パーコレーションモデルを用いた二次元液滴群の燃え広がり特性に関する検討. A study on flame-spread characteristics of two-dimensional droplet cloud using percolation model

Transcription

1 28A01 パーコレーションモデルを用いた二次元液滴群の燃え広がり特性に関する検討 瀬尾健彦,Herman Saputro, 三上真人 ( 山口大 ), 菊池政雄 (JAXA) A study on flame-spread characteristics of two-dimensional droplet cloud using percolation model Takehiko SEO, Herman SAPUTRO, Masato MIKAMI (Yamaguchi Univ.), Masao KIKUCHI (JAXA) Flame spread in fuel spray plays an important role in heat release in the initial stage and in the stability of flame in the internal combustion engine. In order to improve understanding of the flame-spread mechanism, the flame spread along evenlyspaced droplet arrays has been investigated experimentally 1), theoretically 2) and numerically 3). In spray combustion, however, fuel droplets are randomly distributed in the combustion chamber. The findings of researches on flame spread along droplet arrays have not been well utilized for elucidation of spray combustion. The present study investigates the flame-spread characteristics of randomly distributed two-dimensional droplet cloud with the equal initial droplet diameter using a percolation model based on the findings of microgravity experiments on flame spread along droplet arrays. 2. Flame-Spread Model and Calculation Results Umemura et al. have applied the percolation theory to the elucidation of spray combustion 4). In view of the general site percolation, only when two particles exist on adjacent lattice points, these two particles are connected. So, the inter-particle distance is meaningless. When percolation theory is applied to spray combustion, a droplet is characterized as a particle. In flame spread, the dimensionless droplet spacing S/d 0 is important, where S is the droplet spacing and d 0 is the initial droplet diameter. Therefore, the flame-spread-limit distance is the threshold of the connection shown in Fig. 1. In this study, the local-flame-spread-limit distance (S/d 0 ) limit is 14, which was investigated in microgravity at room temperature 1). Droplet number density is characterized by mean droplet spacing (S/d 0 ) m. Figure 2 shows the typical distributions of flames for different mean droplet spacing. When the droplet number density is small enough, the flame spread terminates on its way to the sides of the lattice, leading to incomplete combustion of the droplet cloud. As the droplet number density is increased, the flame area increases. And finally, the flame reaches all the sides of the lattice and the group combustion occurs. Figure 3 shows flame spread rate based on all droplet ignition time vs. (S/d 0 ) m. For (S/d 0 ) m < 8, flame-spread rate is almost constant. In case of near 11.4, the critical mean droplet spacing, flame-spread rate has a wide range of values. As (S/d 0 ) m increased over the critical mean droplet spacing, the number of cases that the flame spread to droplets on the top side decreased. (a) (b) (c) Fig. 2 Unburned droplet Burned droplet Typical distributions of flames for different mean droplet spacing: (a) (S/d 0 ) m =12.0, (b) (S/d 0 ) m =11.4, (c) (S/d 0 ) m =8.0 Unburned droplet (d 0) (a) (b) y Fig. 3 Flame spread rate based on all droplet ignition time (S/d 0) limit (c) Burning droplet (d) vs. mean droplet spacing (S/d 0 ) m 1) M. Mikami, H. Oyagi, N. Kojima, M. Kikuchi, Y. Wakashima and S. Yoda, Combust. Flame 141 (2005) ) A. Umemura, JSME Trans. B 68 (2002) ) M. Kikuchi, T. Arai, S. Yoda, T. Tsukamoto, A. Umemura, M. Uchida and T. Niioka, Proc. Combust. Inst. 29 (2002) Fig. 1 Flame spread calculation procedure 4) A. Umemura, S. Takamori, Combust. Flame 141 (2005)

2 28A02 FLEX-2J における燃料液滴列火炎燃え広がり実験 野村浩司, 須藤久美子 ( 日本大学 ), 三上真人 ( 山口大学 ), 菊池政雄 (JAXA), Daniel L. DIETRICH(NASA) Microgravity Experiments of Flame Spread along a Fuel Droplet Array in FLEX-2J Campaign Hiroshi NOMURA, Kumiko SUTO (Nihon Univ.), Masato MIKAMI (Yamaguchi Univ.), Masao KIKUCHI (JAXA.), Daniel L. DIETRICH (NASA) A fuel droplet array is often employed as a fundamental experiment model in spray combustion researches. In most of investigations focusing on fuel droplet array combustion, the droplet array consists of fixed droplet. However, fuel droplets freely move in real spray combustion. It is important to study effects of the fuel droplet movement on the droplet array combustion. In the former work, one of the authors investigated droplet motion induced by a flame spreading from a fixed droplet to a movable droplet using microgravity conditions. In the present work, a fuel droplet array with plural movable droplets is employed to investigate flamespreading between movable droplets. New technique was employed to realize a fuel droplet array including plural movable droplets. Microgravity experiments will be conducted in the Flame Extinction Experiment 2 (FLEX2) campaign using the Multi-User Droplet Combustion Apparatus (MDCA) of NASA installed into the Combustion Integrated Rack (CIR) on the International Space Station (ISS). In this presentation, a new droplet array suspension system which Japanese science team fabricates is explained. 2. Experimental Apparatus and Procedure Figure 1 illustrates the employed experimental model of a linier fuel droplet array. Three fixed droplets and 3-10 movable droplets are dispensed along a thin SiC fiber which allows suspended droplets to move freely in the fiber direction. The droplet closest to the igniter is referred to as the first droplet. The three fixed-droplets are employed in order to suppress the influence of disturbance due to the ignition process of the first droplet on the motion of movable droplets. Droplet array suspension system consists of a beaded SiC fiber of 78 m in diameter and 235 mm in length and a fiber support arm. The beaded SiC fiber has two experiment regions in order to minimize the number of the fiber arm replacement during on-orbit experiments. Each experiment region has three ceramic beads for fixed droplets and 22 anchor points for movable droplets. Fixed droplets are generated on the ceramic beads. Center-to-center distances of the ceramic beads are set equal at 4.5 mm. The anchor point consists of two etched spots which are slightly thinner than the original SiC fiber as Ceramic beads Weak back light SiC fiber Igniter Fixed droplets Movable droplets shown in Fig. 2. focused CO 2 laser light. The etched spots are irradiation mark of A movable droplet is generated between the two etched spots and very weakly anchored at predetermined location. From results of the preliminary experiments, it was confirmed that anchor points do not increase the drag force between droplets and a fiber. points is equally 2.0 mm. bead and the first anchor point is 2.5 mm. Spacing of anchor Spacing between the third ceramic The beaded-fiber fixed on the fiber support arm is installed into the MDCA on the orbit. A droplet array is formed alternating between droplet generation and slide motion of the fiber support arm in the beaded fiber direction. Ignition of the first fixed droplet caused by the igniter initiates flame-spreading along a fuel droplet array. Behaviors of a spreading flame and movable droplets are recorded with an ultra-violet video camera (LLL-UV camera), a high speed video camera (HiBMS Camera), and an overview video camera (MDCA Camera). From the recorded images, temporal variations of the movable droplet positions and the leading edge position of a spreading flame are measured. Fig. 1 Experimental model. Droplet SiC fiber Etched spot Fig. 2 Schematic of anchor point. Experiments are performed at atmospheric pressure and room temperature. n-heptane is employed as a liquid fuel. Initial diameter of all droplets is regulated within the range of 1.0 mm ± 5%. 3. Conclusion Design of the droplet array suspension system for FLEX2 has been completed. Beaded SiC fibers were fabricated

3 28A03 燃焼限界の統一理論構築のための極低速対向流火炎実験 中村寿, 小林友哉, 髙瀬光一, 手塚卓也, 長谷川進, 丸田薫 ( 東北大学 ),Roman FURSENKO, Sergey MINAEV(SB RAS; Far Eastern Federal University), 勝田真登, 菊池政雄 (JAXA) Low-flow-Speed Counterflow Flame Experiments for Modeling Unified Theory of Combustion Limit Hisashi NAKAMURA, Tomoya KOBAYASHI, Koichi TAKASE, Takuya TEZUKA, Susumu HASEGAWA, Kaoru MARUTA (Tohoku Univ.), Roman FURSENKO, Sergey MINAEV (SB RAS; Far Eastern Federal University), Masato KATSUTA, Masao KIKUCHI (JAXA) Carbon capture and sequestration (CCS) gains increasing attention for the prevention of the global warming. Power generation through oxy-fuel combustion with CCS, which is expected to cover large portion of the future demand for electricity, is one of the promising, realistic CO 2 -free energy solutions. To further develop oxy-fuel combustion technology, characteristics of combustion with a fresh mixture containing large amount of CO 2, which is thermally participating media, need to be clarified. For establishing stable oxy-fuel combustion and its control, flammability limit of oxy-fuel combustion is one of the most important issues in the basic combustion physics to be clarified. It was concluded that flammability limit of thermally nonparticipating mixture is induced by the radiative heat loss from the burned gas to the ambient, that is, radiative heat losses from H 2 O and CO 2 in the burned gas to the circumstances. Meanwhile, a combustible fresh mixture which contains thermally participating gas, such as CO 2, should be strengthened by its own thermal nature that may lead to significant extension or the vanishment of the flammability limit. This would be due to the reabsorption of the radiation energy by the fresh mixture. The space experiments of flame balls by Ronney et al. 1) showed that flammability limit of flame balls do exist for fresh mixtures with large amount of CO 2. Hence, we need to revisit this paradox from the fundamental viewpoint of the combustion science. For this, low-speed counterflow flame technique will be applied for combustion with large amount of radiation participating gas in the present study. Gravity-free, long duration experiments only available in the space will provide a clear picture of this unresolved issue. On top of that, highly reliable computational techniques for the limit prediction, which is also of use for practical combustion technology, would be expected to be developed. Microgravity experiments using low-speed counterflow technique will be conducted in ISS/JEM and preliminary microgravity experiments using an airplane were conducted. This paper presents highlights of the preliminary experiments. 2. Experimental method A pair of counterflow burners with inner diameters of 3.0 cm was set in a combustion chamber. The distance between burner nozzles was kept at 3.0 cm. Methane was used as a fuel. In every experiment, the stretch rate, a, was maintained at a corresponding constant value and the equivalence ratio,, was gradually decreased. A CCD camera with an image intensifier was used to record flame behaviors. CO 2, krypton and xenon were chosen as an inert gas to investigate the effects of Lewis number and radiation on flammability limit. Microgravity environment (20 sec) was provided by a parabolic flight of MU300 operated by Diamond Air Service. 3. Results Figure 1 shows flame images with xenon as an inert gas at a = 3.2 s -1. Twin planar flames were observed at high (Fig. 1a). The distance between the twin planar flames decreased with the decrease of and a single planar flame was observed (Fig. 1b). Immediately after the extinction of the planar flame, a ball-like flame was observed (Fig. 1c) and extinguished with a further decrease of. Further analysis was conducted using numerical simulation 2). In a lower stretch rate (a = 1.7 s -1 ), cellar flames were observed 3). (a) = 8 Time Time (b) = 0.39 (c) = 0.36 (a) (b) (c) Fig. 1 flame = 7images with xenon = 0.38 as an inert at a = s -1 : (a) twin planar flames ( = 7); (b) single planar flame ( = 0.38); (c) ball-like flame ( = 0.35). 1) P. D. Ronney, M. S. Wu, K. J. Weiland and H. G. Pearlman: AIAA J. 36 (1998) ) K. Takase, X. Li, H. Nakamura, T. Tezuka, S. Hasegawa, M. Katsuta, M. Kikuchi, K. Maruta: Combust. Flame 160 (2013) ) R. Fursenko, S. Minaev, H. Nakamura, T. Tezuka, S. Hasegawa, K. Takase, X. Li, M. Katsuta, M. Kikuchi, K. Maruta: Proc. Combust. Inst. 34 (2013)

4 Ignition delay τ s 28A04 燃料液滴群の燃焼モードに及ぼす液滴直径と液滴間隔の影響 多田純輝, 山下宏顕, 板井満生, 森永裕太, 片岡秀文, 瀬川大資, 角田敏一 ( 大阪府立大 ) Effects of Droplet Diameter and Spacing on Combustion Mode of Fuel Droplets Clusters Junki TADA, Hiroaki YAMASHITA, Manaru ITAI, Yuta MORINAGA, Hidefumi KATAOKA, Daisuke SEGAWA, Toshikazu KADOTA (Osaka Prefecture Univ.) In the chemical conversion of energy, combustion of liquid fuels plays an outstanding role. There is still lack of information on combustion of droplet clusters, which bridges between knowledge on droplet combustion and on spray combustion, and additional experiments are needed. This has motivated us to be involved in the experimental study on the evaporation, autoignition and combustion of clusters of multiple droplets. A monodispersed suspended-droplet cluster (MSDC) model has been developed, and autoignition and combustion behavior of the MSDC model in a quiescent high-temperature air has been observed under the microgravity condition 1)-2). Recent results on the effects of droplet diameter and spacing on the beginning combustion mode of fuel droplets clusters are presented in this report. 2. Experimental apparatus and method In this study, microgravity is realized with a drop tower in our university. The drop tower is 10 meters high, and microgravity environment is realized for 1 seconds. The experimental apparatus consists of an electric furnace, a sample transfer unit, a high-speed camera for observation and a sequence controller (Fig. 1). The 3D MSDC model (Fig. 2) is realized hexagonal closest packing structure, with 13 n-eicosane droplets. The distance between neighboring droplets is defined as the droplet spacing s. The droplets are suspended on SiC fibers of the diameter 14 m x 3. In this study, droplet spacing is varied from 3.0 to 11.0 mm, and the initial diameter of each droplet is set to 0 or 0.50 mm. The furnace temperature is set to 973 K. 3. Results and discussion Two types of combustion mode are formed just after ignition. One is named a group flame (Fig. 3a), which envelops the whole model, and the other is named individual flames (Fig. 3b), which envelop each droplet. Figure 4 shows the effects of droplet spacing on the beginning combustion mode and the ignition delay. The ignition delay shows an increasing trend with decreasing the droplet spacing, and the trend shift may occur around the transition region of the beginning combustion mode which exists from to 9 mm of the droplet spacing. 4. Summary The combustion mode just after ignition changes by the droplet spacing. The ignition delay has a tendency to increase with decreasing the droplet spacing. 1) D. Segawa, M. Yoshida, S. Nakaya, and T. Kadota, Microgravity Sci. Tech. 17-3: (2005)23 2) D. Segawa, M. Yoshida, S. Nakaya, and T. Kadota, Proc. Combust. Inst. 31: (2007)2149 (a) Group flame(s = 3 mm) Fig. 3 Combustion mode. (b) Individual flame(s = 11 mm) Group flame Individual flame Microgravity n-eicosane 13 droplets d : 0.50 mm s : 5.0 mm T : 973 K Fig. 1 Experimental apparatus. Fig. 2 The MSDC model Droplet spacing s mm Fig. 4 Effects of droplet spacing on ignition delay

5 28A05 微小重力下における単一液滴の燃焼挙動に与える二酸化炭素を含む高圧雰囲気の影響 瀧本理仁, 小林芳成, 中谷辰爾, 津江光洋 ( 東京大学 ), 瀬川大資 ( 大阪府立大学 ) Influence of High-Pressure Atmosphere including Carbon Dioxide on Burning Behavior of a Single Fuel Droplet under Microgravity Rihito TAKIMOTO, Yoshinari KOBAYASHI, Shinji NAKAYA, Mitsuhiro TSUE (Tokyo Univ.), Daisuke SEGAWA (Osaka Pref. Univ.) Single fuel droplet combustions in microgravity have been widely conducted to obtain fundamental knowledge of spray combustions that are used in internal-combustion engines such as gas turbines and diesel engines. To simulate the droplets in fuel spray using larger droplets, the effect of natural convection must be removed. Free fall method using a drop tower to obtain microgravity condition is generally used. In droplet combustions, the quasi-steady assumption of d 2 -law is widely known. However, there are some cases that don t obey the d 2 -law according to recent researches. In this study, the influence of high-pressure atmosphere and atmosphere composition on the instantaneous burning rate of a single fuel droplet will be observed. 2. Experimental Apparatus Experiments were conducted in a free falling package with a drag shield at drop tower in the University of Tokyo. The schematic of the free falling package is shown in Fig.1. The height of the tower is 20m, and the gravitational level created by this tower is about 10-4 of normal gravity for 1.4 seconds. The instantaneous burning rate was calculated from the measured surface area of a droplet. 3. Results and Discussion The histories of instantaneous burning rate of ethanol at various pressure are shown in Fig. 2. It can be seen that the overall burning rate increases as the ambient pressure rises. For the case of MPa, the instantaneous burning rate slightly decreased with time. This might be due to the water condensation. This phenomenon was not observed for 1.0MPa, where the effect of the radiation from soot was remarkable. Figure 3 shows the comparison of the histories of the instantaneous burning rate between air and CO 2 -rich ambience at MPa. The overall burning rate decreased in CO 2 -rich ambience. Also, the decrease of the instantaneous burning rate became larger in CO 2 -rich ambience. Instantaneous burning rate [mm 2 /s] MPa MPa 1.0MPa time [sec] Fig.2 Histories of instantaneous burning rate at high pressure condition 0.8 Fig.1 Schematic of the experimental setup 1.Combustion chamber, 2.Quartz fiber, 3.Heater, 4.Servo motor, 5.Back light, 6.Mirror, 7.High speed camera, 8.Video camera, 9.Ignition controller, 10.Battery, 11.UPS, 12.Pressure sensor, 13.Vacuum pump, 15.Stirrer, 16.Gas, 17.Syringe Instantaneous burning rate [mm 2 /s] Fig.3 Histories of instantaneous burning rate at MPa in CO 2-rich 4. MPa air 0 MPa CO2 60% time [sec] condition 1) S.Nakaya,D.Segawa,T.Kadota,Y.Nagashima,T.Furuta, Proceedings of the Combustion Institute 33 (2011) ) A.Yozgatiligil,S.Park,M.Y.Choi,A.Kazakov,F.L.Dryer, Proceedings of the Combustion Institute 31 (2007)