in 20.8% O2/Ar and 20.8% O2/0.21% CO2/Ar (equivalence ratio 0.53) at 760 K, 4.9 atm. Laser absorption techniques were used to measure initial fuel and time-resolved temperature, CO2, H2O, and C2H4. Sampled-gas GC analysis was used in conjunction with a variable-test-time shock tube facility to obtain quasi-time-resolved measurements of n-heptane, C2H4, CO, H2, C3H6, and CO2 in the same experiments. Measurements obtained using both techniques are compared to each other, and to initial results predicted by a detailed kinetic model using the experimentally measured pressure trace to constrain the model. Discrepancies between measured and predicted ignition delay times indicate the overestimation of three primary RO2 isomerization reaction rates. The three reaction rates were modified to improve agreement of modeled ignition delay times with the measurements. Final results produced using the modified mechanism are compared to the experimental results; the comparison shows close agreement between the two experimental measurement techniques, and measured species yields confirm the low-temperature reaction pathways that govern n-heptane decomposition and C2H4, CO, H2, and C3H6 production.

The rate constants for the reaction H + O2+ M→HO2+ M were investigated at elevated pressures from 12 to 33 atm using ignition delay time (IDT) measurements behind reflected shock waves in H2/O2/M mixtures with different collision partners M = Ar, H2O, N2 and CO2. The temperature and pressure ranges where the rate constants of the reactions H + O2+ M→HO2+ M and H + O2- > OH + O dominate the IDT sensitivity were selected as optimum test conditions using the detailed H2/O2 mechanism of Hong et al. (2011). The current study thus provides a quantitative and relatively direct method for determining the rate constants for H + O2+ M→HO2+ M using simple IDT measurements. The rate constants found are consistent with earlier studies, but with reduced uncertainties.

Simultaneous, dual-perspective, high-speed imaging is used to provide new insight into the morphology of flames generated at high unburned-gas temperatures behind reflected shock waves. Single-perspective, end-wall imaging left previous applications of the shock-tube flame speed method reliant on an assumption of axial symmetry when interpreting experimental results. Here, we report qualitative and quantitative observations from the first application of simultaneous side- and end-wall emission imaging to characterize the three-dimensional morphology of these flames. Side-wall imaging reveals that, while the expected flame symmetry is observed under static and relatively low temperature post-reflected-shock conditions, symmetry can break down at higher temperatures. These results reveal that the concentric regions of emission previously observed in flame experiments in the negative-temperature-coefficient ignition regime can be explained as the axial integration of emission through a distorted flame. Several physical mechanisms are evaluated towards the goal of identifying the underlying cause of flame distortion. The post-reflected-shock flow field and pressure-wave–flame interactions are both found insufficient to explain the results. Presently, not enough is known about the recently identified local-double-flame structure to assess the likelihood of its relevance in the present experiments. As such, while the dual-perspective imaging results provide important new insight into the morphology of high-temperature flames and inform the correct interpretation of previous experimental observations, the specific mechanism leading to the observed distortion remains a topic requiring further attention.