Investigations on Activated Charcoal, a Burn-Rate Enhancer in Composite Solid Propellant

Investigations on Activated Charcoal, a Burn-Rate Enhancer in Composite Solid Propellant Sumit Verma∗ and P. A. Ramakrishna† Indian Institute of Technology Madras, Chennai 600 036, India DOI: 10.2514/1.B34809 Through careful experimentation, this paper establishes that moisture trapped inside the pores of activated charcoal was the reason for the burn-rate enhancement of composite solid propellant observed in literature. This paper demonstrates that, even with the presence of moisture in activated charcoal, the propellants can not only be cured to be free from blow holes and voids but also have good mechanical properties. Experiments were also carried out to examine the possible mechanism for high burn rates and burn-rate pressure index observed in composite propellants when moisture was present in activated charcoal. Scanning-electron-microscope images show that moisture in activated charcoal reduces the binder melt flow. It is noticed that formation of the bubbly layer is more when moisture is present in the binder. This bubbly layer might be exploding near the surface of the propellant and taking away the binder melt layer. This could be the cause for the observed increase in the burn rate and burn-rate pressure index of aluminized composite propellants. I. Introduction ACTIVATED charcoal (AC) as the burn-rate modifier in acomposite propellant was reported earlier by us [1]. It was shown that the addition of a small percentage (0.4%) of AC increased the burn rate of aluminized composite propellant significantly. Activated charcoal along with iron oxide (IO) exhibited a synergetic effect and increased the burn rate by around five times (54 mm∕s at 70 bar) over the base composition (11 mm∕s at 70 bar). These high burn rates were also accompanied by a higher (0.65) burn-rate pressure index (n). In a more recent work [2], we reported the effect of AC along with copper chromite increasing the burn rate of aluminized composite propellant, but not as dramatically as shown earlier [1].We also found that AC is good at reducing the burn-rate pressure index of composite propellants and not burn rates per se as reported in [1]. Thiswas a very perplexing result, and an effort was made to find out the reason for such a large difference between the two sets of experiments. After having looked at all the possible issues that could have caused this, we realized that AC (the same as that used earlier in [1]), because of its porous nature, has an ability to absorb moisture. In our earlier experiments [1], during the preparation of the propellant, AC was exposed to ambient conditions, which were humid (Chennai, India). There was a possibility that moisture could have been absorbed by AC, but the exact percentage of moisture absorbed by AC is not known. In the later study [2], care was taken to ensure that all of the ingredients, including AC, were free from moisture. From the previous discussions, it is evident that moisture could have a certain role to play in increasing the burn rates of composite solid propellants. This study explores the role ofmoisture content inAC, as a parameter affecting the burn rate and burn-rate pressure index of the aluminized composite solid propellant. It also aims to uncover the mechanism of action of moisture content in AC on increasing the burn rate of a composite propellant. Moisture content is never preferred in the solid propellant because the presence of moisture is one of the causes for the blow holes in a cured propellant. Presence of blow holes lowers the density of the solid propellant and affects the burn rate of the solid propellant. Thus, great care was taken when curing propellants so as to make them free from defects like voids and blow holes when moisture was added to AC. Iqbal and Liang [3] report experiments conducted to examine the mechanical properties of composite propellants with different moisture fraction in ammonium oxalate. They report that moisture in the ammonium oxalate shows adverse affect on the tensile strength and Young’s modulus (E) of the propellant. II. Experiments The ingredients used here in this study for the experiments are the same as described earlier [4,5]. The coarse ammonium perchlorate (AP) with purity of >99.5% and size between 355 and 250 μm was procured from Tamil Chlorates. The purity of other ingredients used here is >99%. Average coarse and fine AP particle size used in this study is 327.5 (AP particles between 45- and 50-number mesh) and 54 (AP particles between 230- and 325-number mesh) μm, respectively. Fine AP was used immediately after the sieving to overcome the agglomeration problem, and the detailed description is given in earlier studies [4,5]. A. Preparation of Propellant Samples Various compositions (indicated in Table 1) were prepared during this study. All the mixtures had 18% aluminum (25 μm). A typical composition had 68%AP, 18% aluminum (Al), and 14%binder, as in a typical high-performance composite propellant. Binder consists of 7.1% of isophorone di-isocyanate (IPDI), 16.2% of di-octyl adipate (DOA), and 76.7%of hydroxyl terminated polybutadiene (HTPB) by weight. The burn-rate additives like AC and IO were added by suitably reducing only the AP content. Most of the composition used here had a AP coarse-to-fine ratio of 1∶1. All propellant formulations used here had a solid loading of around 86%. The reasoning behind the choice of various mixes will be presented in Sec. III. All of the solid ingredients including AC (when studying the case without moisture) were stored in an oven at 333 K for one day to remove the moisture from them. All of the ingredients were hand- mixed for about 15–20 min in a glass beaker before transferring the slurry to the mixer of 200 g capacity to ensure uniform mixing for about 1 h. The detailed description of the mixing and curing procedure is given in our earlier studies [4,5]. All of the propellants were kept under a pressure of around 2.5 bar at 303 K in the oven for curing. All propellant samples took 12–14 days for curing. After curing, x-ray photographs of the cured propellant were taken to check whether the propellants had voids, cracks, or blow holes, and propellants with these imperfections were discarded. The propellants Received 1 October 2012; revision received 17 May 2013; accepted for publication 20 May 2013; published online 22 July 2013. Copyright © 2013 by the American Institute of Aeronautics and Astronautics, Inc. All rights reserved. Copies of this paper may be made for personal or internal use, on condition that the copier pay the $10.00 per-copy fee to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923; include the code 1533-3876/13 and $10.00 in correspondence with the CCC. *Ph.D. Scholar, Department of Aerospace Engineering; ae07d011@smail .iitm.ac.in. †Associate Professor, Department of Aerospace Engineering; parama@ae .iitm.ac.in (Corresponding Author). AIAA Early Edition / 1 JOURNAL OF PROPULSION AND POWER D ow nl oa de d by U N IV O F SO U TH ER N C A LI FO RN IA o n Se pt em be r 2 , 2 01 3 | ht tp: //a rc. aia a.o rg | D OI : 1 0.2 514 /1. B3 480 9 free from voids, cracks, and blow holes were then cut into samples of 5 × 5 × 35 mm. A thin layer of inhibitor (silicon grease) was coated on the four side surfaces (5 × 35 mm) of the samples to ensure that the burning surface area (5 × 5 mm) of the sample remained same during the regression. In the present study, the propellants were to be prepared with moisture present in activated charcoal. For adding the moisture in AC, moisture-free AC (from the hot air oven) was spread thin over a stainless steel plate and kept in an aluminum container along with distilled water. The top surface of the container was then closed with an aluminum plate. The entire container was kept in the oven for one day at 333 K to allow for longer absorption time. After one day, the weight of this AC was measured. It was noticed that AC absorbed around 40% of moisture by weight. Propellants were prepared using this AC (containing 40% moisture). After having ensured that the cured propellant was free from voids and blow holes, they were burned to check whether different sections of the cured propellant gave similar burn rates when experiments are conducted under similar conditions. Upon obtaining satisfactory results, various propellant formulationswere preparedwith 40%ofmoisture inACas shown in Table 1. In addition to x-ray picture, the density of the propellants was measured. Density of the propellant is a very good check for determining whether the propellant has blow holes or void fractions. If the density of the propellant is much less than the theoretical density calculated based on the composition, it indicates probably the presence of voids and blow holes. Theoretical density calculated for a propellant with 68% AP, 14% binder, and 18% Al is around 1778 kg∕m3. Density of a propellant was measured by volume displacement of water. Density of basic propellant (mix 1) was around 1765 kg∕m3 and density of other samples is with in�1% of the basic composition as shown in Table 1. Density of the propellants is in good agreement with the theoretical density and the density reported in literature [6–8] of 1770 kg∕m3 for 86% solid loading. Despite the previous evidence indicating the absence of blowholes or voids in the cured propellant, there was still a lingering doubt that the addition of moisture in AC, which is then added to the propellant, could lead to reduced strength of the solid propellant. To examine this, the tensile strength, percentage elongation, and Young’s modulus were measured for the basic propellant and the propellant composition containing 1.5% AC with 40% moisture in it. The propellant containing 1.5% AC with 40% has the largest amount of moisture content (0.6% of the weight of the propellant) as compared to other compositions used in this study. Experiments to determine the strength of the propellant were carried out using INSTRONUTM (3365) of 0.5 ton capacity with the least count of 0.001 N. For the present study, the strain rate of 5 mm∕minwas used. Propellant samples of 7mm thicknesswere cut into the shape shown in Fig. 1. Two specimenswere tested tomeasure the tensile strength and other mechanical properties. Reported value shown in Table 2 are the average of the two readings. From Table 2, it is evident that the propellant with moisture in AC (mix 5) has properties similar to those of mix 1, which is free frommoisture. This indicates that the presence ofmoisture inAC does not lead to a drastic deterioration of mechanical properties. In fact, it improves both the tensile strength and Young’s modulus while decreasing the percentage elongation slightly. This improvement in mechanical properties could be due to the fact that the moisture introduced in AC remains trapped only inside the pores of AC. Considering both AP and aluminum particle size distribution and DOA content, the mechanical properties presented in Table 2 are within the values reported in literature [7–10] for 86% solid loading propellants. B. Experimental Setup for Burn-Rate Measurement A standard window bomb was used to measure the burn rates of the propellant. The internal volume of the bomb is 7.38 × 10−4 m3, which is very much greater than the sample volume of 0.875 × 10−6 m3. The bomb has two windows made of toughened glass to facilitate the recording of the burning process. The bomb also has a provision for measuring the burning rate using a timer circuit, and the detailed description is given in [4,5]. The present setup allows tomeasure two burn rates for the same sample at a particular pressure. The least count of the timer circuit used here is 10−6 s, and the maximum time it can measure is 9.999999 s. The uncertainty to measure the burn rate and burn-rate pressure index in all of the cases is within 2.5%. C. Experimental Procedure Nichromewire of 0.5mmdiameter was used to initiate the burning of the propellant sample. Commercially available nitrogen was used to build up the pressure in the window bomb by using a pressure regulator. A timer circuit was used to measure the burn time of the propellant sample for the distance between 9.8 and 10 mm. Two samples were burned at each pressure to ascertain the repeatability of the process. The reported burn rate was the average of all four readings. The dispersion in all cases from the average value was less than 2% of the average value. III. Results and Discussions Having confirmed that propellant samples even with moisture present inAC are free from voids and blow holes and have reasonable mechanical properties, tests were carried out to evaluate their burn- rate characteristics. Experiments were conducted at three different pressures of 10, 40, and 70 bar. First, mixes 1 to 5were prepared to examine the role of moisture in AC on the burn rate and n. Mix 1 was the base composition without any burn-ratemodifier ormoisture content.Mixes 2 and 3 had 0.4 and 1.5% dry AC (moisture-free), respectively. Mixes 4 and 5 were prepared with 0.4 and 1.5% AC, respectively, with 40% moisture of the weight of AC in them. Burn-rate variation with pressure of these mixes is shown in Fig. 2. It is seen in Fig. 2 that the burn rates are Table 1 Composition of the various mixes Mix Coarse/fine AP ratio AP, % Al, % AC, % Moisture in AC, % IO, % Binder, % Density, kg∕m3 1 1∶1 68 18 0 0 0 14 1765 2 1∶1 67.6 18 0.4 0 0 14 1762 3 1∶1 66.5 18 1.5 0 0 14 1760 4 1∶1 67.6 18 0.4 40 0 14 1760 5 1∶1 66.5 18 1.5 40 0 14 1762 6 1∶1 67.6 18 0 0 0.4 14 1765 7 1∶1 67.2 18 0.4 40 0.4 14 1760 8 1:2 67.6 18 0.4 40 0 14 1768 9 Fine AP only 68 18 0 0 0 14 1770 10 Fine AP only 67.6 18 0.4 40 0 14 1778 11 Fine AP only 66.5 18 1.5 0 0 14 1775 Fig. 1 Bone-shapedpropellant specimen (all dimensions inmillimeters). 2 AIAA Early Edition / VERMA AND RAMAKRISHNA D ow nl oa de d by U N IV O F SO U TH ER N C A LI FO RN IA o n Se pt em be r 2 , 2 01 3 | ht tp: //a rc. aia a.o rg | D OI : 1 0.2 514 /1. B3 480 9 higher formixes 4 and 5 as compared to the basic composition (mix 1) and the propellant with 0.4% (mix 2) and 1.5% (mix 3) dry AC. The burn-rate pressure index for mixes 4 and 5 are 0.39 and 0.34, respectively, which is higher compared to mixes 1–3 (indicated in Fig. 2). This result shows that a very small percentage (0.16 and 0.6% of theweight of the propellant) ofmoisture trapped inside the pores of AC is causing an increase in the burn rate and n of composite propellant. This is qualitatively similar to the results reported in [1]. The burn rate of propellant containing 0.4% AC with moisture is higher compared to the propellant containing 1.5% AC with moisture. As seen in Fig. 2, the burn rate and n are lower for mix 3 (1.5% dry AC) compared to mix 2 (0.4% dry AC). Lower burn rates for mix 5 compared to mix 4 could be because of the higher percentage of the AC present in mix 5. Further compositions in this study with mositure were prepared with 0.4% AC along with 40% moisture. The reason for the higher burn rate and n observed with moisture in AC will be discussed later in this study. It is seen in Fig. 2 that the presence of moisture in AC enhances the burn rate of the solid propellant. It has been reported in [1] that the burn rate increases signifcantly when AC is mixed along with IO in the composite solid propellant. Now, mixes 6 and 7 were prepared to examine if AC with moisture present in it when mixed along with IO causes an increase in burn rate of the composite propellant.Mix 6 had 0.4% iron oxide alone, and mix 7 contained 0.4% of iron oxide along with 0.4% AC with 40% moisture in AC (Table 1). The burn-rate results for thesemixes alongwithmix 1 are shown in Fig. 3. The burn rate and n are higher for mix 7 containing IO and AC (with moisture) as compared to mix 6 containing 0.4% IO alone. From the preceding discussion, it is clear that ACwith moisture present in it and IO act in tandem to increase the burn rate of the aluminized composite propellant, again qualitatively similar to those reported in [1]. AC withmoisturewhenmixed alongwith another burn-rate modifier like copper chromite shows a similar behavior as seen with IO. The burn rate for mix 7 (17.5 mm∕s at 70 bar) in Fig. 3 is lower compared to that reported in [1] (54 mm∕s at 70 bar) for the sameAP coarse-to-fine ratio and the content of burn-rate modifiers. One of reason for this difference in burn rates, as alreadymentioned in Sec. I, could be the difference in the moisture content in AC in the two experiments (exact amount of moisture in [1] is not known). In addition to the moisture content, the density of these two propellants is also very different. The density of the cured propellant in [1] was 1600 kg∕m3 and is lower compared to the density of the propellant used in this study (∼1765 kg∕m3). It has been noticed in an earlier study by us [5] that density of the propellant could be an important parameter in determining the burn rates. It was demonstrated in [5] that the reason for the lower density of composite propellant is a larger fraction of coarse AP as compared to fine AP that remains attached to the mixer surface and the blades, hence altering the coarse-to-fine AP ratio. A decrease in the density of the propellant from 1765 to 1640 kg∕m3 due to the increase in fine AP content compared to the coarse AP content leads to an increase in burn rate of more than 20% [5]. Here, mixes 8 to 10 were prepared to examine the effect of moisture inACwith reduction in averageAP particle size (Table 1) or change in the coarse-to-fine ratio of AP. Both mixes 8 and 10 have 0.4% activated charcoal with 40% in activated charcoal, but both have different average particle size AP.Mix 8 has a coarse-to-fine AP ratio of 1:2, and mix 10 has only fine AP particles (54 μm). The comparison of the burn-rate variation with pressure of these mixes along withmixes 1 and 4 is shown in Fig. 4. Reduction in average AP particle size in mix 8 further increases the burn rates and n (0.49) as compared to mix 4 (0.39). This shows that the effectiveness of moisture in AC increases with reduction in the average AP particle size. Burn rates for mix 10 were recorded only at 10 and 15 bar, and the burn-rate pressure index (1.17) for mix 10 is very high as compared to the burn-rate pressure index for the base composition with fine AP (mix 9). At pressures greater than 15 bar, the experiments were not repeatable (possiblly because of high n leading to detonation) and are not reported here. Similar results were observed in [1], where burn rates were not repeatable at 70 bar for the propellant with all fineAP (no coarse AP) and 0.4%AC.Mix 11with 1.5% moisture-free AC (dry) was chosen because, as seen earlier in Fig. 2, 1.5% moisture-free AC in composition with a coarse-to-fine AP ratio of 1∶1 has lowern. The burn-rate pressure index recorded for this mix even with all fine AP (54 microns) and no coarse AP content is 0.32, equal to the n obtained with mix 1. The reason for preparing mix 11 will be discussed a little later in this study. Table 2 Mechanical properties of the propellant Comp. Tensile strength, MPa Elongation, % E, MPa Mix 1 0.78 9.6 9.3 Mix 5 0.96 7.5 12.8 Note: Comp. stands for composition. Fig. 2 Burn-rate dependence on pressure for propellant samples with varying percentage of activated charcoal. Fig. 3 Burn-rate dependence on pressure for propellant samples with iron oxide and activated charcoal. Fig. 4 Burn-rate dependence on pressure for varying coarse-to-fine AP ratio with activated charcoal. AIAA Early Edition / VERMA AND RAMAKRISHNA 3 D ow nl oa de d by U N IV O F SO U TH ER N C A LI FO RN IA o n Se pt em be r 2 , 2 01 3 | ht tp: //a rc. aia a.o rg | D OI : 1 0.2 514 /1. B3 480 9 Moisture present in AC increases the burn rate and the burn-rate pressure index of the composite propellant. It is acting like a burn-rate enhancer in itself, and it alsoworks quitewell in tandemwith ironoxide. To identify the possible mechanism of action of activated charcoal in a composite solid propellant, several experiments were carried out. To examine if moisture is acting on the condensed phase, Thermo- gravimetric Analysis (TGA) and Differential Scanning Calorimeter (DSC) were carried out using TA instruments Q 500 TGA and Q 200 Modulated Differential Scanning Calorimeter, respectively. The samples used here were mixes 1, 2, and 4. The results (not shown here) were similar for all the threemixes. This indicated that moisture in AC is not acting on the condense phase of the composite solid propellant. From the previous discussion, it is apparent thatmoisturemight not be acting on the propellant through a chemical pathway but probably through a physical one. There are several studies [11–17] describing the importance of the binder melt flow in reducing the n of a composite solid propellant. Fong and Hamshere [11] and Jayaraman et al. [12] have observed different n when propellants were prepared with different curatives (dimeryl di-isocyanate, IPDI, and toluene di- isocyanate) with all of the other ingredients being similar. It has been observed [11,12] that larger binder melt flow leads to lower n. The results presented in Figs. 2–4 indicate an increase in the n, when moisture is present in AC. It was conjectured that moisture in AC might be acting on the binder melt flow, thereby increasing n. To explore this, scanning-electron-microscope (SEM) pictures of quenched samples had to be taken. Burning samples were quenched by rapid depressurization in a standard quench bomb setup (similar to the one used in [18]). First, three samples (mixes 1, 2, and 4) were quenched at 70 bar. Two samples of each of the compositionswere quenched for studying the surface of the quenched propellant. SEM pictures were taken at different locations andmagnifications of the quenched samples using a Quanta 200 manufactured by FEI. The binder melt layer was noticed in all of these samples. The images are not shown here because they all looked similar. It was reasoned that, because the separation in n between the various mixes (1, 2, and 4) was not large, there was no tangible outcome from the previous exercise. It was thought that, if there is a large difference in n between two propellants, it might be possible to make out the differences from the pictures of these samples. The propellant with only fine AP and 0.4% AC along with 40%moisture in AC (mix 10) was quenched at 70 bar. It was difficult to get a quenched sample of this propellant because it had very high n (1.17). The other candidate propellant was mix 11 containing only fine AP and 1.5% moisture-free AC. This had lower n of 0.32, as seen in Fig. 4.Both propellants (mixes 10 and 11) had the same AP particle size distribution (all fine), with only the percentage of AC and moisture content being different. This indicates the stark change inn that is possiblewith addition ofmoisture toAC.The SEM pictures for these two propellant mixes (10 and 11) are shown in Fig. 5. As seen in Fig. 5a, very little binder melt flow is noticed for mix 10, and the quenched surface clearly shows large number of particulatematter. In contrast to this, from the SEMpicture in Fig. 5b, it is seen that the surface has copious binder melt flow. The quenched surface appears smooth, and particulatematter is hardly visible. From these two SEMphotographs (Fig. 5), it is possible to explain the large difference in the n values reported earlier for mixes 10 and 11. For mix 10, because the quenched surface ismostly free frombindermelt, there is a possibility of premixing of fuel and oxidizer because AP particle sizes are small [18]. This leads to the higher n observed with mix 10. For mix 11, because of the presence of binder melt, the premixing is reduced even though AP particle sizes are small and diffusion seems to be the dominant phenomenon. This leads to the lower n observed with mix 11. This result is in conformity with the literature [11,12], wherein it has been observed that propellants with binder melt flow usually have a low burn-rate pressure index. The question that remains unanswered still is, “How does the presence of moisture in AC contribute to lower binder melt flow?”To understand this, we should relook at the role of moisture in the condensed phase. Earlier in this study, TGA and DSC analyses were carried out on mixes 1, 2, and 4 to determine if moisture is acting on the condensed phase of the composite propellant to enhance the burn rate. During this, the heating rate usedwas of the order of 10°C∕min . This heating rate is very low compared to that encountered in an actual propellant combustion situation (∼104°C∕s). This has been adequately discussed in literature [19,20]. Arisawa and Brill [19] pointed out that, at lower heating rates, the pyrolysis kinetics for polybutadiene is quite different from the pyrolysis kinetics at high heating rates. They also noticed that the activation energy required is lower when higher heating rates are used. Based on this, it is thought that analysis of the role of moisture on the condensed phase must be explored through a different route. Earlier, while curing propellants with moisture present in them, it was noticed that, if the propellants were to be cured at an elevated temperature of 333 K (for shorter curing time), there would be blow holes in the cured propellant sample. Hence, these propellants were cured at a lower temperature of 303 K (for longer curing time). Based on this, it was conjectured that reactions could be taking place in the binder during curing of the propellant, and moisture might be catalyzing these reactions and creating the blow holes. Reactions similar to this might be occurring in the cured propellant when it burns (subsurface temperature is being raised).With this thought that mainly reactions are taking place in the binder, two HTPB and IPDI compositions, with water present and without it, are prepared (as indicated in the Table 3). Both compositions were cured so as to obtain a sample of thickness of around 6 mm. Fig. 5 SEM pictures for the quenched sample for all fine AP propellants: a) with 0.4% AC along with 40% moisture in AC, and b) with 1.5% dry AC (no moisture). 4 AIAA Early Edition / VERMA AND RAMAKRISHNA D ow nl oa de d by U N IV O F SO U TH ER N C A LI FO RN IA o n Se pt em be r 2 , 2 01 3 | ht tp: //a rc. aia a.o rg | D OI : 1 0.2 514 /1. B3 480 9 To examine this, a muffle furnace (maximum temperature of 1200K) is used tomaintain the samples at a higher temperature. First, compositions 1 and 2 (thickness 6 mm) were kept in the furnace at 623 K. At 623 K, decomposition and formation of bubbles was noticed in both the samples. Thus, it was thought to keep the samples initially at lower temperature (373 K) and then to raise the temperature (in increments of 50 K for temperature range of 373– 473 K and 25 K for temperature range of 473–573 K) to examine the effect of moisture present in the binder. The mass loss for both samples during this exercise was negligible and was of the order of 0.05 g (initial mass of 8 g). A calculation was made based on thermal diffusivity (calculated based on thermal conductivity of 0.3 cal∕cm · s · K, specific heat capacity of 0.61 cal∕g · K, and density of 920 kg∕m3 for HTPB [21]), and it was noticed that about 3 min is required for a binder sample of 6 mm thick to reach a set temperature (573 K) in the muffle furnace (heating rate ∼100K∕min). Samples were taken out of the furnace after every Table 3 Various HTPB and IPDI compositions used Composition HTPB, % IPDI, % Water, % 1 90 10 0 2 89.5 10 0.5 a) At room temperature e) At room temperature b) At 473 K f) At 473 K Bubbles formation No sign of bubbles Sample free from trapped gases Sample free from trapped gases 25 25 c) At 498 K g) At 498 K d) At 573 K h) At 523 K No sign of bubbles Bubbly layer grows Sample gets fluffy Sample gets fluffy 40 25 Fig. 6 Pictures of a–d) binder (90%HTPB and 10% IPDI), and e–h) binder (90%HTPB and 10% IPDI) with 0.5%water, at different temperatures (all dimensions in millimeters). AIAA Early Edition / VERMA AND RAMAKRISHNA 5 D ow nl oa de d by U N IV O F SO U TH ER N C A LI FO RN IA o n Se pt em be r 2 , 2 01 3 | ht tp: //a rc. aia a.o rg | D OI : 1 0.2 514 /1. B3 480 9 10 min, and pictures were taken, which are shown in Fig. 6. After taking the pictures at particular temperatures, the sample was kept in the ambient temperature to cool down. Then, the sample was kept in the muffle furnace maintained at a higher temperature compared to the previous experiment (in steps of either 25 or 50 K, as explained earlier). As seen in Fig. 6d, formation of a bubbly layer is noticed at 573K for composition 1 (without water). In composition 2with 0.5% water in it, formation of a bubble is noticed at a lower temperature of 473 K, as shown in Fig. 6f. For composition 2, the bubbly layer size increases with an increase in the temperature from 473 to 498 K, and at 523 K, the sample gets very fluffy, as seen in Fig. 6h, similar to composition 1 at 573 K (Fig. 6d). Thus, it can be deduced that there are reactions in the binder that are taking place, and the presence of moisture seems to catalyze these reactions in the binder, enabling them to occur at a lower temperature. This bubbly layer (seen more when moisture is present) might be exploding at the burning surface, throwing away the binder melt layer, and in the extreme case (large n), it could lead to a situation as shown in Fig. 5a. In caseswhere there was not such large increase in burn-rate pressure index, evenwith the addition of moisture in AC (like in mixes 4, 5, and 8), the binder melt layer could have been reduced but not completely eliminated. The previous explanation could be a possible reason to explain the increase in burn rates observed at high pressure with a small amount of moisture present in activated charcoal. IV. Conclusions Burn-rate studies of aluminized composite solid propellant were carried out to check the role of moisture in activated charcoal on the burn rate and burn-rate pressure index. Experiments were carried out using awindow bomb for a pressure range of 10–70 bar, and burn rate for the propellant sample was measured by using a timer circuit. 1) Addition of moisture in activated charcoal did not lead to a large change in the mechanical properties of the propellant. 2) Moisture-free activated charcoal (AC) does not increase the burn rate of the composite solid propellant, while a small percentage of moisture (∼0.16% of propellant weight) trapped in the pores of activated charcoal is the reason for the enhancement of the burn rates and burn-rate pressure index of the composite solid propellant. 3) Activated charcoal with moisture along with a known burn-rate modifier like iron oxide acts in tandem and increases the burn rate and burn-rate pressure index of the composite solid propellant. 4) Scanning-electron-microscope pictures indicate that presence of moisture in AC reduces the binder melt flow significantly, thereby increasing the burn-rate pressure index of propellants. 5) The presence of moisture seems to catalyze reactions taking place in the binder, leading to the formation of a bubbly layer. 6) This bubbly layer is greater when moisture is present and might be exploding at the burning surface, throwing away the binder melt layer, and increasing the burn rate of the propellant. Acknowledgments We sincerely thank Subbu Krishna (Combustion, Gasification & PropulsionLaboratory, Indian Institute of Science, Bangalore),M.C. Dattan (Director, Satish Dhawan Space Center, Sriharikota Range), R. Velmurugan, S. R. Chakravarthy (Department of Aerospace Engineering, Indian Institute of Technology Madras), and V. Subramanya Sarma (Department of Metallurgical Materials Engineering, Indian Institute of Technology Madras) for helping us to complete this study. References [1] Verma, S., and Ramakrishna, P. A., “Activated Charcoal as a Burn Rate Enhancer in Composite Solid Propellant,” Combustion and Flame, Vol. 157, No. 6, 2010, pp. 1202–1210. doi:10.1016/j.combustflame.2009.11.017 [2] Verma, S., and Ramkrishana, P. A., “Development of Low Burn Rate Pressure Index Composite Solid Propellant,” 8th Asia Pacific Conference on Combustion, The Combustion Institute – Indian Section (CIIS), Hyderabad, India, 2010, pp. 619–626. [3] Iqbal, M. M., and Liang, W., “Modeling the Moisture Effects of solid Ingredients on Composite Propellant Properties,” Aerospace Science and Technology, Vol. 10, No. 8, 2006, pp. 695–699. doi:10.1016/j.ast.2006.07.003 [4] Verma, S., and Ramkrishana, P. A., “Role of Moisture in Activated Charcoal in Composite Solid Propellant,” 47th AIAA/ASME/SAE/ASEE Joint PropulsionConferenceandExhibit, AIAAPaper 2011-5558, 2011. [5] Verma, S., and Ramkrishana, P. A., “Effect ofMixer Size onDensity and Burn Rate of Composite Solid Propellant,” 47th AIAA/ASME/SAE/ASEE Joint PropulsionConferenceandExhibit, AIAAPaper 2011-5557, 2011. [6] Caveny, L. H., and Glick, R. L., “Influence of Embedded Metal Fibers on Solid-Propellant Burning Rate,” Journal of Spacecraft and Rockets, Vol. 4, No. 1, 1967, pp. 79–85. doi:10.2514/3.28813 [7] Krishnamurthy, V. N., “Some Issue in the Development and Production of HTPB Propellants,” 2nd International High Energy Materials Conference and Exhibit, Indian Inst. of Technology Madras, Chennai, India, 1998, pp. 18–27. [8] Jawalkar, S. N., Mehilal, R. K., Radhakrishnan, K. K., and Bhattacharya, B., “Studies on the Effect of Plasticizer and Addition of Toluene Diisocyanate at Different Temperatures in Composite Propellant Formulations,” Journal of Hazardous Materials, Vol. 164, Nos. 2–3, 2009, pp. 549–554. doi:10.1016/j.jhazmat.2008.08.064 [9] Patil, M. S., and Singh, H., “Ballistic and Mechanical Properties of HTPB Based Composite Propellant,” Journal of Hazardous Materials, Vol. 19, No. 3, 1988, pp. 271–278. doi:10.1016/0304-3894(88)80026-8 [10] Krishnan, S., Chakravarthy, S. R., and Athithan, S. K., Propellants and Explosives Technology, 1st ed., Allied Publ., New Delhi, India, 1998, p. 157, Chap. 5. [11] Fong, C. W., and Hamshere, B. L., “The Mechanism of Burning Rate Catalysis in Composite HTPB-AP Propellant Combustion,” Combus- tion and Flame, Vol. 65, No. 1, 1986, pp. 61–69. doi:10.1016/0010-2180(86)90073-8 [12] Jayaraman, K., Anand, K. V., Chakravarthy, S. R., and Sarathi, R., “Effect of Nano-Aluminum in Plateau-Burning and Catalyzed Composite Solid Propellant Combustion,” Combustion and Flame, Vol. 156, No. 8, 2011, pp. 1662–1673. doi:10.1016/j.combustflame.2009.03.014 [13] Boggs, T. L., and Zurn, D. E., “The Temperature Sensitivity of the Deflagration Rates of Pure and Doped Ammonium Perchlorate,” Combustion Science and Technology, Vol. 4, No. 1, 1972, pp. 279–292. doi:10.1080/00102207108952494 [14] Handley, J. C., and Strahle, W. C., “Behavior of Several Catalysts in the Combustion of Solid Propellant Sandwiches,” AIAA Journal, Vol. 13, No. 1, 1975, pp. 5–6. doi:10.2514/3.49620 [15] Price, E. W., Sambamurthi, J. K., Sigman, R. K., and Panyam, R. R., “Combustion of Ammonium Perchlorate-Polymer Sandwiches,” Combustion and Flame, Vol. 63, No. 3, 1986, pp. 381–413. doi:10.1016/0010-2180(86)90007-6 [16] Boggs, T. L., Derr, R. L., and Beckstead, M. W., “Surface Structure of Ammonium Perchlorate Composite Propellants,” AIAA Journal, Vol. 8, No. 2, 1970, pp. 370–372. doi:10.2514/3.5675 [17] Varney, A. M., and Strahle, W. C., “Experimental Combustion Studies of Two-Dimensional Ammonium Perchlorate-Binder Sandwiches,” Combustion Science and Technology, Vol. 4, No. 1, 1971, pp. 197–208. doi:10.1080/00102207108952486 [18] Williams, F. A., Barrere,M., andHuang, N. C.,Fundamental Aspects of Solid Propellant Rockets, Technivision Services, Slough, England, U.K., 1969, p. 532. [19] Arisawa, H., and Brill, T. B., “Flash Pyrolysis of Hyroxyl-Terminated Polybutadiene (HTPB), 2. Implications to theKinetics toCombustion of Organic Polymers,” Combustion and Flame, Vol. 106, Nos. 1–2, 1996, pp. 144–154. doi:10.1016/0010-2180(95)00254-5 [20] Esker, D. R., and Brewster, M. Q., “Laser Pyrolysis of Hydroxyl- Terminated Polybutadiene,” Journal of Propulsion and Power, Vol. 12, No. 2, 1996, pp. 296–301. doi:10.2514/3.24027 [21] Zanotti, C., Volpi, A., Bianchessi, M., and De Luca, L., Nonsteady Burning and Combustion Stability of Solid Propellants, edited by De Luca, L., Price, E. W., and Summerfield, M., AIAA,Washington, D.C., 1992, pp. 145–196, Chap. 5. S. Son Associate Editor 6 AIAA Early Edition / VERMA AND RAMAKRISHNA D ow nl oa de d by U N IV O F SO U TH ER N C A LI FO RN IA o n Se pt em be r 2 , 2 01 3 | ht tp: //a rc. aia a.o rg | D OI : 1 0.2 514 /1. B3 480 9