Open Access
Add to BibSonomy
Abstract
To reduce the energy of an individual laser pulse, dual-pulse laser ignitions (LIs) at various pulse intervals were investigated in a Mach 2.92 scramjet engine fueled with ethylene. For comparison, experiments on a single-pulse LI were also performed. Schlieren visualization and high-speed photography were employed to observe the ignition processes simultaneously. The results indicate that the energy of an individual laser pulse can be reduced by half via a dual-pulse LI method as compared with a single-pulse LI with the same total energy. The reduction of the individual laser pulse energy degrades the requirements on the laser source and the beam delivery system, which facilitates the practical application of LI in hypersonic vehicles. A pulse interval shorter than 40 μs is suggested for dual-pulse LI in the present study. Because of the intense heat loss and radical dissipation in high-speed flows, the pulse interval for dual-pulse LI should be short enough to narrow the spatial distribution of the initial flame kernel.
© 2018 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
Laser ignition (LI) has brought extensive research attentions because of its potential benefits compared with conventional spark ignition system [1,2]. These merits can be concluded as non-intrusive [3], flexible in selection of ignition location and timing [4,5], reduction of overall burning time et al. [6,7], which was reviewed by Morsy [8] and recently by Dearden and Shenton [9]. Laser ignition was early proposed in internal combustion engines [1,7,10–12], then gradually extended to the application in the high-speed flow environments such as gas turbine engines [13,14] and rocket engines [15–18] and recently studied in air-breathing supersonic combustion jet (scramjet) engines [19–21]. The ignition condition in the latter engine is more critical because of the high incoming flow speed (~1000 m/s) [22,23], subsequent short residence time, inhomogeneous fuel-air mixing, and relative low total temperature at take-over Mach number.
Brieschenk et al. [21,24] experimentally investigated LI of hydrogen fuel using a Q-switched ruby laser in hypersonic flows. The single-pulse energy was as high as 700 mJ, and the ignition position was further placed inside the hydrogen orifice to reduce the ignition energy. For hydrocarbon fuel such as ethylene fuel and liquid kerosene, which were commonly used in the scramjet engine, Yang et al. [22] and Li et al. [25] successfully achieved LI in a Mach 2.52 scramjet combustor using a single-pulse Nd:YAG laser with a pulse energy of approximate 200 mJ. Parameter studies such as the impacts of focusing location and laser energy on the ignition process were further studied by An et al. [26] and Li et al. [27] using single-pulse LI.
However, the high single-pulse energy requirement of LI in scramjet engines leads to larger laser unit volume, more energy consumption [4], and challenges in laser beam delivery using optical fibers [28], which may prohibit its practical application in the hypersonic vehicles. An alternative way to reduce the individual pulse energy is using dual-pulse or multi-pulse laser instead of single-pulse laser while maintaining the same total energy. Bak et al. [29,30] found that plasma induced by the first laser pulse enhanced the energy absorption of the successive laser pulse significantly with a pulse interval shorter than 250 ns. Pulse intervals shorter than 200 ns (the electron lifetime) or longer than ~100 μs (the chemical delay time) are suggested for dual-pulse LI of a diffusion burner [31]. Hsu et al. [32] successfully reduced the minimal ignition energy of individual laser pulse by approximately 10 times while using a high-repetition-rate laser up to 100 kHz. Lorenz et al. [33] proved that the flame kernel resulting from different laser pulses merged together when the pulse interval was short enough. Li et al. [34] experimentally investigated the transient characteristics of the ignition process by dual-pulse LI in a model scramjet combustor fueled by kerosene. The pulse interval was kept at 200 µs for all the tests. Although previous studies indicated that dual-pulse and multi-pulse approaches were promising alternatives to single-pulse laser ignition in low-speed conditions, more detailed investigation is necessary for assessing the performance of dual-pulse LI method in the supersonic combustor.
In this paper, the impacts of pulse interval that varies from 100 ns to 340 μs on dual-pulse LI in a model scramjet combustor were assessed. For comparison, single-pulse LI experiments were also performed. High-speed photography and Schlieren visualization were employed to capture the ignition processes simultaneously.
2. Experimental setup
The present experiments were performed on a direct-connect test facility. The high-enthalpy supersonic flow was produced by simultaneous combustion of air/oxygen/ethanol in the air heater and accelerated by a Mach 2.92 Larval-nozzle simulating the combustor entrance flow conditions. The total temperature, stagnation pressure and velocity of the mainstream were 1650 K, 2.6 MPa and 1515 m/s, respectively. The inflow was mainly composed of O2, H2O, CO2 and N2. The mole fractions of the compositions were O2 23.3%, H2O 6.2%, CO2 10.2% and N2 60.3%, based on thermodynamic calculation.
Figure 1 shows the schematic of the model scramjet combustor. The entrance of the 512-mm-length combustor was 40 mm in height and 50 mm in width. The lower wall of the combustor diverged at 2.25°. A cavity with a depth of 15 mm, a length of 105 mm and an aft angle of 45° was adopted as the flameholder. The cavity was installed 55 mm downstream of the combustor entrance. Ethylene was injected from an orifice with a diameter of 2 mm located 10 mm upstream of the cavity leading edge. The global equivalence ratio was maintained at 0.23 for all the tests. More detailed information about the test facility was presented in [26].
The optical layout used for the LI, Schlieren visualization and high-speed photography is shown in Fig. 2. The second harmonic of the fundamental output of a dual-cavity Q-switched Nd:YAG laser (Beamtech, Vlite-500) at 532 nm with maximum pulse energy of 350 mJ was applied for ignition. Laser pulses were focused at a position 45 mm downstream of the cavity leading edge and 5 mm above the cavity bottom in the flameholder using a convex lens with a focal length of 150 mm. The static pressures were measured to be 64.5 kPa and 85.9 kPa from leading edge to trailing edge in the cavity and it was determined to be 72.7 kPa at the ignition point. The laser beam diameter was approximately 12 mm, and the pulse width was 10 ns. Taking the energy attenuation caused by the scattering and absorption of the convex lens and combustor window, the laser energies of dual-pulse LI and single-pulse LI were 129 + 129 mJ and 258 mJ, respectively. The variation of the laser energy was less than 3%. Note that the successful ignition could not be achieved by only one 129-mJ laser pulse because the minimum ignition energy in single-pulse LI was approximate 198 mJ. The dual-pulse LI method was investigated at an ignition energy of 129 + 129 mJ, instead of 99 + 99 mJ, because successful ignition could be achieved at longer pulse interval. Consequently, more experimental data could be obtained for analysing.
Assuming that the focal region was cylindrical, the radius (r) and length (l) of the focal region were given by [35]:
Here, λ = 532 nm was the wavelength of the laser, f = 150 mm was the focal length of the convex lens, D0 = 12 mm was the beam diameter, θ = 0.5 mrad was the beam divergence. In this study, the r and l were 4.23 µm and 388.3 µm, respectively. As the energies of the individual laser pulses were 258 mJ and 129 mJ for the single-pulse LI and dual-pulse LI, respectively. The energy densities at the focal region were 1.18 × 1013 J/m3 and 5.91 × 1012 J/m3, respectively.The pulse interval between successive laser pulses was controlled by a Digital Delay/Pulse Generator (Stanford Research Systems, DG535). The evolution of the flame kernel and flame propagation were observed by recording the flame chemiluminescence, mainly from the spectral emission of CH*, C2*, and soot [36], using a high-speed camera (Photron, SA-X2) with a Nikon 85 mm, f/1.8 lens with frame rate of 50 kHz and exposure time of 15 μs. Additionally, a Z-type Schlieren system was applied for visualizing the flow field using a research arc lamp as light source (Newport, 66485-500HX-R1). Another high-speed camera (Photron, SA-Z) with a Nikon 200 mm, f/4 lens was employed to capture Schlieren images. The frame rate and exposure time were set at 50 kHz and 248 ns, respectively. These two high-speed cameras were synchronized with the laser system by a Digital Delay/Pulse Generator (Stanford Research Systems, DG645). The laser scattering signal was detected by a Si-biased photodiode (Thorlabs, DET10A/M) and exposure-time signals of the high-speed cameras were monitored simultaneously by an oscilloscope (Tektronix, DPO4054B) during all the tests.
3. Results and discussion
3.1 Temporal evolution of the flame kernel of single-pulse LI and dual-pulse LIs
The ignition behaviors of single-pulse LI and dual-pulse LIs with pulse intervals of 100 ns and 40 μs are displayed in Fig. 3. For brevity, these three cases are denoted as “Case Single”, “Case D1”, and “Case D2”, respectively. The flame chemiluminescence images after 105 µs are enhanced by 2 times to provide better contrast. To determine the edge of the flame, the color chemiluminescence images were converted into grayscale images. The pixel with an intensity larger than 5% of the saturation intensity (65,535 for the 16-bit image) was identified as flame. Then, the edge of the flame which was denoted by the white curve was detected by an in-house MATLAB code and marked in the Schlieren images.
Fig. 3 Ignition processes of single-pulse LI and dual-pulse LIs with pulse intervals of 100 ns and 40 μs.
A bow shock and a separation shock resulting from the interaction between the fuel jet and the supersonic mainstream are clearly observed by Schlieren visualization. The bow shock is then reflected by the combustor upper-wall. The ethylene is carried downstream by the mainstream. A small portion of it entrances the cavity from the aft wall, and is then entrained to the upstream due to the cavity recirculation flow. The vortical structures in the cavity shear-layer also contribute to enhancing the mass-exchange between the mainstream and the cavity recirculation flow [37]. The middle of the cavity is selected for ignition because of low strain rate and suitable local equivalence ratio.
As shown in Fig. 3, the plasma volume of Case Single is the largest, second by Case D1, at 5 μs after the breakdown. The plasma induced by dual-pulse approach decays faster because the energy density of it is inferior to that of single-pulse laser-induced plasma with the same incident energy [38]. The plasma volume of Case D2 is the smallest, since only the first laser pulse (129 mJ) is released at 5 μs. Accompanying with moving toward the cavity leading edge, the flame kernel of Case Single is elongated in stream-wise at 45 μs because of the vertical velocity gradient in the cavity. However, the flame kernel of Case D1 is larger at height, since the second laser pulse is mainly absorbed at the anterior of the plasma [38]. For Case D2, both of the laser pulses are released at 45 μs, and two individual flame kernels are observed. As the distance between the flame kernels is relatively short, these two kernels merge together initially. At 145 μs after releasing the first laser pulse, the flame kernels of these three cases reach the cavity leading edge and reside there. Ma et al. [39] claimed that reasonable agreement could be achieved between the flow speed derived from the trajectory of the flame kernel and the velocity field obtained via particle image velocimetry. Therefore, based on the positions of the flame kernel at different moments, the velocity of the flow in the fore part of cavity is determined as approximate 300 m/s.
After reaching the cavity leading edge, the ignition processes of Case Single, D1, and D2 are almost the same. Taking the test with a pulse interval of 40 μs (Case D2) as an example, the flame kernel resides at the fore part of the cavity for about 440 μs. During this period, fuel and fresh air are transported to the flame kernel by the recirculation flow and the vortical structures in cavity shear layer. The growth of the flame kernel is not significant because of the high strain rate and feeble combustion. However, the hot products resulting from the flame kernel might increase the temperature of the fluid in the cavity and the flame propagation speed. Therefore, the flame starts propagating downstream rapidly at 585 μs, and reaches the cavity aft wall in 180 μs. In this stage, the morphology of the flame is strongly affected by the cavity shear layer [40] because the velocity of the fluid is significantly higher than the flame propagation speed [41]. Finally, stable combustion is achieved at 805 microseconds, as the flame spreads into the mainstream.
Moreover, the pulse intervals of dual-pulse LIs are prolonged to 240 μs and 340 μs, and the results are displayed in Fig. 4. The ignition tests are denoted as “Case D3” and “Case D4”, respectively. The initial flame kernels of Case D3 and D4 are the same at 5 μs. The flame kernel resulting from the first laser pulse has arrived at the cavity leading edge at 265 μs, and the second flame kernel is induced in the middle of the cavity when the pulse interval is 240 μs. Therefore, two separate flame kernels are observed. The second flame kernel then enhances the first flame kernel significantly when they merges with each other. When the pulse interval is 340 μs, the first flame kernel decays more obviously before releasing the second laser pulse, compared with Case D3, because of longer pulse interval and consequently more intense heat loss and radical dissipation.
As the first flame kernel of Case D4 is almost extinguished at 405 μs, the merged flame kernel of Case D3 is much stronger than its counterpart of Case D4 at 605 μs. The flame kernel of Case D3 then fulfills the cavity and spreads to the mainstream in the following 400 μs. In contrast, additional 160 μs is necessary for Case D4 to establish stable combustion. Although the ignition process of Case D3 is faster than Case D4, it is slower than Case Single, D1, and D2 by 240 μs. Dual-pulse LI with a pulse interval longer than 340 μs is not presented in this paper because successful ignition cannot be achieved.
3.2 Discussion
In further study, the color flame chemiluminescence images were converted into grayscale images. The overall chemiluminescence intensity in the observation area was calculated according to the grayscale images, and was normalized by the average overall chemiluminescence intensity after successful ignition. The average integrated flame chemiluminescence intensities of Case Single, D1, D2, D3, and D4 are displayed in Fig. 5. To suppress the effects of random factors, images from eight tests were used to calculate one curve.
Fig. 5 Average integrated flame chemiluminescence intensities of single-pulse LI and dual-pulse LIs with pulse intervals of 100 ns, 40 μs, 240 μs and 340 μs during ignition processes.
As shown in Fig. 5, the normalized intensity curves of Case Single, D1 and D2 overlap together. The main difference is that the normalized intensity of Case D2 peaks at 40 μs as a result of the second laser pulse. In general, the normalized intensities of these three cases decrease exponentially in the first 200 μs because of the rapid decay of the plasma. Then the intensity curves reach the minimum at 300 μs and start to increase significantly after 600 μs. The average flame intensities of Case Single, D1, and D2 during 300-600 μs when the flame kernels reside at the cavity leading edge are 0.132, 0.143, and 0.139, respectively. A sharp increase is witnessed after 600 μs. The moment when the intensity curve reaches 1 for the first time after 500 μs is defined as the ignition time. It means that the current flame shares a similar luminosity with the flame at stable combustion stage. The ignition times of Case Single, D1, and D2 are 792 μs, 787 μs, and 785 μs, respectively.
When the pulse intervals of dual-pulse LIs are set at 240 μs and 340 μs, the normalized intensities keep diminishing until the arrival of the second laser pulse. During this period, the plasma rapidly decays from ions at ~10,000 K to reactive radicals at ~1000 K [13], and forms the initial flame kernel via igniting the surrounding combustible gases. Then the flame kernel shrinks gradually because the heat release of the fresh fuel-air mixture from the ambient gases cannot compensate for the losses of heat and radicals caused by the vortical structures. After peaking at 240 μs, the flame intensity of Case D3 decreases again, and eventually rises from the minimum at 900 μs. The average flame intensity during 600-900 μs is 0.105, and the ignition time is 1049 μs. The normalized intensity of Case D4 shares a similar behavior with Case D3. However, the ignition time is 172 μs longer because the merged flame kernel is inferior to Case D3.
Based on the individual ignition times of 32 independent tests (eight tests for each case), the pulse interval dependence of the ignition times of dual-pulse LIs is presented in Fig. 6. The ignition time and its standard deviation of dual-pulse LI with a pulse interval shorter than 40 μs are approximately 790 μs and 17%, respectively. While, further extension of the pulse interval leads to longer ignition time and degraded ignition reliability.
Fig. 6 The ignition times of dual-pulse LIs with pulse intervals of 100 ns, 40 μs, 240 μs, and 340 μs.
The reasons for the above-described results are as follow. Previous study [38] indicated that the energy absorptions and shock losses of dual-pulse laser-induced plasma (LIP) with a pulse interval shorter than 200 ns and single-pulse LIP with the same total energy were similar. Therefore, Case Single and D1 share the same ignition behavior.
In this study, the velocity of the fluid varies from ~1000 m/s in the mainstream to ~300 m/s in the cavity. The steep velocity gradient leads to high strain rate and shear stress. The flame kernel is more likely to be stretched into smaller pieces, which results in larger flame kernel surface and more serious heat and radical losses. Therefore, the flame kernel with more centralized spatial distribution is preferred for ignition. For dual-pulse LI with a pulse interval of 40 μs, the first flame kernel propagates to the cavity leading edge by approximately 12 mm when the second flame kernel is induced, as the velocity of the recirculation flow in cavity is approximate 300 m/s. Since the length of the first flame kernel of Case D2 is approximate 13 mm at 40 µs, the second flame kernel merges with the first one initially. Therefore, the spatial distribution of the merged flame kernel of Case D2 is relatively centralized, which facilitates swift and successful ignition.
When the pulse interval between successive laser pulses is 240 μs or 340 μs, the first flame kernel has resided at the cavity leading edge for approximate 100 μs. The size and strength of it decays rapidly because of the heat loss and radical dissipation. Consequently, the ignition process is delayed. Therefore, an overlong pulse interval should be avoid for dual-pulse LI in the supersonic combustor. To narrow down the spatial distribution of the merged flame kernel, the pulse interval should be shorter than the ratio between the length of the first flame kernel at when the second laser pulse is released and the velocity of the cavity recirculation flow. Moreover, the energy of individual laser pulse could be further reduced by high-repetition rate LI method if the flame kernels result from different laser pulses can merge together initially, which has been verified in low-speed flows [32,33].
4. Conclusion
Focusing on reducing the energy of individual laser pulse, the performances of single-pulse LI and dual-pulse LIs with various pulse intervals were investigated in detail in a supersonic combustor. Compared with single-pulse LI, the energy of individual laser pulse decreases from 258 mJ to 129 mJ via dual-pulse LI method, which facilitates the practical application of LI in scramjet engines. Since the heat loss and radical dissipation of the flame kernel are proportional to the local strain rate, the optimal pulse interval is related to the size of the first flame kernel and the local flow velocity. The pulse interval could extend to 40 μs in the present study without delaying the ignition process.
Funding
National Natural Science Foundation of China (NSFC) (11502293).
References and links
1. P. D. Ronney, “Laser versus conventional ignition of flames,” Opt. Eng. 33(2), 510–521 (1994). [CrossRef]
2. T. X. Phuoc, “Laser-induced spark ignition fundamental and applications,” Opt. Lasers Eng. 44(5), 351–397 (2006). [CrossRef]
3. D. Bradley, C. G. W. Sheppard, I. M. Suardjaja, and R. Woolley, “Fundamentals of high-energy spark ignition with lasers,” Combust. Flame 138(1-2), 55–77 (2004). [CrossRef]
4. J. V. Pastor, J. M. García-Oliver, A. García, and M. Pinotti, “Effect of laser induced plasma ignition timing and location on Diesel spray combustion,” Energy Convers. Manage. 133, 41–55 (2017). [CrossRef]
5. J. L. Beduneau, N. Kawahara, T. Nakayama, E. Tomita, and Y. Ikeda, “Laser-induced radical generation and evolution to a self-sustaining flame,” Combust. Flame 156(3), 642–656 (2009). [CrossRef]
6. D. K. Srivastava and A. K. Agarwal, “Comparative experimental evaluation of performance, combustion and emissions of laser ignition with conventional spark plug in a compressed natural gas fuelled single cylinder engine,” Fuel 123, 113–122 (2014). [CrossRef]
7. J. D. Dale, P. R. Smy, and R. M. Clements, “Laser ignited internal combustion engine: An experimental study,” SAE paper, 780329 (1978). [CrossRef]
8. M. H. Morsy, “Review and recent developments of laser ignition for internal combustion engines applications,” Renew. Sustain. Energy Rev. 16(7), 4849–4875 (2012). [CrossRef]
9. G. Dearden and T. Shenton, “Laser ignited engines: progress, challenges and prospects,” Opt. Express 21(S6Suppl 6), A1113–A1125 (2013). [CrossRef] [PubMed]
10. N. Pavel, M. Tsunekane, and T. Taira, “Composite, all-ceramics, high-peak power Nd:YAG/Cr4+:YAG monolithic micro-laser with multiple-beam output for engine ignition,” Opt. Express 19(10), 9378–9384 (2011). [CrossRef] [PubMed]
11. D. K. Srivastava, M. Weinrotter, H. Kofler, A. K. Agarwal, and E. Wintner, “Laser-assisted homogeneous charge ignition in a constant volume combustion chamber,” Opt. Lasers Eng. 47(6), 680–685 (2009). [CrossRef]
12. D. H. McNeill, “Minimum ignition energy for laser spark ignition,” Proc. Combust. Inst. 30(2), 2913–2920 (2005). [CrossRef]
13. G. C. Gebel, T. Mosbach, W. Meier, and M. Aigner, “Optical and spectroscopic diagnostics of laser-induced air breakdown and kerosene spray ignition,” Combust. Flame 162(4), 1599–1613 (2015). [CrossRef]
14. J. Griffiths, M. Riley, A. Kirk, A. Borman, J. Lawrence, and C. Dowding, “Lean burn limit and time to light characteristics of laser ignition in gas turbines,” Opt. Lasers Eng. 55, 262–266 (2014). [CrossRef]
15. C. Manfletti and G. Kroupa, “Laser ignition of a cryogenic thruster using a miniaturised Nd:YAG laser,” Opt. Express 21(S6Suppl 6), A1126–A1139 (2013). [CrossRef] [PubMed]
16. C. Manfletti and M. Börner, “Laser ignition for space propulsion systems,” in Laser Ignition Conference, (Optical Society of America, 2015), p. Th2A.3.
17. S. Soller, “Laser ignition application to cryogenic propellant rocket thrust chambers,” in Laser Ignition Conference, (Optical Society of America, 2017), p. LFA4.3. [CrossRef]
18. M. Börner, J. C. Deeken, C. Manfletti, and M. Oschwald, “Determination of the minimum laser pulse energies for ignition in a subscale rocket combustion chamber,” in Laser Ignition Conference, (Optical Society of America, 2017), p. LThA4.2. [CrossRef]
19. H. Horisawa, S. Tsuchiya, J. Negishi, Y. Okawa, and I. Kimura, “Laser ignition and flameholding characteristics in supersonic airstreams,” Proc. SPIE 5448, 586–595 (2004). [CrossRef]
20. S. Brieschenk, H. Kleine, and S. O’Byrne, “Laser ignition of hypersonic air-hydrogen flow,” Shock Waves 23(5), 439–452 (2013). [CrossRef]
21. S. Brieschenk, S. O’Byrne, and H. Kleine, “Laser-induced plasma ignition studies in a model scramjet engine,” Combust. Flame 160(1), 145–148 (2013). [CrossRef]
22. L. Yang, X. Li, J. Liang, Y. Yu, and X. Yu, “Laser-induced plasma ignition of hydrocarbon fuel in supersonic flows,” in 20th AIAA International Space Planes and Hypersonic Systems and Technologies Conference, (AIAA, 2015), pp. AIAA-2015–3544. [CrossRef]
23. X. Li, W. Liu, Y. Pan, L. Yang, and B. An, “Experimental investigation on laser-induced plasma ignition of hydrocarbon fuel in scramjet engine at takeover flight conditions,” Acta Astronaut. 138, 79–84 (2017). [CrossRef]
24. S. Brieschenk, S. O’Byrne, and H. Kleine, “Ignition characteristics of laser-ionized fuel injected into a hypersonic crossflow,” Combust. Flame 161(4), 1015–1025 (2014). [CrossRef]
25. X. Li, L. Yang, J. Peng, X. Yu, J. Liang, and R. Sun, “Cavity ignition of liquid kerosene in supersonic flow with a laser-induced plasma,” Opt. Express 24(22), 25362–25369 (2016). [CrossRef] [PubMed]
26. B. An, Z. Wang, L. Yang, X. Li, and J. Zhu, “Experimental investigation on the impacts of ignition energy and position on ignition processes in supersonic flows by laser induced plasma,” Acta Astronaut. 137, 444–449 (2017). [CrossRef]
27. X. Li, W. Liu, Y. Pan, L. Yang, and B. An, “Experimental investigation on laser-induced plasma ignition of hydrocarbon fuel in scramjet engine at takeover flight conditions,” Acta Astronaut. 138, 79–84 (2017). [CrossRef]
28. J. D. Mullett, R. Dodd, C. J. Williams, G. Triantos, G. Dearden, A. T. Shenton, K. G. Watkins, S. D. Carroll, A. D. Scarisbrick, and S. Keen, “The influence of beam energy, mode and focal length on the control of laser ignition in an internal combustion engine,” J. Phys. D 40(15), 4730–4739 (2007). [CrossRef]
29. M. S. Bak, S. Im, and M. A. Cappelli, “Successive laser-induced breakdowns in atmospheric pressure air and premixed ethane-air mixtures,” Combust. Flame 161(7), 1744–1751 (2014). [CrossRef]
30. M. S. Bak, L. Wermer, and S. Im, “Schlieren imaging investigation of successive laser-induced breakdowns in atmospheric-pressure air,” J. Phys. D 48(48), 485203 (2015). [CrossRef]
31. L. Wermer, J. Hansson, and S. Im, “Dual-pulse laser-induced spark ignition and flame propagation of a methane diffusion jet flame,” Proc. Combust. Inst. 36(3), 4427–4434 (2017). [CrossRef]
32. P. S. Hsu, S. Roy, Z. Zhang, J. Sawyer, M. N. Slipchenko, J. G. Mance, and J. R. Gord, “High-repetition-rate laser ignition of fuel-air mixtures,” Opt. Lett. 41(7), 1570–1573 (2016). [CrossRef] [PubMed]
33. S. Lorenz, M. Bärwinkel, R. Stäglich, W. Mühlbauer, and D. Brüggemann, “Pulse train ignition with passively Q-switched laser spark plugs,” J. Engine Res. 17(1), 139–150 (2016). [CrossRef]
34. X. Li, W. Liu, Y. Pan, L. Yang, B. An, and J. Zhu, “Characterization of ignition transient processes in kerosene-fueled model scramjet engine by dual-pulse laser-induced plasma,” Acta Astronaut. 144, 23–29 (2018). [CrossRef]
35. J. L. Beduneau, B. Kim, L. Zimmer, and Y. Ikeda, “Measurements of minimum ignition energy in premixed laminar methane/air flow by using laser induced spark,” Combust. Flame 132(4), 653–665 (2003). [CrossRef]
36. Y. Wang, L. Zheng, R. Woolley, and Y. Zhang, “Investigation of ignition process from visible to infrared by a high speed colour camera,” Fuel 185, 500–507 (2016). [CrossRef]
37. H. Wang, Z. Wang, M. Sun, and N. Qin, “Large-Eddy/Reynolds-averaged Navier-Stokes simulation of combustion oscillations in a cavity-based supersonic combustor,” Int. J. Hydrogen Energy 38(14), 5918–5927 (2013). [CrossRef]
38. B. An, Z. Wang, L. Yang, G. Wu, J. Zhu, and X. Li, “Experimental investigation of the shock loss and temporal evolution of hot plume resulting from dual-pulse laser-induced breakdown in quiescent air,” J. Appl. Phys. 122(19), 193301 (2017). [CrossRef]
39. L. Ma, Q. Lei, Y. Wu, T. M. Ombrello, and C. D. Carter, “3D measurements of ignition processes at 20 kHz in a supersonic combustor,” Appl. Phys. B 119(2), 313–318 (2015). [CrossRef]
40. L. Ma, Q. Lei, Y. Wu, W. Xu, T. M. Ombrello, and C. D. Carter, “From ignition to stable combustion in a cavity flameholder studied via 3D tomographic chemiluminescence at 20 kHz,” Combust. Flame 165, 1–10 (2016). [CrossRef]
41. F. W. Barnes and C. Segal, “Cavity-based flameholding for chemically-reacting supersonic flows,” Prog. Aerosp. Sci. 76, 24–41 (2015). [CrossRef]
