Inhibition Effect of Water Injection on Afterburning of Rocket Motor Exhaust Plume

Chinese

Journal of Aeronautics

Chinese Journal of Aeronautics 23(2010) 653-659

www.elsevier.com/locate/cja

Inhibition Effect of Water Injection on Afterburning of Rocket Motor Exhaust Plume Jiang Yi*, Ma Yanli, Wang Weichen, Shao Liwu Department of Aerospace Engineering, Beijing Institute of Technology,Beijing,100081, China Received 13 January 2010; accepted 19 June 2010

Abstract One of the most important characteristic signatures of the exhaust plume from rocket motor is the afterburning phenomenon, and the injected water into the plume could inhibit the afterburning. The calculation model for the gas-liquid multiphase flow field with chemical reaction in the plume is built. By inducing the energy source terms caused by the vaporization of liquid water, condensation of the vapor and chemical reaction in the energy equation, the gas-liquid multiphase flow field and the afterburning phenomenon are calculated in a coupling way. Mixture multiphase flow model is used to calculate the gas-liquid flow field, and the vaporization mechanism is used to investigate the water vaporization process. The temperature contours are obtained and accord well with the experimental photos. The mass fraction contours of primary species are obtained, which can indicate the extent of inhibition effect of water injection on the afterburning phenomenon in the plume. When water is injected into the plume, the region of afterburning reduces a lot, and temperature on the ground wall declines rapidly, which can decrease the ablation of the combustion gas to the launch ground. Key words: exhaust plume; finite rate chemistry model; afterburning; water injection; mixture multiphase flow; vaporization; coupling solution

1. Introduction1 During the launch process of the missile, the fuel rich combustion gases are ejected from the nozzle, which are of high temperature. After they are injected into the air, some oxygen is entrained into the plume and then the afterburning reaction occurs in the mixing layer of the plume[1]. The temperature and radiation intensity of the plume increase because of afterburning. When the liquid water is injected into the plume, the temperature of the flow field can decrease, and then the afterburning phenomenon can decline to some extent[2]. Up to now, lots of works have been done on the afterburning in the plume and water injection into the plume separately. W. H. Calhoon Jr[3] studied the inhibition methods of afterburning in the exhaust plume by numerical simulation. A. Bounif, et al.[4] investigated the differences between the frozen plume and the reaction plume. A. Roblin, et al.[5] studied the afterburning phenomenon and the cessation mechanism in fuel rich *Corresponding author. Tel.: +86-10-68942676. E-mail address: [email protected] 1000-9361/$ - see front matter © 2010 Elsevier Ltd. All rights reserved. doi: 10.1016/S1000-9361(09)60267-3

exhaust plume through numerical simulations. J. Troyes, et al.[6] calculated the two-phase flow field of exhaust plume of solid rocket motor. P. L. Viollet, et al.[7] studied the diffusion process of the two-phase flow field. S. M. Dash, et al.[8] investigated the two-phase process of the flow field in the nozzle. L. Gustavsso, et al.[9] studied the afterburning in the application of N2O reduction in the industrial fluidized-bed. K. B. Galitseyskiy[10] studied the turbulent chemistry interaction mechanism in detail. W. A. Engblom, et al.[11] studied the cooling effect of the liquid water to the high speed exhaust plume through three different kinds of experimental methods. van A. Foreest, et al.[12] studied the water vaporization mechanism and the cooling effect on the aerodynamic heating by using the water cooling system. P. Giordan, et al.[13] and M. Miller, et al.[14] studied the influence of water injection on the characteristics of the exhaust plume flow field. M. Molnar, et al.[15] studied the radiation signature of the plume with and without water injection using the two-time step method. M. Kandula[16] studied the inhibition effect of water injection on noise radiation of the exhaust plume. Although the afterburning mechanism and the water injection into the exhaust plume have been widely studied, the inhibition effect of the water injection on

· 654 ·

Jiang Yi et al. / Chinese Journal of Aeronautics 23(2010) 653-659

the afterburning phenomenon has not been fully explored in the literature available. It is this idea that instigates the authors to propose a prediction of the inhibition effect of water injection on the afterburning of rocket motor exhaust plume. The water vaporization and vapor condensation model is studied. Furthermore, the finite rate chemistry model and mixture multiphase flow model are used in a coupling way in the calculations. Finally, the calculated results are compared to the experimental data allowing the validation of the model. By inducing the energy source terms caused by the vaporization of liquid water, condensation of the vapor and chemical reaction in the energy equation, the gas-liquid multiphase flow field and the afterburning phenomenon are calculated in a coupling way. The Fluent software is used to calculate the flow field and nine species are considered, i. e. H2O, CO, CO2, H2, N2, O2, OH, H and O. The mixture multiphase flow model is used to calculate the gas-liquid flow field, and the water vaporization mechanism is used to investigate the water vaporization process. The H2/CO reaction system is used to simulate the afterburning phenomenon. The temperature contours and the mass fractions contours of primary species are obtained. The inhibition effect of water injection on the afterburning is analyzed. 2. Physical Models 2.1. Finite rate chemistry model The chemical reaction mainly near field. Arrhenius law can be detailed chemical mechanism in haust plume[17]. The rth reaction form as follows: N

kf

N

occurs in the plume used to describe the the afterburning exis written in general

¦ vicM i œ ¦ viccM i k i 1

b

(1)

i 1

where N is the number of chemical species in the system, vic the stoichiometric coefficient for reactant i in reaction r, vicc the stoichiometric coefficient for product i in reaction r, and Mi symbol denoting species i. The forward rate constant kf for reaction r is computed using the Arrhenius expression: kf ArT n exp( Er /( RT )) (2) where Ar is the pre-exponential factor, n the temperature exponent, Er the activation energy for the reaction, R the universal gas constant and T the temperature. Ar and kf have the same dimension of cm3·s/mole. 2.2. Vaporization equation During the interaction of liquid water and the combustion gas of high temperature, the influence of the water vaporization on the characteristics of flow field is great. The vaporization and condensation process are calculated in the finite cell. The vaporization rate

No.6

of water is calculated according to its saturation temperature. When the temperature of mixture is higher than the saturation temperature of water, the water transforms into vapor by absorbing energy, otherwise, the vapor liquefies to water by releasing energy. The thermal equilibrium is assumed between the two phases in the calculations. The transforming between vapor phase and liquid phase in each cell of the calculation domain is solved by the following formulas[18]. The formulas of water vaporization is Tmix ! Tsat ­ 0.1Vl Ul | Tl  Tsat | / Tsat (3) m l ® Tmix  Tsat ¯0.1Vv U v | Tmix  Tsat | / Tsat The formulas of vapor condensation is Tmix ! Tsat ­ 0.1Vl Ul | Tmix  Tsat | / Tsat m v ® (4) 0.1 V U | T T | / T Tmix  Tsat   v v mix sat sat ¯ The energy generated by vaporization is ­ Q m ʳʳ Tmix > Tsat (5) Sk ® lat l ¯Qlat m v ʳ ʳTmix < Tsat

where the subscript l is the liquid phase, and v the gas phase. m l and m v are the vaporization rate of liquid and the condensation rate of vapor in kg/(m3·s), Vl and Vv the volume fractions of liquid and gas, and ȡl and ȡv the density of liquid and gas. Tl is the temperature of liquid and Tsat the saturation temperature of water, which is determined by the local pressure. The relationship between the saturation temperature of water and the flow field pressure is given in Ref.[19]. Tmix is the temperature of the mixture, and the mixture includes both the gas phase species including water vapor and the liquid water. Sk is the absorbed or released energy, and Qlat the vaporization latent heat at different temperatures. The vaporization process of the water occurs instantaneously, and the time of vaporization has little influence on the calculation, so it is not considered in the calculations. The pressure of liquid water in the water pipes is low, so there is no obvious atomization phenomenon around the exit of the water pipes, and the diameter of the droplet may be very large. The atomization phenomenon of liquid water and the diameter distribution of droplets are not considered in the calculations. 2.3. Energy equation The vaporization and reaction energy source terms are induced in the energy equation. The chemical reaction model and water-fuel multiphase flow are calculated in the coupling way. The energy equation is given as follows: m w m (D q Uq Eq )  ’ ˜ ¦ [D qX q ( Uq Eq  p )] ¦ wt q 1 q 1 ª º Sk  Sh  ’ ˜ « keff ’Tmix  ¦ h j J j  (W eff v ) » j «¬ »¼

(6)

No.6

Jiang Yi et al. / Chinese Journal of Aeronautics 23(2010) 653-659

· 655 ·

where, Eq=hp/Uq + vq2 /2, h=6Yjhj, Yj is the mass fracT

tions of species j, hj= ³ cpj dTˈin which cp,j is the Tref

constant pressure specific heat of species j, and Tref = 298.15 K. D is the volume fraction of each phase. keff = (k+cpPt/Prt), k is the thermal conductivity, cp the constant pressure specific heat of mixture, Prt Prandtl number, Pt the turbulence viscosity coefficient. J is the diffusion flux, Xq the velocity of phase q and Weff the deviator stress tensor, Sh the chemistry energy source term. The subscript q represents the phase q and m is the total number of phases.

Fig.1

Schematic of computational domain.

3. Calculation Method and Cases

3.1. Flow field calculation method The RNG k-H turbulence model[20] is used in the calculations. The turbulent vortex and streamline curvature are considered in the model which could improve the accuracy of the calculations of exhaust plume flow field. The RNG k-H turbulence model provides an analytic expression of the low Reynolds number. So the RNG k-H turbulence model combined with the adiabatic standard wall function[21] could simulate the velocity gradient of the near wall such as the nozzle, the water pipes and the ground wall. The finite volume method is used to discretize the governing equations. The gas-liquid multiphase flow field with vaporization and the finite rate chemistry model are calculated coupled with the mixture multiphase flow model[22]. 3.2. Calculation cases Four water injection pipes are located around the nozzle exit in the near plume field. The angle between the water pipes and the centerline axis is 60°. A quarter of three-dimensional cylindrical domain is used to calculate the flow field shown in Fig.1(a). Fig.1(b) shows the calculation domain and the boundary conditions, in which Line BG is the launch ground wall. Points C, D, E and F are the temperature monitor points. The distances are 0.2 m, 0.3 m, 0.4 m and 0.5 m to the Axis AB separately. The distance from the nozzle exit to the launch ground is AB=1.75 m. Fig.2 shows the symme-

Fig.2

Mesh diagram of nozzle exit and water pipes.

try vicinal meshes of the nozzle exit and the water pipes. The domain inlet is located at two times of the nozzle diameters upstream of the nozzle exit. The inlet boundary type of the nozzle is pressure inlet, and the chamber pressure is 7 MPa. The temperature is 3 000 K. The pressure outlet boundary with general non-reflecting boundary conditions[23] is adopted at the pressure outlet shown in Fig.1(b) and the pressure is 101 325 Pa, and the outlet temperature is 300 K. Several calculation domains of different scales have been used to seek a sufficient length to attain the ambient condition. The length of the domain selected in the calculations proves to be sufficient. The acceleration of gravity is 9.8 m/s along x axis direction, which is the real direction of the gravity in the experiments. Nine species are considered (H2O, CO, CO2, H2, N2, O2, OH, H and O) and the mass fractions of the species are shown in Table 1. The mass flow rate boundary is used at the exit of the water pipes. The mass flow rate is three times of the mass flow rate of gas from the nozzle exit. The velocity of the water at the pipes outlet is about 25 m/s. Table 1

Mass fractions of species in the combustion chamber (species less than 104 are neglected)

Species

H2O

CO

H2

CO2

N2

Mass fraction

0.16

0.27

0.004

0.26

0.305

The H2/CO reactive system[24] is used to simulate

· 656 ·

Jiang Yi et al. / Chinese Journal of Aeronautics 23(2010) 653-659

the chemical reaction. The reaction mechanism is shown in Table 2. Table 2

Chemical reaction model in plume

Reaction formula

Chemistry rate constant

H2+O=OH+H

3.0u1014Texp(4 480/T)

H+ O2=OH+O

2.4u1010exp(8 250/T)

H+H+M= H2+M

3u1030 T 1

OH+OH= H2O +O

1u1011exp(550/T)

O+O+M= O2+M

3u1034exp(900/T)

O+H+M=OH+M

1u1029 T 1

H2+OH= H2O +H

1.9u1015T1.3exp(1 825/T)

CO+OH= CO2+H

2.8u1017T1.3exp(330/T)

H+OH+M= H2O +M

1u1025 T 2

CO+O+M= CO2+M

7u1033exp(2 200/T)

No.6

with and without water injection are calculated. Fig.4 shows the temperature distribution in the centerline axis of the frozen and reaction plumes without water injection. The nozzle exit plane is at about x=0.2 m. As seen in Fig.4, the temperatures of frozen plume and reaction plume are nearly vicinal before x=0.9 m, indicating that the reaction does not occur in this region. The temperature of reaction plume is 100-400 K higher than the frozen plume after x=0.9 m, caused by the energy released from the chemical reactions of the afterburning. The variation trend of the temperature accords well with that in Ref.[25]. The plume is blocked by the ground wall, so the temperature near the ground increases a lot, and then the chemical reactions are enhanced. In summary, it is necessary to consider the afterburning in the plume field calculation.

4. Results and Discussion

Experiments of solid rocket motor firing with and without water injection are carried out. The Phantom v10 high-speed camera and AGEMA-900 infrared camera are used to observe the flow field structure and the radiation intensity emitted from the exhaust plume. The observation distance is about 10 m to the test stand. The E12-1-C-U temperature sensors are used to detect the temperature on points C, D, E and F shown in Fig.1. The detecting bands of AGEMA-900 infrared camera are 2-5.6 ȝm. The sampling frequency is 15 images/s. and the sampling frequency of the Phantom v10 high-speed camera is 480 images/s. The measuring limit of E12-1-C-U temperature sensor is 3 300 K which could satisfy the demands of exhaust plumes temperature test. Fig.3 shows the test bench with water injection before firing, in which four water pipes can be seen. For safety, the motor has not been installed in the test bench.

Fig. 3

Fig.4

Temperature distribution in centerline axis of frozen and reaction plumes without water injection.

Fig.5 shows the temperature distribution in the centerline axis of reaction plume with and without water injection. The temperature decreases to about 400 K around the exit plane of the water pipes with water injection. There is a temperature peak value at x=0.85 m with water injection due to the block by the water. Temperature is a critical factor for species reaction and the temperature after x=0.9 m is too low to activate the chemical reactions. The reaction could be inhibited by the water injection.

Test bench with water injection before firing.

To study the inhibition effects of the water injection on the afterburning in the plume, the exhaust plume

Fig.5

Temperature distribution in centerline axis of reaction plume with and without water injection.

No.6

Jiang Yi et al. / Chinese Journal of Aeronautics 23(2010) 653-659

Fig.6 presents the temperature distribution in the symmetrical plane of reaction plume with and without water injection. As seen in Fig.6, the temperature after the water pipe is lower than that without water injection. The length of plume core region decreases to 0.7 m by the endothermic process of water vaporization. The temperature before water pipes is a little higher than that without water injection. This may be because the exhaust plume is blocked by the liquid water ejected from the pipes. The decrease of temperature may inhibit the occurrence of the chemical reactions in the plume.

Fig.6

Reaction flow temperature distribution with and without water injection.

Figs.7-8 show the experimental photographs taken by the high-speed camera and infrared camera. As the Fig.7(a) and Fig.8(a) show, the exhaust plume boundary region is brighter than the core region because of the afterburning. The ground temperature is also higher because the exhaust plume is blocked by the ground wall. The exhaust plume brightness scope reduces after water injection which indicates the decline of the temperature. The region before water pipes is brighter than that

· 657 ·

Fig.7

Experimental IR images.

Fig.8

High-speed photographs.

without water injection because of the water blockade. The flow field structure of the simulation results accords well with the experimental results. To validate the quantitative inhibition effect of water injection on the exhaust plume temperature and launch platform ablation, Table 3 lists the simulation and experimental temperature measured by the temperature sensors on points C, D, E and F shown in Fig.1(b). The temperature decreases greatly with water injection which suggests the inhibition of afterburning with water injection. The biggest relative error of temperature between the simulation and experiment is 11.5%. Table 3 proves reliability of the simulation models. The neglect of atomization of the water jets may decline the surface area of vaporization, so this may be the primary cause of the deviation of the numerical results from the experimentally measured temperatures. Besides, the ground wall is adiabatic, which may be another cause of the deviation. Table 3

Computational and experimental temperature on test points with and without water injection

Monitor

C

D

E

F

Simulation & without water injection/K

1 130

1 040

870

730

1 000

940

830

681

11.5

9.6

4.5

6.7

467

441

433

426

422

415

410

402

9.6

6.2

5.3

5.6

Experiment & without water injection /K Relative error & without water injection /% Simulation & with water injection/K Experiment & with water injection/K Relative error & with water injection/%

· 658 ·

Jiang Yi et al. / Chinese Journal of Aeronautics 23(2010) 653-659

Fig.9 separately shows H2 and CO2 mass fractions in the symmetrical plane of reaction plume with water injection shown in the upper portion while the one without water injection shown in the lower portion. The distance of the core region becomes shorter after water injection which causes the H2 and CO2 mass fractions decrease greatly after the water pipes. H2 is the reactant of the reaction, and the decrease of it in the core region may inhibit the extent of the reactions. CO2 is the product of the reactions, and the decrease of it shows that the reactions have been inhibited.

Fig.9

No.6

Fig.12 shows the contours of the volume fractions

Fig.10

Contrast of temperature contour of reaction plume and frozen plume with water injection.

Fig.11

Mass fractions contours of H2 and CO2 in reaction plume and frozen plume with water injection.

Contrasts of H2 and CO2 mass fractions of reaction plume with and without water injection.

To specify the region of the chemical reaction in the plume with water injection, the temperature contours of the reaction plume and the frozen plume with water injection are shown in Fig.10. As shown in Fig.10, the temperature of the reaction flow is slightly higher around the exit of the water pipes than that of the frozen flow, indicating that the afterburning reactions occur in this region. Fig.11 separately show H2 and CO2 mass fractions with reaction flow shown in the upper portion, while the one with frozen flow shown in the lower portion. The decrease of H2 is caused by the consumption of the afterburning reactions, and the increment of CO2 is caused by the production of the afterburning. The change of the two species indicates the extent of the afterburning reactions.

Fig.12

Contours of volume fractions of liquid water.

No.6

Jiang Yi et al. / Chinese Journal of Aeronautics 23(2010) 653-659

of the liquid water. The liquid water moves to the axis of the plume and concentrates in the core region, thus the gas plume is blocked by the liquid water, which could increase the contact area of the gas and the liquid water. The temperature of the gas plume would decline because of the water vaporization, and the chemical reactions may be inhibited in the plume. 5. Conclusions

(1) By inducing the energy source terms caused by the vaporization of liquid water and chemical reaction in the energy equation, the gas-liquid multiphase flow field and the afterburning phenomenon are calculated in a coupling way. The calculation results accord well with the experimental results, which indicate that the coupling solution is reliable. (2) The temperature and the mass fractions of the reactant species decrease with water injection which induces the afterburning inhibition. The afterburning reactions mainly occur in the core region after x=0.9 m in the centerline axis without water injection, while in the reaction plume with water injection, the afterburning occurs mainly in the region around the exit of the water pipes, indicating that water injection has an inhibition effect on afterburning in the plume. (3) The temperature on the ground wall declines a lot with water injection, which shows that water injection to the exhaust plume could decrease the ablation of the ground wall. Acknowledgement

[8] [9] [10] [11] [12] [13] [14] [15]

[16] [17] [18] [19] [20]

The authors would like to thank Hao Jiguang for his valuable help in the experiments.

[21]

References

[22]

[1]

[2] [3] [4]

[5] [6] [7]

Buckley F T , Myers R B. Studies of solid rocket exhaust afterburning with a finite difference equilibrium chemistry plume model. JANNAF 6th Plume Technology Meeting. 1971; 59-80. Kennedy R E , Olmsted G S, Ryan R E, et al. Cooling apparatus for a missile launching system. U.S. Patent No.5847307, 1998. Calhoon W H Jr. Evaluation of afterburning cessation mechanisms in fuel rich rocket exhaust plumes. ADA404895, 1998. Bounif A, Aris A, Gokalp I. Pressure effects on the spectral behavior of the thermal field in non-reacting and low Damkohler reacting flows. International Journal of Thermal Sciences 1999; 38 (9) : 819-831. Roblin A, Dubois I, Grisch F, et al. Comparison between computations and measurements of a H2/LOX rocket motor plume. AIAA-2002-3107, 2002. Troyes J, Dubois I, Borie V, et al. Multi-phase reactive numerical simulations of a model solid rocket exhaust jet. AIAA-2006-4414, 2006. Viollet P L, Simonin O. Modeling dispersed two-phase flows: closure, validation and software development. Applied Mechanics Reviews 1994; 47 (6): 80-84.

[23] [24] [25]

· 659 ·

Dash S M, Wolf D E, Beddini R A, et al. Analysis of two-phase flow processes in rocket exhaust plumes. Journal of Spacecraft 1985; 32 (3) : 367-380. Gustavsson L, Glarborg P, Leckner B. Modeling of chemical reactions in afterburning for the reduction of N2O. Combustion and Flame 1996; 106(3): 345-358. Galitseyskiy K B. Modeling afterburning of high-speed turbulent jets. Combustion, Explosion, and Shock Waves 2006; 42(2): 125-130. Engblom W A, Fletcher B, Georgiadis N. Validation of conjugate heat-transfer capability for water-cooled high-speed flows. AIAA-2007-4392, 2007. van Foreest A, Giilhan A, Esser B, et al. Transpiration cooling using liquid water. AIAA-2007-4034, 2007. Giordan P, Fleury P, Guidon L. Simulation of water injection into a rocket motor plume. AIAA-1999-2517, 1999. Miller M, Kneer M, Ramsey C, et al. Enhancement of ablative performance by injecting water into the exhaust plume. FMC Report E-4209, 1992. Molnar M, Marek C J. New reduced two-time step method for calculating combustion and emission rates of Jet-A and methane fuel with and without water injection. NASA/TM-2004-213046, 2004. Kandula M. Broadband shock noise reduction in turbulent jets by water injection. Applied Acoustics 2009; 17(7): 1009-1014. Miao R S. Launching fuel dynamics. Beijing: National Defense Industry Press, 2006. [in Chinese] Rock B S. Thermal-fluid analysis of a lithium vapourizer for a high power magnetoplasmadynamic thruster. PhD thesis, Worcester Polytechnic Institute, 2006. Lu Z Q. Two-phase flow and boiling heat transfer. Beijing: Tsinghua University Press, 2002. [in Chinese] Yakhot V, Orszag S A. Renormalization group analysis of turbulence: I. basic theory. Journal of Scientific Computing 1986; 1(1): 1-51. Launder B E, Spalding D B. Numerical computation of turbulent flows. Computer Methods in Applied Mechanics and Engineering 1974; 3(2): 269-289. Manninen M, Taivassalo V, Kallio S. On the mixture model for multiphase flow. VTT Publications 288, Technical Research Centre of Finland, 1996. Thompson K W. Time dependent boundary conditions for hyperbolic systems. Journal of Computational Physics 1987; 68: 1-24. Jensen D E, Jones G A. Reaction rate coefficients for flame calculations. Combustion and Flame 1978; 32: 1-34. Jiang Y, Fu D B. Numerical simulation for non-equilibrium chemically reacting fluid field of the solid rocket motor exhaust plume. Journal of Astronautics 2008; 29(2): 615-620. [in Chinese]

Biographies: Jiang Yi Born in 1965, he is an associate professor in Beijing Institute of Technology. His main research interest lies in launching jet flow and design of launching equipment. E-mail: [email protected] Ma Yanli Born in 1985, she is a Ph.D. candidate in Beijing Institute of Technology. Her main research interest lies in the gas-liquid multiphase flow. E-mail: [email protected]