Laser ignition of a cryogenic thruster using a ... - OSA Publishing

Laser ignition of a cryogenic thruster using a miniaturised Nd:YAG laser Optics Express Chiara Manfletti,1,? and Gerhard Kroupa2 1 German

Aerospace Center (DLR), Institute of Space Propulsion, 74239 Hardthausen, Germany 2 Carinthian Tech Research AG (CTR), Villach, Austria ? [email protected]

Abstract: An experimental study has been conducted to assess the feasibility of implementing laser ignition in cryogenic reaction and control and orbital manouvering thrusters. A experimental thruster with a single-coaxial injector element combustion chamber for testing with liquid oxygen/gaseous hydrogen and liquid oxygen/gaseous methane was designed for this purpose. Mapping tests conducted using a standard table top laser revealed that the minimum incident energies required for 100% reliable laser plasma and laser ablation ignition of liquid oxygen/gaseous hydrogen are 72 mJ and 14.5 mJ respectively. In addition, the miniaturised HIPoLas® laser was mounted directly on the thruster and used as ignition system. This paper reports locations of energy deposition, levels of delivered energy and associated ignition probabilities obtained. The results indicate the feasibility of using a laser system for the direct ignition of reaction and control and orbital manouvering thrusters and highlight further investigations and developments necessary for the implementation of miniaturised laser systems for vacuum igntion of cryogenic propellants. © 2013 Optical Society of America OCIS codes: (140.0140) Lasers and laser optics; (140.3540) Lasers, Q-switched.

References and links 1. European Chemical Agency agreement, ”Agreement of the memeber state committee on the identification of hydrazine as a substance of very high concern” (European Chemical Agency, 2011). http://echa.europa.eu/documents/10162/b3561467-aa0f-4551-9dff-dcdc5ba60eec. 2. C. Scharlemann, “GRAPS - analysis of green propellant candidates,” presented at the 62nd International Astronautical Congress, Cape Town, South Africa, 3 - 7 October 2011. 3. A. Woschnak, “Development of a green bi-propellant hydrogen peroxide thruster for attitude control on satellites,” presented at the 4th European Conference for Aerospace Sciences, Saint Petersburg, Russia, 4 - 8 July 2011. 4. W. H. M. Welland, B. M. J. Brauers, and E. J. Vermeulen, “Future igniter technologies,” presented at the Space Propulsion Conference, Bordeaux, France, 7 – 10 May 2010. 5. P. James, C. Fiorentino, P. Caisso, J. N. Caruana, and J. M. Bahu, “Technological readiness of the Vinci expander engine,” presented at the 59th International Astronautical Congress, Glasgow, Scotland, 29 Sept. – 3 Oct. 2008. 6. M. Weinrotter, H. Kopecek, M. Tesch, E. Wintner, M. Lackner and F. Winter, “Laser ignition of ultra-lean methane/hydrogen/air mixture at high temperature and pressure,” Experimental Thermal and Fluid Science 29, 569–577 (2005). 7. D. L. McIntyre, A Laser Spark Plug Ignition System for a Stationary Lean-Burn Natural Gas Reciprocating Engine (West Virginia University, 2007).

#194445 - $15.00 USD Received 23 Jul 2013; revised 20 Sep 2013; accepted 14 Oct 2013; published 4 Nov 2013 (C) 2013 OSA 4 November 2013 | Vol. 21, No. S6 | DOI:10.1364/OE.21.0A1126 | OPTICS EXPRESS A1126

8. B.E. Forch and A.W. Miziolek, “Ultraviolet laser ignition of premixed gases by efficient and resonant multiphoton photochemical formation of microplasmas,” J. Combustion, Science and Technology 52, 151–159 (1987). 9. B.E. Forch, “Resonant laser ignition of reactive gases,” in Laser Applications in Combustion and Combustion Diagnostics II, R. J. Locke, eds., Proc. SPIE 2122, 1118–1128 (1994). 10. G. M. Weyl, “Physics of laser-induced breakdown: an update,” in Laser-Induced Plasmas and Applications, L. J. Radziemski and D. A. Cremers, eds. (Academic, 1989), pp. 1–67. 11. T. X. Phuoc and F. P. White, “An optical and spectroscopic study of laser-induced sparks to determine available ignition energy,” in Proceedings of the Combustion Institute 29, 1621–1628 (2002). 12. D. Bradley, C. G. W. Sheppard, I. M. Suardjaja, and R. Woolley, “Fundamentals of high-energy spark ignition with lasers,” J. Combustion and Flame 138, 55–77 (2004). 13. L. I. Sedov, Similarity and Dimensional Methods in Mechanics (Academic, 1959). 14. C. Manfletti, “Low ambient pressure injection and consequences on ignition of liquid rocket engines,” presented at the 48th Joint Propulsion Conference, Atlanta, Georgia, 29 July – 1 August 2012. 15. H. El-Rabii, G. Gaborel, J.-P. Lapois, D. Thevenin, J. C. Rolon and J.-P. Matin, “Laser spark ignition of two-phase monodisperse mixtures,” Optics Communications 256, 495–506 (2005). 16. T. Marchione, S. F. Ahmed and E. Mastorakos, “Ignition of turbulent swirling n-heptane spray flames using single and multiple sparks,” Combustion and Flame 156, 166–180 (2005). 17. G. Gebel, T. Mosbach, M. Aigner, W. Meier and S. Le Brun, “An experimental investigation of kerosene droplet breakup by laser-induced blast waves,” presented at the ASME Turbo Expo 2012, Power for Land, Sea and Air, Copenhagen, Denmark, 11 – 15 June 2012. 18. R. G. Root, “Modeling of post-breakdown phenomena,” in Laser-Induced Plasmas and Applications, L. J. Radziemski and D. A. Cremers, eds. (Academic, 1989), pp. 69–103. 19. K. Hasegawa, K. Kusaka, A. Kumakawa, M. Sato, and M. and Tadano, “Laser ignition characteristics of GOX/GH2 and GOX/GCH4 propellants,” presented at the 39th Joint Propulsion Conference, Huntsville, Alabama, 20 – 23 July 2003.

1.

Introduction

In the now over a decade-old search for alternative propellant combinations for reaction and control system (RCS) and orbital and manoeuvring system (OMS) thrusters which move away from the classic but toxic hypergolic propellants monomethylhydrazine (MMH) and dinitrogen tetroxide (NTO) [1], alternative ignition systems have been subject of great interest due to the need of developing a new generation of ignition systems for this new generation of thrusters. Despite the fact that such RCS and OMS thrusters are characterised by low combustion chamber pressures, typical of pressure-fed engines, the number of cycles that they need to withstand is significant. Table 1 gives an overview of the application-dependent lifecycles required. Table 1. Engine cycle numbers depending on application

Application Launcher RCS, Ariane 5 Automatic Transfer Vehicle RCS Satellite apogee OMS Herschel/Planck satellites OMS

Engine Thrust 400 N 220 N 400 N 20 N

Cycles ≈ 6000 ≈ 160000 ≈ 100 93130

In addition to new propellants [2, 3], classic propellant combinations, such as oxygen/hydrogen (Ox/H2) and oxygen/methane (Ox/CH4), are potential candidates. The advantage inherent of these propellant combinations is their performance (specific impulse, Isp) and nontoxicity. Beyond performance considerations and structural weight considerations linked to the propellant storage systems required for these alternative green propellants, another important issue is ignition. As current thrusters make use of either monopropellants or hypergolic bipropellants, no igniter technology currently exists for such low thrust systems. Current igniters for non-hypergolic bipropellants are for engines of much greater thrust, massflows and therefore


size and weight. These igniters therefore cannot be directly implemented on small thrusters. Pyrotechnical and electrical igniters, which are small combustion chambers whose hot gases act as pilot flames in the main combustion chamber, are the two most frequently used igniter types for high thrust engines. Electrical igniters are more commonly used in the US and in Japan whereas, with the exception of the in-development Vinci® engine which also makes use of an electrical igniter, European engines commonly use pyrotechnical igniters. Each igniter system has its advantages and disadvantages [4]. Whereas pyrotechnical igniters are compact, extremely reliable, and require little energy for the ignition of the solid propellant they contain, they do not allow for re-ignition, either after a failed ignition or for mission purposes. Electrical igniters on the other hand, despite their higher complexity, allow for re-ignition and, depending on their architecture, the number of re-ignitions is either limited to just a few or is simply a function of the lifetime and transients phases of the engine. Electrical igniters function as small combustion chambers. One or more spark plugs are implemented to ignite a gaseous fuel and gaseous oxidiser. In American and Japanese engines, the igniter feedlines are bypasses of the engine lines downstream of the main pumps, whereas the in-development Vinci® engine electric igniter is fed using two gas bottles [5]. The first system, whilst maintaining a lower weight, couples the igniter transients to the engine transients. The second system which decouples the two systems, must however, in addition to the higher weight associated with an additional feed system, also ensure that the pressure in the feed bottles is sufficiently high that the feed pressure for the last re-ignition satisfies the ignition pressure requirements. This effectively means that more gas must be stored in the igniter feed tanks at higher pressures than actually required and that the igniter is overdimensioned for all but the last ignition. A downsizing of the igniter systems currently in use by cryogenic high thrust engines is certainly possible but brings with it added weight which is undesirable for flight systems. Furthermore, downsizing is limited to the geometrical constraints of components such as spark plugs and feedlines. A more versatile and lightweight ingition system is desirable. Direct, in-chamber ignition, of propellants in the thruster chamber without an igniter is advantageous as it avoids the added weight and complexity of a separate igniter chamber. Potential solutions for the direct ignition of RCS and OMS thrusters are direct spark plug or laser ignition. Ignition via a laser beam is an external ignition method which potentially offers a number of advantages when compared to other ignition methods. A non-exhaustive list includes: high temporal and spatial precision and accuracy, therefore minimal ignition delay (defined as the time between valve opening and ignition), no need for premixing and an increased ignition probability for a wider range of mixture ratios and initial chamber conditions [6, 7] (from vacuum to high pressure). In addition to the increased probability of ignition for a wider range of mixture ratios, the advantage of direct laser igntion over direct spark ignition is that whereas the location of the focal spot in laser ignition can be defined by using a lens with the desired focal length allowing for the lens to be placed away from hot combustion gases, the discharge location resulting from a spark plug is pre-detemined and lies close to the electrodes. This implies that the spark plug would have to be screwed directly onto the thruster chamber wall at locations where mixture ratios are suitable for ignition. The challenge associated with this configuration is the exposure of the spark plug electrodes to hot combustion gases and the need to protect the plug with an additional film cooling. Film cooling for the sole purpose of spark plug protection is undesirable as it adds complexity to the system and reduces thruster performance. There are four laser ignition methods which can be theoretically implemented: thermal, photochemical, resonant and non-resonant laser ignition. Each method differs in the energy levels required and wavelength region in which they operate. For photochemical ignition the ioni-


sation potentials of oxygen, hydrogen and methane are however such that photons capable of initiating photochemical ignition are of the Vacuum-UV range or of higher frequencies (λ = hc E, where E is the activation energy). For resonant multi-photon ignition the wavelengths required are in the UV-C range [8, 9]. Laser that are therefore required for photochemical and for resonant ignition methods are complex and large lasers that would not survive the vibrational loads experienced during a launch. Thermal laser ignition implements a low energy laser beam, generally in the infra-red range (either a CO2 laser or a Nd:YAG laser), which is directed towards a metal target and which thus absorbs the incoming laser energy and when heated transfers the energy to the propellants which are heated above their autoignition temperature. In thermal ignition no ablation of the metal targets takes place and is thus comparable to ignition using a glow plug. Alternatively a low energy beam can be used to heat up a gas volume directly. This is not a vaiable option for direct in-chamber ignition where propellants are not stationary but rather fast flowing. The challenge with thermal ignition is the time required for the heat to transfer to the propellants, which increases the ignition delay and thus increases the potential for harder ignitions. Non-resonant ignition is the more viable laser ignition method. A high energy beam is focused in a small volume creating a local electric field which, interacting with gas molecules, causes their breakdown [10]. A large portion of the energy absorbed is used up in the form of blast wave losses and radiation and convection losses [11, 12]. A number of studies have shown that only a small percentage (in the range of < 10%) is available for ignition [11]. When the laser beam is focused solely in a gas volume, the method is known as plasma ignition. If instead the beam is focused on a metal target, the ignition method is known as ablation ignition. In both cases plasma is created and a shock wave results. The advantage of ablation ignition is that, due to the fact that the breakdown threshold of solids is significantly lower than that of gases, it requires smaller ignition energies. Given the significant hardware-related limitations associated with photochemical and resonant laser ignition methods, and the time constraints of thermal ignition, only the non-resonant method is a viable solution for space applications with oxygen, hydrogen and methane as potential propellants. Furthermore, due to the intensive use made of Nd:YAG lasers in many scientific fields, they have seen a tremendous miniaturisation in the last decade allowing their integration into weight-sensitive applications such as space propulsion. The investigations conducted, aimed at assessing the feasibility of implementing laser plasma and laser ablation ignition methods for the ignition of a cryogenic RCS engine using liquid oxygen/gaseous hydrogen (LOx/GH2) and liquid oxygen/gaseous methane (LOx/GCH4) as propellant combinations. Additional goals of the investigations were to: • Establish the overall technical readiness of laser ignition for novel RCS and OMS systems based on cryogenic propellants • Determine the existence of and identify optimal locations of ignition on the basis of ignition energies required, ignition probabilities and the overall ensuing ignition behaviour and repeatability • Establish ignition probabilities of both the flame kernel, Pker , and of the combustion chamber as a whole, Pign , for a range of focusing locations and energies • Explore minimum incident energies, Emin , required based on propellants, focusing locations, and chamber conditions prior to ignition • Qualify the ignition both in terms of repeatability as well as ignition overpressure whilst considering the effect of ignition delay


• Examine the technical readiness of the HiPoLas® laser for use as ignition system of a cryogenic RCS experimental thruster and identify development needs This paper will address both plasma and ablation ignition methods for LOx/GH2 and examine the performance of the HiPoLas® laser in this context. 2.

Experimental apparatus

In order to fulfil all pre-set goals, a dedicated 400N experimental thruster complete with laser ignition adapter and laser ignition system was designed and manufactured. The sections below describe the thruster, laser systems implemented, and the optical diagnostics used. 2.1.

Experimental thruster

The experimental thruster used for testing is a semi-cylindrical 60 mm combustion chamber. A single coaxial-injector element without recess or tapering was implemented to inject the propellants in the previously evacuated combustion chamber. Table 2 provides the main characteristics 8 6 4 3 of the experimental7 thruster as depicted in Fig. 1. 5

2

1

F

F

Table 2. Experimental thruster geometry

Combustion Chamber Geometry ratio, εc Chamber diameter, dc Throat diameter, dt Chamber length, lc Chamber length (cylindrical), lcc

CWContraction E

5.1 60 mm 26.6 mm 138 mm 78 mm

Injector Geometry LOx post innter CW diameter, d0 LOx post outer diameter, d1 H2 outer diameter, d2 CH4 outer diameter, d2 Fuel sonic nozzles diameter, dsnl

2.4 mm 3.2 mm 6.0 mm 5.0 mm 2.0 mm

E

CW-CW ( 1 : 1 ) D

D

Injector Detail Chamber length, lc C

d1

d0

Throat diameter, dt

Chamber diameter, dc

Injector Detail

d2

C

Cylindrical Chamber length, lcc

B

B

Deutsches Zentrum für Luft-und Raumfahrt e.V.

Fig. 1. Experimental thruster geometry. A

a roh

2.2.8

Laser systems7 implemented

unbeh

Rz100

b

c

5

4

Rz25

Rz6,3

Material:

Maßstab:

DLR

>120° ±5'

Gezeichnet

Datum 06.08.2010

Name C. Manfletti

Kontrolliert

A

Norm

55 20130717_316.18_RCS400N-Cryo A2 Status

6

Oberfläche b

Lampoldshausen 74239 Hardthausen

Für nicht tolerierte Maße und Winkel gilt 6...30 30...120 >120 ...50° 50°...120° ...6 ±0,05 ±0,05 ±0,1 ±0,1 ±0,15 ±5' Rauhheitsangaben nach DIN ISO 1302

Änderungen

Datum

Name

3

Two laser systems were used during the experimental investigation: a table-top Quantel YG981 E10 flashlamp pumped Neodymium YAG solid state laser with a passive Chromium YAG Qswitch and a miniaturised diode-pumped solid state laser HiPoLas® delivering a 30 mJ 2.5 ns

2

1

pulse a 1064 nm. The table-top laser was implemented to deliver a 6.5 ns pulse at 532 nm at various energy levels for mapping purposes. The miniaturised HiPoLas® laser was implemented to assess the feasibility of using a directly mounted laser for ignition of a cryogenic RCS experimental thruster. Table 3 summarises the main characteristics of the two laser systems used. Whereas the miniaturised laser system was mounted directly onto the battleship thruster, the table-top laser was located far from the thruster and the laser beam was focused into the combustion chamber via a system of 3 mirrors, 1 lens (60 or 80 mm) and a window, as shown in Fig. 2. Using a powermeter (Scientech 380101), power losses along the optical path were measured. Using this method it was determined that the power delivered into the combustion chamber is 87% of that measured after the first mirror. During testing the internal Q-switch delay was varied using a delay generator (Stanford Research Systems DG535) to obtain the desired energy. In this manner the energy delivered into the combustion chamber could be varied from a maximum of 160 ± 5 mJ to a minimum of 10 ± 5 mJ per pulse. In this investigation the two lasers are compared only in terms of the ignition probabilities associatd with their respective energies delivered despite the fact that the differing pulse durations and wavelengths will lead to different breakdown intensity thresholds and will results in different plasma kernel shapes and fractions of energy absorbed. This limited comparison is due to the difficulties associated with the quantification of the flow field in the chamber prior to ignition due not only to its stochastic nature but also to the highly transient nature of the flow prior to ignition. Furthermore, the highly turbulent flow in the chamber hinders the visualisation of the resulting plasma in the chamber making it impossible to perform an quantitative and in-depth analysis of the plasma kernel and plasma kernel development. Table 3. Laser system characteristics

Testing purpose Wavelength, λ Pumping Pulse length Beam width Energy Pulse rate Burst rate Focusing lens Focal spot size (calculated) Lens material

2.2.1.

Quantel YG981 E10 mapping 1064 nm, 532 nm flashlamp 6.5 ns 8 mm 10-180 mJ ± 5 mJ 10 Hz 60 mm, 80 mm ≈ 80 µm Borosilicate (BK7)

HiPoLas® technology readiness assessment 1064 nm diode pumped 2.5 ns 2.5 mm 30 mJ ± 3 mJ 40 Hz 100 Hz (max. 5 pulses) 36 mm ≈ 50 µm Sapphire

HiPoLas® laser

The CTR HiPoLas® laser is a miniaturized, side-pumped solid state laser using a passive Qswitch. The laser was designed with a monolithic Nd:YAG crystal optically bonded with a Cr:YAG Q-switch as resonator. The crystal was manufactured with the end mirror and output coupling mirrors directly coated on the end faces. The laser pulse energy and pulse duration are inherently given by the design of this resonator and cannot be influenced after the manufacturing of the rod. The advantage of this monolithic resonator design is the lack of any optical components needing alignment and consequently this type of laser is highly robust against environmental influences as high temperatures and vibration.


Mirror2

Mirror3

Laser Beam

Mirror Lens

Laser Window Laser Beam y

Lens Powermeter Mirror1

Optical Window x

Tt-Laser Injector Element

Fig. 2. Optical path: table-top laser to experimental thruster.

A hexagonal arrangement of laser diode bars, mounted on a copper heat spreader ring, is located around the laser rod and provides the necessary pump light to operate the laser. The mechanical setup with this ring-type pump chamber offers the possibility to stack several pump rings and thus increase laser power and to use a liquid loop to cool pump diodes and the laser rod. Figure 3 (left) shows the laser resonator with two pump rings and the laser with protective housing (right) compared to a standard size computer mouse.

Fig. 3. Left: laser chamber; right: laser with protective housing.

The laser generates output pulses with 30 mJ, a pulse duration of 2.5 ns and a collimated beam diameter of 2.5 mm. A power driver unit using a capacitor array as energy storage was developed to operate the laser at a maximum continuous repetition rate of 40 Hz and a burst rate of up to 100 Hz. An optical breakdown in air can be generated using focusing lenses with focal lengths of up to 50 mm. In this test campaign a focal length of 36 mm was used. Figure 4 depicts a cut-view of the experimental thruster equipped with the HiPoLas® laser. A tubus with the 36 mm sapphire lens soldered at its end was used to mount the laser onto the experimental thruster. 2.3.

Optical diagnostics

A standard Z-set-up, as shown in Fig. 5, was used for schlieren imaging. Two high-speed CCD video cameras (Ultima 1024 C and Fastcam SA-X) at 4000 and 19200 fps and with a resolu#194445 - $15.00 USD Received 23 Jul 2013; revised 20 Sep 2013; accepted 14 Oct 2013; published 4 Nov 2013 (C) 2013 OSA 4 November 2013 | Vol. 21, No. S6 | DOI:10.1364/OE.21.0A1126 | OPTICS EXPRESS A1132

Fig. 4. Experimental thruster equipped with the HiPoLas® laser.

tion of 512 x 128 and 1024 x 368 pixel respectively were implemented for image recording purposes. Spontaneous OH emission was recorded by an intensified high-speed CCD video camera (Fastcam APX I2) with 10000-24000 fps and a band-pass filter (310 nm ± 5 nm). The camera was placed at an angle α ≈ 20◦ to the combustion chamber. The complete combustion chamber was visualised with a resolution of 512 x 128 pixel.

Lamp Spherical Mirror

Combustion Chamber

Spherical Mirror

Tt‐Laser

Knife Edge

Flow Direction

Lens

OH Camera

Schlieren Camera Fig. 5. Optical diagnotic set-up.


3.

Results and discussion

The experimental results obtained encompass observations regarding the effect of pre-ignition phenomena on the dynamics and character of the ignition processes as well as observations which are more strictly related to laser ignition, namely the required energy levels for different ignition locations. Using the table-top laser mapping tests were conducted. Initial tests with high incident energy, to identify potential locations of energy deposition, were followed by tests in which the delivered laser beam energy was lowered to identify minimum incident energies required for ignition at the various deposition locations. Tests at sucessful ignition locations identified with the table-top laser were repeated using the HiPoLas® laser. Lessons learned from the HiPoLas® tests include the use of an inert gas, helium, to avoid water condensation or ice formation on the outer surface of the focusing lens located inside the tubus adapter used to install the laser onto the experimental thruster. Without such gas, no plasma could be created. There were no other issues regarding thermal or mechanical loads. The following sections will discuss the chamber pre-ignition conditions, energy deposition locations and resulting ignition probabilities. The probabilities obtained using the HiPoLas® are compared to those obtained with the table-top laser for the same focusing locations but for higher delivered energies. 3.1.

Pre-ignition conditions

Pre-ignition conditions and flow fields play a fundamental role in determining ignition success [14]. Table 4 summarises the feed conditions as well as the pre-ignition chamber pressures and flow velocities for LOx/GH2. The conditions are representative of a future RCS thruster implemented on a launcher and fed from the main engine tanks with a global mixture ratio, MR = 5. Table 4. Propellant feed conditions and combustion chamber pressure levels during testing

Feed/Chamber Condition Oxygen feed pressure, p fox Oxygen feed temperature, T fox Hydrogen feed pressure, pGH2 Hydrogen feed temperature, TGH2 Liquid oxygen injection velocity vox Gaseous hydrogen Mach number, MGH2 Gaseous hydrogen flow speed at MGH2 = 2.8, vGH2 Velocity ratio (fuel/oxidiser), vR Gaseous hydrogen flow sonic speed, vGH2s Chamber pressure prior to priming, pcc1 Chamber pressure prior to ignition - LOx/GH2, pcc2 Steady state combustion pressure pcc3

Average Value 2.5 bar 90 K 12 bar 224 K 8 m/s 2.8 3206 m/s 407 1145 m/s approx. 20-75 mbar 450 mbar 1.5-1.7 bar

The most important characteristic of the flow field observed in the combustion chamber is the Mach disk structure caused by the supersonic flow of the gaseous hydrogen located as depicted in Fig. 6. The gaseous hydrogen is underexpanded at the exit of the annular slit and the resulting system of shock and expansion waves is a complex one. Expansion waves travelling from the inner edges of the annular slit deflect the gaseous fuel towards the liquid oxygen jet thus causing the latter to converge. A shock wave subsequently deflects the gaseous fuel outwards allowing the oxygen to diverge. It has also been observed that the angle of the spray downstream of #194445 - $15.00 USD Received 23 Jul 2013; revised 20 Sep 2013; accepted 14 Oct 2013; published 4 Nov 2013 (C) 2013 OSA 4 November 2013 | Vol. 21, No. S6 | DOI:10.1364/OE.21.0A1126 | OPTICS EXPRESS A1134

the Mach disk is affected by oblique shocks reflected from its surface. When the interaction is weak, the spray takes the form of an expanded jet. However, when the interaction is stronger, changes in the gradient of the spray can be clearly identified at distinct inflection points and the slope becomes increasingly more shallow.

Fig. 6. Spray image LOx/GH2 prior to ignition.

A number of important characteristics dictating which regions of energy deposition will lead to successful ignition can be identified in Fig. 6: • The most prominent feature is the inner liquid oxygen core jet which displays two very different regions of spray breakup: an upstream region in which no breakup takes place and in which the jet at first converges, and a second downstream region in which a growing shear layer may be identified and in which significant spray breakup occurs • A Mach disk and oblique shock structure resulting from the supersonic gaseous hydrogen which induce the two regions of liquid oxygen spray breakup • A large recirculation zone which extends for the whole length of the cylindrical portion of the chamber and into the contraction portion 3.2.

Energy deposition locations

Mapping tests conducted using the table-top laser examined a number of different energy deposition locations. These locations are grouped into four major regions: near-faceplate (A), shear layer (B1 and B2 ), downstream region (C), and near-faceplate recirculation zone (D) as shown in Fig. 7. The investigation determined that, despite very high energies delivered (approximately 160 mJ), focusing of the laser beam in the near-faceplate regions (Region A and Region D) and upstream of the Mach disk (Region B1 ) does not lead to ignition. The local LOx/GH2 flow field, which is characterised by supersonic flow at MGH2 = 2.8 and by a contained liquid oxyen jet at short distances from the injector head seem to be inhibiting parameters in that they result in fuel-rich mixture ratios at short ignition delay times. In all tests, the energy was delivered as soon as the propellants entered the combustion chamber, i.e. at low ignition delay times. Ignition was therefore initiated not during steady-state cold flow conditions but rather during the priming transients. An increase of the ignition delay would lead to more favorable mixture ratios as oxygen is carried to these regions via the recirculating flow but would also increase the probability of hard ignitions with high ignition peaks and is therefore avoided. Furthermore, the high hydrogen velocities result in an increase in the convective heat losses of initial plasma kernels effectively increasing the probability of quenching [15]. Flame kernel stretching due


to velocity fluctuations within or in the vicinity of the shock structures also result in quenching [16]. Successful regions of ignition identified with the table-top laser are B2 and C located downstream of the Mach disk along the shear layer region and in the larger breakup region respectively. Investigations using the HiPoLas® laser were therefore conducted in these two regions only. Figure 7 depicts the exact (x,y) locations of energy deposition for the HiPoLas® laser on the background of the pre-ignition flow field.

Fig. 7. Combustion chamber flow field with HiPoLas® laser plasma ignition regions for LOx/GH2.

3.3.

Oxygen-hydrogen plasma ignition using the HiPoLas® laser

Ignition tests conducted were aimed at determining locations of successful ignition and the probability of success for a certain delivered energy level. For this reason all tests performed were examined to determine whether a plasma or flame kernel resulted after energy deposition and whether such a kernel then led to a successful ignition of the entire combustion chamber. Repetition tests were performed to determine the repeatability of ignition success or failure. Table 5 provides a summary of the regions of energy deposition in the HiPoLas® laser plasma ignition experiments for LOx/GH2, the energies delivered and associated ignition probabilities achieved as well as the number of tests conducted in each region. For comparison purposes, Table 5 also summarises the minimum incident energies for 100% reliable ignition which were measured using the table-top laser. Table 5. Oxygen-hydrogen energy deposition and ignition probabilities

Region

Laser

Position, (x, y)

Region B2 Region B2

HiPoLas® table-top

(11-15 mm, 4-7 mm) (12-16 mm, 5-8 mm)

Region C Region C

HiPoLas® table-top

(32-34 mm, 8 mm) (57-60 mm, 10-17 mm)

Incident Energy 30 ± 3 mJ 72 mJ ± 5 mJ 54 mJ ± 5 mJ 30 ± 3 mJ 66 mJ ± 5 mJ

Ignition Probability 17% 100% 50% 33% 100%

Sample Size 30 53 9 7

Ignition probabilities were very low with the energy level of 30 mJ ± 3 mJ provided by the HiPoLas® laser. Of the 39 tests conducted only 6 resulted in ignition of the combustion chamber. In Region B2 , the ignition probability is incrementally reduced from 100% to 17% #194445 - $15.00 USD Received 23 Jul 2013; revised 20 Sep 2013; accepted 14 Oct 2013; published 4 Nov 2013 (C) 2013 OSA 4 November 2013 | Vol. 21, No. S6 | DOI:10.1364/OE.21.0A1126 | OPTICS EXPRESS A1136

as the energy delivered is lowered from 72 mJ ± 5 mJ with the table-top laser to 30 mJ ± 3 mJ with the HiPoLas® . Interesting however, is the fact that the probability of ignition achieved is significantly higher at locations further downstream (Region C) despite the more stochastic nature of the spray in these downstream regions. Although the statistical basis for an extensive analysis of the HiPoLas® results is missing, a number of flow characteristics can be identified that may be responsible for this. Firstly, the shear layer thickness increases with distance downstream. Given that the initial plasma volume created by the HiPoLas® laser is very small (of the order of 50 µm), a thiner shear layer is associated with a higher probability that energy will be deposited in regions where mixing is not optimal and rather in fuel-rich or oxidiser-rich portions of the shear layer. Furthermore, small variations in the focal spot location and/or small variations in the radial position of the shear layer due to variations in mass flow which are possible during start-up transients will also affect the relative position of energy deposition. As Fig. 7 shows, at positions further downstream, the mixing region is not only larger but, due to the fact that spray breakup is more, small oxygen droplets, which are important for sustaining combustion, exist rather than only a unified jet. Studies have shown that, for a given energy level, ignition is more likely to occur in regions where smaller propellant droplets are localised [16] effectively leading to a reduction in the minimum ignition energy [15]. In non-resonant laser ignition a large portion of the absorbed energy is used up in the form of blast wave losses [11, 12]. Despite the fact that it is frequently argued that the blast wave energy is wasted, the blast wave itself has a significant impact on the flow field and may play an important role in the preparation of injected propellant for combustion and specifically in droplet break-up [17]. The impact of the blast wave on the oxygen jet surface at locations further downstream where the propellant velocities are slower due to the increased distance from the injector head needs to be investigated further to determine if it enhances an already initiated break-up. 3.4.

Continuous pulse and burst operation of the HiPoLas® laser and kernel probability

All tests conducted were executed following testing sequences that aimed at reducing the ignition delay whilst ensuring that at the precise moment of energy deposition, a suitable mixture would be present at the focusing location with stable fluid flow conditions to avoid a potential flame extinguishing due to mass flow oscillations. Thus, after opening of the main propellant run valves, the laser was triggered such as to have energy deposition at t = 0ms. Successful ignition results in a pressure increase in the combustion chamber only 1-2 ms after energy delivery. Initial tests conducted using the HiPoLas® laser had only one energy deposition event. As mentioned above, the laser beam was focused into the chamber at t=0ms. As these tests were not successful, subsequent tests were conducted with the laser set to pulse at a frequency of 5 Hz for the entire duration of the test (2 seconds) to assess whether the chamber would ignite at a later time and thus whether t = 0ms was too early. Following this approach, it was found that in some cases a later energy deposition indeed led to ignition of the combustion chamber. Ignition in the 6 successful test runs took place at differing times in the range 2-1130 ms, with only one ignition (t = 2 ms) taking place after the first energy deposition. No correlation has been found between the time of ignition and the overall conditions in the combustion chamber. A detailed analysis of the local mixture ratio would be necessary. OH images recorded during all tests show that energy depositions which resulted in a plasma kernel did not necessarily lead to ignition of the entire combustion chamber. OH images reveal that in all successful test runs mentioned, with the exception of that which occurred at 2 ms, plasma kernels were observed for energy deposition events prior to the one that finally led to ignition. This indicates that the energy was sufficient to create a plasma kernel at the deposition


location but that the latter was not always energetic enough to compensate for energy losses. Due to the high injection velocities (8 m/s for liquid oxygen and > 1145 m/s for gaseous hydrogen) it is clear that with a pulsing frequency of 5 Hz no energy accumulation occurs, each plasma kernel is created in a completely renewed region. For the same reason, tests conducted with a short duration burst of laser pulses at 100 Hz (i.e. 5 bursts) did not increase the ignition probability of the chamber is all cases. 3.5.

Oxygen-hydrogen ablation ignition using the table-top laser

Mapping tests conducted with the table-top laser also included laser ablation tests. The laser beam was focused onto a metal copper target located at (x,y) = (0.8-0.9 mm, 5.9-6.0 mm) from the centre of the injector element using a 80 mm lens as depicted in Fig. 8. In terms of ignition probability and energy levels the ablation ignition tests using the table-top laser were the most successful with 14.5 ± 5 mJ incident energy leading to 100% ignition success. Incident energies required for LOx/GH2 for ablation ignition were much lower than those for plasma ignition. No ablation tests were conducted using the HiPoLas® laser. However, the results obtained suggest that ablation ignition is a viable option for a miniaturised laser in terms of energy levels. Laser Beam

Metal Target

Plasma Propagation

GH2 LOx

Fig. 8. Metal target location, laser beam focusing for ablation testing and plasma propagation for low energies delivered.

A total of 24 ablation tests were made. No degradation of the metal target was observed. 4.

Conclusion

The investigation conducted has demonstrated that ignition of a cryogenic LOx/GH2 thruster under altitude conditions, i.e. low pressure conditions, using laser ignition is a viable concept. It was however also shown that the energy required which will ensure 100% reliable ignition in the case of laser plasma ignition (≈ 72 mJ) is well above the energy levels currently provided by miniaturized lasers. Energy levels required for ablation ignition of LOx/GH2 (≈ 14.5 mJ), on the other hand, are well in the range of the provided energy output. Furthermore, this study has demonstrated that the direct mounting of a laser ignition system onto a thruster is a viable option. The results on the one hand have demonstrated that for a future implementation of a miniaturised laser for plasma ignition of cryogenic propellants under low pressure conditions the energy output of miniaturised lasers has to been significantly increased, and on the other hand spur for further research to be conducted in the field of laser ablation, which in terms of energy #194445 - $15.00 USD Received 23 Jul 2013; revised 20 Sep 2013; accepted 14 Oct 2013; published 4 Nov 2013 (C) 2013 OSA 4 November 2013 | Vol. 21, No. S6 | DOI:10.1364/OE.21.0A1126 | OPTICS EXPRESS A1138

alone seems to be more the more viable option. Investigations are required into issues related to the physical phenomena associated with ignition and propellant management prior to ignition but also to more hardware related issues. Investigations are required into the minimum necessary incident energy as a function of target placing and geometry and resulting target lifetime. Further more, additional experimental studies need to be performed to further understand the relationship between the local flow dynamics and resulting ignition probabilities, as well as the relationship between the dynamics of flame kernel development and ignition overpressure. Acknowledgments The experimental work presented in this paper was performed in the frame of the ESA TRP Cryogenic RCS Thruster Technology - Laser Ignition. The authors thanks ESA for the financial and technical support provided.
