An overview on combustion researches by means of microgravity fields is described. The first microgravity experiment was conducted by Prof. Kumagai for a fuel droplet burning and its sphere flame strongly impressed combustion researchers in the world. Since then Japan is in a position of leadership in this field. Many marvelous research achievements in recent years, what piece of Japanese research is expected by using the space-station from now on, and our scientific attitude or principle we should prepare with experiments in the space-station near at hand, are discussed.

A Japan-U.S. joint research on liquid-fuel combustion experiment titled “Elucidation of Flame Spread and Group Combustion Excitation Mechanism of Randomly-distributed Droplet Clouds (Group Combustion)” is being prepared for the first combustion experiment in the Japanese Experimental Module “KIBO” on ISS. The objective of this experiment is to obtain flame-spread data to improve the percolation model to well describe the group combustion excitation through the flame spread. This research is expected to contribute to bridging the gap between the simplified droplet combustion in microgravity and complicated spray combustion. Percolation theory predicts that a transition occurs at the critical dropletnumber density between partial combustion and group combustion of fuel droplets in flame spreading over randomlydistributed droplet clouds. This transition is possibly identical to the transition between incomplete combustion and stabilized combustion in practical combustors. Application of percolation theory to flame spread of droplet clouds requires information on the flame spread between droplets. The present space experiments will investigate the effect of droplet interaction and free droplet on flame spread in a scale of droplet size order and effect of radiation heat loss, which becomes important in flame spread from a droplet cluster to another droplet cluster. This paper reports the background and current status of the preparation of the experiment.

The new atomization concept derived from the drop tower experiments claims that a liquid issued from an injector is destabilized by itself in a deterministic way. Therefore, we can predict where the jet disintegrates or onsets turbulent atomization, if the undergoing self-destabilizing mechanism is found. On the other hand, all conventional theories assume the presence of ” unknown” nozzle flow noises as the sources of excited jet instabilities. Therefore, what we can know is only the possibility of excitation of a specific instability. If an observed convectively unstable wave is produced by an “unknown” mechanism operating within the liquid jet region, one might attribute the instability excitation to nozzle flow noises. Our investigations show that this is the case in the conventional theories because they have overlooked the effects of nonlinear phenomena occurring in a downstream jet region. In the present paper, how various self-destabilizing mechanisms are being found and how new knowledge derived from these studies may be integrated into a turbulent atomization sub-grid model for LES are described, along with the space experiment purpose.

Two research projects relating to solid material combustion in microgravity are introduced with their scientific background and expected outcome. Both of them include space flight experiments in ISS/Kibo, which will provide key information of these projects, map of limiting condition for ignition and extinction. The first project called as “Solid Combustion” aims to clarify the differences of ignition limits of overloaded wire and extinction limit of spreading flame over solid material in normal and micro-gravity and to provide scientific understanding of the differences. The second project called as “FLARE (Flammability Limit at Reducing Gravity)” attempts to propose new fire safety standard for materials intended for use in habitable environment in spacecraft after the combination of the scientific understanding of the Solid Combustion project and extended data on practical materials and additional effective parameters to be taken in the second project. In the present article the overview of those projects and examples of tentative progress of the research are described and finally it is emphasized that the extension of fundamental knowledge on combustion is the crucial element to attain the goal of the projects.

Low-speed counterflow flame experiments will be conducted at Japanese Experiment Module (JEM), Kibo, in the International Space Station (ISS) for constructing unified flammability theory. The subject was selected as an experimental theme in the latter half of the second utilization planning phase of Kibo. This paper presents theoretical background of the present topic and the current progress of experiments and numerical simulations. First, fundamental theories on combustion limits of 1-D planar propagating flames and flame balls are described. Previous works which examined the effects of stretch rate and Lewis number on flames were introduced. The goal of this study is to construct unified theory of combustion limits of deflagration flames and flame balls. Recent microgravity experiments using the airplane, and relevant computations were presented. Both experiments and computations showed the existence of ball-like flames near limits in counterflow field of CH₄/O₂/CO₂ and CH₄/O₂/Xe mixtures.

As a fundamental study on the droplet-interaction effect in the spontaneous ignition of fuel spray, spontaneous ignition of an n-decane droplet pair rapidly inserted into hot air was experimentally studied in the ambient temperature range where the low-temperature oxidation reactions are active. In order to exclude the effect of buoyancy, the experiments were performed in microgravity. Two droplets suspended on 14 μm SiC fibers initially at room temperature were inserted into a hot furnace. Droplet diameter was 1 mm. First, temperature near the droplets were measured by thermocouples, and cool-flame and hot-flame ignition delays were evaluated. Cool-flame ignition delay increased with decreasing inter-droplet distance. This is supposed to be mainly caused by the mutual cooling effect. On the other hand, the duration between coolflame appearance and hot-flame appearance (second induction time) decreased with decreasing inter-droplet distance. This is supposedly because of higher cool-flame temperature caused by the enhanced fuel supply through duplicated fuel sources. Next, density field around a droplet pair was qualitatively observed by interferometry with a high-speed camera, and the locations of cool-flame and hot-flame appearances were detected. Cool flame appeared on the outer side of the droplet pair, and hot flame appeared on the inner side of the droplet pair, which corresponds with the discussion on ignition delays.

The flame characteristics and structure in pool fires are of current interest because of both fundamental curiosity and practical concerns related to flame size and shape, smoke production, radiant emission and fuel regression. These parameters are important in model for various types of fires. The pool fire is mainly controlled by buoyancy, so that the characteristics of pool fire such as flame height and puffing (oscillation) frequency change depending on both scale and gravitational environments. In this paper, the effects of scale in normal gravity on 1) fuel regression rate, 2) puffing frequency and 3) flame height are firstly discussed related to the buoyancy controlled flame structure. In recent years, the human activity range is expanding to varied gravity environments. A fire safety assessment must be assessed under the different gravity level to normal. So that the effect of gravity on flame behavior which varies depending on buoyancy is secondary discussed related to flow structure around the flame.

With use of the ejector effect of tangentially injected high-velocity gas streams, a self-recirculation type tubular flame burner has been newly designed and its NOx emission characteristics have been determined. The burner has eight tangential injectors, each of which a recirculation pass is connected. The extent of recirculation of the burned gas is varied by changing the number of the recirculation path opened. Results show that with an increase of the number of the recirculation path opened, the NOx emission is decreased. For the stoichiometric premixed combustion, the NOx concentration without recirculation is 84 ppm, which is reduced to 46 ppm when all the eight recirculation passes are opened. For the stoichiometric rapidly-mixed combustion, in which methane is injected through two injectors while air is injected through any of the other six injectors, the NOx concentration without recirculation is 75 ppm, which is reduced to 31 ppm for the other six recirculation paths opened. This value is further reduced to 9 ppm at an overall equivalence ratio of 0.8 (the air excess ratio of 1.25). The temperature distribution is determined along the burner axis and it is found that with an increase of the number of the recirculation path opened, the high temperature zone shrinks and its distribution is flattened below 1200°C for premixed and rapidly-mixed combustion. This suppression of high temperature zone may lead to significant reduction of NOx emission in the present burner.

Performance and accuracy of several time integration methods for chemical reaction equations are comprehensively investigated, aiming at an efficient reacting flow simulation with large detailed chemical kinetics. In this study, a modified CHEMEQ2, dynamic multi-timescale method (MTS), and two Runge-Kutta-based methods (R-K-Chebyshev and R-K-Fehlberg) are considered as currently available and possible explicit time integration methods, while VODE is used as a reference implicit time integration method. Ignition problems for three hydrocarbon systems (CH₄, n-C₇H₁₆, and n-C₁₀H₂₂) and an internal combustion engine model with n-C₇H₁₆ are simulated. The results for both problems show that the modified CHEMEQ2 shows the best performance for all the conditions in not only the accuracy, but also the robustness, while MTS gives less performance and the two Runge-Kutta-based methods cannot work even for the ignition problems with hydrocarbon systems. It is also found that the two explicit time integration methods (CHEMEQ2 and MTS) reduce nearly 20-100 times computational time compared to the reference implicit time integration method.

A high-sensitivity laser absorption spectroscopic technique has been employed for measurements of the stable carbon isotope ratio of carbon dioxide (CO₂), ¹³CO₂/¹²CO₂, in automobile exhaust. The amounts of ¹²CO₂ and ¹³CO₂ were detected through wavelength modulation spectroscopy using a distributed feedback laser diode in the 2-μm wavelength region. A 0.15-L one-pass optical cell with an effective optical path length of 0.5 m was used for continuous measurements of automobile exhaust, with the intention of maintaining the residence time to less than 5 s. In addition, a 0.9-L Herriott-type multi-pass optical cell with an effective optical path length of 29.9 m was used for precise measurements of the stable carbon isotope ratio of CO₂ in automobile exhaust, which was collected in a sampling bag. Continuous measurements of the stable carbon isotope ratio were performed with a reproducibility of 0.28 ‰ (1σ) in 14.8 % CO₂ for every 1-s signal measurement without signal integration over an 1830 s period. The feasibility of the developed system was evaluated for continuous measurements of the CO₂ stable carbon isotope ratio in automobile exhaust. Resultantly, we successfully performed continuous measurements of the stable carbon isotope ratio of CO₂ in automobile exhaust and found that the stable carbon isotope ratio of CO₂ changed in accordance with the driving conditions.

The damage caused by an accidental gas explosion, for example the maximal blast pressure, is strongly influenced by the flame propagation velocity during the explosion. The flame propagation during an explosion is significantly affected by flame instability, and it is known that the flame propagation velocity has certain scale dependency. This study investigates scale effects of diffusive-thermal and hydrodynamic instability by numerically solving the Sivashinsky equation. The following four conditions are considered: a purely diffusive-thermally unstable flame, a purely hydrodynamically unstable flame, a flame that is diffusive-thermally stable but hydrodynamically unstable, and a flame that is unstable both diffusive-thermally and hydrodynamically. It is found that diffusive-thermal instability mainly influences the flame wrinkle structure of a specific wavelength, while hydrodynamic instability influences the largest structure. Thus, hydrodynamic instability shows scale dependency. The fractal dimensions of the flames, which can be used to estimate the flame propagation velocity during an explosion, are computed by two different methods: a Fourier analysis and a method based on the scale dependency of flame propagation velocity. The both methods yield consistent results, and it is found that the fractal dimension mainly depends on the thermal expansion ratio; its dependency on the Lewis number is rather weak.