Performance Evaluation of Ammonium Dinitramide-Based Monopropellant in a 1N Thruster

: The development of propulsion systems based on green propellants, as an alternative to hydrazines, has been gaining interest within the space community. The study of Ammonium Dinitramide (ADN)-based liquid monopropellant, which is low-toxic and can deliver high performance, is the focal point of interest for Space Solutions Co., Ltd., Daejeon, Republic of Korea. A 1N ADN-based propulsion system was designed to evaluate the performance of the propellant. By combining a thermal and catalytic bed in a reactor, the performance of the propellant was examined in a designed thruster (chamber pressure of 11 bar). A total of 16 tests, with pulse mode experiments, were conducted; the accumulated ﬁring time was 285 s. The preheating temperatures were maintained between 350 and 400 ◦ C to achieve steady-state combustion. Notably, the maximum combustion efﬁciency was 91%. Test 9 recorded the highest decomposition temperature of propellant in the catalyst bed (1422 ◦ C). Interestingly, the combustion instability observed throughout this study was ≤ 0.5%. This study could assist in the further development of ADN-based propulsion systems.


Introduction
The Ammonium Dinitramide (ADN)-based green monopropellant propulsion system has been commercialized since the first flight demonstration in 2010 [1].In this system, a high-performance green propellant (HPGP) thruster was installed in the PRISMA satellite.The initial test results concluded that the specific impulse was 6 to 12% higher and had a 31% higher density-specific impulse compared with a hydrazine thruster [2].Since then, several ADN-based green monopropellant thrusters have been used for small satellite propulsion systems, including SkySat [3], CubeSat [4], ESLA-d [5], etc.The HPGP thruster, utilizing the LMP-103S formulation (63% ADN, 14% water, 18.4% methanol, 4.6% ammonia), has emerged as a preferred choice due to its low toxicity, ease of handling, and superior performance over hydrazine [6]; furthermore, it does not require the use of a SCAPE (Self-Contained Atmospheric Protective Ensemble) suite while loading the propellant.Additionally, LMP-103S can easily decompose on the catalyst surface at specific conditions [7-10]; hence, it is more suitable for monopropellant applications.Hydrazine, the traditional propellant, has good performance but produces toxic vapor, is carcinogenic, and poses risks to both humans and the environment.In recognition of the advantages of ADN-based propellants, extensive research has been undertaken in the past decades [11][12][13][14][15][16][17].This has led to several companies, such as ECAPS, the Lithuanian company, Nano-Avionics, Beijing Institute of Control Engineering (BICE), etc., continuously commercializing green propulsion systems for satellites [4,18].
While the satellite market continues to expand, the interest in green propulsion systems has extended, encompassing both developed and developing countries.One such example, Space Solutions Co., Ltd., South Korea, is commercializing a green propulsion system based on hydrogen peroxide and ADN-based monopropellant systems.We successfully tested a 50 N hydrogen peroxide green monopropellant propulsion system for the altitude control system of the three-stage Korean Space Launch Vehicle (KSLV) during model flight launches [19].Although hydrogen peroxide is a good option for latecomers to space technology who find it difficult to use hydrazine, it has issues with storage stability and requires passivation for aerospace components.As a result, it is not preferable for long-term missions.However, it remains highly valuable for short-term operations or early-stage space technology development.On the other hand, ADN-based green monopropellants offer good storability and are better suited for long-term space missions.Hence, here, we investigate a 1 N ADN-based monopropellant propulsion system.In our previous work, different types of reactors were used in order to study the combustion of 1 N ADN-based propellant [12].The composition of the propellant used for the study was similar to that of LMP-103S.The use of a thermal bed in addition to a catalyst bed is advantageous because it helps to describe the improvement in combustion efficiency.Moreover, the further improvement of a suitable catalyst to combust the propellant was studied in [10].However, there remains a need to refine the propulsion system to meet flight model specifications.
Hence, in the present study, a new reactor with a 1N thruster was fabricated, utilizing a stainless steel ball as the thermal bed and Pt-LHA (platinum-lanthanum hexa-aluminate) as the catalyst bed.The combustion of ADN-based propellant was thoroughly investigated using pulse mode operations.Sixteen experiments were conducted, and the collected data, including pressure and temperature parameters, were systematically analyzed.This study aims to delineate suitable combustion conditions and contribute to the ongoing development of environmentally friendly propulsion systems.

Thermal and Catalyst Bed Materials
Stainless steel (STS) balls with a diameter of 1.0 mm were used as the thermal bed material in the reactor.Furthermore, 20 wt% Pt on LHA (Pt-LHA) was used as the catalyst material, after the thermal reactor, to decompose the propellant.Figure 1 presents images of the thermal and catalyst bed materials.
The synthesis methods for the catalyst are reported in our previous work [10].The reduction of Pt-LHA was performed in a tubular furnace at a heating rate of 2 • C/min and an isothermal temperature of 600 • C for 2 h.A 4% H 2 in N 2 gas mixture was used (flow rate of 0.4 L/min) to reduce Pt (II) to Pt (0).The surface area and density of the catalyst were 8.9 m 2 /g and 4.7 g/cm 3 , respectively.

Reactor Design
For the thruster design, a theoretical analysis of the propellant was performed using NASA CEA [20,21].The nozzle expansion ratio (A e /A t ) and chamber pressure (P c ) were set to 100 and 11 bar, respectively.The heat of formation of ADN in solution was calculated using a reported method, resulting in a value of −110.22 kJ/mol [16].This value, along with the percentage composition of each molecule was utilized in the CEA program.The rocket performance, such as characteristic velocity, specific impulse, and adiabatic decomposition temperature, was computed by considering vacuum conditions (Table 2).The specific impulse (I sp ) of the propellant was calculated by varying the nozzle expansion ratio (Figure 2).It was found that the specific impulse reached 268.6 s when the nozzle expansion ratio was set to 100.Observations indicated that beyond an expansion ratio of 100, the specific impulse showed minimal increases with an increase in the expansion ratio.Therefore, it was concluded that an expansion ratio of 100 was appropriate for a 1N-class thruster.In order to design the 1N thruster, the mass flow rate of the propellant was determined by considering the theoretical specific impulse of 268.6 s.However, it was noted that the I sp efficiency may not be high when considering the ECAPS 1N thruster.The commercialized 1N-class ADN-based monopropellant thruster of ECAPS has specific impulse of 231 s.It can be inferred that the 1N-class has relatively low combustion efficiency.Hence, the specific impulse efficiency according to the thrust level was calculated by considering theoretical I sp and ECPAS thruster I sp (Table 3).By considering the similar propellant composition used by ECAPS, the estimated I sp efficiency for the 1N thruster was 86%.The mass flow rate of the propellant was determined by considering the specific impulse efficiency.The mass flow rate was calculated using Equation ( 1).
where F is the thrust, ṁ is the mass of the propellant, I sp is the theoretical specific impulse, g o is the standard acceleration of gravity, and η I sp is the specific impulse efficiency.Furthermore, the nozzle throat size was calculated by considering the parameters such as chamber pressure (p c ), mass flow rate ( .m), and characteristic velocity (c * ).The nozzle throat area (A t ) can be determined using Equation (2), Table 4 mentions the design parameters of the reactor.The diameter and length of the reactor were specified as 10 mm and 40 mm, respectively.Further, it was segmented into a heat bed and catalyst bed, based on the findings of a previous study [12].Following the injector and preceding to the catalyst bed, a 10 mm thermal bed of STS balls (1.0 mm) was installed.A 30 mm catalyst bed, containing Pt-LHA (16-20 mesh), was installed.A thermal barrier and a feed tube were placed between the reactor and the propellant control valve to reduce heat transfer from the reactor to the control valve.The feed tube, functioning as a capillary tube, with a diameter of 0.25 mm was utilized, and the outlet of the capillary tube served as an injector.

Experimental Setup
The thruster test setup was configured as shown in Figure 4.The propellant was pressurized with gaseous nitrogen, and the mass flow rate of the propellant was measured using a coriolis mass flow meter (OVAL, ALTI Mass Type U).The propellant was supplied to the reactor via a solenoid valve.The pressure and temperature were measured using a sensor port installed on the reactor.Pressure was measured using a Kulite ETM-375 (1000 psia), while temperature was measured with an R-type thermocouple.Both the pressure sensor and temperature sensor were connected to the data acquisition (DAQ) system, with a sample rate of 1000 Hz for the pressure sensor and 10 Hz for the temperature sensor.The accuracy for both the pressure and temperature sensors was ±0.5% FSO (max) and ±0.1%, respectively.

Combustion Efficiency
Combustion efficiency (ⴄ * ) was calculated using Equation (3). is the experimental characteristics velocity calculated using Equation (4).To ensure accurate estimation, the chamber design parameters, experimental propellant mass, and chamber pressure for each test value were considered.The theoretical characteristics velocity ( * ) was calculated using NASA-CEA.
where  is the chamber pressure,  is the area ratio, and ṁ is the mass flow rate of the propellant.

Combustion Instability
Combustion instability can be calculated by considering the root mean square of chamber pressure ( ), as shown in Equation ( 5).

𝑝 𝑝
where  , is the chamber pressure of ith value,  , means the average chamber pressure, and  is the number of ith values.

Combustion Efficiency
Combustion efficiency (ⴄ * ) was calculated using Equation (3). where . * is the experimental characteristics velocity calculated using Equation (4).To ensure accurate estimation, the chamber design parameters, experimental propellant mass, and chamber pressure for each test value were considered.The theoretical characteristics velocity ( * ) was calculated using NASA-CEA.
where  is the chamber pressure,  is the area ratio, and ṁ is the mass flow rate of the propellant.

Combustion Instability
Combustion instability can be calculated by considering the root mean square of chamber pressure ( ), as shown in Equation ( 5).

𝑝 𝑝
, where  , is the chamber pressure of ith value,  , means the average chamber pressure, and  is the number of ith values.
where c * expt. is the experimental characteristics velocity calculated using Equation (4).To ensure accurate estimation, the chamber design parameters, experimental propellant mass, and chamber pressure for each test value were considered.The theoretical characteristics velocity (c * ideal ) was calculated using NASA-CEA.
where p c is the chamber pressure, A t is the area ratio, and ṁ is the mass flow rate of the propellant.

Combustion Instability
Combustion instability can be calculated by considering the root mean square of chamber pressure (p c−RMS ), as shown in Equation ( 5).
where p c,i is the chamber pressure of ith value, p c,avg means the average chamber pressure, and n is the number of ith values.

Results and Discussion
A 1N ADN-based monopropellant thruster was designed by considering the specific impulse efficiency of 86% and a theoretical propellant mass flow rate of 0.44 g/s.To decrease void volume and increase contact points with the propellant in a reactor bed, 1.00 mm stainless steel balls were used in the thermal bed.The purpose of the thermal bed was to evaporate the propellant before reaching the catalyst bed.The Pt-LHA catalyst bed (mesh size of 16-20 mm) was installed after the thermal bed to decompose the propellant.
As previously discussed in our published work, the thermal bed plays a significant role in the reactor [12].Therefore, the total length of the reactor was 45 mm, including the thermal bed (10 mm), the catalytic bed (30 mm), and the chamber (5 mm) (Figure 5).A shorter thermal bed was necessary to prevent the thermal decomposition of the propellant.In addition, temperature and pressure sensors were installed to obtain accurate data during the thruster test.Pressure sensors P1 and P2 were installed to measure the pressure drop (∆P) across the injector.The injector was calibrated to precisely measure the mass flow rate of the propellant.The designed mass flow rate was 0.44 g/s at a tank pressure of 22 bar.
A total of sixteen hot firing tests were conducted, including pulse mode operation (Tables 5 and 6).The preheating temperature, which heats the reactor before propellant injection, is essential to initiate the decomposition reaction of ADN-based propellants on the catalyst.In this study, the rocket engine was started with a preheating temperature of around 350-400 • C. The rocket fired 148 g of the propellant over an accumulated firing time of 285 s.During the firing test, the minimum temperature (T1 Min ) was measured after the propellant was injected into the thermal bed (Tables 5 and 6).The highest temperature of the catalyst bed for each pulse was measured at T2 Max , T3 Max , and T4 Max .Initially, the first three tests were performed with a pulse duration of 1 s each and a total of five pulses.The preheating temperature was similar for tests 1 to 3.However, the temperature generated in the reactor was different after combustion.A low combustion temperature was (at T2) and a high mass flow rate was recorded in test 1.The propellant mass was consumed initially by an empty feeding line rather than reaching the reactor at the first pulse.As a result, excess mass (1.097 g/s) was recorded on the mass flow controller.This also resulted in a low temperature in the reactor.However, in the consecutive pulses of test 1, the propellant mass was stabilized, improving the reactor temperature at T2 in the range of 800-900 • C (Table 5).This was still lower than tests 2 and 3, which could be due to the heat generated during the initial pulses in test 1 being used to heat the reactor.Once the reactor heated up enough, most of the energy could be used for combustion reactions in tests 2 and 3, rather than heating the system.Furthermore, the average chamber pressures were not able measured due to the low pulse time.Hence, the maximum pressure (P Max ) and temperature (T Max ) of each pulse were noted for comparison.The maximum chamber pressure of each pulse was considered to calculate the combustion efficiency of tests 1 to 3. It was found that the combustion efficiency improved with consecutive pulses (Table 5).Additionally, the combustion performance was also improved in tests 1 to 3 with consecutive pulses.∆P: injector pressure drop, ï c* : combustion efficiency calculated by considering average chamber pressure.
Figure 6a,b presents representative pressure and temperature curve data of test 3.When the thermal bed was preheated at 353 • C (T1), the catalyst bed was also heated to 700 • C in an uncontrolled manner.However, in our future investigations, we will aim to maintain a constant temperature throughout the reactor.During the firing, the temperature at T1 dropped to 164 • C in the first pulse and remained in the range of 147 to 153 • C for subsequent pulses.This suggests that the propellant absorbed heat without decomposition within a specific time period.Interestingly, the temperature at T2 gradually increased, indicating that the propellant may sufficiently evaporate before reaching the catalyst bed and begin to decompose on the catalyst.As the short pulse mode operation continued, the temperature at T2 gradually increased until the third pulse and then stabilized.Additionally, the temperature was higher at T2 than at T3 and T4, suggesting that the propellant was fully decomposed and produced maximum heat at T2 (T2 Max : 1270 • C).
Table 6 mentions the experimental results of tests 4 to 16.For tests 4 and 5, three consecutive pulses of 5 s each were performed.The preheating temperature was about 383-394 • C and the mass flow rate of the propellant was 0.54-0.56g/s.The temperature and pressure results were similar for both tests with a maximum temperature of 1346 • C at T2.The firing time of 5 s was sufficient to stabilize the chamber pressure, and the average chamber pressure of a stabilized pressure curve region was noted.The chamber pressure for all pulses varied around 11 bar, and the combustion efficiency ranged from 81 to 83%.
Figure 7a,b shows the pressure and temperature curves of test 6.The preheating temperature (T1) was 400 • C. Further, the temperature (T1) of the thermal bed was stabilized at around 200 • C over the period of 10 s firing, indicating that the temperature was sufficient to heat or evaporate the propellant.Moreover, it was also revealed that the pressure was slowly increased from 9 to 11 bar until 4 s and then stabilized.In a similar way, the maximum temperature at T2 (T2 Max : 1376 • C) was achieved after 4 s and remained stable over time.Hence, it seems that the temperature and pressure were correlated.The complete combustion of the propellant generated maximum temperature and pressure at an average mass flow rate of 0.53 g/s; the pressure-stabilized curve region was considered for the analysis (Figure 7a).The average chamber pressure was 11.12 bar, resulting in a combustion efficiency of 85%.The chamber temperature also steadily increased to 1031 • C. It can be concluded that the combustion of the propellant may be completed at the T2 region, and the hot gases pass to the chamber via the catalyst bed, lowering the temperature at T4 (Figure 7b).However, the temperatures at T3 and T4 gradually increased over time.Furthermore, two consecutive pulses (10 s each) were performed in test 7. Similar test results were obtained, with a maximum combustion efficiency of 88%.From tests 8 to 10, the firing time of a single pulse increased from 15 to 25 s without any changes in preheating temperature (Table 6).Figure 8 shows a representative example of a long firing pulse.A steady-state combustion with an average chamber pressure of 11.12 bar was achieved at a propellant mass flow rate of 0.51 g/s.The pressure and temperature curves were similar to those in Figure 7. Interestingly, the combustion efficiency improved to 88% as the mass flow rate of the propellant decreased compared with the previous test results (see Table 6).The initial chamber pressure rising curve was similar to Figure 6a and stabilized after 4 s.However, it was found that the temperature of the heat bed and the catalytic bed reactor was very stable over the firing period (Figure 8b).In tests 11 to 16, multiple pulses were performed (of 10 s each).The propellant mass flow rate was ~0.51 g/s.The preheating temperature was set at 400 • C, except for test 13.In test 13, a low preheating temperature (370 • C) was used to find the effect on propellant heating during combustion.As a result, the heat bed temperature (T1) was reduced to 177 • C during the first pulse.However, no significant effect on the temperature at T2, T3, or T4 was observed.
Figure 9 presents the experimental results of test 16.A steady-state combustion was observed with a continuous firing time of 10 s each (total of four pulses).The time lapse between the two pulses was 3 s, and the duty ratio was 77%.Initially, the temperature was not steady over the time at T1, fluctuating around 400 • C. The chamber pressure, the mass flow rate of the propellant, and the firing temperature values were similar to the previous test.In the first pulse, a significant time delay was observed to raise pressure from 9 to 11 bar.However, the results, from the second pulse onward were not similar to those of the first pulse, as a negligible time delay was observed to raise the pressure.This phenomenon was similar to the research reported in [23].Although propellant was supplied for a longer time of approximately 40 s (with an on-time of 10 s and off-time of 3 s), the decomposition remained stable.Interestingly, the average combustion efficiency was 91% at a pressure drop of 8 bar.Furthermore, the combustion stability of test 16 was calculated by considering P c-RMS , Equation ( 5), resulting in a value of ~0.5%.This low combustion instability indicates a steady-state combustion of the ADN-based propulsion system.Overall, it was observed that there was a decrease in combustion efficiency during initial or short pulses and an increase in combustion efficiency during consecutive pulses or longer burns.This may be due to thermal leakage, as the system takes time to heat up during the initial period and once it is heated, more energy may be used in the combustion reaction rather than heating the system.

Conclusions
The present study focused on evaluating the performance of an ADN-based propellant in a 1N-class rocket engine.The preheating temperature for the thermal reactor was varied within the range of 350 to 400 • C, and a temperature above 370 • C was found to be suitable for the ADN combustion process.The use of STS balls in the thermal bed revealed efficient primary evaporation of the cooled propellant, as evidenced by the relatively stable temperature at T1 during firing tests, with the exception of test 16.The maximum decomposition temperature of the propellant was observed at T2 in the catalyst bed (test 9, T2 Max of 1422 • C), confirming the efficient decomposition of the propellant on the catalyst bed at the initial phase.In addition to pre-heating the thermal bed, the catalyst bed was also heated above 400 • C, thus contributing to the immediate decomposition of the propellant at T2.As the gaseous species passed from T2 to T4, the temperature decreased by more than 200-300 • C, likely due to heat loss in the catalyst bed.This suggests that the size of the catalyst bed could be minimized to maintain the combustion temperature.
Steady-state combustion was maintained even when the thruster operated in pulse mode within the range of 5 to 25 s.However, a time delay was observed in achieving a steady pressure, which may be correlated with thermal leakage.This issue could potentially be addressed by reducing the length of the catalyst bed.Test 16 demonstrated the highest average combustion efficiency of 91%, with an average chamber pressure of 11.24 bar and a propellant mass flow rate of 0.50 g/s.The pressure drop in the injector consistently remained at 8-9 bars throughout the experiment.Notably, the combustion instability was 0.5%, indicating the stable combustion of the ADN-based monopropellant system.
In summary, this study provides valuable insights for improving the development model of ADN-based monopropellant thrusters.In a further study, our focus will be on improving the combustion efficiency of monopropellant thrusters based on the findings from the current investigation.

Figure 1 .
Figure 1.Images of the thermal (STS balls) and catalyst (Pt-LHA) bed materials.

Figure 2 .
Figure 2. Specific impulse of the ADN-based monopropellant by the nozzle expansion ratio.

Figure 3
Figure3presents a 3D view of the 1N-class thrusters used in this study, where the exact position of sensor ports is shown.The sensors were used for the simultaneous measurement of temperature and pressure.

Figure 5 .
Figure 5. Pictorial representation of the reactor (P: pressure sensor, T: temperature sensor, TB: thermal bed; CB: catalytic bed; P1: pressure sensor before solenoid valve connected to the injector, T1 and P2: temperature and pressure sensor near to an injector in thermal bed; T2: temperature sensor position in an initial 20% of the catalyst bed after the thermal bed; T3: temperature sensor position in the catalyst bed near to the chamber).

Figure 6 .
Figure 6.Results of hot fire test 3: (a) pressure and mass flow rate of the propellant with time and (b) temperature with time.

Figure 7 .
Figure 7. Thruster test 6 with a heating temperature of 400 • C and valve operation of 10 s: (a) pressure results and (b) temperature results.

Figure 8 .
Figure 8. Thruster test 10 with a heating temperature of 400 • C and valve operation of 25 s: (a) pressure results and (b) temperature results.

Figure 9 .
Figure 9. Thruster test 16 with a heating temperature of 400 • C and valve operation of 10 s: (a) pressure results and (b) temperature results.

Table 4 .
Design parameters of the 1N-class thruster.

Table 5 .
Hot firing test results of tests 1 to 3.

Table 6 .
Experimental results of the 1N thruster test (tests 4 to 16)