http://www.diva-portal.org
Preprint
This is the submitted version of a paper presented at Space Propulsion Conference 2018,
organized by 3AF and ESA.Citation for the original published paper:
Fridh, J., Östlund, J. (2018)
ON THE USE OF METHANE IN ROCKET NOZZLE COOLING CHANNELS: BENCH SCALE TRIALS
In:
N.B. When citing this work, cite the original published paper.
Permanent link to this version:
http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-294136
1
SP2018_ 00089
ON THE USE OF METHANE IN ROCKET NOZZLE COOLING CHANNELS
BENCH SCALE TRIALS Jens Fridh (1), Jan Östlund (2)
(1) Royal Institute of Technology (KTH), Dept. of Energy Technology, 100 44 Stockholm, Sweden, Email: jens@energy.kth.se
(2) GKN Aerospace, Dept. of Nozzles, 461 38 Trollhattan, Sweden, Email:
jan.ostlund@gknaerospace.com
KEYWORDS: methane propellant, nozzle, heat transfer, cooling channels
ABSTRACT:
A combination of conjugate heat transfer calculations and experiments in a dedicated test rig at Royal Institute of Technology (KTH), Sweden are performed in order to characterize methane’s impact on the nozzle cooling channels in terms of heat transfer, coking and pressure loss. The design procedure includes development of a numerical conjugate heat transfer model as well as inert trials, not only in order to validate model but more importantly to gain experience in order to reduce uncertainties in the final design. In the final design, the methane can be pre-heated to 655 K at pressure levels between 10 to 160 bars and enters the final heater that simulates the heat load from the flame- side with electric cartridges heating a well-insulated copper block. The heat flux is between 1 to 7 MW/m2 for cooling channel flows representative to a full- scale nozzle of an upper stage engine.
1. NOMENCLATURE
A area [m2]
Dh hydraulic diameter [m]
Ė internal energy power [J/s]
M Mach number [-]
Nu Nusselt number [-]
Pr Prandtl number [-]
Q heat energy [J]
𝑄̇ heat power [W]
Re Reynolds number [-]
T temperature [K]
𝑊̇ work power [Nm/s]
c velocity [m/s]
cp specific heat capacity at constant pressure
[J/kgK]
h specific enthalpy or heat transfer coefficient
[J/kg] or [W/m2K]
k thermal conductivity [W/mK]
𝑚̇ mass flow [kg/s]
p pressure [Pa]
q heat flux [W/m2]
r recovery factor [-]
t time [s]
𝛼 thermal diffusivity [m2/s]
𝛾 ratio of specific heats [-]
𝜇 dynamic viscosity [Pa s]
𝜌 density [kg/m3]
𝜈 kinematic viscosity [m2/s]
subscripts
1,2 positions (inlet, outlet) cold the cold side of the wall hot the hot side of the wall r reduced
∞ Infinity, here “core” flow x,y,z coordinates
2. INTRODUCTION
A propulsion system using hydrocarbons, liquid or hybrid, is a challenge for today’s rocket and space propulsion systems as demonstrated by [1] to [13].
As a consequence of the good performance in specific thrust in combination with operating benefits such as low toxicity, availability, storage stability and low production cost liquid biogas/natural gas (LNG) with a high content of methane is one of the most interesting future propellants for rocket engines with liquid fuel. It is of strategic importance that the industry and academia develop necessary knowledge of the heat transfer characteristics and material influence at relevant operating conditions. Here the focus is on methane in gas phase as a coolant for the rocket nozzle walls.
Before combustion the fuel is lead through cooling channels in the rocket nozzle in order to cool it to withstand the high heat load from the rocket flame.
Fuels with a high content of methane can contain impurities such as sulfur that leaves deposits on the metal surface that in turn increases the pressure loss and decreases the heat transfer, which ultimately can lead to failure of the rocket nozzle.
Furthermore, methane undergoes thermal decomposition at high temperatures that leads to coking of cooling channels, which is affected by the amount of nickel in the surrounding material.
Cryogenic experiments with methane have been performed in some countries (foremost the USA) but in other parts of the engine where copper materials are used, [14] to [17].
Higashino et al. [18] performed tests on nickel- alloys under moderate pressures that shows a material dependent thermal cracking of pure methane (99.99 %). The gas temperature when cracking occurs decreases at increased nickel content from about 800 °C for 0 % Ni and down to about 600 °C for 78 % Ni for the studied materials.
Material analysis of the steel surfaces shows
2 evidence of carbon deposits (coking) of the studied materials that the authors connect to the thermal cracking of the methane. When the cracking occurs and what kind of deposits there are when a lower fuel grade is used, and how this is affecting the heat transfer and pressure loss is not published in open literature.
Methane will be used as propellant of the near future space vehicles and rocket engines. Cooling characteristics of engines, especially CH4 thermal cracking characteristics depend on material candidate for the nozzle and the combustion chamber cooling passage material and on the gas temperature. For increased performance and operability of next generation rocket engines a better understanding of the thermal stability of methane in the cooling channels is required. This investigation focuses on nickel-alloy steels and typical cooling geometries used in the rocket nozzles where the methane is in gas phase. The overall objective with the study is, for different relevant nickel-alloys and typical cooling channel geometries and operating conditions determine the heat transfer characteristics.
3. THEORY
The conjugate heat transfer problem, schematically shown in Fig. 1 includes the conductive heat transfer that takes place in the cooling channel walls (body) as well as the convective heat transfer in the fluid domain. Where the boundaries are the wall conditions and the gas inlet and outlet. Distributed over one of the walls of the rectangular cooling channel there is a heat flux (addition) and the other walls are treated adiabatically in initial calculations but in numerical simulations and tests associated with a heat loss.
The heat distribution, or temperature distribution in an isotropic material can be described with the parabolic differential heat equation, Eq.1 that can be derived from Fourier’s Law of heat conduction Eq.2 here stated one-dimensional, and the conservation of energy. T is temperature, t time, x-y-z space coordinates and is the thermal diffusivity, Eq.3, where k is the thermal conductivity of the material, cp the specific heat capacity at constant pressure and the density of the material, [19]. In Eq.2 q denotes the heat flux and A is the area.
Figure 1: cooling channel and conjugate heat transfer, generic case
𝜕𝑇
𝜕𝑡= 𝛼 (𝜕2𝑇
𝜕𝑥 +𝜕2𝑇
𝜕𝑦 +𝜕2𝑇
𝜕𝑧) Eq.1
𝑄
𝐴= 𝑞 = −𝑘𝑑𝑇
𝑑𝑥 Eq.2
𝛼 = 𝑘
𝑐𝑝𝜌 Eq.3
The fluid domain can be solved by applying conservation laws; energy (power), momentum (Newton’s second law) and continuity that can be expressed in one dimensional form with Eq.4 to 6 for a Newtonian fluid and steady flow neglecting any potential energy change. The convective heat transfer, in general a boundary layer phenomenon is related to the temperature difference between the fluid and the body via Newton’s cooling law, Eq.7 here one-dimensional where the heat transfer coefficient h can be expressed with the Nusselt number that is the ratio of convective to conductive heat transfer normal to a boundary. Tr is the adiabatic recovery temperature and Tw the wall temperature on the fluid side. The recovery temperature is a function of the free-stream velocity, the free-stream temperature and the Prandtl number, Eq.8. At low Mach numbers the recovery
3 temperature is close to the fluid stagnation temperature. The recovery factor r can be approximated with ∛Pr for engineering purposes.
The Prandtl number Pr is a non-dimensional fluid flow parameter that defines the ratio of momentum diffusivity to thermal diffusivity according to Eq.9.
𝑑𝐸̇ = 𝑄̇ − 𝑊̇ = {𝑊̇ = 0} = 𝑚̇ ∙ ∆ℎ̇ 0 Eq.4
∑ 𝐹𝑥=𝜕(𝑚 ∙ 𝑐𝑥)
𝜕𝑡 = 𝑚̇(𝑐𝑥1− 𝑐𝑥2) Eq.5 𝑚̇ = 𝜌𝑥∙ 𝑐𝑥∙ 𝐴𝑥= 𝑐𝑜𝑛𝑠𝑡𝑎𝑛𝑡 Eq.6
𝑄
𝐴 = ℎ(𝑇𝑤− 𝑇𝑟) Eq.7
𝑇𝑟= 𝑇∞(1 + 𝑟 (𝑀∞2𝛾 − 1
2 )) Eq.8
𝑃𝑟 =𝜈
𝛼= 𝜇 𝜌⁄
𝑘 (𝑐⁄ 𝑝𝜌)=𝑐𝑝𝜇
𝑘 Eq.9
In order to analytically estimate the metal temperatures on the hot side of the cooling channel for a preset of boundary conditions (geometry, material, heat flux, fluid flow, inlet pressure, inlet temperature) an approach with correlations found in open literature was used. For the calculations herein homogenous material was assumed and turbulent steady flow that is incompressible at the inlet. All surfaces except where the heat flux is, the
“hot” side, are considered adiabatic. Gas properties for pure methane were taken from NIST REFPROP [20].
Given the BCs at the inlet (p1, T1) the remaining fluid properties are determined and by assuming fully developed turbulent flow in the channel together with a friction coefficient for smooth surfaces the pressure drop can be estimated by a friction loss correlation, assuming an average stream-wise inlet velocity, and thereby the outlet pressure can be estimated. The outlet enthalpy is calculated as the inlet enthalpy plus the absorbed specific heat energy according to Eq.10. With p2 and h2 the remaining quantities at the outlet are determined. In order to estimate the convective heat transfer, a correlation for the Nusselt number (Dittus-Boelter), right-hand side of Eq.11 that gives similar results as the Prandtl-Hoffman equation, Eq.11, typically used for forced convection at turbulent pipe flow. The heat transfer coefficient h can be calculated from the Nusselt number together with the thermal conductivity of fluid and a characteristic length (hydraulic diameter), Eq.12. By using Eq.7 with the bulk gas temperature, Tgas, the fluid exposed wall
temperature, Tw,cold can be calculated at the inlet and outlet. For the heat conduction in the cooling channel wall exposed to heat flux the heat conductivity is approximated with a linear correlation for the relevant temperature range. The hot wall temperature is calculated via Fourier’s law, Eq.2, and can be expressed as Eq.13.
ℎ2= ℎ1+𝑄
𝑚̇ Eq.10
𝑁𝑢 = 0.0395 ∙ 𝑅𝑒3 4⁄ ∙ 𝑃𝑟
1 + (1.5 ∙ 𝑅𝑒−1 8⁄ ∙ 𝑃𝑟−1 6⁄ ) ∙ (𝑃𝑟 − 1)≈
≈ 0.023 ∙ 𝑅𝑒0.8∙ 𝑃𝑟0.4
Eq.11
ℎ = 𝑘 ∙𝑁𝑢
𝐷ℎ Eq.12
𝑞 = − 𝑘
𝑡ℎ𝑖𝑐𝑘𝑛𝑒𝑠𝑠∙ (𝑇𝑤,ℎ𝑜𝑡− 𝑇𝑤,𝑐𝑜𝑙𝑑) Eq.13
4. CONCEPTUAL HEATER DESIGN
Due to the flammable nature of the planned tests and the high heat loads present and pressures, up to 160 bars, it is considered necessary to elaborate on a conceptual design of the test rig and to perform heater trials with inert gas in order to validate the numerical model and be able to judge the viability of the final heater design.
4.1. Inert Trials
The inert tests are conducted with air as fluid and a test cooling channel with high Nickel content, where the geometric parameters and operating points can be seen in Tab. 1. A Cu-block (Cu-OF) is heated up by six electric cartridges of 400 W, resistively controlled by an adjustable transformer. A mass flow controller, Bronkhorst with a max capacity 180 kg/h, is used to control and thermally measure the air mass flow with an uncertainty of ±0.1% Full Scale (FS) + 0.5% of read value. The heater power is measured by a power meter (± 1% of FS) and gas temperatures and Cu-block temperature are measured with type K thermocouples. The heat losses from the insulation varies between 2…4 % of added power depending on the cooling mass flow, the insulation material is calcium silicate. Here, for the inert trials the Cu-block max temperature is limited to 973 K due to electric cartridge limitations.
Furthermore, the outlet pressure is limited to 10 bars due to instrument limitations. The inlet and outlet pressures where measured with UNIK 5000 GE absolute sensors with a full scale of 20 bars and an uncertainty of 0.1% of FS. Fig. 2 shows the setup
4
Table 1: Inert trial parameters Norm. mass flow
[-]
Heat flux [MW/m2]
Outlet pressure [bar]
Inlet temp.
[K]
w / h ratio [-]
t / h [-]
Length / h [-]
0.2…1 1.1…2.1 10 293 0.6 0.125 37.5
Figure 2: Left-hand picture: inert trial setup. Right-hand picture: assembled in the test rig, without top insulation. Total insulation 600*480*380 mm (l*w*h)
To verify the quality of the brazing a piece was cut post-trials and was step-wise heated up to 1223 K and then cooled down in room temperature.
Although the thermal expansion is different for the Cu compared to the coolant channel material no visible signs of defect brazing could be detected.
From a heat transfer perspective, it is important with a conductive coupling or fixation of the cooling channel onto the heater block. Therefore, brazing is selected, however, this means that it is also important to purge the cooling channel during pre- heating to avoid material degradation of the flow exposed surfaces.
4.2. Numerical Model
The inert test rig is modelled with conjugate heat transfer simulations in ANSYS CFX by Ristic [21].
The results obtained are then compared with initial calculations and measurements, in order to validate the model to be used in the subsequent design phase. Fig. 3 shows the CAD model of the inert test rig.
Figure 3: Inert test rig CAD model [21]
To ensure a high quality mesh in the fluid domain, a selective and multi-zone meshing approach is implemented. The fluid domain mesh, hexahedral mesh is generated initially, followed by the test
channel and the rest of the rig depicted if Fig. 4. The bottom heater block and insulation are assigned unstructured meshes.
A mesh quality study is conducted and the fluid boundary layer is modelled with inflation layers starting with a sufficient thickness at the wall to have a y+ < 1, and thereafter increased thickness with a factor of 1.2 for each subsequent layer where the number of layers are adapted to ensure a smooth transition to the core fluid elements.
Figure 4: Inert test rig mesh. Adapted from [21]
The fluid is modelled as a continuum and ideal gas with subsonic flow at the inlet and outlet, and the heat transfer simulated with the total energy model.
The turbulence model is set to Shear Stress Transport (SST). The mass flow and bulk total temperature are given as inlet boundary conditions with an isotropic turbulence intensity of 5%, and at the outlet the static pressure is set. Moreover, after setting appropriate material properties for the solid domains, an average heat flux input is set as a boundary condition on the solid cylindrical cavity surfaces for the electrical cartridge’s effective length. The external insulation boundary condition
A-A A A
5 is set to 293 K and with a heat transfer coefficient of 10 Wm-2K-1.
4.3. Results
The fluid temperature increase is compared in Fig. 5 for the Experimental, Analytical and Numerical results.
Due to spatial fluid gradients and single point measurement two numerical solutions exist, TM1 where the outlet temperature is mass-averaged for the cross-section and TM2 where the area corresponds to the thermocouple probe position.
Figure 5: fluid temperature increase, T vs.
supplied heat flux
There are a number of uncertainties in the experimental results. Firstly, the recovery temperature has been assumed to be ∛Pr.
Secondly, the conduction along the thermocouple shield going through the heated insulation material showed to have a great influence for the inlet temperature measurement. Thirdly due to non- homogenous flow field at the outlet the position of the thermocouple is sensitive and was difficult to verify in the experiments due to the setup. Another known difference between the analytical result and the averaged numerical result (dashed lines) is the local heat sinks due to the vertical channel side walls which is not modelled in the analytical model.
This effect becomes greater at lower mass flows when the temperature differences are greatest, as is apparent in Fig. 5. Fig. 6 shows the temperature contours for one case that reveals that the maximum copper block temperature is on the bottom surface close to the second from last heater element where the cooling effect from the downstream insulation is less. It should be mentioned that the numerical modelling also has its limitation with adiabatic pipe surface boundary assumed at the inlet and outlet, something that in reality gets affected by the surrounding insulation that shifts the absolute temperatures. This effect is more pronounced at the inlet and increases with larger heat flux.
Figure 6: Temperature contours in the solid block domain at a heat flux of 1.1 MW/m2 and a
normalized flow of 0.2 [21]
All the above observations and experiences conducting the inert trials are valuable inputs for reducing the total uncertainty of the final design, e.g.
the outlet bulk gas temperature will be measured with a special design temperature rake and placed further down at mixed out flow.
5. FINAL DESIGN
Although other heating options such as inductive heating and liquid metal were studied, electrical cartridges were selected due cost, control and simplicity reasons.
5.1. Final Heater Design
Different heater block geometries where conceptually studied starting from the numerical model. The goal is to maximize the heat pickup and to minimize the maximum block temperature. A decreased material volume will promote the heat pickup and stacking the electric cartridges standing will minimize the heater block volume. Shaping the upper part of the block to gradually decrease the material width towards the cooling channel had a small beneficial effect. The positive effect compared to a flat plate was not great and therefore from a manufacturing point of view a flat upper surface was chosen. Figs. 7-8 show the final design for one of the channels going to be tested, where also the inlet nozzle and outlet diffuser has an aerodynamic optimized design in the flow path creating a smooth transition going from rectangular to circular by employing wire EDM manufacturing.
Figure 7: Final heater design, cross sectional side- view
6 Figure 8: Final heater design, 3D view
Fig. 9 shows the maximum block temperature for several operating points revealing a much lower heating power is needed than installed. 21 electric cartridges, á 400W placed in three by seven matrix supplies a total effect of 8.4 kW that is then used to preheat the block before methane is supplied where a much lower power is needed to maintain the required fluid temperature increase.
Figure 9: maximum block temperature Tmax at an inlet temperature of 655 K (pout = 160 bar) [21]
In Fig. 10 a transient run is seen where the low over-shoot and relatively short pre-heating time is the main benefit.
Figure 10: transient study of the block heating [21]
5.2. Infrastructure
The methane gas supply system is an addition to the infrastructure of the laboratory and consists of gas-packs of 12 x 50 liters CH4 bottles placed on ground level in a gas cage. The pipe work, certified for 200 bars is externally lined to the roof test area.
The inlet gas flow is measured with a Coriolis flow meter and controlled by a control valve. The cooling channel inlet pressure level is controlled with a downstream pressure reduction valve. Both the massflow and back-pressure are PID-controlled.
The gas is flared after the experiments with an automatic flare system rated for 1.8 MW of thermal power and is integrated on top of the test cell container, Fig. 11. It only operates at low pressures and moderate temperatures meaning that pressure reduction has to be performed in two steps with an intermediate gas cooling in a plate heat exchanger.
The N2 gas for pipe/channel purge and the CO2 for test cell purge are provided via the existing lab infrastructure, approved for 60 bars.
7 Figure 11: Test area
The flow chart of the test facility in its present design is shown below in Fig. 12. The gas supply systems, pre-heaters and flare system are delivered from
AGA, ExHeat Ltd. and Euromekanik AB (C-deg design), respectively, which all are third-party audited systems. The remaining components except the final heater (in-house design) are purchased with third-party audit requirements but system-wise designed and selected by KTH. The ongoing risk assessment goes through the different risks and guides in the risk management. Important to realize is that the risk assessment is a continuous work that will continue throughout the life span of the test rig with necessary updates and modifications. There are currently 30 risks identified ranging from potentially very low human, equipment or project consequences to very serious ones with the most serious one being a total deflagration of the test rig. Due to strict safety authority requirements regarding combustible gases, high pressure and high temperature environments it has been judged necessary with a stand-alone safety control system, SIL3 class. The approach has been to have dual safety sensors for the test cell itself, third-party audited components and operating hot pressurized parts in an inert surrounding with CO2. Details like block-and-bleed valve arrangements, have been preferred over regular check valves due to safety and reliability reasons. The establishment of the process flow system has been an ordeal and preceded with a number of design selections.
Figure 12: Test facility process flow chart
8 6. COMMISSIONING
The gas supply and flare systems have been commissioned and the flow test indicates that pressures down to 10 bars can be used at the maximum mass flow, which will be important for the endurance tests in order to detect eventual coking.
Fig. 13 shows a capacity flow test with N2 from one gas pack.
Figure 13: flow capacity test
7. TEST AND MEASUREMENT PLAN
The objectives with the investigations are, for different relevant nickel-alloys and typical channel geometries and operating conditions determine for a specific hydrocarbon fuel:
heat transfer coefficient (HTC)
degree of coking and corrosion in the cooling channel
pressure loss
as a function of supplied heat load, wall temperature, Reynolds number and pressure level.
Four final cooling channels are going to be tested that have hydraulic diameters between 1.6 and 3.5 mm and a Nickel content between 6 and 72%. The overall possible operating envelopes for the different cooling channels with maximum pre- heating, 655 K, are summarized in Fig. 14 where the maximum allowable temperature on the hot side (heater block) sets the limit.
Figure 14: Operating envelopes for #1…3 configurations at maximum pre-heating, T1=655 K
and p1=160 bar. Config #4 is at a pre-heating of T1=655 K and p1=40 bar. Based on 1D calculation.
The maximum allowable outlet bulk gas temperature will be 795 K. The heat transfer is characterized with arrays of thermocouples in the gas flow, on the hot wall and on the cooled wall. The pressure loss is measured with a differential pressure transducer, single crystal silicon resonant sensor, with an adjustable range.
8. SUMMARY AND CONCLUDING REMARKS A combination of conjugate heat transfer calculations and experiments in a dedicated test rig at KTH, Sweden are performed in order to characterize methane’s impact on the nozzle cooling channels in terms of heat transfer, coking and pressure loss. In the viable design finally chosen for the test rig, the methane can be pre- heated from 273 to 655 K at pressure levels between 10 to 160 bars and enters the final heater that simulates the heat load from the flame-side with electric cartridges heating a well-insulated copper block. The heat flux is between 1 to 7 MW/m2 for cooling channel flows representative to a full-scale nozzle of an upper stage engine. This paper has presented the design of the test facility with initial trials outcomes.
The numerical model was validated with experimental data from inert trials with reasonable agreement. It was used to determine the final heater design and extract operating conditions and outlay the measurement plan.
9. FUTURE WORKS
Although the ambition for this paper was to include initial trial results from the final test rig assembly operating with methane it was not managed due to higher complexity of the infrastructure than first anticipated. Thus, it will have to be published in future works. This also includes material and gas analyzes.
10. ACKNOWLEDGEMENTS
The authors would like to acknowledge the Swedish National Space Board and the European Space Agency's Future Launcher Preparatory Programme (FLPP) for the financial support. Many thanks to MSc thesis student Richard Ristic for the excellent contribution of the numerical model. Also special thanks to Laboratory engineers Leif Pettersson and Göran Arntyr that have contributed a great deal regarding the infrastructure realization.
11. REFERENCES
1. Haeseler D., Gotz, A., Mading C., Roubinski V., Gorokhov V., and Khrissanfov S., 2002,
“Testing of LOX/Hydrocarbons Thrust Chambers for Future Reusable Launch Vehicles”, AIAA Paper 2002-3845, July 2002, 38th AIAA/ASME/SAE/ASEE Joint Propulsion Conference.
9 2. Preclik D., Hagemann G., Knab O., Mading C.,
Haeseler D., Haidn O., Woschnak A., and DeRosa M., 2005, “LOX/Hydrocarbon Preparatory Thrust Chamber Technology Activities in Germany”, AIAA Paper 2005-4555, July 2005, 41st AIAA/ASME/SAE/ASEE Joint Propulsion Conference.
3. Ierardo N., Biagioni M., and Pirrelli P., 2008,
“Development of the LM10-MIRA LOX-LNG Expander Cycle Engine for the LYRA Launch Vehicle”, Proceedings of the 2nd conference space propulsion 2008, May 2008.
4. Arione L., Ierardo N., Rudnykh M., Caggiano G., Lobov S., Shostak A., De Lillis A., and D’Aversa E., 2010, “Development status of the LM10-MIRA LOX-LNG Expander Cycle Engine for the LYRA Launch Vehicle”, Proceedings of space propulsion 2010, May 2010.
5. Pempie P., Frohlich T., and Vernin H., 2001,
“LOX/Methane and LOX/Kerosene High Thrust Engine Trade-Off,” AIAA Paper 2001-3542, July 2001, 37th AIAA/ASME/SAE/ASEE Joint Propulsion Conference.
6. Preuss A., Preclik D., Mading C., Gorgen J., Soller S., Haidn O., Oschwald M., Clauss W., Arnold R., and Sender J., 2008, “LOX/Methane Technology Efforts for Future Liquid Rocket Engines,” Proceedings, May 2008,
Proceedings of the 2nd Conference Space Propulsion 2008.
7. Crocker A. M. and Peery S. D., 1998, “System Sensitivity Studies of a LOX/Methane
Expander Cycle Rocket Engine”,AIAA Paper 1998-3674, July 1998, 34th
AIAA/ASME/SAE/ASEE Joint Propulsion Conference.
8. Brown C., 2004, “Conceptual Investigations for a Methane-Fueled Expander Rocket Engine,”
AIAA Paper 2004-4210, July 2004, 40th AIAA/ASME/SAE/ASEE Joint Propulsion Conference.
9. Schuff R., Mayer M., Sindiy O., Ulrich C., and Fugger S., 2006, “Integrated Modeling and Analysis for a LOX/Methane Expander Cycle Engine: Focusing on Regenerative Cooling Jacket Design,” AIAA Paper 2006-4534, July 2006, 42nd AIAA/ASME/SAE/ASEE Joint Propulsion Conference.
10. Strunz R., et al., 2006, “Main Stage Liquid Propulsion Activities within Europe's Future Launcher Preparatory Programme FLPP”, AIAA Paper 2006-4698, 42nd Joint Propulsion Conference, July 9-12, 2006, Sacramento, California
11. Coppinger R., 2015, "Airbus’ Adeline Project Aims to Build Reusable Rockets and Space Tugs". Space.com. Retrieved 2015-06-11.
12. Amos J., 2015, "Airbus unveils Adeline re-
usable rocket concept", BBC News. Retrieved 2015-06-05.
13. de Selding P. B., 2015, "Meet Adeline, Airbus’
Answer To SpaceX Reusability". Space News.
Retrieved 2015-06-05.
14. Maas E. et al. (2004). A High Heat Flux Facility Design for Testing of Advanced Hydrocarbon Fuel Thermal Stability, Air Force Research Laboratory Propulsion, Edwards AFB, CA 93524
15. Billingsley M. C. et al. (2007). Experimental and Numerical Investigations of RP-2 Under High Heat Fluxes, 54th JANNAF Propulsion Meeting / 5th MSS / 3rd LPS / 2nd SPS Joint Meeting, Denver, CO, 14-17 May 2007
16. Trejo A. et al. (2011). Development of a Carbo- Thermal Rig to Study Cryogenic Heat Transfer Characteristics, AIAA 2011-6018, 47th
AIAA/ASME/SAE/ASEE Joint Propulsion Conference & Exhibit 31 July - 03 August 2011, San Diego, California
17. Votta R. et al. (2014). Experimental Investigation on Methane in Transcritical Conditions, AIAA conf. Paper July 2014, DOI:
10.2514/6.2014-4005
18. K. Higashino et al. (2008). Fundamental Study on Coking Characteristics of LNG Rocket Engines, AIAA 2008-4753, 44th
AIAA/ASME/SAE/ASEE Joint Propulsion Conference & Exhibit 21 - 23 July 2008, Hartford, CT
19. Lakshminarayana, B. (1996). Fluid Dynamics and Heat Transfer of Turbomachinery, ISBN 0- 471-85546-4
20. Lemmon, E.W. et al., 2014, "NIST Standard Reference Database 23”, Beta version 9.11, July 29 2014, US Secretary of commerce.
21. Ristic, R. (2017). A High Heat Flux Facility Design for Testing of Methane in Rocket Cooling Channels, MSc thesis at KTH Royal Institute of Technology, EGI-2017-0031MSC EKV-1186.
