08 Paper PDF

Download as pdf or txt
Download as pdf or txt
You are on page 1of 16

archives

of thermodynamics
Vol. 40(2019), No. 4, 151–166
DOI: 10.24425/ather.2019.131432

A thermodynamic study on catalytic


decomposition of hydrazine in a space thruster

SHAHRAM PAKDEHI ∗
FATEMEH SHIRVANI
REIHANEH ZOLFAGHARI

Malek Ashtar University of Technology, Iran

Abstract Most satellites stationed in space use catalytic propulsion sys-


tems for attitude control and orbit adjustment. Hydrazine is consumed ex-
tensively as liquid monopropellant, in the thrusters. Catalytic reactor is
the most important section in the catalytic thruster. Ammonia and nitro-
gen gases are produced as a result of complete catalytic decomposition of
hydrazine in the reactor, causing an increase in temperature and a rise in
specific impulse. Ammonia is subsequently decomposed, leading to nitrogen
and hydrogen gases. Decomposition of ammonia leads to a decrease in tem-
perature, molecular weight and specific impulse. The latter phenomenon
is unavoidable. The effect of ammonia decomposition on the reactor tem-
perature, molecular weight of gaseous products and conclusively on specific
impulse was studied in this article. At adiabatic state, thermodynamic anal-
ysis revealed that the maximum and minimum temperatures were 1655 K
and 773 K, respectively. The highest molecular weight was obtained at
ammonia conversion of zero and the lowest when ammonia conversion was
100%. The maximum specific impulse (305.4 S) was obtained at ammonia
conversion of zero and completely conversion of ammonia, the minimum spe-
cific impulse (about 213.7 s) was obtained. For specific impulse, the result
of thermodynamic calculation in this work was validated by the empirical
results.

Keywords: Catalytic thruster; Hydrazine; Ammonia; Specific impulse; Thermodynamic


analysis


Corresponding Author. Email: sh− [email protected]
152 S. Pakdehi, F. Shirvani and R. Zolfaghari

Nomenclature
A – empirical constant (Eq. (12))
B – empirical constant (Eq. (12)), 1/K
C – empirical constant (Eq.(12)), K2
Cp – heat capacity for outlet gas mixture at constant pressure, J/mol K
C p,i – heat capacity for component i, J/mol K
Cv – heat capacity for outlet gas mixture at constant volume, J/mol K
Din – diameter of orifices of showerhead injector, m
g – gravitational acceleration, m/s2
H – enthalpy, J
∆H p,I – enthalpy change for component i, J
o
∆Hr1,298 – standard heat for hydrazine decomposition, J/mol
o
∆Hr2,298 – standard heat for ammonia decomposition, J/mol
I sp – specific impulse for hydrazine monopropellant, s
I sp,thruster – specific impulse for thruster, s
k – - proportional constant in Eq. (9)
L – length of the catalytic bed, m
M – outlet gas molecular weight, g/mol
Mi – molecular weight for component i, g/mol
ni – mole of component i, mol
no,NH3 – initial mole of ammonia, mole
nNH3 – present mole of ammonia, mole
Pc – gas pressure in the reactor, Pa
P in – injection pressure, Pa
Pe – outlet gas pressure from the reactor, Pa
Q – amount of heat exchanged, J
R – universal gas constant, 8.314 J/mol K
T – temperature, K
Tc – exit gas temperature from the catalytic reactor (= T2 ), K
u – internal energy, J
Ws – shaft work, J
X – mole conversion of ammonia, %
yi – mole fraction of component i
z – altitude, m

Greek symbols

γ – heat capacity ratio (= C p /C v )

Subscripts

ig – ideal gas
A thermodynamic study on catalytic decomposition. . . 153

1 Introduction
Hydrazine (N2 H4 ) is consumed extensively in space thrusters for attitude
control and orbit adjustment [1–3]. Catalytic decomposition of hydrazine
produces hot gases, which contribute to the propulsion power for satel-
lite [4]. The catalytic reactor is the most important section in a hydrazine
thruster. Shell 405 (30–45 wt% Ir over γ-Al2 O3 ) is the most appropriate
catalyst for decomposition of hydrazine [3,5]. A schematic drawing of the
catalytic reactor is presented in Fig. 1.

Figure 1: Schematic diagram for a catalytic reactor in hydrazine decomposition


thruster [3].

Hydrazine monopropellant is decomposed into ammonia and nitrogen ac-


cording to following reaction as [6]

3N2 H4 → 4NH3 + N2 , ∆Ho298 = −336.330 kJ . (1)

A portion of the produced ammonia is decomposed as [6]

4NH3 → 2N2 + 6H2 , ∆Ho298 = +184.440 kJ . (2)

All substances, except hydrazine are in gaseous state. Based on general


equation for specific impulse (Isp = k(Tc /M )0.5 [2]), decomposition of am-
monia is effective on outlet gaseous products, hence the effect on specific
154 S. Pakdehi, F. Shirvani and R. Zolfaghari

impulse. The control of ammonia decomposition in maximizing the specific


impulse is very important.
With respect to reaction thermochemistry in Eqs. (1) and (2) and the
general equation for Isp , there exists a paradox. Equation (1) increases
the temperature of catalytic reactor, leading to an increase in Isp , while
Eq. (2) decreases the gas molecular weight, temperature of the reactor
and Isp . This is a problem in determining how much Eq. (2) should be
proceeded in order to obtain an appropriate Isp . In other words, it is nec-
essary to study the changes in I sp with respect to ammonia conversion.
Since hydrazine is carcinogen [7], the prediction of NH3 decomposition will
be helpful prior to thruster experimental tests. So, the hydrazine decom-
position should be studied thermodynamically for prediction of maximum
temperature, ammonia conversion, gas molecular weight and the maximum
obtainable Isp,thruster . The solution to the problem would assist researchers
to design better catalytic reactor in hydrazine thruster.
Since 1950 the literature on this issue has been and is in progress. The
studied aspects in this field are: hydrazine decomposition kinetics [8–12],
hydrazine decomposition catalysts [13–17], designing and manufacturing of
hydrazine thrusters [18–20] and performance of hydrazine thrusters [21–25].
In all studies, the researchers have dealt with the hydrazine thruster and
catalytic decomposition of hydrazine with respect to their dynamics or
kinetics but not its thermodynamic analysis. The thermodynamics of
hydrazine monopropellant decomposition represents the maximum perfor-
mance or specific impulse. Studying the thermodynamic aspect is the ob-
jective of this article, the results of which will be validated through the
experimental work.

2 Extracting equations
Hydrazine is decomposed exothermically into ammonia and nitrogen, Eq. (1),
and the produced ammonia is decomposed endothermically in a simultane-
ous manner into nitrogen and hydrogen, Eq. (2). To analyze this process,
the reactions are assumed to occur in two separate sections or reactors
(Fig. 2).
Hydrazine enters the reactor 1 at 289 K and is decomposed into ammo-
nia and nitrogen adiabatically, Eq. (1). In reactor 1, the hydrazine conver-
sion is 100% [26]. In reactor 2, ammonia is decomposed into nitrogen and
hydrogen with different conversion percentages. The following assumptions
A thermodynamic study on catalytic decomposition. . . 155

Figure 2: Virtual process for catalytic decomposition of hydrazine.

were considered for analysis of reaction Eq. (1) in virtual reactor 1:


1. Hydrazine decomposition, Eq. (1), occurs in such a rapid manner
that there is no opportunity for heat exchange, hence, reactor 1 is
adiabatic (Q = 0) which is confirmed by experimental observations
[27–30].
2. Due to short length of the reactor (less than 0.05 m in conventional
thrusters [31–34]), the changes in kinetic energy (∆u2 /2) and po-
tential energy (g∆z) are negligible. Also, there is no shaft work in
the catalytic bed (Ws = 0). According to general energy balance in
reactor 1, ∆H will be zero:
∆u2
∆H + + g∆z = Q − Ws . (3)
2

3. The temperature in hydrazine thrusters is high [27–30]. Therefore, it


is assumed that the all gases are in ideal state.
4. It is assumed no changes in the catalyst morphology or no deactiva-
tion in catalyst.
5. It is assumed gravity acceleration to be constant (g = 9.8 m/s2 ).
6. All produced gages are stable.
Since the change in enthalpy is independent on the reaction path [35],
equation
o
∆Hr1,298 + ∆HPo = ∆H = 0 (4)
will be predominant in reactor 1 for hydrazine decomposition, Eq. (1).
The heat generated by hydrazine decomposition reaction enhances the
temperature of the products in reactor 1. It is expressed in
∆Hr01 ,298 ∆Hp1
Z T1
= = nNH3 Cp,NH3 dT +
3 3 298.15
4 T1 1 T1
Z T1 Z Z
n N2 Cp,N2 dT = Cp,NH3 dT + Cp,N2 dT . (5)
298.15 3 298.15 3 298.15
156 S. Pakdehi, F. Shirvani and R. Zolfaghari

The temperature of products in reactor 1, T 1 , will be determined. The same


assumptions hold true for reactor 2. Due to unknown level of conversion
in ammonia, the mole of consumed ammonia will be a variable for second
reaction in reactor 2. According to Fig. 2, the following equation will be
applied:
o
∆Hp1 + ∆Hr2,298 = ∆Hp2 ,
(6)
o
∆Hp1 + ∆Hr2,298 = ∆Hp2 .

Here the energy balance is applied as

4
Z T1 1 T1
Z
X∆Hr02 ,298
Cp,NH3 dT + Cp,N2 dT − =
3 298.15 3 298.15 3
 1 + 2X  T2
Z Z T2
Cp,N2 dT + 2X Cp,H2 dT +
3 298.15 298.15
 4 − 4X  Z T2
Cp,NH3 dT , (7)
3 298.15

where T2 is the outlet gas temperature from reactor 2 or thrust chamber


(Tc ), and X is the mole conversion of ammonia defined as reacted or con-
sumed mole of ammonia per initial mole of ammonia (an amount between
0 and 1)
(no,NH3 − nNH3 )
X= . (8)
no,NH3
Specific impulse is expressed in thermodynamic form [2,36] as follows:
s s s s
Tc Tc 2γRTc 2Cp Tc
Isp ∝ =k = = , (9)
M M (γ − 1)g2 M g2 M

where γ = Cp /Cv , R is the universal gas constant (Cp − Cv = R = 8.314


J/mol K) and g is the gravitational acceleration. As a thermodynamic
parameter C p is expressed as
X
Cp = yi Cp,i , (10)

and the gas molecular weight as


X
M= yi Mi . (11)
A thermodynamic study on catalytic decomposition. . . 157

3 Results and discussion


General equation for heat capacity for NH3 , N2 , and H2 in Eqs. (5) and
(7) is presented in

Cpig = 8.314(A + BT + CT −2 ‘) (12)

in which A, B, and C are material dependent constants. This relation is


valid for ammonia up to 1800 K, nitrogen up to 2000 K, and hydrogen up
to 3000 K [35]. The constants for the heat capacity equations are given in
Tab. 1.

Table 1: Constants for specific heat capacity equation for NH3 , N2 , and H2 [35].

Chemical A [–] B×103 [K−1 ] C ×10−5 [K2 ]


NH3 3.578 3.020 -0.186
N2 3.280 0.593 0.040
H2 3.249 0.422 0.083

Temperature T1 is calculated through Eq. (5) as 1654.4 K. By inserting T1


in Eq. (7), and rearranging the equation would result in X versus T2 as

0.704 × 105
46497.9 − 17.592T2 − 6.3365 × 10−3 T22 −
T2
X= . (13)
−3 2 1.322 × 105
19498.47 + 11.742T2 − 4.181 × 10 T2 −
T2

The plot of the thrust chamber temperature versus ammonia conversion


according to this equation is shown in Fig. 3. As observed here, the maxi-
mum temperature is 1655 K. This temperature is obtained at no ammonia
conversion (X = 0). In the presence of the catalyst, a portion of ammonia
is eventually decomposed. This fact suggests that catalysts must be found
to inhibit the ammonia decomposition reaction.
Due to endothermic decomposition of ammonia, an increase in ammo-
nia conversion leads to a decrease in gas temperature. As shown in Fig. 3,
the temperature of thrust chamber decreases linearly as the ammonia con-
version increases (a slope of 830 K per conversion unit). Since ammonia,
nitrogen and hydrogen are stable molecules and no other molecule will be
158 S. Pakdehi, F. Shirvani and R. Zolfaghari

Figure 3: Changes in gas temperature in the thruster versus ammonia conversion.

formed, so the thrust chamber contains only these gases. Outlet gas molec-
ular weight, as a thermodynamic property, is calculated through following
relation, where MNH3 = 17 g/mol, MN2 = 28 g/mol, and MH2 = 2 g/mol:
X 1−X 1 + 2X 6X
M= yi Mi = MNH3 + MN2 + MH2
1.25 + X 5 + 4X 5 + 4X
148 + 68X
= . (14)
6.25 + 10X + 4X 2
The molecular weight changes of the outlet gas mixture versus ammonia
conversion is shown in Fig. 4, where the highest molecular weight of outlet
gas mixture is obtained when there is no ammonia conversion (3N2 H4 →
4NH3 + N2 ). The lowest molecular weight is for complete decomposition of
ammonia (3N2 H4 → 3N2 + 6H2 ). In other words, hydrogen with the lowest
molecular weight and the highest mole fraction reduces the total molecular
weight.
Composition of the outlet gas from the reactor is one of the major pa-
rameters in evaluating the reactor performance (Fig. 5). A decrease in
mole fraction of ammonia is proportional to the increase in mole fraction
of nitrogen and hydrogen. As observed in figure, when hydrazine is de-
composed into ammonia and nitrogen, in the absence of hydrogen the mole
fractions of ammonia and nitrogen are 0.8 and 0.2, respectively. At full
A thermodynamic study on catalytic decomposition. . . 159

Figure 4: Variation of molecular weight for gas mixture in the reactor versus ammonia
conversion.

Figure 5: Variation in outlet gas composition from the reactor versus ammonia conversion
(--NH3, -N-H2 , --N2 ).

ammonia conversion (3N2 H4 → 3N2 + 6H2 ), the nitrogen mole fraction is


3/9 or 0.34 and that of the hydrogen is 6/9 or 0.66.
With respect to Eq. (10), the heat capacity of the gas mixture at
160 S. Pakdehi, F. Shirvani and R. Zolfaghari

constant pressure is expressed through


X 1−X 1 + 2X 6X
Cp = yi Cp,i = Cp,NH3 + Cp,N2 + Cp,H2 . (15)
1.25 + X 5 + 4X 5 + 4X

The heat capacity for each component is calculated at T2 or Tc and in


this relation Cp depends on temperature, while the resulting temperature
depends on ammonia conversion, indicating that the Cp for the gas mixture
will depend on the ammonia conversion in a complicated equation, the
calculations of which will be carried out by Matlab (version 2015b) software
[38]. The variation in Cp for outlet gas mixture is shown in Fig. 6, where
the increase in ammonia conversion leads to a decrease in heat capacity of
gas mixture. This is because Cp for ammonia is higher than nitrogen and
hydrogen at any operational temperature interval [35]. Thus increase in
ammonia conversion means lower composition of ammonia and therefore,
a decrease in heat capacity of the gas mixture.

Figure 6: Variation in heat capacity for outlet gas mixture from the reactor versus am-
monia conversion.

The horizontal axis (abscissa or x-axis) may represent the catalytic re-
actor length. In Figs. 3–6 are presentedthe values of temperature of thrust
chamber, molecular weight, gas composition and heat capacity, respectively
which vary with respect to ammonia conversion.
Specific impulse for the hydrazine monopropellant is calculated through
A thermodynamic study on catalytic decomposition. . . 161

Eq. (9) by applying Eqs. (13)–(15) (see the results in Fig. 7). The highest
specific impulse (305.4 s) is obtained when there is no ammonia conver-
sion. If ammonia conversion is 100%, then the Isp will be 213.7 s. The
experimental results confirmed that ammonia is certainly decomposed in
the presence of the catalyst [27–30], that is, the value of 305.4 s will not be
obtained at any time.

Figure 7: Variation of specific impulse for hydrazine versus ammonia decomposition.

As observed in Figs 3, 4 and 6, the change in Tc is much more than that


of M or Cp (the slope of Tc versus X is 830 K/conversion unit but the
slopes for M and Cp are 21.7 g/mol per conversion unit and 58.5 J/mol K
per conversion unit, respectively). Therefore, it is expected that Isp for
hydrazine monopropellant, Eq. (9), be more sensitive to Tc than M or
Cp . Conclusively, ammonia conversion strongly decreases Isp for hydrazine
monopropellant.
The specific impulse is a character of hydrazine monopropellant, that is,
Eq. (9) is provided merely for decomposition of hydrazine monopropellant.
The most important advantage of Eq. (9) is that when ammonia conversion
is accomplished by it, the Isp value will be obtained, where Cp , Tc , and M
depend on ammonia conversion. If a nozzle is connected to the end of the
reactor, the effects of pressure in the chamber and the exit of the nozzle
162 S. Pakdehi, F. Shirvani and R. Zolfaghari

will be inserted in Isp equation as [2]


s
(γ−1)/γ i
2Cp Tc h Pe

Isp,thruster = 1− . (16)
g2 M Pc

Here, the Isp,thruster is a specific impulse for the thruster, not for hydrazine
monopropellant. That is, Isp,thruster includes Isp for hydrazine monopro-
pellant and pressure effects in the chamber and the nozzle
s
(γ−1)/γ i
Pe
h 
Isp,thruster = Isp 1− . (17)
Pc

If the pressure in the thrust chamber is much more than the exit pres-
sure, then the Isp,thruster will approach Isp of hydrazine monopropellant
(Isp,thruster is always less than Isp ).
An outstanding point deduced from Eqs. (9) and (16) is that Eq. (9)
represents the maximum Isp for hydrazine monopropellant. By adjusting
and designing modified throat and exit diameters for the nozzle, maximum
specific impulse might be actuated. In other words, the maximum value of
specific impulse for the thruster is the same as hydrazine monopropellant.
By knowing the value of ammonia conversion, the performance condition of
the reactor and the thruster will be predictable, which is a matter of con-
cern. Moreover, Isp is measured by a gas chromatograph (equipped with a
thermal conductive detector) in an experimental manner while Isp,thruster
is measured by a load cell (to determine the thrust) with hydrazine flow
meter (to determine hydrazine mass flow rate) in the experimental manner.
Based on the design type and the comments of thruster designer, the
term ( PPec )(γ−1)/γ will have different values. For example, an increase in
the chamber pressure through a reduction in the throat cross section area
leads to increase in Isp,thruster , conclusively, thermodynamic analysis will
determine the maximum Isp,thruster .
The obtained results here were validated through experimental work due
to Adler et al. [37]. The specifications of design for Adler et al. thruster
are presented below:

• injection pressure, Pin , between 20 and 80 kPa;


• diameter of orifices of showerhead injector, Din , between 0.0008 and
0.005 m;
• length of the catalytic bed, L, from 0.0525 to 0.105 m;
A thermodynamic study on catalytic decomposition. . . 163

• exhaust nozzle diameter, De , from 0.01 to 0.015 m.

Adler et al. obtained the optimum I sp,thruster at P e /Pc = 0.0175 and am-
monia conversion of 55% [37]. At this optimum conditions, they obtained
γ = 1.271, T c = 1221 K, M = 13.33g/mol and I sp,thruster = 201.1 s as
well. Comparison of the results between Adler et al. experimental work
and this work is presented in Tab. 2, indicating the confirmation of this
thermodynamic analysis with only a slight difference.

Table 2: Comparison between the quantities for hydrazine thruster in this work and Adler
et al. [37] at Pe /Pc = 0.0175 and XNH3 = 55%.

Quantities Unites Results of Adler Results of Error (%)


et al. [37] this work

γ – 1.271 1.268 0.2


Tc K 1221 1231 0.8
M g/mol 13.33 14.3 0.7
Isp,thruster s 201.1 197.3 1.8

4 Conclusions
Hydrazine, as a carcinogenic monopropellant, is consumed in space thrusters.
The most important section of the thruster is the catalytic reactor. In
designing and manufacturing the reactor which works on hydrazine de-
composition basis, it is necessary to have the reactor temperature, out-
let gas molecular weight and the specific impulse. These quantities are
obtainable through thermodynamic studies run on hydrazine monopropel-
lant. At adiabatic state of the reactor, thermodynamic analysis indicated
that those parameters depend on ammonia conversion. The highest and
lowest temperatures in the reactor were obtained as 1655 K and 773 K,
respectively. The maximum and minimum molecular weights for outlet gas
from decomposition of hydrazine were obtained at zero and complete con-
versions of ammonia, respectively. The results from the provided analysis
show that the specific impulse has its maximum (as 305.4 s) when there
is no ammonia conversion. It is obtained by regulating the hydrazine flow
rate into the reactor. Also, the lowest amount of the specific impulse (as
213.7 s) is obtained for the full conversion of ammonia. The results between
164 S. Pakdehi, F. Shirvani and R. Zolfaghari

this thermodynamic analysis and that of the Adler experimental work at


Pe /Pc = 0.0175 and XNH3 = 55% indicate that there exists a 1.8% er-
ror between the thermodynamic calculations and experimental results for
I sp,thruster .

Received 30 June 2018

References
[1] Li L., Wang X., Zhao X., Zheng M., Cheng R., Zhou L., Zhang T.: Mi-
crocalorimetric studies of the iridium catalyst for hydrazine decomposition reaction.
Thermochimica Acta 434(2005), 1-2, 119–124.
[2] Sutton G.P., Biblarz O.: Rocket Propulsion Elements (8th Edn.). Wiley & Sons,
New York 2010.
[3] Eckert E.W.: Hydrazine and its Derivatives (8th Edn.). Wiley & Sons, New York
2001.
[4] Brown C.D.: Spacecraft Propulsion. AIAA, Washington 1995.
[5] Cho S.J., Lee J., Lee Y.S., Kim D.P.: Characterization of iridium catalyst for
decomposition of hydrazine hydrate for hydrogen generation. Catal. Lett. 109(2006),
3-4, 181–186.
[6] Zheng M., Chen X., Cheng R., Li N., Sun J., Wang X., Zhang T.: Catalytic
decomposition of hydrazine on iron nitride catalysts. Catal. Commun. 7(2006), 3,
187–191.
[7] Agrawal J.P.: High Energy Materials: Propellants, Explosives and Pyrotechnics.
Wiley-VCH, Weinheim 2010.
[8] de Medeiros J.E., Valenca G.P.: Kinetic analysis of the catalytic decomposition
of hydrazine. Braz. J. Chem. Eng. 15(1998), 2, 1–8.
[9] Smith O.I., Solomon W.C.: Kinetics of Hydrazine Decomposition on Iridium and
Alumina Supported Iridium Catalysts. AFRPL-TR-73-59, Edwards 1973.
[10] Sayer C.F.: The heterogeneous decomposition of hydrazine. Part 5. The kinetics
of the decomposition of liquid hyrazine on a supported ruthenium catalyst. Rocket
Prop. Establ. 72(1972), 1–26.
[11] Khomenko A., Apelbaum L.O.: Study of the kinetics of the catalytic decomposi-
tion of hydrazine vapors on paladium. Kinet. Catal. 17(1976), 600–607.
[12] Pakdehi S.G., Salimi M., Rasoolzadeh M.: A review on decomposition of hy-
drazine and its kinetics as a novel approach for CO-free H2 production. Res. Appl.
Mech. Eng. (RAME) 3(2014), 21–25.
[13] Williams P.H.: Modification of Shell 405 Catalyst. AFRPL-TR-72-7, Edwards
1972.
[14] Armstrong W.E., La France D.S., Voge H.H.: Hydrazine decomposition and
other reactions. United States patent US 4122671, 1978.
A thermodynamic study on catalytic decomposition. . . 165

[15] Balcon S., Mary S., Kappenstein C., Gengembre E.: Monopropellant decom-
position catalysts II. Sintering studies on Ir/Al2 O3 catalysts, influence of chloride
anions. Appl. Catal. A 196(2000), 2, 179–190.
[16] Fan C., Wu T., Kaden W.E., Anderson S.L.: Cluster size effects on hydrazine
decomposition on Irn/Al2 O3 /NiAl(110). Surf. Sci. 600(2006), 2, 461–467.
[17] Yu M.-J., Lee K.-H., Kim S.-K., Choi J.-M., Cho S.-J.: Method of manufac-
turing high-crush-strength iridium catalyst for hydrazine decomposition reaction in
spacecraft thrusters using bauxite. United States patent US 7651972 B2, 2010.
[18] Thunnissen D., Engelbrecht C., Weiss J.: Assessing model uncertainty in
the conceptual design of a monopropellant propulsion system. In: Proc. 39th
AIAA/ASME/SAE/ASEE Joint Propul. Conf. Exhibit, 2003, Huntsville, 1–17.
[19] Rocket Research Company: Development of Design and Scaling Criteria for Mono-
propellant Hydrazine Reactors Employing Shell 405 Spontaneous Catalyst: Vol II.
RRC-66-R-76, 1967.
[20] Grant Jr. A.F.: Basic Factors Involved in the Design and Operation of Catalytic
Monopropellant-Hydrazine Reaction Chambers. Report 20-77, Jet Propulsion Labo-
ratory, Pasadena 1954.
[21] Makled A.E., Belal H.: Modeling of hydrazine decomposition for monopropellant
thrusters. In: Proc. 13th Int. Conf. Aerospace Sci. and Aviation Technol. ASAT- 13,
May 26 – 28, 2009,
[22] Kesten A.S.: Analytical study of catalytic reactors for hydrazine decomposition.
Report H910461-38, United Aircraft Research Laboratories, 1969.
[23] Shankar V., Ram K.A.: Experimental investigations of the 10 N catalytic hy-
drazine thruster. Acta Astronaut. 12(1985), 4, 237–249.
[24] Hwang C.H., Lee S.N., Baek S.W., Han C.Y., Kim S.K., Yu M.J.: Effects of
catalyst bed failure on thermochemical phenomena for a hydrazine monopropellant
thruster using Ir/Al2 O3 catalysts. Ind. Eng. Chem. Res. 51(2012), 15, 5382-5393.
[25] Meng P.R.: A Life Test of a 22-Newton (5-lbf) Hydrazine Rocket. NASA Technical
Memorandum, Washington, D.C., 1987.
[26] Vieira R., Bastos-Netto D., Ledoux M.J., Pham-Huu C.: Hydrazine decompo-
sition over iridium supported on carbon nanofibers composite for space applications:
Near actual flight conditions tests. Appl. Catal. A 279(2005), 1-2, 35–40.
[27] Zolfaghari R.: Effective Parameters on Synthesis of Ir/γ-Al2 O3 for Hydrazine
Decomposition and Optimizing Them. MSc thesis, Malek Ashtar University of Tech-
nology, Tehran 2013.
[28] Shirvani F.: Modeling of Classic Fixed Bed Catalytic Reactor for Hydrazine De-
composition. MSc thesis, Malek Ashtar University of Technology, Tehran 2014.
[29] Salimi M.: Evaluation of Monolith Nanocatalyst in Hydrazine Decomposition Re-
actor. MSc thesis, Malek Ashtar University of Technology, Tehran 2014.
[30] Asadi A.: Synthesis and Evaluation of Bimetallic Nanocatalyst for Hydrazine De-
composition. Malek Ashtar University of Technology, Tehran, 2013.
[31] Shankar V., Anantha Ram K., Bhaskaran K.A.: Experimental investigations
of the 10 N catalytic hydrazine thruster. Acta Astronaut. 12(1985), 237–249.
166 S. Pakdehi, F. Shirvani and R. Zolfaghari

[32] Soares Neto T.G., Gobbo-Ferreira J., Cobo A . G., Cruz G M.: Ir-
Ru/Al2 O3 catalysts used in satellite propulsion. Braz. J. Chem. Eng. 20(2003), 3,
273–282.
[33] Soares Neto T.G., Cobo A.J.G., Cruz G.M.: Textural properties evolution of Ir
and Ru supported on alumina catalysts during hydrazine decomposition in satellite
thruster. Appl. Catal. A 250(2003), 2, 331–340.
[34] de Almeida Coelho N.M., Furtado J.L.B., Pham-Huu C., Vieira R.: Carbon
nanofibers: a versatile catalytic support. Mater. Res. 11(2008), 3, 353–357.
[35] Smith J.M., Van Ness H.C.: Introduction to Chemical Engineering Thermody-
namics (7th Edn.). McGraw-Hill, New York 2010.
[36] Fortini A.J., Wright M.J.: Self-adjusting catalyst for propellant decomposition.
United States patent US 20080064913A1, 2008.
[37] Adler D., Dubrov E., Manheimer-Timnat Y.: The performance of a hydrazine
engine with an improved catalyst. Acta Astronaut. 2(2005), 7-8, 613–625.
[38] Matlab, version 2015b.

You might also like