Aerospace Science and Technology 91 (2019) 337–356

Contents lists available at ScienceDirect

Aerospace Science and Technology

www.elsevier.com/locate/aescte

Alcohol and alkane fuel performance for Gas Generator cycle Air Turbo
Ramjet Engine
Ryojiro Minato
Department of Mechanical and Aerospace Engineering, Muroran Institute of Technology, 27-1 Mizumoto-cho, Muroran, Hokkaido, 050-8585, Japan

a r t i c l e i n f o a b s t r a c t

Article history: The present study investigates the feasibility of alcohol fuels for a Gas Generator cycle Air Turbo Ramjet
Received 9 March 2018 (GG-ATR) Engine in comparison with alkane fuels. The alkane and alcohol fuels in the present study
Received in revised form 2 March 2019 range from C1 species (CH4 and CH3 OH) to C4 species (n-C4 H10 and 1-C4 H9 OH).
Accepted 8 May 2019
The present study analytically evaluates Specific thrust and Specific impulse (Isp) of GG-ATR engine
Available online 17 May 2019
as functions of gas generator combustion temperature. For both alcohols and alkanes, the peak Gas
Keywords: Generator (GG) combustion temperatures exist for specific thrust and ram combustor temperature.
Air turbo ramjet engine Comparing alcohol to alkane in the lower GG combustion temperature, specific thrusts and Isp of alcohol
Supersonic UAV tend to be higher than those of C2 H6 , C3 H8 , and 1-C4 H10 . GG combustion gases for alcohol fuels
Gas generator contain higher mole fractions of H2 O and CO2 than those for alkanes. Because of high mole fractions
Alcohol of H2 O and CO2 , the ratio of GG combustion gas for alcohols to air flow rate is fortunately close to
Alkane the stoichiometric ratio, resulting in higher ram combustion temperature. The high ram combustion
temperature can contribute to the higher specific thrust and Isp. Lower GG combustion temperature is
preferable because of a thermal limit of a turbine blade. Thus, alcohols have superiorities to alkanes for
the GG-ATR engine application.
© 2019 Elsevier Masson SAS. All rights reserved.

1. Introduction

An air turbo-ramjet (ATR) engine is a turbine-based combined

cycle (TBCC) engine and considered for future spaceplane or hy-
personic vehicle application. The ATR engine has greater specific
thrust than turbojet engines and higher specific impulse (Isp) than
rocket engines. Thus, it is suitable for a propulsion engine of a
hypersonic vehicle or a missile [1–3]. Different from the turbo-
jet engine, propellant gas of the ATR engine is utilized for turbine
driving gas. A gas generator (GG, Gas generator cycle ATR Engine,
GG-ATR) or a heat exchanger (ATREX engine, [4]) can provide en-
thalpy for turbine work. For bipropellant GG-ATR engine, not only
fuel but also oxidizer must be on-board, resulting in lower specific
impulse (Isp). However, the GG-ATR engine can be expected to be
lighter than the ATREX engine because the GG can be designed to
be more compact than the heat exchanger. The author considers
that the GG-ATR engine is the propulsion engine for the super-
sonic unmanned aerial vehicle (UAV), which is now under devel-
opment in Muroran Institute of Technology [5]. Fig. 1 and 2 show Fig. 1. GG-ATR Engine developed in Muroran Institute of Technology.
the GG-ATR engine developed at Muroran Institute of Technology
and the conceptual schematic of the GG-ATR engine, respectively.
High-temperature, fuel-riched GG combustion gas can drive a tur-
bine and a compressor. The compressor increases the incoming air
E-mail address: r-minato@mmm.muroran-it.ac.jp. pressure. GG combustion gas mixes with the air and burns at a

https://doi.org/10.1016/j.ast.2019.05.023
1270-9638/© 2019 Elsevier Masson SAS. All rights reserved.
338 R. Minato / Aerospace Science and Technology 91 (2019) 337–356

Nomenclature

A Cross section area, m2 π Total Pressure Ratio at Engine Component

C∗ Characteristic velocity as represent in eq. (6), m/s τ Total Temperature Ratio at Engine Component
CP Heat Capacity at constant pressure, J/(kg K)
Subscript
F Thrust, N
f Ratio of propellant to air in eq. (3) air Airflow
g Gravity acceleration (=9.80665 m/s2 ) comp Compressor
h Specific heat release of fuel, J/(kg K) cor Corrected value
Isp Specific Impulse, s de On design condition
ṁ Mass flow rate, kg/s ex Exit condition
N Rotational speed, rpm fuel Fuel
P Pressure, Pa in Inlet
rOF Oxidizer-to-Fuel Ratio at gas generator GG Gas Generator
T Temperature, K nozzle Nozzle
U Rotational speed of turbine blade, m/s out Outlet
V Velocity, m/s ram Ram combustor
ε Nozzle Expansion Ratio S Static Condition
η Efficiency T Stagnation condition
γ Specific Heat Capacity Ratio turb Turbine

larger hydrogen atoms/carbon atoms ratio (H/C ratio). For exam-

ple, H/C ratios of methane (CH4 ) and ethane (C2 H6 ) are 4 and 3,
respectively. However, boiling temperatures of lower alkane than
normal-pentane (n-C5 H12 ) are generally lower than room temper-
ature. Those alkanes have to be stored at cryogenic state in a fuel
tank. Moreover, the densities of liquefied those alkanes are smaller
than aviation kerosene. Low density and cryogenic properties in-
duce handling difficulties and require a large volume of the fuel
tank, resulting in disadvantages for atmospheric flight.
On the other hand, alcohol is the chemical compound, which
has hydroxyl group to saturated alkane species. From the view-
Fig. 2. Conceptual schematic of GG-ATR engine.
point of organic chemistry, alcohol can be considered the organic
derivatives of water, whose one of the hydrogen atoms is replaced
ram combustor after driving the turbine. This ram combustor tem- by an alkyl group. Hydroxyl group of alcohol has a similar structure
perature will be higher than the afterburner temperature for the to water. Because of this characteristic, alcohol has higher boiling
turbojet engine. Thus, greater thrust is obtainable for GG-ATR en- temperature and lower heat release than alkane. For example, the
gine. boiling temperature of methanol and ethanol are 337.8 K and 351.5
In the previous studies, many researchers investigated GG-ATR K, respectively.
engine performance analytically and experimentally. Most of those However, specific impulse (Isp) of an aerospace propulsion en-
studies focused on liquefied hydrogen (LH2) fuel with liquefied gine is usually proportional to the heat release of fuel. Thus, alco-
oxygen (LOX) [1]. LH2 is one of the most promising fuels because hol fuels have scarcely used for aerospace propulsion fuels except
of its high specific heat release and large cooling capability. How-
for some rocket fuel applications. The heat release of ethanol is
ever, the density of LH2 is only one-tenth of that of kerosene and
only two-thirds of that of aviation kerosene. On the other hand,
a large volume of a fuel tank is required, which induces greater
soot formation in GG can be expected to be less than kerosene fu-
aerodynamic drag. Low-density characteristics of LH2 cause a sig-
els. Moreover, the author clarified that Isp of GG-ATR engine with
nificant drawback to an atmospheric flight vehicle.
ethanol is superior to that with kerosene [8]. This result indicated
Besides of LH2 fuel, Sullerey et al. compared LH2 with propane
that the ethanol has the feasibility for the GG-ATR engine [5]. Isp
fuel for GG-ATR engine application [6]. Lilly et al. conducted the
of GG-ATR engine is affected by the molecular weight of GG com-
ground test of the GG-ATR engine with hydrazine [7]. Hydrazine or
bustion gas rather than specific heat release of the fuel. GG com-
its derivatives can be utilized as monopropellant by its exothermic
decomposition. Hydrocarbon fuels are also expected for promising bustion gas of LOX-ethanol has lower molecular weight than that
fuel of GG-ATR engine because they have large density and are of LOX-normal dodecane (n-C12 H26 ) because ethanol has a higher
easy handling. Hydrocarbon compounds, which have only single H/C ratio than n-C12 H26 . The lower molecular weight of GG com-
bonds, are categorized into alkanes. Those compounds are often bustion gas is, the higher specific turbine power is available. That
called the saturated hydrocarbon or paraffin. Besides cycloalka- is the reason why ethanol has a higher Isp than n-C12 H26 .
nes, a general formula of the saturated straight-chain alkane is However, it is still unknown whether ethanol has the supe-
represented as Cn H2n+2 . For GG-ATR engine application, the ra- riority to lower alkanes, such as methane (CH4 ), ethane (C2 H6 )
tio of hydrogen atoms number to carbon atoms number in fuel and propane (C3 H8 ). The aim of the present study focuses on
molecule (H/C ratio) is preferable to be larger, because molecu- the feasibility of alcohol fuel for GG-ATR engine application. The
lar weight of Gas Generator (GG) combustion gas can be reduced, present study treats from C1 (CH4 and CH3 OH) to C4 (n-C4 H10
resulting of higher Isp. With the same carbon atoms number, alka- and 1-C4 H9 OH) species. With LOX oxidizer, GG-ATR engine per-
nes have more hydrogen atoms than alkenes, alkynes, cycloalka- formances for alcohol fuel analytically evaluated and compared to
nes or aromatic compounds. Lower molecular weight alkanes have those for alkane fuel.
 R. Minato / Aerospace Science and Technology 91 (2019) 337–356 339

2. Analytical methods and condition (Compression work per unit airflow rate) is proportional to the
square of N COR [12],
2.1. GG-ATR engine analytical method γ −1
2
ṁair ∝ N COR , (πcomp ) γ − 1 ∝ N COR (4)
The specific thrust is defined as thrust per unit airflow rate.
If N COR is specified, mCOR and πcomp can be determined. The
The Isp is thrust per unit propellant flow rate. Those are funda-
compressor exit static pressure, P S ,comp,out , can also be calcu-
mental parameters to evaluate air-breathing engine performance.
lated by πcomp .
Equations (1) and (2) represent the specific thrust and the Isp of
4. Chemical equilibrium calculation is conducted at the GG in
the GG-ATR engine, respectively, on-design condition. Oxidizer-to-Fuel (O/F) ratio at the GG is

    γnozzle −1  determined to set T T ,GG , to a specified value. The obtained O/F

F 1  P nozzle γnozzle ratio is constant in off-design condition analysis.
= (1 + f ) 2C P ,ram T T ,ram 1 − 5. Turbine adiabatic efficiency is given as a function of velocity
ṁair g P T ,nozzle
ratio, U /C 0 , where U and C 0 are circumferential velocity of a
V ( P nozzle − P atm ) A nozzle
− + (1) turbine rotor and gas spouting velocity from a turbine nozzle,
g ṁair g respectively [13]. Gas flow is usually choked flow at the tur-

 
    γnozzle −1  bine nozzle. Thus, the following formulas are derived,
1 1  P nozzle γnozzle
I SP = 1+ 2C P ,ram T T ,ram 1 − ∗
ṁGG C GG P T ,GG
∗ ) = P
g f P T ,nozzle (5)
(ṁGG C GG de T ,GG,de
V ( P nozzle − P atm ) A nozzle 
− + (2)    γGG +1
fg ṁair f g 
∗  R GG T T ,GG γGG + 1 γGG −1
where f is the Propellant-to-Air Ratio (PAR) and defined as equa- C GG = (6)
γGG 2
tion (3).
∗ in equation (6) is referred to as the characteristic veloc-
C GG
γ −1
ṁGG C P , Air T T , Air (π
γ
− 1) ity at the turbine nozzle. The iterative procedure is employed
comp
f = = γGG −1 (3) to obtain GG combustion pressure, P T ,GG from eq. (3), (5),
ṁair γGG and (6). The initial value of P T ,GG is assumed at first. mGG
ηcomp ηturb C P ,GG T GG (1 − π turb
)
can be given from eq. (5). P T ,GG and the turbine exit static
PAR in equation (3) can be derived from compressor-turbine pressure, P S ,turb,out , can determine the turbine expansion ratio,
power balance and corresponds to Fuel-to-Air Ratio of a turbo- πturb . P S ,turb,out is equal to the compressor exit static pressure,
jet engine. Different from the turbojet engine, estimation of Isp P S ,comp,out . By using πturb , P T ,GG is revised by the correlation
and PAR must include consideration for a mass flow rate of oxi- between mair and mGG in eq. (3).
dizer. Thus, the author refers f as Propellant-to-Air Ratio to avoid 6. P T ,GG is calculated iteratively until the relative difference of
the confusion. As shown in Eq. (2), PAR can affect significantly on P T ,GG is converged to be less than 10−5 .
the Isp of the GG-ATR engine. GG-ATR engine performance anal- 7. Total pressure and mixture ratio at ram combustor can be
ysis considers turbomachinery characteristics in conjunction with given from πcomp and PAR, respectively. The ram combustor
chemical equilibrium calculation. Especially, chemical equilibrium temperature can be calculated under those conditions. Finally,
calculation is necessary for combustion at GG and a ram combus- the exhaust gas velocity from the nozzle can be determined.
tor in GG-ATR engine. The author developed Gas generator cycle-
Air Turbo Ramjet Engine Analysis Code (GATRAC), which contains The turbine expansion process is treated as chemical frozen flow,
chemical equilibrium analysis. Comparing with NASA-CEA [9], the because of simplicity. Christensen indicated that analytical result
author validated the accuracy of the chemical equilibrium calcula- in frozen flow condition at a turbine is identical to those in the
tion in GATRAC thoroughly. GATRAC can analytically obtain thrust chemical equilibrium condition [14]. Sullerey et al. also employed
and Isp of GG-ATR engine in a given altitude and flight Mach num- the chemical frozen flow assumption for the GG-ATR engine anal-
ber and is capable of treating on/off-design performance analysis. ysis and could obtain reasonable results [6].
Moreover, GATRAC can also calculate total pressure and total tem-
perature at GG-ATR engine components. The computational pro- 2.2. Alcohol and alkane fuels properties
cedure for GATRAC is the same one with reference [8], and its
summary is described as follows. In the present study, CH4 , C2 H6 , C3 H8 and 1-C4 H10 are se-
lected as alkane fuels of GG-ATR engine performance analyses. On
1. GG-ATR engine parameters in on-design condition are given the other hand, methanol (CH3 OH), ethanol (C2 H5 OH), 1-propanol
from the input data. Those parameters contain pressure/heat (1-C3 H7 OH), and 1-butanol (1-C4 H9 OH) are selected for alcohol fu-
loss coefficients at each engine components, compressor pres- els. Table 1 indicates the physical and the chemical properties of
sure ratio, πcomp , corrected airflow rate, mair,COR , turbine ex- those alcohols and alkanes, which are referred from the database
pansion ratio, πturb , GG combustion pressure, P T ,GG , and tem- of NIST [15]. In general, alkanes have more specific heat release
perature, T T ,GG , and so on. and less specific heat capacity than alcohols if those have the same
2. Atmospheric pressure and temperature are given as functions number of carbon atoms.
of altitude [10]. Total pressure and temperature at compres- The isomers are compounds that have the same numbers and
sor inlet are calculated from altitude, flight Mach number, and kinds of atoms but different structure. The present analysis treats
pressure loss at an air intake duct. The intake pressure loss for straight-chain alkanes and primary alcohols because of the sim-
supersonic flight is estimated by MIL-E-5008B [11]. plification of analytical conditions. Among alcohols and alkanes
3. Based on the assumption that a velocity diagram can keep its in Table 1, C4 H10 and C3 H7 OH have two isomers individually,
similarity during turbomachinery operation, corrected air flow and C4 H9 OH has four isomers. Differences of GG combustion gas
rate, mair,COR , is considered to be linearly proportional to cor- characteristics among isomers depend on their heat release. For
rected rotational speed, N COR . The specific compression work C4 H10 and C3 H7 OH isomers, differences of specific heat release
340 R. Minato / Aerospace Science and Technology 91 (2019) 337–356

Table 1
Alkane and alcohol fuels for GG-ATR engine performance analysis.

Alkane C1 species C2 species C3 species C4 species

Chemical Species CH4 C2 H6 C3 H8 n-C4 H10
Molecular Weight kg/kmol 16.0425 30.069 44.0956 58.1222
Density at liquid phase g/cm3 0.415 0.5465 0.582 0.601
Heat of Formation at 298.15 K kJ/(mol K) −74.6 −84 −104.7 −125.6
Specific Heat Release MJ/kg 50.03 47.51 46.33 45.72
Heat Capacity at 298.15 K kJ/(kg K) 2.229 1.746 1.669 1.695
Boiling Temperature at 101.3 kPa K 111.2 184.6 231.1 273
Critical Temperature K 190.6 305.3 369.9 425
Critical Pressure MPa abs 4.61 4.91 4.25 3.8

Alcohol C1 species C2 species C3 species C4 species

Chemical Species CH3 OH C2 H5 OH 1-C3 H7 OH 1-C4 H9 OH
Molecular Weight kg/kmol 32.0419 46.0684 60.095 74.1216
Density at liquid phase g/cm3 0.7918 0.789 0.803 0.81
Heat of Formation at 298.15 K kJ/(mol K) −238.4 −276 −302.54 −328
Specific Heat Release MJ/kg 19.935 26.840 30.706 33.124
Heat Capacity at 298.15 K kJ/(kg K) 2.481 2.440 2.396 2.386
Boiling Temperature at 101.3 kPa K 337.8 351.5 370.3 390.6
Critical Temperature K 513 514 536.9 562
Critical Pressure MPa abs 8.1 6.3 5.2 4.5

Table 2 Table 3
GG-ATR engine component performance parameters. Nozzle expansion ratio and gas generator combustion
pressure.
Pressure Loss at Intake Duct 0.96
Heat Loss Coefficient at Intake Duct 1 πcomp ε P T ,GG [MPa abs]
Compressor Pressure Ratio (Total-to-Total) 2.0, 2.5, 3.0, 4.0
2.0 1.40 1.08
Adiabatic Compressor Efficiency (Total-to-Total) 0.78
2.5 1.60 1.35
Gas Generator Combustion Temperature [K] 900∼1400
3.0 1.75 1.62
Turbine Expansion Ratio (Total-to-Total) 5
4.0 2.10 2.16
Pressure Loss at Ram Combustor 0.97
Heat Loss Coefficient at Ram Combustor 1

value of total-to-total, ηturb , is about 70% in on-design condition.

In the present study, the author conducts the parametric stud-
among them are less than 1.0%. For butanol isomers, the maximum
ies concerned with T T ,GG , and compressor pressure ratio, πcomp .
difference for the heat release is 1.27% between 1-C4 H9 OH and
As mentioned previously, T T ,GG is the most crucial parameter for
tert-C4 H9 OH. The engine performance differences among those
the GG-ATR engine because its turbine power is proportional to it.
isomers are discussed in the later. The present study focuses on the
Moreover, a thermal limitation of a turbine blade strictly restricts
effects of the carbon atoms number in fuel molecules and the dif-
T T ,GG . Practical T T ,GG must be less than 1100 K. The author analyt-
ferences between alkanes and alcohols. Performance of a GG-ATR
ically evaluates specific thrust and Isp as functions of T T ,GG , which
engine depends on temperature and molecular weight of GG com-
is ranged from 900 K to 1400 K with a temperature increment of
bustion gas [8]. GG combustion gas temperature, T T ,GG , is usually
50 K. The compressor pressure ratio, πcomp , for the GG-ATR engine
restricted by a thermal limitation of a turbine blade. Oxidizer-to-
in Muroran Institute of Technology is equal to 2.5 in on-design
Fuel (O/F) ratio at GG is determined to set T T ,GG to a specified
condition. In the present study, four values of πcomp are selected
value. If fuel, oxidizer and its O/F ratio are specified, T T ,GG and as 2.0, 2.5, 3.0 and 4.0 to investigate its effect on the engine per-
chemical composition of GG combustion gas can be determined. formances.
The chemical composition of GG combustion gas can determine Turbine expansion ratio, πturb , is considered to be constant
the molecular weight and the specific heat ratio of GG combustion regardless of πcomp . Therefore, GG combustion pressure, P T ,GG ,
gas. Those results are necessary to evaluate turbine power. should vary depending on πcomp because turbine exit pressure,
P T ,turb,out , must be larger than the ram combustor pressure, P T ,ram .
2.3. Analytical condition of GG-ATR engine performance As well as P T ,GG , nozzle expansion ratio, ε , should also vary with
πcomp because πcomp can affect static pressure at a nozzle exit,
Table 2 lists the analytical parameters for GG-ATR engine com- P nozzle . The present analysis eliminates the effects of pressure dif-
ponents. Pressure loss coefficients at each engine components are ference on the nozzle exit from the influences of πcomp on specific
defined as the ratio of outlet to inlet total pressure. On the other thrust and Isp. Based on the above, P T ,GG and ε are listed in Ta-
hands, heat loss coefficients are the ratio of outlet to inlet total ble 3. The values of ε in Table 3 give the static pressure at the
temperature at the engine components. For example, pressure and nozzle exit as about 40 kPa in sea level static condition. If the noz-
heat loss coefficients at a ram combustor, π ram , and τ ram , are rep- zle exit pressure is lower than this pressure level, flow separation
resented as equation (7), will occur within a nozzle. Moreover, for ε in Table 3, the nozzle
flow will be optimum expansion when flight Mach number and al-
T T ,ram,out P T ,ram,out titude are 1.3 and 11 km, respectively, which correspond to typical
τram = , πram = (7)
transonic flight condition.
T T ,ram,in P T ,ram,in
The parameters in Table 2 are based on those of the GG-ATR 2.4. Code validation with experimental data
engine developed at Muroran Institute of Technology. Adiabatic
compression efficiency, ηcomp , is given as 78% in on-design con- The author has already confirmed the accuracy of GATRAC in
dition. Turbine adiabatic efficiency, ηturb , is determined by a veloc- reference [8] in comparison with the experimental data. However,
ity ratio, U /C 0 , as described in the previous section. The typical the present paper describes those results of the code validation of
 R. Minato / Aerospace Science and Technology 91 (2019) 337–356 341

engine was called Build 1. The latest one was Build 5, which em-
ploys the improved ram combustor mixer. The rotational speed of
their GG-ATR engine is 68000 rpm in on-design condition. The
analytical results by the present code are compared with the ex-
perimental data in the range from 40000 to 68000 rpm, where the
specific rotational speed, N S , corresponds from 0.588 to 1.0.
Fig. 3 indicates the comparison between the analytical and the
experimental thrust and Isp of the hydrazine-fueled GG-ATR en-
gine. The engine performance parameters are described in ref.
[7], although the ram combustor pressure recovery, π ram , is not
mentioned. If π ram is equal to 0.88, the analytical predictions of
thrust and Isp show good agreement with the experimental results.
Fig. 4 shows the comparison between the analytical prediction
and the experimental result for the ram combustor temperature,
T T ,ram . This figure also indicates that the analytical predictions
agree well with the experimental data. The results in the code
Fig. 3. Analytical and experimental thrust and Isp of hydrazine fueled GG-ATR en- validation imply the reliability of GATRAC for the GG-ATR engine
gine. (For interpretation of the colors in the figure(s), the reader is referred to the
web version of this article.)
analyses.

3. Results and discussion

3.1. Specific thrust, Isp and ram combustor temperature

In general, a GG-ATR engine has higher specific thrust than tur-

bojet engine because of its high exhaust gas temperature. The typi-
cal specific thrust of an air-breathing engine is about 100 kgf/(kg/s)
at the maximum. Fig. 5 shows the specific thrusts of the GG-ATR
engine for four alkanes and four alcohols as functions of T T ,GG .
πcomp in Fig. 5 are 2.0, 2.5, 3.0, and 4.0. The results in Fig. 5 are
the ones in the Sea Level Static condition. In general, a larger πcomp
leads to a higher specific thrust. For all of those alkanes and alco-
hols, those specific thrusts have the maximum value about T T ,GG .
The peak T T ,GG for the specific thrust moves to higher T T ,GG con-
dition as the number of carbon atoms in fuels increases. Moreover,
the peak T T ,GG for the specific thrusts of alcohols are less than
those of alkanes.
Alcohols have higher specific thrust than alkanes in lower
Fig. 4. Analytical and experimental ram combustor temperature of hydrazine fu eled T T ,GG condition. This result indicates the advantage of alcohol over
GG-ATR engine. alkane in GG-ATR engine application because higher T T ,GG cannot
be acceptable. To explore the mechanism of these results, Fig. 6
GATRAC [8]. For code validation, the author utilizes the experimen- shows the ram combustor temperature, T T ,ram , for these fuels, as
tal data in reference [7], where Lilly et al. tested the hydrazine- functions of T T ,GG . The behaviors of T T ,ram are very similar to
fueled GG-ATR engine. In their tests, they conducted the GG-ATR those of specific thrust in Fig. 5. The peak T T ,GG for T T ,ram are a
engine tests for five different engine configurations. The original little higher than those for specific thrusts. Combustion in a ram

Fig. 5. Specific thrust of GG-ATR engine at sea level static condition.

342 R. Minato / Aerospace Science and Technology 91 (2019) 337–356

Fig. 5. (continued)

combustor at those peak T T ,GG corresponds to a stoichiometric higher. An increase of πcomp means a growth of the compression
condition. Thus, if T T ,GG is lower than those peak ones, GG com- work. Thus, higher T T ,GG is necessary to keep a propellant-to-air
bustion gas burns under fuel rich condition in the ram combustor. ratio to stoichiometric one at the ram combustor, if πcomp in-
Those peak T T ,GG for specific thrust increase as πcomp becomes creases.
 R. Minato / Aerospace Science and Technology 91 (2019) 337–356 343

Fig. 6. Ram combustor temperature at sea level static condition.

Fig. 7 shows Isp of the GG-ATR engine for four alkanes and four eq. (3), higher T T ,GG can decrease PAR, resulting in the improve-
alcohols with LOX oxidizer in Sea Level Static condition. For all ment of Isp. In general, lower molecular weight alkane leads to
those eight fuels, higher T T ,GG leads to higher Isp. As indicated in higher Isp. The differences of Isp among those four alkanes grow
344 R. Minato / Aerospace Science and Technology 91 (2019) 337–356

Fig. 6. (continued)

Fig. 7. Isp at sea level static condition.

 R. Minato / Aerospace Science and Technology 91 (2019) 337–356 345

Fig. 7. (continued)

in lower T T ,GG condition, and those shrink in higher T T ,GG . With 3.2. GG combustion gas properties
those four alkanes, the increments of Isp with T T ,GG show steep
increase under lower T T ,GG condition and turn to be moderate in PAR and ram combustor temperature, T T ,ram , are the main pa-
rameters to influence on Isp, as indicated in eq. (2). Thus, PAR is
higher T T ,GG . Higher T T ,GG is not realistic in the actual GG-ATR en-
analytically evaluated to investigate its effect on Isp. PAR is pro-
gine because of a thermal limitation of a turbine blade. Moreover,
portional to the molecular weight of GG combustion gas, M W ,GG ,
from the results in Figs. 5 and 6, too much higher T T ,GG is not
as indicated in eq. (3). Fig. 8 shows the molecular weights of GG
necessary for the GG-ATR engine application. combustion gas for πcomp = 2.5, as functions of T T ,GG . Higher πcomp
The behaviors of Isp for alcohols are very contrastive to those requires higher GG combustion pressure, P T ,GG . However, the de-
for alkanes. For example, in the case of πcomp = 2.5 and T T ,GG less pendency of πcomp on the molecular weight of GG combustion gas
than 1050 K, less number of carbon atoms in alcohol molecules has scarcely appeared. Fig. 9 shows the PAR, as functions of T T ,GG ,
leads to higher Isp. The relationship between Isp and the num- for πcomp = 2.5. The dependency of πcomp on PAR mainly comes
ber of carbon atoms in fuel molecule shows the same tendency from the numerator of eq. (3). Thus, the following relation can be
for alkanes. However, this relationship turns to be opposite in derived,
higher T T ,GG condition. Moreover, the more carbon atom num- (γ −1)/γ
f ∝ πcomp −1 (8)
ber in alcohol molecule leads to the higher Isp. Fig. 7 indicates
the significant advantage of alcohols to alkanes. For example, in From eq. (8), the results of PAR for πcomp = 2.0, 3.0 and 4.0 are
roughly equal to 0.72, 1.23 and 1.62 times of PAR for πcomp = 2.5,
the case of πcomp = 2.5, the lower molecular weight alcohol such
respectively. As indicated in Fig. 8, a smaller molecular weight
as CH3 OH or C2 H5 OH have larger Isp than C2 H6 or C3 H8 for of the GG combustion gas can be obtained for lower molecu-
T T ,GG less than 1050 K. This advantage is prominent for higher lar weight alkane in all range from T T ,GG = 900 K to 1400 K.
πcomp . The molecular weights of GG combustion gas for CH4 and C2 H6
346 R. Minato / Aerospace Science and Technology 91 (2019) 337–356

Fig. 8. Molecular weight of GG combustion gas for πcomp = 2.5.

Fig. 9. PAR for πcomp = 2.5.

have the minimum values at T T ,GG = 1200 K and 1350 K, respec- However, the results in Fig. 9 cannot explain that Isp of alcohols
tively. On the other hands, the tendency of molecular weight of GG is higher than those of alkanes. Higher Isp of alcohol arises from
combustion gas for alcohols alters at T T ,GG = 1200 K. Under this higher T T ,ram as shown in Fig. 6.
temperature, lower molecular weight alcohol has a smaller molec- Fig. 10 shows the equivalence ratios at GG as functions of T T ,GG
ular weight of GG combustion gas. However, in T T ,GG more than for πcomp = 2.5. In the range from T T ,GG = 1000 to 1250 K, the
1200 K condition, larger molecular weight alcohol has a slightly equivalence ratios of those four alcohols are roughly equal to that
smaller molecular weight of GG combustion gas. For both alcohols of CH4 and less than the other three alkanes. The GG combustion
and alkanes, the differences among molecular weight of GG com- gases of C2 H6 , C3 H8 and 1-C4 H10 burn under fuel-richer condition
bustion gas also shrink in higher T T ,GG condition. than those of alcohols in this T T ,GG range. For T T ,GG from 1300 K
PAR is inversely proportional to T T ,GG as indicated in the to 1400 K, the equivalence ratios of those four alkanes and four al-
eq. (3). PAR contains the molecular weight of GG combustion gas. cohols are all converged to about 3.0. Fig. 11 shows the chemical
Thus, PAR in Fig. 9 is greatly influenced by it. Therefore, PAR compositions of GG combustion gas for those alkane and alco-
has some common characteristics with the molecular weight of hol fuels in the range of T T ,GG from 900 K to 1400 K. The main
GG combustion gas. Firstly, lower molecular weight alkane has a chemical species in GG combustion gas are hydrogen (H2 ), CH4 ,
lower PAR. Secondary, the behaviors of PAR for alcohols alter at carbon monoxide (CO), carbon dioxide (CO2 ) and water (H2 O). The
T T ,GG = 1200 K. Lower molecular weight alcohols have lower PAR amounts of other species are negligible. H2 , CH4 , and CO are com-
for T T ,GG less than 1200 K. On the other hands, larger molecu- bustible species, and CO2 and H2 O are incombustible ones. In the
lar weight alcohols have a slightly larger PAR for T T ,GG more than chemical species in GG combustion gases, mole fractions of H2 and
1200 K. Moreover, for all those fuels, PAR converges on the same CH4 significantly contribute to the turbine power. High H2 mole
level if T T ,GG increases. For T T ,GG less than 1100 K, in comparison fraction contributes to the reduction of the molecular weight of
of alcohols to alkanes with the same number of carbon, 1-C4 H9 OH GG combustion gas. In lower T T ,GG condition, H2 mole fractions
has a lower PAR and molecular weight than 1-C4 H10 . However, PAR are relatively small, and CH4 mole fractions are higher in GG com-
of CH3 OH is higher than CH4 , and that of C2 H5 OH is also higher bustion gas. This tendency is prominent for C3 H8 and 1-C4 H10 . As
than C2 H6 . The previous study made clear that lower PAR can con- T T ,GG increases, H2 mole fraction also increases and CH4 is de-
tribute to higher Isp in comparison of C2 H5 OH with n-C12 H26 [8]. creased. For T T ,GG more than 1300 K, CH4 mole fraction is nearly
 R. Minato / Aerospace Science and Technology 91 (2019) 337–356 347

Fig. 10. Equivalence ratio at gas generator for πcomp = 2.5.

Fig. 11. Chemical composition of GG combustion gas for πcomp = 2.5.

348 R. Minato / Aerospace Science and Technology 91 (2019) 337–356

Fig. 11. (continued)

 R. Minato / Aerospace Science and Technology 91 (2019) 337–356 349

Fig. 11. (continued)

equal to zero, and H2 mole fractions are about 40 to 50 percent. fractions for alcohol are relatively higher than those for alkane for
In such conditions, the main chemical species in GG combustion T T ,GG less than 1100 K.
gases are H2 , CO, and others. An increase of T T ,GG means to reduce The chemical compositions of GG combustion gases in Fig. 11
the equivalence ratio at the GG combustion. Thus, comparing alco- can explain T T ,ram behaviors. The mole fractions of CO2 and H2 O
hols and alkanes with the same number of carbon atoms, H2 mole for alcohol fuel are generally higher than those for alkane. Those
350 R. Minato / Aerospace Science and Technology 91 (2019) 337–356

(a)

(b)

(c)

Fig. 12. (a) Comparison of specific thrust of 1-C4 H9 OH and t-C4 H9 OH. (b) Comparison of Isp of 1-C4 H9 OH and t-C4 H9 OH. (c) Comparison of ram combustor temperature of
1-C4 H9 OH and t-C4 H9 OH.
 R. Minato / Aerospace Science and Technology 91 (2019) 337–356 351

Fig. 13. Flight trajectory for off design analysis.

two are incombustible species and significantly affect T T ,ram be- tert-C4 H9 OH for the GG-ATR engine application. However, the
haviors. For example, the sum of CO2 and H2 O mole fractions in melting temperature of tert-C4 H9 OH is 298.8 K, and it tends to
the GG combustion gas of C2 H5 OH is about 20% for T T ,GG from be a solid state at room temperature. Thus, it is almost impossible
900 K to 1400 K. On the other hand, those are about 5% or less to utilize pure tert-C4 H9 OH for aviation fuel. It is used as a solvent
for C2 H6 fueled GG combustion gas. Although PAR of C2 H6 is less in ethanol or fuel additives in gasoline.
than that of C2 H5 OH, as indicated in Fig. 9, the ram combustion Fig. 12 shows the comparisons between 1-C4 H9 OH and tert-
of C2 H6 fuel for πcomp = 2.5 and T T ,GG = 1100 K is done under C4 H9 OH. Fig. 12a to 12c indicate specific thrusts, Isp and ram
fuel rich condition because its peak T T ,GG for T T ,ram is equal to combustor temperature, T T ,ram , respectively. In Fig. 12, compres-
about 1200 K. The ram combustion of C2 H5 OH for those πcomp sor pressure ratio, πcomp is 2.0, 2.5, 3.0, and 4.0. T T ,GG is ranged
and T T ,GG is considered to be fortunately done under the stoichio- from 900 K to 1400 K. There are almost no differences in those
metric condition because the peak T T ,GG for T T ,ram is about 1100 results between them. The differences among isomers on GG-ATR
K. This higher T T ,ram for C2 H5 OH can contribute to higher specific engine performance is negligible.
thrust and Isp than those of C2 H6 . Those tendencies are promi-
nent in lower T T ,GG conditions. As mentioned previously, lower 3.4. GG-ATR engine performance along flight trajectory
T T ,GG is preferable from the viewpoint of a thermal limitation of
a turbine blade. In general, alcohol has a disadvantage of its low The previous section treats the GG-ATR engine performance
specific heat release. The previous study described that C2 H5 OH analyses in the sea level static condition. From the results in Fig. 5
has an advantage to n-C12 H26 in the GG-ATR engine application. and 7, the superiorities of alcohols to alkanes is prominent in
The GG combustion gas of ethanol has a lower molecular weight lower T T ,GG condition. Thus, the present section investigates the
and greater turbine specific work than n-C12 H26 . These results are feasibility of alcohol fuels in the off-design condition. The off-
caused by high H/C ratio of ethanol, resulting in lower PAR and design analysis is conducted along the flight trajectory as shown
larger Isp [8]. The results in the present study indicate that al- in Fig. 13. This trajectory is the same one in reference [8]. A ve-
cohol like CH3 OH and C2 H5 OH have the superiority to not only hicle will take off at Mach 0.3, and travel along 25 kPa constant
n-C12 H26 , but also other alkanes like C2 H6 and C3 H8 , especially dynamic pressure trajectory for more than Mach 0.7.
in the case of lower T T ,GG . This advantage is brought about by In the off-design analysis, πcomp is equal to 2.5 and T T ,GG are
high T T ,ram due to nearly stoichiometric ram combustion. The GG- 1050 K and 1400 K. Fig. 14 and 15 indicate specific thrust and Isp
ATR engine in Muroran Institute of Technology utilizes ethanol and along this trajectory, respectively. Comparing Fig. 5(b), Fig. 7(b) and
LOX, as mentioned previously [5]. Thus, the utilization of C2 H5 OH Fig. 14, Fig. 15, the magnitude relations of specific thrust and Isp
for the GG-ATR engine is a reasonable option in comparison to avi- among those eight fuels do not change throughout this trajectory.
ation kerosene or C3 H8 . Both specific thrust and Isp increase in high altitude/high Mach
number conditions. Equation (9) defines the thermal efficiency of
3.3. Effects of isomers on GG-ATR engine performance the GG-ATR engine. This definition is the same one in Ref. [8],
2
V ext V2  
The analyses in the previous sections investigate the GG-ATR ṁ Air [(1 + f ) 2
− 2
] 1 + rOF 2
V ext V2
engine performances for the fuels in Table 1. Among those fuels, ηth = = (1 + f ) −
ṁfuel h fh 2 2
1-C4 H10 and 1-C3 H7 OH have isomers of 2-C4 H10 and 2-C3 H7 OH,
respectively. 1-C4 H9 OH has isomers of 2-C4 H9 OH, iso-butanol
(9)
(2-CH3 -1-C3 H7 OH), and tert-butanol (2-CH3 -2-C3 H7 OH). Differ- where h is heat release of fuel. rOF in eq. (9) is the oxidizer-to-
ences of engine performance among those isomers are considered fuel ratio at a gas generator, and f is PAR defined in eq. (3). The
to arise from a difference of specific heat release among them. thermal efficiency definition is similar to that of a turbojet engine.
The maximum difference of heat release among those isomers is Fig. 16 indicates the thermal efficiencies of the GG-ATR engine
only 1.27%, which is between 1-C4 H9 OH and tert-C4 H9 OH. Thus, along this trajectory. Those behaviors in Fig. 16 are entirely dif-
the present study focuses on the comparison of 1-C4 H9 OH and ferent from those of specific thrust and Isp. For both T T ,GG = 1050
352 R. Minato / Aerospace Science and Technology 91 (2019) 337–356

Fig. 14. Specific thrusts along the flight trajectory.

K and 1400 K, CH3 OH has the highest thermal efficiency in those engine. In the case of a turbojet engine, combustion occurs in fuel
fuels throughout this trajectory. In general, a thermal efficiency for lean conditions, and an enthalpy increment at a burner is roughly
alcohol is higher than that for alkane. Moreover, fuel with lower proportional to heat release of fuel. However, at the ram combustor
molecular weight has higher thermal efficiency. For T T ,GG = 1050 of the GG-ATR engine, fuel is often burnt under stoichiometric or
K, the thermal efficiency of 1-C4 H9 OH is almost equal to that of fuel rich conditions. As indicated in Fig. 6(b), the peak T T ,GG of
CH4 . CH3 OH and C2 H5 OH for T T ,ram are equal to about 1000 K and 1050
For the investigation to the thermal efficiency of the GG-ATR K, respectively. Therefore, the ram combustion for T T ,GG = 1050 K
engine, it is assumed that the kinetic energy of the nozzle exhaust is nearly stoichiometric ones.
gas is proportional to the ram combustion temperature, T T ,ram , On the other hand, the peak T T ,GG of alkanes for T T ,ram are
ranged from 1150 K to 1300 K. Thus, the ram combustion of alka-
2
V ext ∝ T T ,ram (10) nes for T T ,GG = 1050 K corresponds to fuel rich one. In those cases,
a part of fuel supplied to the GG is exhausted from the nozzle
Thus, the assumption in Eq. (10) can derive the following formula without being burned. The unburned fuel does not contribute to
concerned with the thermal efficiency, thrust production. That is why the thermal efficiencies of alkanes
(1 + rOF ) T T ,ram are lower than alcohols in T T ,GG = 1050 K.
ηth ∝ (11) In the case of T T ,GG = 1400 K, the ram combustion for all fu-
fh
els corresponds to fuel lean condition. The variances of thermal
The right-hand side of equation (11) represents the ratio of the efficiencies among alcohols and alkanes are shrunken. However,
ram combustor temperature to heat release of the fuel within the the thermal efficiencies of CH3 OH or C2 H5 OH are still higher than
 R. Minato / Aerospace Science and Technology 91 (2019) 337–356 353

Fig. 15. Isp along the flight trajectory.

those of alkanes. The ram combustion temperature, T T ,ram , of alka- From those analyses and discussions, the alcohol fuel is gener-
nes are much higher than that of alcohols as shown in Fig. 6(b). In ally superior to the alkane fuels for the GG-ATR engine application
higher T T ,ram gas, radical species, such as H atom, O atom, and OH for T T ,GG less than 1100 K. The superiorities of alcohols are es-
are much more produced. Fig. 17 shows the mole fractions of H pecially prominent for CH3 OH and C2 H5 OH. For actual GG-ATR
atom, O atom, and OH in the ram combustor for T T ,GG = 1400 K engine, T T ,GG must be less than 1100 K because of a thermal limi-
and πcomp = 2.5. The mole fractions of radical species for alkanes tation of a turbine blade. Therefore, the application of alcohol fuel
is higher than those for alcohols because of high temperature at is a reasonable choice for the GG-ATR engine.
the ram combustor. The formation of these radical species can ab-
4. Conclusion
sorb the heat release due to combustion. This heat absorption can
reduce the ratio of the combustion temperature to the heat release
In the present study, the alcohols are compared with the alka-
of fuel, which is proportional to the thermal efficiency of the GG-
nes to investigate those feasibilities to GG-ATR engine. The present
ATR engine, as indicated in eq. (11). Therefore, high mole fractions study treats CH3 OH, C2 H5 OH, 1-C3 H7 OH, and 1-C4 H9 OH for al-
of the radical species in the ram combustor result in the reduction cohol fuels. CH4 , C2 H6 , C3 H8 and 1-C4 H10 are selected as alka-
of the thermal efficiencies. nes. The conclusions of the present study are summarized be-
Finally, Fig. 18 indicates the overall efficiencies of the GG-ATR low.
engine for those eight fuels. The thermal efficiencies for those fu-
els can significantly affect the magnitudes of the overall efficiency 1. The specific thrust and ram combustor temperature, T T ,ram , for
among them because the differences of the propulsive efficiency those alcohols and alkanes are evaluated as functions of the
among those fuels are less than those of thermal efficiencies. GG combustion temperature, T T ,GG . Those have the maximum
354 R. Minato / Aerospace Science and Technology 91 (2019) 337–356

Fig. 16. Thermal efficiency along the flight trajectory.

values for T T ,GG . In general, the peak T T ,GG of alcohol is lower 3. The thermal efficiency of the GG-ATR engine largely depends
than that of alkane if the numbers of carbon atoms in fuel on a state of ram combustion. The superiority of alcohol fu-
molecule are the same. The less the carbon atoms number els can also appear in the thermal efficiency of the GG-ATR
is in fuel molecule, the lower peak T T ,GG is. The peak T T ,GG engine. For T T ,GG = 1050 K and πcomp = 2.5, the ram combus-
for T T ,ram is equal to or a little higher than that for specific
tion of alkanes corresponds to fuel-riched one, and some of
thrust.
the fuel does not burn. This unburned fuel reduces the ther-
2. For Isp of alkanes, the lower molecular weight of alkane can
lead the higher Isp in T T ,GG range from 900 K to 1400 K. On mal efficiency of the GG-ATR engine. On the other hand, for
the other hand for alcohol fuels, in lower T T ,GG condition, the the same T T ,GG and πcomp , the ram combustion of alcohol is
relationship between Isp and molecular weight of alcohol fuel nearly under the stoichiometric condition, resulting in higher
shows the same tendency with alkanes. The superiority of al- thermal efficiency.
cohol fuel to alkanes is significantly prominent in lower T T ,GG
condition. However, in higher T T ,GG condition, this relation- Alcohol has been scarcely used for aerospace vehicle fuels be-
ship is altered. cause of its lower specific heat release. However, the present study
For example in the case of πcomp = 2.5, the lower molecular
can indicate the superiority of alcohol fuel to alkane for the GG-
weight alcohols, such as CH3 OH or C2 H5 OH, have higher Isp
ATR engine application, especially in lower T T ,GG condition. Con-
than C2 H6 or C3 H8 for T T ,GG less than 1050 K. This advantage
becomes even more remarkable for higher πcomp . This higher sidering a thermal limitation of a turbine blade and, low T T ,GG is
Isp of alcohols can be explained by higher T T ,ram because the preferable for the actual GG-ATR engine operation. Therefore, the
ram combustion in those conditions corresponds to the stoi- utilization of alcohol is a reasonable option for the GG-ATR en-
chiometric one. gine.
 R. Minato / Aerospace Science and Technology 91 (2019) 337–356 355

Fig. 17. Radical species mole fractions in ram combustor for πcomp = 2.5 and T T ,GG = 1400 K.

Fig. 18. Overall efficiency along the flight trajectory.

356 R. Minato / Aerospace Science and Technology 91 (2019) 337–356

Declaration of Competing Interest [5] K. Mizobata, R. Minato, K. Higuchi, M. Ueba, S. Takagi, D. Nakata, K. Higashino,
N. Tanatsugu, Development of a small-scale supersonic flight experiment ve-
hicle as a flying test bed for future space transportation research, ISTS Special
The author declares no conflict of interest.
Issue: Trans. Jpn. Soc. Aeronaut. Space Sci., Aerosp. Technol. Jpn. 12 (29) (2014),
Po 3 1–Po 3 10.
Acknowledgement [6] R.K. Sullerey, A.M. Pradeep, M. Kedia, Performance Comparison of Air Tur-
borocket Engine with Different Fuel Systems, AIAA-Paper 2003-4417, 2003.
I would like to express my appreciation to Prof. Kazuyuki [7] J.S. Lilly, S.E. Hecht, B.G. Kirkham, C.A. Eadon, Experimental Evaluation of AA
Higashino, Prof. Masaharu Uchiumi, Assist. Prof. Daisuke Nakata Air Turbo Ramjet, AIAA Paper 94-3386.
[8] R. Minato, Advantage of ethanol fuel for gas generator cycle air turbo ramjet
(Aerospace Plane Research Center, Muroran Inst. of Tech.), Prof. engine, Aerosp. Sci. Technol. 50 (2016) 161–172.
Ryoji Imai, Associate Prof. Mitsutomo Hirota (Department of Me- [9] S. Gordon, B.J. McBride, Computer Program for Calculation of Complex Chem-
chanical and Aerospace Engineering, Muroran Inst. of Tech.), for ical Equilibrium Compositions, Rocket Performance, Incident and Reflected
their fruitful discussion, cooperation and assistance in the research Shock and Chapman–Jouguet Detonations, NASA SP-273, 1971.
[10] NOAA, NASA, USAF, U.S. Standard Atmosphere 1976, U.S. Government Printing
and development of Gas Generator Cycle Air Turbo Ramjet Engine.
Office, Washington, DC, Oct. 1976.
[11] Model Specification for Engines Aircraft Turbojet, MIL-SPEC MIL-E-5008B, U.S.
References Department of Defense, Jan. 1959.
[12] J.D. Mattingly, Element of Propulsion, AIAA Educational Series, 2006, p. 459.
[1] C.A. Snyder, A Parametric Study of a Gas-Generator AirTurbo Rocket (ATR), [13] Liquid Rocket Engine Turbine, NASA SP-8110, 1974.
NASA TM 88808, 1986. [14] K. Christensen, Comparison of methods for calculating turbine work in the air
[2] K. Christensen, Air turborocket/vehicle performance comparison, J. Propuls. turborocket, J. Propuls. Power 17 (2) (2001) 256–261.
Power 15 (5) (1999) 706–712. [15] http://webbook.nist.gov/chemistry/.
[3] H. Hasegawa, K. Kitahara, Y. Inukai, Compact and high thrust AirTurbo Ram
engine, J. Jpn. Soc. Aeron. Space Sci. 50 (582) (2002) (in Japanese).
[4] N. Tanatsugu, Development study on air turboramjet, in: Developments in
High-Speed-Vehicle Propulsion Systems, in: Progress in Astronautics and Aero-
nautics, vol. 165, 1996.