Burning characteristics of ammonium perchlorate-based composite propellant supplemented with diatomaceous earth

Combustion and Flame 161 (2014) 620–630

Contents lists available at ScienceDirect

Combustion and Flame j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / c o m b u s t fl a m e

Burning characteristics of ammonium perchlorate-based composite propellant supplemented with diatomaceous earth Shingo Shioya, Makoto Kohga ⇑, Tomoki Naya Department of Applied Chemistry, National Defense Academy, Japan

a r t i c l e

i n f o

Article history: Received 28 July 2013 Received in revised form 3 September 2013 Accepted 20 September 2013 Available online 12 October 2013 Keywords: Ammonium perchlorate Composite propellant Diatomaceous earth Low-thermal-conductivity material Burning characteristics Thermal decomposition

a b s t r a c t For composite propellants, solid-phase thermal conduction is one of the dominant processes of propellant combustion and influences their burning characteristics. In this study, diatomaceous earth (DE) was used as a low-thermal-conductivity material, and the influence of DE on the burning characteristics of an ammonium perchlorate (AP)-based composite propellant was investigated. The ignitability of the propellant was improved by the addition of DE. DE showed both positive and negative effects on the burning rate of the propellant. The negative effect was attributable to the reduction of energy by the addition of DE. The enhancement in the burning characteristics was attributable to the particle shape and size of DE, the catalytic effect of Fe2O3, and the physical effect of SiO2; Fe2O3 and SiO2 are constituents of DE. The mechanism of the physical effect of SiO2 is as follows. The heat conduction in the solid phase is obstructed by SiO2 particles in the propellant matrix, and the temperature in the vicinity of these particles becomes higher. Consequently, a hot spot is formed on the burning surface side of the SiO2 particles, and the burning rate is then increased. Further, the hot spot effect was dependent on the AP interparticle distance in propellant matrix and the specific surface area of AP. Ó 2013 The Combustion Institute. Published by Elsevier Inc. All rights reserved.

1. Introduction Solid propellants are commonly used as solid fuels for rockets and missiles. Composite propellants are solid propellants that consist of oxidizer crystals, binder, curing agent, metal fuel, burning catalyst, and other components. Propellants with a high burning rate that generate a large quantity of combustion gases in a short period of time are required to realize high-performance rocket motors that would enable rockets to fly at high speeds. On the other hand, propellants with a low burning rate generate low thrust, and for example, are used as a gas generator for controlling vehicle flight. Therefore, many studies have been directed toward the development of propellants with wide range of burning rates, especially high burning rates. Ammonium perchlorate (AP) and hydroxyl-terminated polybutadiene (HTPB) have been widely used as an oxidizer and a binder, respectively. This is because AP/HTPB-based propellants have excellent burning and mechanical characteristics. The burning rate of an AP-based composite propellant depends on the AP particle size and AP content; it increases with a decrease in the AP diameter and an increase in the AP content.

⇑ Corresponding author. Address: Department of Applied Chemistry, National Defense Academy, Hashirimizu 1-10-20, Yokosuka, Kanagawa 239-8686, Japan. Fax: +81 046 844 5901. E-mail address: [email protected] (M. Kohga).

The burning process of the AP/HTPB composite propellant begins with the production of the decomposition gases of AP and HTPB at the burning surface with the heat fed back from the flame and the heat generated at the burning surface by AP thermal decomposition. The heat is conducted from the burning surface to the solid propellant and preheats the unreacted solid phase. These decomposition gases diffuse and mix in the gas phase and finally burn. It is important to study the combustion flame structure of a propellant to control its burning rate. The combustion mechanism of AP-based composite propellants has been investigated for many years and is almost established [1]. However, sufficient experimental data required to describe the combustion mechanism have not yet been collected. The thermal conduction process in solid propellants is one of the dominant processes of solid propellant combustion and influences the burning characteristics of the propellants. Metal wires embedded in a propellant increase its burning rate, thereby improving the thermal conduction in the solid phase [2–7]. However, SiC powder, which is a fine and high-thermal-conductivity material, does not affect the burning characteristics of propellants [8]. To study the influence of variations in solid-phase thermal conduction on the burning characteristics of a propellant, it is necessary to investigate the burning characteristics of a propellant supplemented by not only high-thermal-conductivity materials but also by low-thermal-conductivity materials.

0010-2180/$ - see front matter Ó 2013 The Combustion Institute. Published by Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.combustflame.2013.09.019

621

S. Shioya et al. / Combustion and Flame 161 (2014) 620–630

Nomenclature

Chemicals AP ammonium perchlorate DE diatomaceous earth HTPB hydroxyl-terminated polybutadiene Acronyms CAP coarse AP DTA differential thermal analysis FAP fine AP GDE ground DE SEM scanning electric microscopy TG thermogravimetry

Symbols Dw Isp Lp R T0 T1 Tf TP k n

weight mean diameter specific impulse AP interparticle distance in propellant matrix ratio of burning rate of AP/DE propellant to that of corresponding AP propellant temperature of unreacted propellant final combustion temperature adiabatic flame temperature exothermic peak temperature on DTA curve thermal conductivity amount of DE added to propellant

A low-thermal-conductivity additive must be unburnable because a burnable material generates heat by combustion, thereby increasing the burning rate. Kevlar fiber is an unburnable and low-thermal-conductivity material. It is used for increasing the burning rate of propellants by addition in small amounts [9]. This is because Kevlar fibers protruding through the propellant surface and into the gas phase have a ‘‘flame-holding’’ effect due to the shape of Kevlar. Reports of this nature have been hardly published to date. In this study, diatomaceous earth (DE) was used as an unburnable and low-thermal-conductivity material. Furthermore, DE is a powder, is cheap, and can be easily handled. The purpose of this study is to investigate the influence of DE on the burning characteristics of AP-based propellants. The addition of an unburnable material reduces the propellant performance by decreasing the energy density of the propellant. This study is part of a series of studies on the influence of variations in solid-phase thermal conduction on the burning characteristics of propellants, and is not an attempt to obtain high-performance propellants. 2. Experimental

Fig. 1. SEM image of DE.

2.1. Sample ingredients Coarse AP (CAP) and fine AP (FAP) were used as oxidizers in this study. CAP was prepared by grinding commercial AP (Kanto Kagaku) for 5 min in a vibration ball mill. FAP was prepared by the freeze-drying method [10]. The mean particle diameters of CAP and FAP were about 100 and 4 lm, respectively. DE (Kanto Kagaku) was used as a low-thermal-conductivity material. Figure 1 shows an SEM image of DE. The DE sample had a lot of porous particles with a heterogeneous shape. The specific surface area of DE was 27 m2 g 1. The main ingredients of DE are SiO2, Al2O3, and Fe2O3. Therefore, these materials were used as additives. SiO2 was supplied by Denka Kagaku Kogyo, Al2O3 by Association of Powder Process Industry & Engineering, and Fe2O3 by Kanto Kagaku. For a propellant with a high AP content, variations in solidphase thermal conduction due to the addition of DE would not be detected because of a large quantity of heat feedback from the gas phase and heat generated by the exothermic decomposition of AP at the burning surface. In this study, fuel-rich propellant samples with 75%, 70%, and 65% AP were prepared. HTPB was used as a binder and was cured by the addition of 8% isophorone diisocyanate (Tokyo Kasei). DE was added to the propellants in amounts ranging from 0.2% to 12%. The amount of DE added to a propellant is represented by n.

For preparing propellant samples, HTPB binder and additives were first sufficiently mixed in a polyethylene container. AP was then added to this mixture and mixed manually till a viscous slurry of the uncured propellant was obtained. The mixing temperature was 333 K. The uncured propellants were poured into a steel container and were degassed under vacuum. Subsequently, the degassed samples were cured in an oven at 333 K for one week. The density of AP is larger than that of the binder. If the volume of HTPB is larger than that of AP, AP particles would be precipitated at the bottom of the propellant. Therefore, an uncured propellant with a low AP content was rotated slowly during the curing process to avoid the separation of AP and HTPB. Three batches of propellants were prepared with the same propellant formulation. 2.2. Estimation of theoretical propellant performance The specific impulse (Isp), adiabatic flame temperature (Tf), and combustion products of the propellants were theoretically estimated in this study. The theoretical performance of the propellants was calculated using the NASA CEA program [11] with a combustion pressure of 7 MPa, an exit pressure of 0.1 MPa, and an initial temperature of 298 K. The theoretical performance was calculated at frozen equilibrium and optimum expansion. The standard en-

S. Shioya et al. / Combustion and Flame 161 (2014) 620–630

2.3. Measurement of propellant density The densities of all propellant samples were calculated from their volume and weight. Each sample was approximately 10 mm in diameter and 10 mm in length. The weight of a sample was determined with an electric balance with a minimum reading of 0.01 g. Further, the volume of the sample was measured with a pycnometer (Shimadzu Multivolume Pycnometer 1305), for which the minimum reading was 0.01 cm3. Helium was used as the charge gas for this apparatus.

2.5. Measurement of thermal decomposition The thermal decomposition of the propellants was measured by differential thermal analysis (DTA) and thermogravimetry (TG) performed using Rigaku Thermo Plus TG–DTA8120. The equipment was operated under a nitrogen flow (0.5 dm3 min 1) at atmospheric pressure. DTA and TG were carried out at a heating rate of 20 K min 1. The weight of the samples used was 1–2 mg. Reproducible TG–DTA curves of the composite propellants could not be obtained precisely owing to the heterogeneity of the propellants. The TG–DTA measurements were conducted more than four times for each sample, and the typical TG–DTA curve was used for analysis. 2.6. Measurement of burning rates Each propellant strand was 10 mm in diameter and 40 mm in length. Their burning behavior was investigated in a chimney-type strand burner (Kyouwa Kougyo KRS-RG-6085), which was pressurized with nitrogen. Each strand was ignited by applying 12 V to an electrically heated nichrome wire attached at the top. Each propellant strand was combusted in the pressure range 0.5–7 MPa. The burning phenomenon of the propellants was recorded using a high-speed video recorder (Photron FASTCAM-NET1000C). The burning rate was measured from the pictures recorded with this recorder. The pictures were recorded at a shutter speed of 400– 1000 frames s 1. The object was enlarged approximately 10 times and the regression length of the burning surface was measured with a resolution of 0.01 mm. The measurement error was within 0.1%. 3. Results and discussion 3.1. Theoretical propellant performance Figure 2 shows the theoretical Isp and Tf values of the propellant with 75% AP. As mentioned in Section 2.2, the standard enthalpy of formation of SiO2, which is the main ingredient of DE, was used to calculate the theoretical propellant performance. Both Isp and Tf decreased as n increased. The influence of n on Isp and Tf for the pro-

Isp Tf

2500

2400

2300

2600

2500

0

5

10

2400 15

ξ (%) Fig. 2. Theoretical Isp and Tf values of the propellant with 75% AP.

2.4. Measurement of thermal conductivity The thermal conductivity (k) of the propellants was measured with a heat flow meter apparatus (EKO Thermal Conductivity Tester HC-110) on the basis of Japanese Industrial Standard A 1412-2. Samples used in this measurement were 60 mm in diameter and 5 mm in thickness. The temperature range was 283–363 K.

2700

Tf (K)

.

2600

-1

thalpy of formation of DE was not obtained because DE is not a pure substance. DE consists of 88% SiO2, 5% Al2O3, 1% Fe2O3, and very small quantity of other additives. The standard enthalpy of formation of SiO2, which is the main ingredient of DE, was used to calculate the theoretical propellant performance.

Isp (Nskg )

622

pellants with 65% and 70% AP was the same as that for the propellant with 75% AP. For all the propellants, the SiO2 content in the combustion products theoretically agreed with that in the propellant, indicating that SiO2 contained in the propellant did not burn. Therefore, DE would hardly burn and would not entirely generate the heat of combustion. These results suggest that the burning rate would decrease with increasing n from the standpoint of energy balance. 3.2. Propellant density The DE sample had a lot of porous particles, as shown in Fig. 1. It was predicted that bubble contamination in a propellant containing DE would occur inside the porous DE particles because the voids in porous DE were not completely charged with HTPB. The bubble contamination increases the burning rate [12]. The propellant density was measured, and the bubble contamination in the propellants was estimated on the basis of the density. The densities of the propellants prepared with 75% AP are listed in Table 1. The deviation of the density remained within 0.01 g cm 3. The density decreases with increasing n. The densities of AP, cured HTPB, and DE were 1.95, 0.93, and 2.10 g cm 3, respectively. The theoretical density of the propellant with 75% AP was 1.52 g cm 3. These propellant densities almost agreed with their theoretical values. Moreover, the densities of the propellants with 65% and 70% AP were also identical to their theoretical values. These results indicate that these propellants did not have bubble contamination. 3.3. Thermal conductivity Table 2 lists the k values of the propellants. The k values of AP and cured HTPB were 0.371 and 0.184 W m 1 K 1, respectively. These k values were measured at hot and cool plate temperatures of 363 and 283 K, respectively. k was scarcely dependent on the

Table 1 Densities of propellants prepared at 75% AP. n (%)

0 2 4 8 12

Density (g cm

3

)

CAP

FAP

Theoretical value

1.52 1.53 1.54 1.56 1.57

1.52 1.52 1.54 1.55 1.56

1.52 1.53 1.53 1.55 1.56

623

S. Shioya et al. / Combustion and Flame 161 (2014) 620–630

3.4. Thermal decomposition behavior

65% AP

FAP

CAP

FAP

CAP

FAP

0.324 0.332 0.358 0.374

0.317 0.328 0.348 0.363

0.311 0.316 0.343 0.358

0.308 0.321 0.341 0.357

0.292 0.303 0.323 0.344

0.272 0.302 0.320 0.344

Hot plate temperature: 363 K Cool plate temperature: 283 K.

ξ =0%

50

513 500

600 700 Temperature (K) (a)

ξ =2%

Exo.

662

0

Endo.

50 513

500

600 700 Temperature (K) (b)

Exo.

653

Endo.

50 513

500

600 700 Temperature (K) (c)

800

Fig. 3. TG–DTA curves of CAP propellants with 75% AP at (a) n = 0%, (b) n = 2%, and (c) n = 8%.

0

50

100 800

600 700 Temperature (K) (a) 596 612

ξ =2%

0

50

500

100

ξ =0%

DTA TG

513

100 800

(c)ξ =8% 0

659

513 500

100 800

Mass loss (%)

Endo.

DTA TG

0

Mass loss (%)

Exo.

671

Mass loss (%)

plate temperatures in this study. At n = 0%, k of the CAP propellant was slightly higher than that of the FAP propellant. As the AP content reduced, k decreased because of the decrease in the proportion of AP, k of which was higher than that of cured HTPB, in the propellant. The k value increased with increasing n. The k value of SiO2 was approximately 1.3 W m 1 K 1. This is a low value among ceramics but is higher than those of AP and cured HTPB. Consequently, k slightly increased as n increased. k of the AP-based propellant added with SiC, which is a highthermal-conductivity material, slightly increases with increasing SiC content [8]. The increment in k for the propellant added with SiO2 was almost the same as that for the propellant added with SiC. Further, it was found that the variation of k of the solid phase was hardly dependent on the presence of high- or low-thermalconductivity powder in the propellants.

513 500

600 700 Temperature (K) (b)

100 800

ξ =8% 0

579

602

Mass loss (%)

0 2 8 12

70% AP

CAP

Mass loss (%)

75% AP

Figure 3 shows the TG-DTA curves of the CAP propellants with 75% AP at n = 0%, 2%, and 8%. The DTA curves show an endothermic peak at 513 K corresponding to the crystal transition point of AP, and the exothermic decomposition began after the crystal transition point of AP. An exothermic decomposition peak was observed around 650–670 K, and the exothermic peak temperature (Tp) decreased with increasing n. From the TG curves, it is clear that rapid consumption occurred in the region of exothermic decomposition of the propellants, i.e., between 550 and 670 K. This region represents the main decomposition of the propellants. Further consumption occurred in the range from 700 to 760 K. The consumption of HTPB began at approximately 600 K with total consumption around 780 K; in particular, the consumption in the range 700–780 K was large [13]. The consumption in the range 700–760 K was due to the decomposition of HTPB that was not consumed during the main decomposition of the propellant. Figure 4 shows the TG-DTA curves of the FAP propellants with 75% AP at n = 0, 2, and 8%. For the FAP propellant, at n = 0%, an exothermic decomposition peak is observed at 659 K; however, for the propellants containing DE, two exothermic peaks are observed. The temperature at the higher of these two peaks was defined as Tp in this study. Tp is greatly reduced by the addition of DE; furthermore, it also reduces as n increases. From the TG curves, it is clear that rapid consumption occurred in the region of exothermic decomposition of the propellants. The

50

Mass loss (%)

)

Exo.

1

Exo.

K

Endo.

1

Exo.

k (W m

Endo.

n (%)

Endo.

Table 2 Influence of n on k of propellant.

100 600 700 Temperature (K) (c)

800

Fig. 4. TG–DTA curves of FAP propellants with 75% AP at (a) n = 0%, (b) n = 2%, and (c) n = 8%.

624

S. Shioya et al. / Combustion and Flame 161 (2014) 620–630

Table 3 Influence of n on Tp. -1

FAP

CAP

FAP

665 663 655 648

659 589 583 579

667 666 658 648

660 599 578 572

consumption in the range 700–760 K, which was observed in the TG curve of the CAP propellant, was hardly detected for the FAP propellant. HTPB that came in contact with AP particles was decomposed by the exothermic decomposition of AP. The contact surface area of AP and HTPB for the FAP propellant was much larger than that for the CAP propellant because the specific surface area of FAP was larger than that of CAP. Therefore, HTPB contained in the FAP propellant was almost completely consumed by the thermal decomposition of FAP. Table 3 lists the Tp values of the propellants. At a constant AP content and n, Tp of the FAP propellant was lower than that of the CAP propellant. At constant n, Tp was almost independent of the AP content. The value of Tp decreased with increasing n. For the CAP propellant, the differences between Tp values for n = 0% and 12% were 28 K at 75% AP, 17 K at 70% AP, and 19 K at 65% AP. For the FAP propellant, the differences were 87 K at 75% AP, 80 K at 70% AP, and 88 K at 65% AP. The decrease in Tp due to DE addition in the FAP propellant was larger than that in the CAP propellant. These results suggest that with increasing n, the temperature region of main decomposition shifted lower for the FAP propellant than for the CAP propellant. Tp decreased upon the addition of a burning catalyst such as Fe2O3 and the decrease in Tp for the FAP propellant was larger than that for the CAP propellant [14]. As described above, Tp decreased upon DE addition because DE contains a small amount of Fe2O3. 3.5. Burning rate characteristics All measurements were verified at least three times at each pressure, and the burning rate was determined as the average of these values. The burning rate was not recorded if just one of the three batches of propellants did not ignite or combust in a stable manner. The unignited condition implied that the propellant sample did not burn, or barely burned, after the ignition. Self-quenched burning behavior indicated that the sample burned for some time after ignition, before extinguishing. This usually occurred within half of the propellant sample length in this study. The burning phenomena of the propellants were recorded using a high-speed video recorder as described in Section 2.6. The propellants containing Kevlar fibers showed luminous streaks on the combustion gases just above the burning surface [9]. Visual observation of the pictures recorded using this high-speed video recorder revealed that all the propellants prepared in this study showed steady-state combustion and their burning surface was nearly flat. Furthermore, the burning surface regressed with a constant speed. It was also found that the burning phenomena of the propellants supplemented with DE were almost identical to those of the propellants without DE. Figures 5 and 6 show the burning rate characteristics of the CAP and FAP propellants supplemented with DE. The standard deviation of the burning rate was within 0.08, and therefore, their dispersion was small. When the composition of the propellant is not uniform, the combustion of the propellant is unstable and the burning rates are greatly scattered. The propellants prepared

Burning rate

CAP

659 596 579 572

75% CAP

10 5

1

0.5

0.5

1

5

10

Pressure (MPa) (a) 20 -1

FAP

671 662 653 643

(mm s )

CAP

Burning rate

65% AP

70% CAP

10 5

1 0.5

0.5

1

5

10

Pressure (MPa) (b) 20 -1

70% AP

(mm s )

75% AP

0 2 8 12

(mm s )

Tp (K)

Burning rate

n (%)

20

65% CAP

10

0% 0.5% 2%

5

4% 8% 12%

1 0.5

0.5

1

5

Pressure

10

(MPa)

(c) Fig. 5. Burning rate characteristics of CAP propellants supplemented with DE at n = 0%, 0.5%, 2%, 4%, 8%, and 12%. (a) 75% CAP, (b) 70% CAP, and (c) 65% CAP.

in this study had stable combustion and reproducible burning rates, indicating that the AP and DE particles were dispersed uniformly in the propellant. The burning rates increased almost linearly on the logarithmic scale except for the FAP propellant with 65% AP. At n = 8%, the FAP propellant with 65% AP showed plateau burning at 1–2 MPa and mesa burning at 2–3 MPa; further, at n = 12%, this propellant showed plateau burning at 1–2 MPa and mesa burning at 2–5 MPa. At n = 0%, the CAP propellants with 75% and 70% AP burned in the pressure range 0.5–7 MPa, but that with 65% AP did not burn at 0.5 MPa. Further, the FAP propellants with 75% and 70% AP burned above 0.5 MPa and 2 MPa, respectively, but that with 65%

625

S. Shioya et al. / Combustion and Flame 161 (2014) 620–630

3

75% FAP

75% CAP

10 5

R (-)

Burning rate

-1

(mm s )

20

0% 0.5% 2%

1 0.5

4% 8% 12%

2

1

0 0.5

1

5

Pressure

5

10

15

ξ (%)

10

(MPa)

(a)

(a) 3

Burning rate

70% CAP

70% FAP

10

R (-)

-1

(mm s )

20

5

2

1 1 0 0.5

0.5

1

5

Pressure

10

15

(b)

(MPa) 3

65% CAP 65% FAP

10

R (-)

-1

(mm s ) Burning rate

10

ξ (%)

(b) 20

5

5

0.5 MPa 1 MPa 2 MPa

2

3 MPa 5 MPa 7 MPa

1 1 0.5

0

5

10

15

ξ (%) 0.5

1

5

Pressure

10

(MPa)

(c) Fig. 6. Burning rate characteristics of FAP propellants supplemented with DE at n = 0%, 0.5%, 2%, 4%, 8%, and 12%. (a) 75% FAP, (b) 70% FAP, and (c) 65% FAP.

AP did not burn in this pressure range. The CAP propellant with 65% AP and the FAP propellant with 70% AP could be burned at 0.5 MPa by the addition of DE above n = 0.5%. The FAP propellant with 65% AP could be burned above 2 MPa at n = 2% and at 0.5 MPa above n = 4%. Thus, it was established that the ignitability of fuel-rich AP-based propellants can be improved by the addition of DE. As mentioned above, the burning rates of AP-based propellants were increased by the addition of DE. The effect of n on the increase in the burning rate was investigated. The ratio of the burning rate of the AP/DE propellant to that of the corresponding AP propellant (R) was calculated from Figs. 5 and 6. Figures 7 and 8 show the

(c) Fig. 7. Relationship between R and n of CAP propellants. (a) 75% CAP, (b) 70% CAP, and (c) 65% CAP.

relationship between R and n of the CAP and FAP propellants. R of the FAP propellant with 65% AP could not be obtained because this propellant at n = 0% did not burn in the studied pressure range. For the CAP propellant, R increased with increasing n below 2% and decreased above this value, that is, R was maximum at n = 2%. The relationship between R and n was virtually independent of the burning pressure. The maximum R values for the CAP propellants with 75%, 70%, and 65% AP were 1.50 (at 5 MPa), 1.65 (at 7 MPa), and 1.73 (at 5 MPa), respectively. For the CAP propellant with 70% AP and at n = 12%, the R values at 1 and 2 MPa were below 1; in other words, the burning rate of this propellant at n = 12% was lower than that at n = 0%. For the FAP propellant with 75% AP, R increased with increasing n below 2% and decreased above this value. Further, for the FAP propellant with 70% AP, R increased below n = 4% and decreased

S. Shioya et al. / Combustion and Flame 161 (2014) 620–630

3

663 Exo.

2

0

CAP

DTA TG

Endo.

R (-)

75% FAP

50

100

513 500

1

Mass loss (%)

626

600

700

800

Temperature (K) 5

10

(a)

15

ξ (%) 3

0

FAP

50

Endo.

(b) 70% FAP

R (-)

598 619

Exo.

(a)

513

2

Mass loss (%)

0

100

500

600

700

800

Temperature (K)

0.5 MPa 1 MPa 2 MPa

1

0

5

3 MPa 5 MPa 7 MPa 10

(b) Fig. 9. TG–DTA curves of propellants added with GDE. (a) CAP propellant and (b) FAP propellant.

15

ξ (%) (b)

-1

(mm s )

above this value. The relationship between R and n was virtually independent of the burning pressure, similar to the case of the CAP propellant. The maximum R values for the FAP propellant with 75% and 70% AP were 2.32 (at 3 MPa) and 2.11 (at 0.5 MPa), respectively. The R values for all FAP propellants were above 1. It was found that the burning rate was increased by the addition of DE and R had a maximum value. This indicated that the addition of DE to the propellants had both positive and negative effects on the burning rate. The negative effect was attributable to reduction in energy density due to the addition of DE, as described in Section 3.1. The positive effect is discussed in the next section.

20

Burning rate

Fig. 8. Relationship between R and n of FAP propellants. (a) 75% FAP and (b) 70% FAP.

5

ξ =2% GDE 2%

1

0.5

0.5

1

5

10

Pressure (MPa) (a) 20

Burning rate

-1

(mm s )

3.6. Factors affecting positive effect of DE It was speculated that two factors increased R: particle properties of DE and the constituents of DE. However, the heat feedback process in the gas phase was not affected by the presence of DE because DE did not burn and its concentration in the gas phase was very small. The DE sample had a lot of porous particles with heterogeneous shapes. These particle properties were expected to affect the burning characteristics of the propellants. Ground DE (GDE) was prepared by grinding DE with mortar to remove its porosity and heterogeneity. The thermal decomposition behaviors and burning rate characteristics of the propellants prepared with GDE were measured, and the obtained results were compared with those for the propellants with DE. The main ingredients of DE are SiO2 (88%), Al2O3 (5%), and Fe2O3 (1%). These ingredients were expected to influence the burning characteristics of the propellants. Propellants with the constituents

CAP

10

FAP

10 5

1 0.5

0.5

1

5

10

Pressure (MPa) (b) Fig. 10. Burning rate of propellants added with GDE. (a) CAP propellant and (b) FAP propellant.

627

S. Shioya et al. / Combustion and Flame 161 (2014) 620–630

3.6.1. Particle properties Figure 9 shows the TG-DTA curves of the propellants added with GDE. From DTA curves, the Tp values of the CAP and FAP propellants added with GDE were 663 K and 598 K, respectively. These values almost agreed with those of the propellants with DE. Further, the TG curves of the CAP and FAP propellants supplemented with GDE were almost identical to those of the propellants with DE. Thus, the particle shape and size of DE did not affect the thermal decomposition of the propellants. Figure 10 shows the burning rate of the propellants added with GDE. The burning characteristics of the CAP/GDE propellants were almost identical to those of the CAP/DE propellants. On the other hand, the burning rates of the FAP/GDE propellants were lower than those of the FAP/DE propellants. Bubble contamination in a propellant increases its burning rate [12]. The DE sample had a lot of porous particles. If the burning

Exo.

DTA TG

700

800

500

Temperature (K) (a)

700

0

658 (b) Al2O3 0 50

513

800

500

Temperature (K) (b)

700

800

Temperature (K) (c) Fig. 11. TG–DTA curves of CAP propellants added with (a) 1.76% SiO2, (b) 0.10% Al2O3, and (c) 0.02% Fe2O3.

616 Exo.

Fe 2O3 0

100 600

700

800

(c) Fe2O3 0

595 50

Endo.

Endo. 500

600

(b)

50

513

100

Temperature (K)

Mass loss (%)

Exo.

665

800

Exo.

Al 2O3

100 600

700

Endo.

Endo. 500

600

(a)

50

513

100

Temperature (K)

Mass loss (%)

Exo.

669

0

50

513

100 600

SiO 2

656

Mass loss (%)

500

0

50

513

3.6.2.1. Thermal decomposition behavior. Figures 11 and 12 show the TG-DTA curves of the CAP and FAP propellants added with

513 500

Mass loss (%)

Endo.

DTA TG

3.6.2. Constituents of DE DE consists of 88% SiO2, 5% Al2O3, and 1% Fe2O3. As mentioned above, propellants supplemented with the constituents of DE were prepared on the basis of the formulation of the propellant with 75% AP and n = 2%. These propellants contained 1.76% SiO2, 0.10% Al2O3, and 0.02% Fe2O3. The thermal decomposition behaviors and burning characteristics of the propellants supplemented with each constituent were investigated.

Endo.

SiO 2

Mass loss (%)

Exo.

672

rates of the propellants were increased by bubble contamination inside the porous DE particles, the burning rates of the propellants added with GDE should have been lower than those of the propellants with DE. As mentioned above, the burning characteristics of the CAP/GDE propellants were almost identical to those of the CAP/DE propellants, indicating that these propellants did not have bubble contamination. This result supports the one presented in Section 3.2. The burning rate of the FAP propellant was increased because of factors other than bubble contamination. It was found that the particle shape and size of DE did not affect the burning characteristics of the CAP propellants but influenced those of the FAP propellants.

Mass loss (%)

of DE were prepared, and the thermal decomposition behaviors and burning characteristics of these propellants were investigated. As described above, the FAP propellant with 75% AP and at n = 2% showed the maximum R value in this study. For subsequent experiments, the propellants supplemented with GDE and constituents of DE were prepared on the basis of the formulation of the propellant with 75% AP and n = 2%.

100 600

700

800

Temperature (K) (c) Fig. 12. TG–DTA curves of FAP propellants added with (a) 1.76% SiO2, (b) 0.10% Al2O3, and (c) 0.02% Fe2O3.

628

S. Shioya et al. / Combustion and Flame 161 (2014) 620–630

1.76% SiO2, 0.10% Al2O3, or 0.02% Fe2O3. From DTA curves, the Tp values of the CAP propellant with SiO2 or Al2O3 were 672 K and 669 K, respectively, and those of the FAP propellant with SiO2 or Al2O3 were 656 K and 658 K, respectively. These Tp values of the CAP and FAP propellants were almost the same as those of the propellants at n = 0%. Fe2O3 is a burning catalyst, and its addition shifted Tp to a lower value [14]. The Tp values of the CAP and FAP propellants with Fe2O3 were 665 K and 616 K, respectively. These values are lower than those of the propellants with n = 0% but are almost equal to those with n = 2%. This fact suggested that Tp shifted to a lower value even when the quantity of Fe2O3 was very small (0.02%), and the addition of this small amount of Fe2O3 increased the burning rate. 3.6.2.2. Burning rate characteristics. Figure 13 shows the burning characteristics of the propellants supplemented with 1.76% SiO2, 0.10% Al2O3, or 0.02% Fe2O3. The burning rates of the propellants increased upon the addition of Fe2O3 because Fe2O3 acts as a burning catalyst in the propellants even when present in a small quantity. The burning rate of the CAP propellant with Fe2O3 was almost equal to that of the propellant with DE at n = 2%. However, the burning rate of the FAP propellant with Fe2O3 was lower than that of the propellant with n = 2%. The burning rates of the CAP and FAP propellants with Al2O3 were almost equal to that of the propellant with n = 0%. This result indicates that Al2O3 did not influence the burning characteristics. The burning rates of the CAP propellant with SiO2 were almost equal to that of the propellant with n = 0%. On the other hand, the

Burning rate

-1

(mm s )

20

CAP

10 5

1 0.5

0.5

1

5

10

Pressure (MPa) (a)

Burning rate

-1

(mm s )

20

burning rates of the FAP propellant increased upon the addition of SiO2. However, this increase was smaller than that in the propellant supplemented with DE at n = 2%. For the CAP propellant, the increment of the burning rate upon the addition of DE only contributed to the catalytic effect of Fe2O3 contained in DE because the burning rates of the CAP propellant with Fe2O3 were almost equal to those of the propellants with DE. For the FAP propellants, the burning rate of the propellant added with GDE was lower than that of the propellant added with DE, indicating that the particle properties of DE affected the burning characteristics of the FAP propellants. Furthermore, the burning rates were increased not only by the catalytic effect of Fe2O3 contained in DE but also by the presence of SiO2. The burning catalyst of the AP propellant shifted Tp to a lower value [15–20]. As mentioned above, Tp was not decreased by the addition of SiO2, suggesting that SiO2 did not have a catalytic effect. SiO2 is an unburnable material; therefore, the burning rate was increased because of the physical effect of SiO2 addition. This physical effect is discussed in the next section. 3.7. Mechanism of burning rate increase by SiO2 3.7.1. Combustion flame structure The combustion flame structure of a propellant is illustrated in Fig. 14. AP and HTPB decompose at the burning surface. These decomposition gases diffuse into the gas phase and burn. A large quantity of heat is produced by combustion of the decomposition gases and the heat is fed back to the burning surface of the propellant. AP and HTPB at the burning surface are heated by the absorption of the heat flux and the exothermic decomposition of AP. They are decomposed by the heat, and the decomposition gases are emitted to the gas phase. The combustion of the AP/HTPB propellant was maintained by these processes occurring in sequence. The temperature at the burning surface (Ts) is very high as compared to that (T0) of the unreacted solid propellant. Further, the heat conducts from the burning surface to the solid phase. The scheme of heat conduction of the propellant without/with SiO2 is illustrated in Fig. 15. For the propellant without SiO2, the heat feedback from the gas phase and heat generated by the exothermic decomposition of AP smoothly transmitted into the solid phase as shown in Fig. 15a. On the other hand, for the propellant with SiO2, the heat conduction was obstructed by SiO2 particles in the propellant, and the temperature in the vicinity of the SiO2 particles became higher as shown in Fig. 15b. Consequently, a hot spot was formed around the SiO2 particles. This hot spot accel-

FAP

10 5

ξ =0% ξ =2%

1 0.5

0.5

1

SiO2 Al2O3 Fe2O3 5

10

Pressure (MPa) (b) Fig. 13. Burning characteristics of propellants added with 1.76% SiO2, 0.10% Al2O3, or 0.02% Fe2O3. (a) CAP propellant and (b) FAP propellant.

Fig. 14. Combustion flame structure of propellant.

S. Shioya et al. / Combustion and Flame 161 (2014) 620–630

629

Fig. 16. Scheme of propellant matrix.

Fig. 15. Scheme of heat conduction in propellant without/with SiO2. (a) Propellant without SiO2 and (b) propellant with SiO2.

erated the thermal decomposition of AP and binder, and the burning rate was then increased. SiO2 is an unburnable and low-thermal-conductivity material. As mentioned above, the heat conduction from the burning surface to the unreacted solid phase was obstructed by SiO2 particles in the propellant, even though k of SiO2 is higher than that of AP and HTPB. This was because AP and HTPB decompose and burn, while SiO2 is an unburnable material. If the unburnable property improved the burning characteristics, SiC should have increased the burning rate. However, the burning rates of the propellants were not increased upon the addition of SiC [8]. Thus, unburnable and low-thermal-conductivity powders such as those of SiO2 particles act as heat conductors, and are expected to improve the burning characteristics of propellants. 3.7.2. Difference between CAP and FAP propellants As described in Section 3.6.2.2, the burning rate of the FAP propellant was increased by the addition of SiO2, but that of the CAP propellant was not increased. The reason is as follows. The flame of the AP composite propellant was a so-called diffusion flame, and the model of multiple flames is generally accepted as a model for its flame structure [21]. According to this model, the flame structure of the AP composite propellant consists of an AP monopropellant flame, a primary flame, and a final diffusion flame. The AP monopropellant flame, which is composed of AP decomposition products, is not considered to occur at the propellant surface but to extend from the surface. The primary flame is a premixed flame created by the oxidizer and binder decomposition products mixing completely before the reaction occurs. Further, the final diffusion flame follows the primary flame. The distance from the burning surface to these flames was dependent on the AP particle size. The distance decreased with decreasing particle size. Furthermore, the AP interparticle distance in the propellant matrix was one of the dominant factors determining the burning characteristics of the AP propellant [22,23]. The burning characteristics of the AP propellant were greatly affected by the gap (Lp) between the neighboring AP particles in the propellant matrix. As the value of Lp decreased, the flames came closer to the burning surface because the mixing of the decomposition gases of AP and HTPB became easier. Therefore, the heat feedback from the flame to the burning surface increased, and the decomposition of AP and binder at the burning face accelerated, thereby increasing the burning rate.

In the propellant matrix, SiO2 particles became closer to AP particles as the value of LP decreased. Thus, the hot spot effect of SiO2 particles increased with decreasing LP. Furthermore, AP particles with large specific surface areas received greater hot spot effect than the AP particles with small specific surface areas. The increase in the burning rate by the hot spot effect of SiO2 particles became greater with decreasing LP and increasing specific surface areas of AP particles. Figure 16 shows the scheme of the propellant matrix obtained by assuming that the unit cell of AP dispersed in the propellant was a face-centered cubic structure. LP was theoretically calculated using Dw and the AP content. The LP values of the CAP and FAP propellants with 75% AP were 9.9 and 0.3 lm, respectively. The values of the specific surface area of CAP and FAP were 0.06 and 2.0 m2 g 1, respectively. For the FAP propellant, the hot spot effect appeared because LP of the FAP propellant was small and the specific surface area of FAP was large. On the other hand, the burning characteristics of the CAP propellant were not affected by the hot spot effect because LP of the CAP propellant was large and the specific surface area of CAP was small. As described in Section 3.6.1, the burning rate of the FAP propellant with GDE was higher than that of the propellant with n = 0%; however, it was lower than that of the propellant with DE. GDE was prepared by the grinding of DE, and therefore, the particle size of GDE was smaller and the particle number of unit mass was larger than in the case of DE. The obstruction effect of heat conduction by the DE particles decreased with decreasing size of DE. Therefore, even when the number of DE particles, that is, the hot spot in the propellant matrix, increased, the burning rate of the FAP propellant with GDE was lower than that of the propellant with DE. In a previous paper [8], SiC particles, which are unburnable and have high thermal conductivity, did not affect the burning characteristics of propellants. From the results described above, it was found that the hot spot formed around the SiO2 particles improved the burning characteristics of the propellants in this study. It is necessary that the burning characteristics of the propellants supplemented with other low-thermal-conductivity materials should be investigated to clarify the influence of solid-phase thermal conductance on the burning characteristics.

4. Conclusions The thermal conduction process in solid propellants is one of the dominant processes of propellant combustion and influences the burning characteristics of propellants. In this study, DE was used as an unburnable and low-thermal-conductivity material, and the influence of DE on the burning characteristics of propel-

630

S. Shioya et al. / Combustion and Flame 161 (2014) 620–630

lants was investigated. Fine and coarse AP (FAP and CAP) were used as oxidizers for the propellants. The burning rate and ignitability of the AP-based propellants increased upon the addition of DE. DE showed both positive and negative effects on the burning rate. The negative effect was attributable to the reduction in energy density due to the addition of DE. The improvement in the burning characteristics was attributable to the particle shape and size of DE, catalytic effect of Fe2O3, and physical effect of SiO2; Fe2O3 and SiO2 are the main ingredients of DE. The addition of a very small amount of Fe2O3 (0.02%) improved the burning characteristics. For the CAP propellants, the increase in the burning rate upon the addition of DE was only contributed by the catalytic effect of Fe2O3. The particle properties of DE affected the burning characteristics of only the FAP propellants. Furthermore, the burning rates of the FAP propellants were increased by the catalytic effect of Fe2O3 as well as by the physical effect of SiO2. The mechanism of the physical effect of SiO2 is as follows. The heat conduction in the solid phase is obstructed by SiO2 particles in the propellant matrix, and the temperature in the vicinity of the SiO2 particles becomes higher. Consequently, a hot spot is formed around the SiO2 particles, and the burning rate is then increased. The hot spot effect was dependent on the gap between the neighboring AP particles and the specific surface area of AP. Further, it was found that low-thermal-conductivity particles influenced the thermal conduction in unreacted solid propellants.

References [1] K.K. Kuo, M. Summerfield (Eds.), Fundamentals of Solid-Propellant Combustion, Progress in Astronautics and Aeronautics, vol. 90, New York, 1984, pp. 53–119. [2] N. Kubota, M. Ichida, T. Fujisawa, AIAA J. 20 (1982) 115–121. [3] M. Tanaka, K. Morisaki, Kogyo Kayaku (Sci. Tech. Energ. Mater.) 50 (1989) 436– 441. [4] T. Tachibana, H. Kuribayashi, T. Torikai, Kogyo Kayaku (Sci. Tech. Energ. Mater.) 50 (1989) 442–446. [5] M.K. King, J. Propul. Power 7 (1991) 312–321. [6] M. Tanaka, K. Morisaki, Kogyo Kayaku (Sci. Tech. Energ. Mater.) 52 (1991) 270– 277. [7] K.H. Shin, C. Lee, Y. Seon, J.Y. Koo, in: Meeting Papers 43rd AIAA Aerospace Sci. Meeting and Exhibit, 2005, pp. 12293–12302. [8] M. Kohga, Mem. Natl. Def. Acad. Jpn. 48 (2010) 57–62. [9] M.H. Hites, M.Q. Brewster, J. Propul. Power 12 (1996) 616–619. [10] M. Kohga, Y. Hagihara, Kagaku Kogaku Ronbunshu 23 (1997) 163–169. [11] http://www.grc.nasa.gov/WWW/CEAWeb/. [12] M. Kohga, Propell. Explos. Pyrot. 33 (2008) 249–254. [13] D. Saravanakumar, N. Sengottuvelan, V. Narayanan, M. Kandaswamy, T. Varghese, J. Appl. Polym. Sci. 119 (2011) 2517–2524. [14] M. Kohga, Propell. Explos. Pyrot. 36 (2011) 57–64. [15] M. Kohga, Y. Hagihara, Sci. Tech. Energ. Mater. 64 (2003) 110–115. [16] H. Xu, X. Wang, L. Zhang, Powder Technol. 185 (2008) 176–180. [17] K. Fujimura, A. Miyake, J. Therm. Anal. Calorim. 99 (2010) 27–31. [18] D. Saravanakumar, N. Sengottuvelan, V. Narayanan, M. Kandaswamy, T.L. Varghese, J. Appl. Polym. Sci. 119 (2011) 2517–2524. [19] G. Singh, I.P.S. Kapoor, S. Dubey, Propell. Explos. Pyrot. 36 (2011) 367–372. [20] Y. Wang, X. Xia, J. Zhu, Y. Li, X. Wang, X. Hu, Combust. Sci. Technol. 183 (2011) 154–162. [21] M.W. Beackstead, R.L. Derr, C.F. Price, AIAA J. 8 (1970) 2200–2207. [22] M.W. Beckstead, in: Proc. 18th Symposium Combust., 1981, pp. 175–185. [23] M. Kohga, Y. Hagihara, Sci. Tech. Energ. Mater. 64 (2003) 68–74.