About this Journal Submit a Manuscript Table of Contents
Advances in Mechanical Engineering
Volume 2013 (2013), Article ID 928970, 8 pages
http://dx.doi.org/10.1155/2013/928970
Research Article

Experimental Study of Ignition over Impact-Driven Supersonic Liquid Fuel Jet

1Combustion and Jet Application Research Laboratory (CJARL), Department of Mechanical Engineering, Faculty of Engineering, Ubon Ratchathani University, Ubon Ratchathani 34190, Thailand
2Shock Wave Interdisciplinary Application Division, Institute of Fluid Science, Tohoku University, 2-1-1 Katahira, Sendai 980-8577, Japan
3School of Mechanical and Manufacturing Engineering, University of New South Wales (UNSW), Sydney, NSW 2052, Australia

Received 28 September 2012; Revised 30 December 2012; Accepted 1 January 2013

Academic Editor: Jianqiao Ye

Copyright © 2013 Anirut Matthujak et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract

This study experimentally investigates the mechanism of the ignition of the supersonic liquid fuel jet by the visualization. N-Hexadecane having the cetane number of 100 was used as a liquid for the jet in order to enhance the ignition potential of the liquid fuel jet. Moreover, the heat column and the high intensity CO2 laser were applied to initiate the ignition. The ignition over the liquid fuel jet was visualized by a high-speed digital video camera with a shadowgraph system. From the shadowgraph images, the autoignition or ignition of the supersonic liquid fuel jet, at the velocity of 1,186 m/s which is a Mach number relative to the air of 3.41, did not take place. The ignition still did not occur, even though the heat column or the high intensity CO2 laser was alone applied. The attempt to initiate the ignition over the liquid fuel jet was achieved by applying both the heat column and the high intensity CO2 laser. Observing the signs of luminous spots or flames in the shadowgraph would readily indicate the presence of ignitions. The mechanism of the ignition and combustion over the liquid fuel jet was clearly clarified. Moreover, it was found that the ignition over the supersonic liquid fuel jet in this study was rather the force ignition than being the auto-ignition induced by shock wave heating.

1. Introduction

The study of high-speed liquid jets was initiated regarding the impact of rain droplets on high-speed vehicles [1, 2]. This topic was subsequently extended to wide ranges of engineering applications including cavitation research [3, 4], jet cutting technology and material cleaning by jets, mining and tunneling by means of high-speed jet impingements [5, 6], fire extinguishers, diesel fuel injection, and their mixing for direct injection diesel engines [7, 8], direct injection gasoline engine [9, 10], and supersonic combustion ram jet engines [11, 12].

Recently, it is known well that increasing the jet speed into supersonic range may be beneficial in improving autoignition and combustion in combustion applications. In the applications to diesel engines [7, 8], high-pressure injection creates high-speed fuel spray velocity and is believed to promote to effectively atomize fuel droplets and hence to improve ignition and combustion efficiency. Such improvements would enhance fuel consumptions and reduce smoke emissions. It is expected that further increase in jet speeds, say to a supersonic speed, would improve the combustion process of, for example, direct injection diesel engine, gasoline direct injection engine, and supersonic combustion ram (SCRAM) jets engine.

A bow shock wave formed in front of supersonic jet tips significantly elevates the temperature there, which would improve ignition around the fuel jet. Although fundamental studies regarding the droplet breakup and following atomization and mixing are important research topics of combustion [13, 14]; these mechanisms are not explained thoroughly, in particular, in high-supersonic to hypersonic speed ranges, although, Field and Lesser [15] investigated experimentally and analytically supersonic liquid jet flows and predicted self-combustion or autoignition of liquid fuel jets at supersonic jet speeds, because partial autoignition took place in their liquid fuel jets injected even in ambient air.

In 1994, Shi [16] and Shi and Takayama [17] experimentally studied the autoignition of a supersonic diesel jet of 2-3 km/s accelerated by two-stage light gas gun. In this study, the autoignition was visualized by a double exposure holographic interferometry technique. From holographic images, autoignition of the diesel fuel jet at the jet velocity of 2 km/s in normal ambient condition was claimed to have occurred; the result needs to be carefully reexamined as the observations were very subjective. It was supposed that the heat generated by the leading edge shock was able to ignite the liquid diesel fuel jets based upon the gas dynamics theory.

Subsequently, more detail studies on the autoignition feasibility over the supersonic liquid fuel jet have been carried out by Pianthong’s study [18, 19]. Using shadowgraph optical system visualization, the autoignition of supersonic diesel fuel jet at the jet velocity of around 2 km/s was not found at normal ambient conditions, even though the air in test chamber was heated up to be around 110°C using a 600 w electric heater. This conflicts with Shi’s study.

In 2007, the argument of autoignition over supersonic fuel jets was concluded by Matthujak’s study [20, 21]. The autoignition over the fuel jets in the atmospheric air was carefully reexamined using a double exposure holographic interferometry and shadowgraph technique. In this study, it was also found that no autoignition took place in all liquid fuel jets injected into ambient air even though the estimated temperature and pressure at the jet tip across the shock wave to induce autoignition were sufficiently high. This study concluded that in high-speed liquid fuel jets experiments the jet speed alone is not sufficient to initiate the autoignition of the supersonic liquid fuel jet. The discrepancy between the experiments and the estimation is attributable to the unsuitability of the high temperature distribution around the jet, the mixing condition, the air-fuel ratio, and the ignition delay time.

Although, the autoignition over the supersonic liquid fuel jet has been concluded that it does not occur at the ambient condition by Pianthong’s and Mutthujak’s study, no study has presented, visualized, and described the ignition over the liquid fuel jet so far. Hence, this study aims to initiate the ignition of the supersonic liquid fuel jets in order to obtain the initial condition of the ignition. Also, it aims to present, visualize, and describe the mechanism of the ignition over the fuel jet which has never been achieved in the previous studies. In order to enhance the ignition, the n-Hexadecane jet having a cetane number of 100 were used as liquid jets and impinged against the hot column at temperature of 650°C K for enhancing the vaporization process and heating up the air in the test chamber. Moreover, the high intensity CO2 laser was also applied to initiate the ignition.

2. Experimental Setup

In order to generate high-speed jets, Bowden and Brunton [22, 23] enhanced pressures in a liquid filling a container by a momentum transfer created with a sudden impingement of a high-speed projectile. Then shock waves are generated inside the container. The shock compression generates high pressures of a several GPa maintained for a few hundred microseconds, which is even more effective than adiabatic compression. Hence to obtain higher jet speeds, impact speeds should be as high as technically possible.

A light gas gun was designed [2426] for achieving higher projectile speeds. In principle, hydrogen or helium filled in a tube is compressed by the motion of a heavy piston, and the process of compression is performed in a mode between adiabatic compression and shock compression, and then gas pressures and temperatures are highly increased as the higher temperature is essential for obtaining higher sound speed and hence a higher projectile speed. This concept is basically the same as a shock tube performance to achieve higher shock Mach number.

A vertical two-stage gas gun (VTSLGG), as shown in Figure 1, was designed and manufactured in house [2426]. Its main structure consists of a 230 mm diameter and 1.5 m long high pressure gas reservoir, a coaxially positioned 50 mm diameter and 2.0 m long pump tube, a diaphragm section, a 15 mm diameter and 1.0 m long launch tube, a 15 mm diameter and 530 mm long pressure relief section or blast remover, and a container of liquid connected to a nozzle, and a test chamber or a dump tank into which high speed liquid jets are ejected. The test chamber has a cylindrical structure of 305 mm in diameter and 850 mm in length. Two rectangular observation windows made of 20 mm thick polymethylmethacrylate (PMMA) plates are attached on its side.

928970.fig.001
Figure 1: Vertical two-stage light gas gun.

The high pressure gas reservoir, the pump tube, and other massive metal pieces are suspended by a steel wire connected at the gas reservoir’s end flange and supported between two 300 mm diameter steel pillars. To make them more precisely vertical, the whole pieces are put on linear guides. The vertical movement is manually controlled with a winch. Precise positioning was achieved, which is important for various reasons: flow visualization; the piston motion control; and exact normal impacts of projectile on liquid surfaces.

Before each run, a heavy piston, as shown in Figure 2(a), made of high density polyethylene of 50 mm in diameter, 75 mm in length, and 130 grams in weight is inserted at the top end of the pump tube. A small space behind the piston is then evacuated in order to hold it at its starting position. Helium is fed in the pump tube and high pressure nitrogen in the high pressure gas reservoir. As soon as the pressurized nitrogen is fed into the space behind the piston, it starts to move down along the pump tube and is further accelerated. The launch tube is connected at the pump tube’s end by sealing with a Mylar diaphragm. A polycarbonate projectile of 15 mm in diameter, 20 mm in length, and 4.2 gram in weight is placed just behind the diaphragm as shown in Figure 2(b).

fig2
Figure 2: (a) Piston, (b) projectile, and (c) nozzle geometry.

At the launch tube’s end, a container of liquid made of high strength carbon steel is connected to a 0.7 mm diameter nozzle as shown in Figure 2(c). The projectile impacts the liquid surface vertically and hence drives a shock wave in liquid. Although it was not noticed in previous works, longitudinal and transversal waves are simultaneously generated in the metal container which are precursory to the main shock wave in liquid. These stress waves are spontaneously released into liquid forming oblique shock waves, which have very high overpressures enough otherwise to independently drive jets. It is then realized the importance of experimental proof of the contribution of these shock waves to jet formation [20, 21].

To prevent liquids from leakage, a 50 μm thick Mylar film is attached on the liquid container inlet and at the nozzle exit. A projectile ejected from the muzzle accompanies a precursory shock wave known as a muzzle blast. To prevent it from its advanced impact on the liquid surface, a blast relief section is connected at the muzzle as shown in Figure 3.

928970.fig.003
Figure 3: Ignition system of supersonic liquid fuel jet with ignition detection system.

Before restarting the experiment, at first the test section was removed from the initial position, and put down the whole tubes on the floor and then disconnect the launch tube. During this careful but tedious procedure, the linear guide system worked effectively to the reinstallation of the pump tube and others accurately in the right position. The total height of VTSLGG and supporting pillars is about 6.2 m as seen in Figure 1. The position of a container of liquid or the nozzle is shown in Figure 3. The operation procedure of the VTSLGG was described in [2426].

In order to enhance the ignition over supersonic fuel jets, n-Hexadecane having a cetane number of 100 was used instead of diesel fuel (cetane number of 52) as liquid fuel jets. The copper metal column having a diameter of 25 mm and 180 mm long was used as a heat column. The column surface at temperature of 650°C was heated up by the electric wire being spiraled around the column as shown in Figure 3. Heat from the column made the air temperature in the test chamber to be 450°C. The heat column was inserted in two metal cylindrical tubes for heat insulation. Moreover, the high intensity CO2 laser (Synrad, type j48-5SW), which the output energy is 50 watts, was applied to initiate the ignition. In order to precisely detect the ignition, not only the high-speed video digital camera with shadowgraph optical system, but also pressure sensor (Kistler type 5011) was used in this experiment.

3. Visualization

Figure 4 shows a high-speed digital video camera with shadowgraph optical arrangement used for visualization in this study. A flash lamp is used as light source. The source light is collimated passing through a circular slit and a concave lens. The flash light interval is 2 ms and rise time of 250 μs. The laboratory space is so limited that plane mirrors of diameter 150 mm, 200 mm, and 300 mm and a rectangular plane mirror of 240 mm × 600 mm were combined. Two paraboloidal schlieren mirrors of diameter 500 mm were used for collimating source light beam passing the test section area. A convex lens was used to focus the object image on the camera screen. The high-speed digital video camera is a Shimadzu HPV-1 at frame rate of 1,000,000 f/s, exposure time of 1/4 of interframe time, and the total number of images of 104. The test section has 20 mm thick acrylic windows, and its view field is 150 mm × 650 mm.

928970.fig.004
Figure 4: Arrangement of shadowgraph optical and high-speed digital video system.

4. Results and Discussion

In this study, n-Hexadecane was used as a liquid fuel jet. Its fire and explosion properties are shown in Table 1. Four experimental conditions were conducted and investigated as shown in Table 2. Using a high-speed digital video camera, it could record shadowgraph images with frame rate of 1,000,000 f/s with 104 frames in series. Therefore, it is possible to visualize the whole jetting process from its emergence from the nozzle orifice until it passes from the test scene. Since not all frames can be displayed in the paper, six sequential ones are selected to represent the stages in jet development, as shown in Figure 5. The physical width of each frame is always 150 mm, the width of the test window, which provides a scale in the vertical direction as well.

tab1
Table 1: Fire and explosion properties of n-Hexadecane used in the experiment.
tab2
Table 2: Experimental conditions.
fig5
Figure 5: Various stages of supersonic n-Hexadecane jet for (a) condition A, (b) condition B, (c) condition C, and (d) condition D.

Figure 5(a) shows various stages of supersonic n-Hexadecane jet for condition A. The n-Hexadecane jet shows the slimmest width and looks more elongated to be over 350 mm at 295 μs. Its averaged speed at 295 μs is 1,186 m/s which is a Mach number relative to the air of 3.41 in room temperature air. The jet motion is supersonic so that oblique shock waves are created over its top part and also the jet’s nodes. The multiple jet pulse being attributable to the high pressure generation created by wave interactions inside the nozzle is more obviously seen before the emerging time of 234 μs (further detail of the multiple jet pulse has been clarified in [20, 21]).

Base upon normal shock theory [23], temperature behind normal shock as shown in Figure 6 is estimated by; where is ambient temperature (300 K for this experiment), is temperature behind normal shock (K), is speed of sound (m/s), is jet or shock velocity (m/s), is adiabatic index (1.4 for air), is gas constant (287.058 J/kgK for air), and is Mach number.

928970.fig.006
Figure 6: Position of temperature calculation.

In this experiment, temperature behind the normal shock at the jet tip over supersonic n-Hexadecane jet tips at its velocity of 1,186 m/s is estimated by (1) to be 687.46°C, which is much higher than the autoignition point of n-Hexadecane fuel at 205°C as shown in Table 1. However, no sign of ignition took place even though the estimated temperature at the jet tip behind the shock wave was sufficiently high enough to induce ignition. Hence, it can be concluded that the jet velocity alone cannot be a sufficient condition of ignition or autoignition. There are four possible reasons for this. Firstly, the temperature and pressure conditions behind the leading edge shock wave are not uniform, and they decrease progressively behind it. Secondly, the normal shock wave is not retained along the jet body where autoignition usually starts in spray combustion. Therefore, normal shock wave theory overestimates the conditions, these existing at the jet tip only. When using oblique shock theory [27], temperature behind the oblique shock as shown in Figure 6 would be only 32.40°C, which is estimated by where is temperature behind oblique shock (K). Autoignition will not then readily occur as the ignition delay above would be about 2-3 s even at an equivalence ratio of 1.0 [18, 19]. Thirdly, the leading edge shock wave dissipates very quickly as the jet velocity reduces, with the temperature and pressure behind it being no longer maintained. Finally, the equivalence ratio in this case might not be appropriate for autoignition.

Therefore, a high intensity CO2 laser was shot continuously on the jet at 125 mm below from the nozzle tip to initiate ignition over the jet for condition B. Still no sign of ignition took place as shown in Figure 5(b). The jet development in Figure 5(b) was similar to that in Figure 5(a) even though the jet received the heat energy from CO2 laser. It can be implied that the high intensity CO2 laser alone could not initiate any ignition over the jet.

Further ignition enhancement over the jet was conducted by installing a heat column inside the test chamber. For condition C, the jet was injected and impinged against the heat column as shown in Figure 5(c). In order to extensively investigate the ignition after the jet being heated up from the heat column, the duration time of visualization was extended by changing the interval time of the high-speed digital video camera from 4 μs to 4 ms (millisecond). From the images, there was no ignition over the jet even though the temperature of heat column and air being 650°C and 350°C, respectively, was much higher than the autoignition point of n-Hexadecane at 205°C. It can be implied that the heat column alone could not initiate any ignition over the jet.

From the previous experiments for conditions B and C, the CO2 laser or the heat column alone could not initiate the ignition over the fuel jet. Hence, in condition D, both of the CO2 laser and the heat column were applied to enhance the ignition as shown in Figure 5(d). The ignition over the jet was successfully achieved in this condition. Signs of luminous spots or flames and readily indicated the presence of ignitions, were started to be observed from the elapsed time of 388 ms. Even though the ignition took place, its pressure being built up due to the ignition or the combustion was not high enough for being sensed by the pressure sensor. It was found that this ignition is force ignition, not autoignition, because the ignition was initiated or started at the passage position of the laser beam at 388 ms in Figure 5(d). The energy of laser beam provided the activation energy for ignition and combustion process. Hence, the laser beam is as an igniter for this experiment. Moreover, no ignition in conditions A, B, and C can imply that the autoignition cannot take place, and the ignition of high-speed jet cannot occur without the outer activation energy.

Therefore, based upon combustion theory for liquid fuel and condition D, the mechanisms of the ignition over the high-speed fuel jet for this study can be described as follows. After the fuel was injected from the nozzle, it was broken up from 4 ms to 64 ms in Figure 5(d), being so-called atomization process. Then, the liquid fuel was enhanced to be fuel vapor by heat from the heat column until 384 ms, being so-called vaporization process. During such process, the fuel vapor was mixed with the hot air to appropriate air/fuel ratio, being so-called mixing process. Finally, the air-fuel vapor was ignited by CO2 laser beam at 392 ms, being so-called ignition process. The combustion process continuously took place, which it can be observed the flame propagation from 388 ms to 488 ms, until the fuel vapor was completely burned.

5. Concluding Remarks

The ignition over supersonic n-Hexadecane jet was examined using a high-speed digital video camera with shadowgraph system. Results obtained are summarized as follows.

The temperature behind the shock waves over supersonic n-Hexadecane jet tips could not initiate the ignition over the jet even though it was much higher than the autoignition point.

The autoignition over supersonic jet has not occurred because of the unsuitability of the high temperature distribution around the jet, the atomization condition, the vaporization condition, the mixing condition, the air-fuel ratio, and the delay time.

The heat column or the high intensity CO2 laser alone cannot initiate the ignition over the jet.

The ignition over the fuel jet was achieved using both the heat column and the high intensity CO2 laser, in which the heat column enhanced the vaporization process while the CO2 laser initiated the ignition process.

The occurance of the ignition over the fuel jet was confirmed by signs of luminous spots or flames; these indicate the presence of ignitions obviously observed in the shadowgraph images.

The ignition over the jet in this study is the force ignition, not the autoignition.

Acknowledgments

The authors wish to acknowledge Mr. S. Haysaka of Tohoku University for his encouragement and devotion in conducting the present experiments and all staff members of the Interdisciplinary Shock Wave Research Laboratory. The first author would like to express acknowledgement to the Faculty of Engineering, Ubon Ratchathni University, and the Royal Thai Government for financial support. This paper was also supported by the Grant-in-Aid for Science Research no. 12 CE 2003 offered by the Ministry of Education, Culture, Sport, Science and Technology, Japan.

References

  1. T. J. Methven and B. Fairhead, “A correlation between rain erosion of perspex specimens in flight and on a ground rig,” Wear, vol. 2, no. 6, p. 498, 1959.
  2. A. Déom, R. Gouyon, and C. Berne, “Rain erosion resistance characterizations Link between on-ground experiments and in-flight specifications,” Wear, vol. 258, no. 1–4, pp. 545–551, 2005. View at Publisher · View at Google Scholar · View at Scopus
  3. M. Lesser and J. Field, “Studies in shock waves, liquid impact, jets and cavitation,” in Proceedings of the 18th International Symposium on Shock Waves (ISSW '91), K. Takayama, Ed., pp. 63–72, Springer, Sendai, Japan, July 1991.
  4. N. K. Bourne and J. E. Field, “A high-speed photographic study of cavitation damage,” Journal of Applied Physics, vol. 78, no. 7, pp. 4423–4427, 1995. View at Publisher · View at Google Scholar · View at Scopus
  5. N. K. Bourne, “On stress wave interactions in liquid impact,” Wear, vol. 258, no. 1–4, pp. 588–595, 2005. View at Publisher · View at Google Scholar · View at Scopus
  6. F. P. Bowden and J. E. Field, “The brittle fracture of solids by liquid-impact, by solid impact, and by shock,” Proceedings of the Royal Society of London, vol. 282, pp. 331–352, 1964.
  7. T. Nakahira, M. Komori, N. Nishida, and K. Tsujimura, “A study of shock wave generation around high pressure fuel spray in a diesel engine,” in Proceedings of the International Symposium on Shock Waves (ISSW '91), K. Takayama, Ed., pp. 1271–1276, Sendai, Japan, 1991.
  8. B. E. Milton and K. Pianthong, “Pulsed, supersonic fuel jets—a review of their characteristics and potential for fuel injection,” International Journal of Heat and Fluid Flow, vol. 26, no. 4, pp. 656–671, 2005. View at Publisher · View at Google Scholar · View at Scopus
  9. J. Alle and G. Hargrave, “Fundamental study of in-nozzle fluid flow and its effect on liquid jet break-up in Gasoline Direct injectors,” in Proceedings of the Annual Conference on Liquid Atomization and Spray System (ILASS Europe '00), pp. I4.2–I4.5, Darmstadt, Germany, 2000.
  10. Z. Wang, S. J. Shuai, J. X. Wang, and G. H. Tian, “A computational study of direct injection gasoline HCCI engine with secondary injection,” Fuel, vol. 85, no. 12-13, pp. 1831–1841, 2006. View at Publisher · View at Google Scholar · View at Scopus
  11. A. Ferri, “Review of problems in application of supersonic combustion,” Journal of the Royal Aeronautical Society, vol. 68, pp. 575–597, 1964.
  12. F. S. Billig, “Research on supersonic combustion,” Journal of Propulsion and Power, vol. 9, no. 4, pp. 499–514, 1993. View at Publisher · View at Google Scholar
  13. M. Arai, M. Shimizu, and H. Hiroyasu, “Break-up length and spray formation mechanism of a high speed liquid jet,” in Proceedings of the 4th International Conference of Liquid Atomization and Spray Systems (ICLASS '88), pp. 177–184, London, UK, 1988.
  14. S. P. Lin and Z. W. Lian, “Mechanisms of the breakup of liquid jets,” AIAA Journal, vol. 28, no. 1, pp. 120–126, 1990. View at Scopus
  15. J. E. Field and M. B. Lesser, “On the mechanics of high speed liquid jets,” Proceedings of the Royal Society of London. Series A, vol. 357, no. 1689, pp. 143–162, 1977. View at Scopus
  16. H. H. Shi, Study of hypersonic liquid jets [Ph.D. thesis], Tohoku University, Sendai, Japan, 1994.
  17. H. H. Shi and K. Takayama, “Generation of hypersonic liquid fuel jets accompanying self-combustion,” Shock Waves Journal, vol. 9, no. 5, pp. 327–332, 1999. View at Scopus
  18. K. Pianthong, Supersonic liquid diesel fuel jets, generation, shock wave, characteristics, auto-ignition Feasibilities [Ph.D. thesis], The University of New South Wales, Sydney, Australia, 2002.
  19. K. Pianthong, S. Zakrzewski, M. Behnia, and B. E. Milton, “Characteristics of impact driven supersonic liquid jets,” Experimental Thermal and Fluid Science, vol. 27, no. 5, pp. 589–598, 2003. View at Publisher · View at Google Scholar · View at Scopus
  20. A. Matthujak, Experimental study of impact-generated high-speed liquid jet [Ph.D. thesis], Tohoku University, Sendai, Japan, 2007.
  21. A. Matthujak, S. H. R. Hosseini, K. Takayama, M. Sun, and P. Voinovich, “High speed jet formation by impact acceleration method,” Shock Waves, vol. 16, no. 6, pp. 405–419, 2007. View at Publisher · View at Google Scholar · View at Scopus
  22. F. P. Bowden and J. H. Brunton, “Damage to solids by liquid impact at supersonic speeds,” Nature, vol. 181, no. 4613, pp. 873–875, 1958. View at Publisher · View at Google Scholar · View at Scopus
  23. F. P. Bowden and J. H. Brunton, “The deformation of solids by liquid impact at supersonic speeds,” Proceedings of the Royal Society of London, Series A, vol. 263, pp. 433–450, 1961.
  24. K. Pianthong, K. Takayama, B. E. Milton, and M. Behnia, “Multiple pulsed hypersonic liquid diesel fuel jetsdriven by projectile impact,” Shock Waves, vol. 14, no. 1-2, pp. 73–82, 2005. View at Publisher · View at Google Scholar · View at Scopus
  25. K. Ohashi, Experimental characterization of flow fields [M.S. thesis], Tohoku University, Sendai, Japan, 2002.
  26. A. Matthujak, K. Pianthong, M. Sun, K. Takayama, and T. Ikohagi, “Characteristics of high-speed liquid fuel jets,” in Proceedings of the The 16th Japanese Symposium of Shock Wave, 2005.
  27. J. D. Anderson, Modern Compressible Flow, McGraw-Hill, New York, NY, USA, 3rd edition, 2004.