- Email: [email protected]
EDPM insulation
Journal of Hazardous Materials 229–230 (2012) 251–257
Contents lists available at SciVerse ScienceDirect
Journal of Hazardous Materials journal homepage: www.elsevier.com/locate/jhazmat
Migration kinetics and mechanisms of plasticizers, stabilizers at interfaces of NEPE propellant/HTPB liner/EDPM insulation Zhi-ping Huang ∗ , Hai-ying Nie, Yuan-yuan Zhang, Li-min Tan, Hua-li Yin, Xin-gang Ma The 42nd Institute of the Fourth Academy of CASC, Xiangfan 441003, China
h i g h l i g h t s
g r a p h i c a l
a b s t r a c t
Migration appeared in the interfaces of NEPE propellant/HTPB liner/EPDM insulation was studied. The migration occurred within 1 mm to the interfaces. The apparent migration activation energy (Ea) of mobile components is among 15 and 50 kJ/mol. The average diffusion coefficients were in the range of 10−19 m2 s−1 to 10−16 m2 s−1 . The migration kinetics is affected by the property of a mobile component and the based material.
a r t i c l e
i n f o
Article history: Received 8 February 2012 Received in revised form 4 May 2012 Accepted 30 May 2012 Available online 6 June 2012 Keywords: Migration Plasticizer Stabilizer NEPE propellant HTPB liner EDPM insulation
a b s t r a c t Migration appeared in the interfaces of nitrate ester plasticized polyether (NEPE) based propellant/hydroxyl-terminated polybutadiene (HTPB) based liner/ethylene propylene terpolymer (EPDM) based insulation was studied by aging at different temperatures. The migration components were extracted with solvent and determined by high performance liquid chromatography (HPLC). The migration occurred within 1 mm to the interfaces, and the apparent migration activation energy (Ea) of nitroglycerin (NG), 1,2,4-butanetriol trinitrate (BTTN) and a kind of aniline stabilizer AD in propellant, liner and insulation was calculated respectively on the basis of HPLC data. The Ea values were among 15 and 50 kJ/mol, which were much less than chemical energy, and almost the same as hydrogen bond energy. The average diffusion coefficients were in the range of 10−19 m2 s−1 to 10−16 m2 s−1 . It seemed the faster the migration rates, the smaller the apparent migration activation energy, the larger the diffusion coefficient and the less the amount of migration. It could be explained that the migration rate and energy were affected by the molecular volume of a mobile component and its diffusion property, and the amount of migration was resulted from the molecular polarity comparability of a mobile component to the based material. © 2012 Elsevier B.V. All rights reserved.
1. Introduction Solid rocket motor in its simplest form consists of a combustion chamber, which serves as a pressure vessel and houses the propellant, non-combustible material insulation to protect
∗ Corresponding author. Tel.: +86 710 3219203; fax: +86 710 3219111. E-mail address: [email protected] (Z.-p. Huang). 0304-3894/$ – see front matter © 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.jhazmat.2012.05.103
the motor case from the extreme heat generated during propellant combustion, and liner to bind propellant and insulation [1]. There are mainly three kinds of propellant: double base (DB) propellant composed of two basic components: nitrocellulose as a matrix and nitroglycerin as a plasticizer and blasting oil; hydroxylterminated polybutadiene (HTPB) based propellant which used a great deal of ammonium perchloride (AP) as oxidant and energetic component, and nitrate ester plasticized polyether (NEPE) propellant which uses polyether, such as polyethylene glycol (PEG) or
252
Z.-p. Huang et al. / Journal of Hazardous Materials 229–230 (2012) 251–257
ethylene oxide-tetrahydrofuran co-polyether, as binder and use mixed nitrate (usually using nitroglycerin (NG) and 1,2,4butanetriol trinitrate (BTTN)) as plasticizer and energetic component [2,3]. Though it is hard to study a propellant’s property by its complex composition, scientists never stop their research. Mechanical property is widely used to evaluate usage and storage property of a propellant or a propellant agglutinated with a liner and insulation by high temperature aging to accelerate their property changes, and kinetic method to study the mechanism. The kinetic data obtained, from the variation of both the storage modulus (E ) and loss factor or damping efficiency (tan ı), decomposition heat through different accelerated thermal aging programs can be analyzed according to Arrhenius methodology [4–7]. For example, the decomposition apparent activation energies for 81 kinds of propellants were obtained by thermogravimetry, derivative thermogravimetry (TG–DTG) and differential scanning calorimetry (DSC) through a multi-temperature artificial accelerated aging test to be among 100 and 160 kJ/mol [5]. HTPB/AP propellant modulus was used to obtain inputs for the Arrhenius relationship between temperature and degradation rate and gave an activation energy value of 71.0 kJ/mol [7]. Wu et al. found the primary failure mode of NEPE propellant/liner was changed with the mechanical quality and nitrogen element content in the interface by microcosmic mechanical performance study [8]. Other methods are used to study the property of a propellant, either. Nuclear magnetic resonance (NMR) imaging was used to study changes in the crosslinking density and molecular degradation during aging of propellant so as to explain the aging mechanisms like change in mechanical properties [9]. Using rigorous homogenization theory for composite materials, Xu et al. propose a general 3-D nonlinear macroscopic constitutive law that models microstructural, damage of particle cracking, dewetting along particle/polymer interfaces, void nucleation and growth, evolution upon straining through continuous void formation and growth [10]. By using X-ray photoelectron spectroscopy (XPS) analysis, Wu et al. found that interface failure is resulted from the decrease of active functional group in the interfacial areas of NEPE propellant/liner after aging [11]. Li et al. studied the phase separation of macromolecular binder and small molecular plasticizer in NEPE propellant using mesoscale simulation method of dissipative particle dynamics to find phase separation takes place later with temperature decreasing, and the length of polymer chains is larger, the occurrence of phase separation is earlier [12]. Migration and leakage of some mobile components in solid propellant is another property scientists focus on. Migration produces an inhomogeneous composite on region at which migration takes place, which can lead to premature detonation, changes in ballistic characteristics, and so on. Yin et al. found that the change of bond strength in HTPB/toluene diisocyanate (TDI) liner affected the ingredients and contents of migration in NEPE propellant after aging by high performance liquid chromatography (HPLC) and inductively coupled plasma atomic emission spectroscopy (ICP–AES), some broke the chemical bonds, some decreased the mechanical property in the storage of HTPB/TDI liner [13]. Ünver et al. found migration of a burning-rate catalyst acetyl ferrocene (AcF) in HTPB based elastomer in the presence of a plasticizer (dioctyl adipate) depending on time and temperature, and influencing by crosslink density of the barrier layer [14]. Suceska et al. studied a very early stage of evaporation of nitroglycerin from a DB propellant by applying isothermal thermogravimetry experiments. The activation energy of nitroglycerin evaporation was calculated to be 81.9 kJ/mol [15]. Gottlieb and Bar studied plasticizer migration across bonded HTPB/AP propellant interfaces during cure and evaluated the diffusion coefficient to show that the curing period is significant to the migration phenomenon and direct influence on tensile strength for short aging
periods [16]. Libardi et al. determined the dioctyl azelate (DOZ) plasticizer diffusion coefficient (D) for samples containing the interfaces of rubber, liner and solid composite propellant based on HTPB to be 0.70, 2.29, 6.27 × 10−11 m2 s−1 in liner and propellant at different layer resulted from Fick’s second law [17]. Grythe and Hansen determined diffusion coefficients for isocyanates and plasticizers in ethylene propylene terpolymer (EPDM), HTPB, hydroxyl-terminated polyethylene glycol (HTPE) or glycidyl azide polymer (GAP) to be vary between 10−11 and 10−17 m2 s−1 by the weight of uptake method in polymer materials [18]. Our previous research found that mobile components, that is, plasticizers NG and BTTN, a kind of aniline stabilizer AD as well as the burning catalyzer of octyl dicyclopentadienyl iron (ODCI) in the propellant, and plasticizer dioctyl sebacate (DOS) in the liner consumption, migration, and decomposition appeared in the interfaces of NEPE propellant/HTPB liner/EPDM insulation during storage period [19]. We also quantified the main migrating components to find that NG, BTTN and AD in the propellant could migrate to the liner and insulation, and DOS in the liner migrated only to the insulation, not to the propellant [20]. Migration can reduce the allowable physical properties, weakening the grain, so that it will crack at stress-concentrated points and cause unfavorable increases or fluctuations in the burning rate and performance of the rocket motor [21]. It is, therefore, important to be able to predict the behavior of low-molecular-weight mobile additives and to control the leakage of them from the propellant. In this paper, we try to estimate the migration kinetic parameters of some components of a NEPE propellant migrated to liner and insulation through Arrhenius relationship to understand their behavior of migration. 2. Experimental 2.1. Materials HPLC-grade methanol was purchased from Sigma Chemical (USA). Analytic reagent methenyl trichloride and acetonitrile were purchased from Shanghai Shiyi Chemicals Reagent Co., Ltd. (China). Nitroglycerin (NG), 1,2,4-butanetriol trinitrate (BTTN) and a kind of aniline stabilizer AD were produced by the 42nd Institute of the Fourth Academy of CASC (China). Polyethylene glycol (PEG), hydroxyl-terminated polybutadiene (HTPB) and hexamethylene biuret polyisocyanate (N100) were purchased from Liming Research Institute of Chemical Industry (China). Ethylene propylene terpolymer (EPDM) were purchased from Du Pont China Holding Co., Ltd. (China). 25 mm cylindrical specimens were prepared with 20 mm NEPE based propellant in the middle, which was made from PEG/N100 binder system containing NG and BTTN plasticizers, capped with metal covers which were pre-covered with a 1.5–2.5 mm EPDM based insulation and a 0.5–1.0 mm HTPB based liner at both ends. The specimens were cured for 7 days at 50 ◦ C and placed inside aluminum bags and aged at temperatures 40 ◦ C and 50 ◦ C for 20 months, 60 ◦ C for 10 months, and 70 ◦ C for 4 and a half months and analyzed for their tensile properties and contents of NG, BTTN and AD at designated intervals. All chemical reagents were of chemical grade unless mentioned specially and commercially available and they were used without further treatment. 2.2. Mechanical analysis The bond strength of the specimen was measured on an electronic universal material testing machine (Model 5567, Instron, USA) through maximal tensile strength ( m ) at a rate of 20 mm/min at 25 ◦ C until failure. The tensile strength is calculated as m = F/S where F is the force, S is the transect area of sample at direction
Z.-p. Huang et al. / Journal of Hazardous Materials 229–230 (2012) 251–257
of the force. Each value is an average of five measurements with a relative standard deviation of ±3%.
For the determination of the plasticizers and stablizer in propellant, liner or insulation, the propellant, liner and insulation of a specimen was separated and cutted into 1 mm × 1 mm × 10 mm pieces with a knife manually. 25 mL of methanol was added to accurately weighed propellant pieces of approximately 300–400 mg, 50% methenyl trichloride/50% acetonitrile (v/v) to liner or insulator pieces and extracted for 8 h at room temperature. The extraction was removed to a 50 mL volumetric flask, and the residue was washed with methanol for several times and the washed methanol was also removed to the volumetric flask, adding methanol to 50 mL. The prepared solution samples were transferred to sampling vials for analysis via a high performance liquid chromatograph (HPLC, Alliance 2695, Waters, USA) under following conditions: isocratic mobile phase of methanol at 1.0 mL/min, Shodex DE-613 column (5 m, 4.6 mm × 300 mm), diode array detector at 230 nm, 10 L sample injections. Under these conditions, NG, BTTN and AD were eluted as three separate peaks [20]. A component concentration in the sample is calculated from: cx =
1000cxd Vx mx Mx
Table 1 Migration ratios of the components migrated in different interface after aging for one month at different temperatures. T (◦ C)
2.3. Plasticizers and stabilizer analysis
(1)
where cx is a component concentration in tested sample, mol/kg; 1000 is the ratio of kg/g; cxd is the component concentration determined by HPLC external standard method, g/mL; Vx is total solution volume for a sample solution, mL; mx is the mass of a sample, g; Mx is relative molecular mass of a component, g/mol. Each value is an average of three measurements with a relative standard deviation of ±1%. The relative migration ratio of a component migrated from propellant to liner was calculated as cL /cP , and that of which migrated from liner to insulation was calculated as cI /cL Where cL , cP and cI was concentration of a component in liner, propellant and insulation at the same condition, respectively. 3. Results and discussion If the mobile components are not chemically bonded to a substrate material, they will migrate in the direction of lower concentrations [15]. Fig. 1 illustrates the components concentration in propellant, liner and insulation of a bonded specimen without aging at different distance to liner interfaces. It showed that
Fig. 1. The component concentration in propellant, liner and insulation at different distance to liner interfaces.
253
40 50 60 70
cL /cP (%)
cI /cL (%)
NG
BTTN
AD
NG
BTTN
AD
24.7 21.9 20.4 11.3
22.7 19.0 18.9 19.7
281.8 228.6 292.9 272.7
8.7 9.0 13.1 5.6
6.5 7.6 11.9 10.0
35.5 36.7 31.4 32.1
concentrations of all migrated components were almost constant in propellant and insulation at distance to liner interface more than 1 mm, and those within 1 mm to liner interface were different from propellant and insulation. It revealed that migration appeared within 1 mm of the interfaces. The concentrations of migrated NG and BTTN in liner were less than those in propellant and more than those in insulation, but AD in liner was more than that in propellant and insulation. NG and BTTN were trace in insulation and much less than those in propellant, and AD was almost the same as in propellant although it was trace, too. It indicated that migration occurred during curing of the specimen and the migration of AD was different from that of NG and BTTN. To determine the difference of migration quantificationally, Table 1 listed the data of relative migration ratio of NG, BTTN and AD migrated from propellant to liner, and that of which migrated from liner to insulation after aging for one month at different temperatures. All the migration ratios of NG and BTTN were less than 25%, and migration ratios of AD were more than 200% from propellant to liner, and those from liner to insulation were less than those from propellant to liner. The migration ratios of AD were more than those of NG and BTTN at the same condition, and all the migration ratios decreased with temperature increasing from propellant to liner except AD at 60 ◦ C, and they were reverse from liner to insulation except those aging at 70 ◦ C and AD at 60 ◦ C. The results suggested that migration was changed with temperature and decomposition occurred at high temperature such as 70 ◦ C because NG decreased obviously at 70 ◦ C. It revealed that NG, BTTN migrated differently from AD, and liner could enrich AD and prevent NG and BTTN migrating from propellant to insulation in a way, just as the work we did previous [20]. Fig. 2 illustrates the changing tendency of contents of NG, BTTN and AD in propellant, liner, and insulation, and the bond strength of the propellant/liner/insulation specimen with time during aging at 70 ◦ C, respectively. It showed that all components present in propellant kept almost constant first, then decreased with time increasing at aging time after 3 × 106 s, that is to say, there were significant loss of plasticizer and stabilizer after 3 × 106 s of aging, while they increased first and decreased afterwards in liner and insulation at similar time. The maximal tensile strength ( m ) of the specimen remained almost constant at first period and decreased fast at the very time 5 × 106 s when components decreased in liner and insulation. The migration phenomenon occurs mainly due to the differential concentration of plasticizers and stabilizer between the additive part and the nonadditive part. All these components were added in propellant originally, and none was in liner, neither was in insulation, so they migrated from propellant to liner and insulation. It is well known that the maximal tensile strength is a kind of characterization for the linkage of a polymer network or a splice specimen, broken in polymer network or splice specimen linkage would result in sharp decrease of its maximal tensile strength. The relation between the maximal tensile strength and contents of components in liner and insulation suggested that migration appeared during the first period and decomposition occurred during the second period at 70 ◦ C.
254
Z.-p. Huang et al. / Journal of Hazardous Materials 229–230 (2012) 251–257
Fig. 2. The component concentration in propellant, liner and insulation, respectively during aging at 70 ◦ C and corresponding mechanic property of the specimen.
Fig. 3 illustrated chromatograms of propellant extractor aging at before and after 3 × 106 s. There are mainly four peaks before 3 × 106 s, and a new peak appears after 3 × 106 s. It proved the suggestion that decomposition occurred during the second period
70 ◦ C
Fig. 3. Chromatograms of propellant extractor before and after aging time 3 × 106 s.
at 70 ◦ C mentioned above. So migration was not the fatal factor to result in the invalidation of the specimen, while the chemical changes are the reason of their decrease in tensile strength. Kinetic parameters of migration could predict the tendency of migration of a mobile component, from which the mechanism of migration might be understood, so as some action might be taken to forbid the migration. The changing tendency of the specimen in contents of plasticizers and stabilizer aged at three other temperatures was similar as that aged at 70 ◦ C. The only difference was the migration rate. The higher the temperature, the faster the rate of migration is. From relation of contents of the components in propellant, liner and insulation with time at 70 ◦ C in Fig. 2 and those obtained at 40 ◦ C, 50 ◦ C and 60 ◦ C (not shown), the migration concentration versus time relations were obtained as c = k1 t + k2 t2 + b, and the constants k1 , k2 and b are listed in Table 2. It can be seen that k2 was much smaller than k1 , which indicated that the mechanism of migration of these compounds is zero order. Suceska et al. found that at a very early stage of evaporation of nitroglycerin from a double base rocket propellant, the evaporation can be described by the zero-order reaction model [15]. Though migration is not the same as evaporation, they are comparable because they are both physical changes in complex materials. The absolute value of k1 for NG is about 5 times that for BTTN and AD in the same condition. It corresponded to their molecular structure because NG and BTTN are homolog, and AD is aromatic compound, the molecular volume of NG is smaller than that of BTTN and AD, so the migrating rate of NG was faster than that of BTTN and AD. From above mentioned, the mechanism of migration of these compounds is zero order, so we calculated the apparent migration activation energy for the migration of NG, BTTN and AD by Arrhenius relation written as follows:
log10 [k1 ] = A −
Ea RT
(2)
4.43 4.76 4.75 3.92 – – – –5.41 0.40 0.56 1.22 2.31 0.015 0.014 0.013 0.015
−0.61 −12.3 2.86 5.43 0.82 −6.49 − −46.6 4.80 40.9 19.1 46.8 0.055 – 0.054 0.091
5.9 −0.087 5.6 −0.81 −0.39 −6.06 − −9.01 9.27 35.5 18.1 10.1 0.077 − 0.070 0.032
k1 (10−10 mol/(kg s))
255
–0.39 – –1.02 –15.0
– – – –
21.6 – 40.4 113
31.4 – 41.3 64.0
−0.64 −2.47 −4.89 −13.6
0.39 0.37 0.39 0.32
0.32 0.32 0.30 0.32
0.029 0.027 0.025 0.027
mol/(kg s))
−0.27 – −1.32 −14.0
k1 (10 b (mol/kg)
k2 (10
−16
mol/(kg s))
Liner
−10
−2.33 −5.08 −8.49 −27.3
−2.07 −5.54 −11.1 −29.2
40 50 60 70
40 50 60 70
BTTN
AD
–11.6 –24.1 –31.6 –38.7 40 50 60 70
k1 (10
NG
−10
T (◦ C)
Propellant
Fig. 4. The plots of lg [k1 ] against −1/RT of three components in propellant, liner and insulation.
Component
Table 2 The migration rate constants k1 , k2 and b of three components in propellant, liner and insulation at different temperatures.
2
mol/(kg s ))
b (mol/kg)
Insulation
k2 (10−17 mol/(kg s2 ))
b (10−3 mol/kg)
Z.-p. Huang et al. / Journal of Hazardous Materials 229–230 (2012) 251–257
where k1 is the migration rate constant at a given temperature, A is an empirically derived constant, Ea is the apparent activation energy for migration, R is the ideal gas constant, which is 8.3145 J mol−1 K−1 , T is the temperature of migration in Kelvin. The plots of log10 [k1 ] against −1/RT are illustrated in Fig. 4. It was found that the plots were good straight lines and the correlation coefficients were more than 95%. The apparent migration activation energy of three mobile components in propellant, liner and insulation were obtained by slopes of the lines and they are listed in Table 3. The Ea values were among 15 and 50 kJ/mol, which were much less than chemical energy, and almost the same as hydrogen bond energy, which was correspond to migration. The Ea of NG and BTTN were decreased in the direction of propellant to liner and insulation, and their difference was small. The Ea of AD in
256
Z.-p. Huang et al. / Journal of Hazardous Materials 229–230 (2012) 251–257
Table 3 The apparent migration activation energy of three materials in propellant, liner and insulation. Material
Ea (kJ/mol)
Propellant Liner Insulation
NG
BTTN
AD
20.9 19.9 17.4
30.5 29.7 28.9
36.4 46.1 20.5
liner was the largest, while that in insulation was the smallest, and their difference was higher than that of NG and BTTN. It suggested that the less the Ea, the faster and easier the migration, which was correspond to the migration rate constants. Diffusion coefficients of the migration components in propellant, liner and insulation may give important information for their migration [16–18]. The diffusion of a single substance from the interface into the isolation may be considered as ordinary diffusion into a semiinfinite slab and is described by Fick’s second law of diffusion. The equation assumes that there is only one diffusing component and that the diffusion coefficient is constant. The flux at the interface is given by Fick’s first law. The propellant is thought to be infinite for a thin layer of liner, so is for insulation. All components are assumed to be independent to migrate from propellant to liner and from liner to insulation. The migrating surface area is only on one side. So we can express the diffusion as following [18]:
D=
m
m0
2 i di 2
−1
(3)
t
Table 4 The average diffusion coefficients of migrating components in propellant, liner and insulation at 40 ◦ C.
Propellant Liner Insulation
has phenyl group was similar to that of liner and insulation which has long alkyl groups, and was different to that of propellant which has ether and urethane links. Further detail study about polarity of the materials will be done in our next work. It concluded that migration rate and energy were affected by the molecular volume of a mobile component and its diffusion property, and the amount of migration was resulted from the molecular polarity comparability of a mobile component to the based material. A schematic depiction of the specimen and the migration of NG, BTTN and AD is shown in Fig. 5. 4. Conclusions
where D is an average diffusion coefficient, m is the mass of the isolation sheet including the absorbed substance, m0 is the initial mass of the sheet, is the density of a diffused substance, i is the initial density of the sheet, di is the sheet thickness, t is the time when m is determined. Table 4 lists the average diffusion coefficient of the three components in propellant, liner and insulation. The average diffusion coefficients were in the range of 10−19 m2 s−1 to10−16 m2 s−1 , which was accordant to those determined by Grythe and Hansen [18]. The average diffusion coefficient of NG, BTTN were much higher than that of AD, and the value change tendency of all the components in the direction of propellant to liner and insulation was almost the same as Ea. The larger the diffusion coefficient is, the less the activation energy. It could be concluded that the migration activation energy was resulted from the diffusion property. From Table 1, the migration ratio of AD was much larger than that of NG and BTTN, which indicated that the final concentration was not resulted from migration rate, migration activation energy Ea or diffusion coefficient D. Why the observed behavior is the opposite of what is expected from the values of the diffusion coefficients and activation energies? It is because the migration activation energy and diffusion coefficient are kinetic parameters, which are affected by properties of the materials, temperature and time, and the migration ratio is a thermodynamic parameter, which is affected by properties of the materials and temperature, ignoring the time. Thus, it could be explained that the polarity of AD which
Material
Fig. 5. Schematic depiction of the specimen and the migration of NG, BTTN and AD.
¯ (10−18 m2 s−1 ) D NG
BTTN
AD
650 34.9 2.67
632 33.8 2.02
0.19 0.80 0.22
Migration appeared in the interfaces of NEPE propellant/HTPB liner/EPDM insulation, the smaller the apparent migration activation energy, the faster the migration rates, the larger the diffusion coefficient, and the less the amount of migration. The migration rate was affected by the molecular volume, and the amount of migration was resulted from the molecular polarity comparability. Acknowledgment We gratefully acknowledge financial support from the major Grant of the National Basic Research Program of China (973 Program 613142). References [1] M.S. Sureshkumar, C.M. Bhuvaneswari, S.D. Kakade, M. Gupta, Studies on the properties of EPDM-CSE blend containing HTPB for case-bonded solid rocket motor insulation, Polym. Adv. Technol. 19 (2008) 144–150. [2] A.M. Pang, J. Zheng, Prospect of the research and development of high energy solid propellant technology, J. Solid Rocket Techn. 27 (2004) 289–293. [3] W. Zhang, X. Fan, H. Wei, J. Li, Application of nitramines coated with nitrocellulose in minimum signature isocyanate–cured propellants, Propell. Explos. Pyrotech. 33 (2008) 279–285. [4] J.L. de la Fuente, O. Rodríguez, Dynamic mechanical study on the thermal aging of a hydroxyl-terminated polybutadiene-based energetic composite, J. Appl. Polym. Sci. 87 (2003) 2397–2405. [5] F. Zhao, S. Heng, R. Hu, H. Gao, F. Han, A study of kinetic behaviours of the effective centralite/stabilizer consumption reaction of propellants using a multi-temperature artificial accelerated ageing test, J. Hazard. Mater. 145 (2007) 45–50. [6] J.H. Yi, F.Q. Zhao, W.L. Hong, S.Y. Xu, R.Z. Hu, Z.Q. Chen, L.Y. Zhang, Effects of Bi-NTO complex on thermal behaviors, nonisothermal reaction kinetics and burning rates of NG/TEGDN/NC propellant, J. Hazard. Mater. 176 (2010) 257–261. [7] M.D. Judge, An investigation of composite propellant accelerated ageing mechanisms and kinetics, Propell. Explos. Pyrotech. 28 (2003) 114–119. [8] F.J. Wu, S. Peng, X.H. Chi, Study on microcosmic mechanics performance/structure of NPPE propellant/liner bonded interface, J. Solid Rocket Techn. 33 (1) (2010) 81–85. [9] J. Götz, Characterization of the structure in highly filled composite materials by means of MRI, Propell. Explos. Pyrotech. 27 (2002) 179–184. [10] F. Xu, N. Aravas, P. Sofronis, Constitutive modeling of solid propellant materials with evolving microstructural damage, J. Mech. Phys. Solid. 56 (2008) 2050–2073.
Z.-p. Huang et al. / Journal of Hazardous Materials 229–230 (2012) 251–257 [11] F.J. Wu, S. Peng, X.H. Chi, F.C. Zhong, XPS characterization of NEPE propellant/line bondline, J. Solid Rocket Techn. 32 (2009) 192–196. [12] S. Li, Y. Liu, X. Tuo, X. Wang, Mesoscale dynamic simulation on phase separation between plasticizer and binder in NEPE propellants, Polymer 49 (2008) 2775–2780. [13] H.L. Yin, Y. Wang, D.F. Li, Ingredient migration and their effect on NEPE propellant bonding system, J. Solid Rocket Techn. 32 (2009) 527–530. [14] A. Ünver, N. Dilsiz, M. Volkan, G. Akovalı, Investigation of acetyl ferrocene migration from hydroxyl-terminated polybutadiene based elastomers by means of ultraviolet–visible and atomic absorption spectroscopic techniques, J. Appl. Polym. Sci. 96 (2005) 1654–1661. [15] M. Suceska, S.M. Musanic, I.F. Houra, Kinetics and enthalpy of nitroglycerin evaporation from double base propellants by isothermal thermogravimetry, Thermochim. Acta 510 (2010) 9–16. [16] L. Gottlieb, S. Bar, Migration of plasticizer between bonded propellant interfaces, Propell. Explos. Pyrotech. 28 (2003) 12–17.
257
[17] J. Libardi, S.P. Ravagnan, A.M.F. Morais, A.R. Cardoso, Diffusion of plasticizer in a solid propellant based on hydroxyl-herminated polybutadiene, Polímeros 20 (2010) 241–245. [18] K.F. Grythe, F.K. Hansen, Diffusion, Rates and the role of diffusion in solid propellant rocket motor adhesion, J. Appl. Polym. Sci. 103 (2007) 1529–1538. [19] Z.P. Huang, J.J. Liu, S.L. Xu, J. Bai, X.G. Ma, Qualitative analysis of migration components in the interface of NEPE propellant, J. Solid Rocket Techn. 33 (2010) 41–44. [20] Z.P. Huang, L.M. Tan, Q.W. Cao, X.G. Ma, Quantitative analysis of migrating components in the interface of NEPE propellant/liner/insulation, Chin. J. Energ. Mater. 18 (2010) 330–334. [21] G.P. Sutton, Rocket Propulsion Elements: An Introduction to the Engineering of Rockets, 6th ed., Wiley, New York, 1992.
Contents lists available at SciVerse ScienceDirect
Journal of Hazardous Materials journal homepage: www.elsevier.com/locate/jhazmat
Migration kinetics and mechanisms of plasticizers, stabilizers at interfaces of NEPE propellant/HTPB liner/EDPM insulation Zhi-ping Huang ∗ , Hai-ying Nie, Yuan-yuan Zhang, Li-min Tan, Hua-li Yin, Xin-gang Ma The 42nd Institute of the Fourth Academy of CASC, Xiangfan 441003, China
h i g h l i g h t s
g r a p h i c a l
a b s t r a c t
Migration appeared in the interfaces of NEPE propellant/HTPB liner/EPDM insulation was studied. The migration occurred within 1 mm to the interfaces. The apparent migration activation energy (Ea) of mobile components is among 15 and 50 kJ/mol. The average diffusion coefficients were in the range of 10−19 m2 s−1 to 10−16 m2 s−1 . The migration kinetics is affected by the property of a mobile component and the based material.
a r t i c l e
i n f o
Article history: Received 8 February 2012 Received in revised form 4 May 2012 Accepted 30 May 2012 Available online 6 June 2012 Keywords: Migration Plasticizer Stabilizer NEPE propellant HTPB liner EDPM insulation
a b s t r a c t Migration appeared in the interfaces of nitrate ester plasticized polyether (NEPE) based propellant/hydroxyl-terminated polybutadiene (HTPB) based liner/ethylene propylene terpolymer (EPDM) based insulation was studied by aging at different temperatures. The migration components were extracted with solvent and determined by high performance liquid chromatography (HPLC). The migration occurred within 1 mm to the interfaces, and the apparent migration activation energy (Ea) of nitroglycerin (NG), 1,2,4-butanetriol trinitrate (BTTN) and a kind of aniline stabilizer AD in propellant, liner and insulation was calculated respectively on the basis of HPLC data. The Ea values were among 15 and 50 kJ/mol, which were much less than chemical energy, and almost the same as hydrogen bond energy. The average diffusion coefficients were in the range of 10−19 m2 s−1 to 10−16 m2 s−1 . It seemed the faster the migration rates, the smaller the apparent migration activation energy, the larger the diffusion coefficient and the less the amount of migration. It could be explained that the migration rate and energy were affected by the molecular volume of a mobile component and its diffusion property, and the amount of migration was resulted from the molecular polarity comparability of a mobile component to the based material. © 2012 Elsevier B.V. All rights reserved.
1. Introduction Solid rocket motor in its simplest form consists of a combustion chamber, which serves as a pressure vessel and houses the propellant, non-combustible material insulation to protect
∗ Corresponding author. Tel.: +86 710 3219203; fax: +86 710 3219111. E-mail address: [email protected] (Z.-p. Huang). 0304-3894/$ – see front matter © 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.jhazmat.2012.05.103
the motor case from the extreme heat generated during propellant combustion, and liner to bind propellant and insulation [1]. There are mainly three kinds of propellant: double base (DB) propellant composed of two basic components: nitrocellulose as a matrix and nitroglycerin as a plasticizer and blasting oil; hydroxylterminated polybutadiene (HTPB) based propellant which used a great deal of ammonium perchloride (AP) as oxidant and energetic component, and nitrate ester plasticized polyether (NEPE) propellant which uses polyether, such as polyethylene glycol (PEG) or
252
Z.-p. Huang et al. / Journal of Hazardous Materials 229–230 (2012) 251–257
ethylene oxide-tetrahydrofuran co-polyether, as binder and use mixed nitrate (usually using nitroglycerin (NG) and 1,2,4butanetriol trinitrate (BTTN)) as plasticizer and energetic component [2,3]. Though it is hard to study a propellant’s property by its complex composition, scientists never stop their research. Mechanical property is widely used to evaluate usage and storage property of a propellant or a propellant agglutinated with a liner and insulation by high temperature aging to accelerate their property changes, and kinetic method to study the mechanism. The kinetic data obtained, from the variation of both the storage modulus (E ) and loss factor or damping efficiency (tan ı), decomposition heat through different accelerated thermal aging programs can be analyzed according to Arrhenius methodology [4–7]. For example, the decomposition apparent activation energies for 81 kinds of propellants were obtained by thermogravimetry, derivative thermogravimetry (TG–DTG) and differential scanning calorimetry (DSC) through a multi-temperature artificial accelerated aging test to be among 100 and 160 kJ/mol [5]. HTPB/AP propellant modulus was used to obtain inputs for the Arrhenius relationship between temperature and degradation rate and gave an activation energy value of 71.0 kJ/mol [7]. Wu et al. found the primary failure mode of NEPE propellant/liner was changed with the mechanical quality and nitrogen element content in the interface by microcosmic mechanical performance study [8]. Other methods are used to study the property of a propellant, either. Nuclear magnetic resonance (NMR) imaging was used to study changes in the crosslinking density and molecular degradation during aging of propellant so as to explain the aging mechanisms like change in mechanical properties [9]. Using rigorous homogenization theory for composite materials, Xu et al. propose a general 3-D nonlinear macroscopic constitutive law that models microstructural, damage of particle cracking, dewetting along particle/polymer interfaces, void nucleation and growth, evolution upon straining through continuous void formation and growth [10]. By using X-ray photoelectron spectroscopy (XPS) analysis, Wu et al. found that interface failure is resulted from the decrease of active functional group in the interfacial areas of NEPE propellant/liner after aging [11]. Li et al. studied the phase separation of macromolecular binder and small molecular plasticizer in NEPE propellant using mesoscale simulation method of dissipative particle dynamics to find phase separation takes place later with temperature decreasing, and the length of polymer chains is larger, the occurrence of phase separation is earlier [12]. Migration and leakage of some mobile components in solid propellant is another property scientists focus on. Migration produces an inhomogeneous composite on region at which migration takes place, which can lead to premature detonation, changes in ballistic characteristics, and so on. Yin et al. found that the change of bond strength in HTPB/toluene diisocyanate (TDI) liner affected the ingredients and contents of migration in NEPE propellant after aging by high performance liquid chromatography (HPLC) and inductively coupled plasma atomic emission spectroscopy (ICP–AES), some broke the chemical bonds, some decreased the mechanical property in the storage of HTPB/TDI liner [13]. Ünver et al. found migration of a burning-rate catalyst acetyl ferrocene (AcF) in HTPB based elastomer in the presence of a plasticizer (dioctyl adipate) depending on time and temperature, and influencing by crosslink density of the barrier layer [14]. Suceska et al. studied a very early stage of evaporation of nitroglycerin from a DB propellant by applying isothermal thermogravimetry experiments. The activation energy of nitroglycerin evaporation was calculated to be 81.9 kJ/mol [15]. Gottlieb and Bar studied plasticizer migration across bonded HTPB/AP propellant interfaces during cure and evaluated the diffusion coefficient to show that the curing period is significant to the migration phenomenon and direct influence on tensile strength for short aging
periods [16]. Libardi et al. determined the dioctyl azelate (DOZ) plasticizer diffusion coefficient (D) for samples containing the interfaces of rubber, liner and solid composite propellant based on HTPB to be 0.70, 2.29, 6.27 × 10−11 m2 s−1 in liner and propellant at different layer resulted from Fick’s second law [17]. Grythe and Hansen determined diffusion coefficients for isocyanates and plasticizers in ethylene propylene terpolymer (EPDM), HTPB, hydroxyl-terminated polyethylene glycol (HTPE) or glycidyl azide polymer (GAP) to be vary between 10−11 and 10−17 m2 s−1 by the weight of uptake method in polymer materials [18]. Our previous research found that mobile components, that is, plasticizers NG and BTTN, a kind of aniline stabilizer AD as well as the burning catalyzer of octyl dicyclopentadienyl iron (ODCI) in the propellant, and plasticizer dioctyl sebacate (DOS) in the liner consumption, migration, and decomposition appeared in the interfaces of NEPE propellant/HTPB liner/EPDM insulation during storage period [19]. We also quantified the main migrating components to find that NG, BTTN and AD in the propellant could migrate to the liner and insulation, and DOS in the liner migrated only to the insulation, not to the propellant [20]. Migration can reduce the allowable physical properties, weakening the grain, so that it will crack at stress-concentrated points and cause unfavorable increases or fluctuations in the burning rate and performance of the rocket motor [21]. It is, therefore, important to be able to predict the behavior of low-molecular-weight mobile additives and to control the leakage of them from the propellant. In this paper, we try to estimate the migration kinetic parameters of some components of a NEPE propellant migrated to liner and insulation through Arrhenius relationship to understand their behavior of migration. 2. Experimental 2.1. Materials HPLC-grade methanol was purchased from Sigma Chemical (USA). Analytic reagent methenyl trichloride and acetonitrile were purchased from Shanghai Shiyi Chemicals Reagent Co., Ltd. (China). Nitroglycerin (NG), 1,2,4-butanetriol trinitrate (BTTN) and a kind of aniline stabilizer AD were produced by the 42nd Institute of the Fourth Academy of CASC (China). Polyethylene glycol (PEG), hydroxyl-terminated polybutadiene (HTPB) and hexamethylene biuret polyisocyanate (N100) were purchased from Liming Research Institute of Chemical Industry (China). Ethylene propylene terpolymer (EPDM) were purchased from Du Pont China Holding Co., Ltd. (China). 25 mm cylindrical specimens were prepared with 20 mm NEPE based propellant in the middle, which was made from PEG/N100 binder system containing NG and BTTN plasticizers, capped with metal covers which were pre-covered with a 1.5–2.5 mm EPDM based insulation and a 0.5–1.0 mm HTPB based liner at both ends. The specimens were cured for 7 days at 50 ◦ C and placed inside aluminum bags and aged at temperatures 40 ◦ C and 50 ◦ C for 20 months, 60 ◦ C for 10 months, and 70 ◦ C for 4 and a half months and analyzed for their tensile properties and contents of NG, BTTN and AD at designated intervals. All chemical reagents were of chemical grade unless mentioned specially and commercially available and they were used without further treatment. 2.2. Mechanical analysis The bond strength of the specimen was measured on an electronic universal material testing machine (Model 5567, Instron, USA) through maximal tensile strength ( m ) at a rate of 20 mm/min at 25 ◦ C until failure. The tensile strength is calculated as m = F/S where F is the force, S is the transect area of sample at direction
Z.-p. Huang et al. / Journal of Hazardous Materials 229–230 (2012) 251–257
of the force. Each value is an average of five measurements with a relative standard deviation of ±3%.
For the determination of the plasticizers and stablizer in propellant, liner or insulation, the propellant, liner and insulation of a specimen was separated and cutted into 1 mm × 1 mm × 10 mm pieces with a knife manually. 25 mL of methanol was added to accurately weighed propellant pieces of approximately 300–400 mg, 50% methenyl trichloride/50% acetonitrile (v/v) to liner or insulator pieces and extracted for 8 h at room temperature. The extraction was removed to a 50 mL volumetric flask, and the residue was washed with methanol for several times and the washed methanol was also removed to the volumetric flask, adding methanol to 50 mL. The prepared solution samples were transferred to sampling vials for analysis via a high performance liquid chromatograph (HPLC, Alliance 2695, Waters, USA) under following conditions: isocratic mobile phase of methanol at 1.0 mL/min, Shodex DE-613 column (5 m, 4.6 mm × 300 mm), diode array detector at 230 nm, 10 L sample injections. Under these conditions, NG, BTTN and AD were eluted as three separate peaks [20]. A component concentration in the sample is calculated from: cx =
1000cxd Vx mx Mx
Table 1 Migration ratios of the components migrated in different interface after aging for one month at different temperatures. T (◦ C)
2.3. Plasticizers and stabilizer analysis
(1)
where cx is a component concentration in tested sample, mol/kg; 1000 is the ratio of kg/g; cxd is the component concentration determined by HPLC external standard method, g/mL; Vx is total solution volume for a sample solution, mL; mx is the mass of a sample, g; Mx is relative molecular mass of a component, g/mol. Each value is an average of three measurements with a relative standard deviation of ±1%. The relative migration ratio of a component migrated from propellant to liner was calculated as cL /cP , and that of which migrated from liner to insulation was calculated as cI /cL Where cL , cP and cI was concentration of a component in liner, propellant and insulation at the same condition, respectively. 3. Results and discussion If the mobile components are not chemically bonded to a substrate material, they will migrate in the direction of lower concentrations [15]. Fig. 1 illustrates the components concentration in propellant, liner and insulation of a bonded specimen without aging at different distance to liner interfaces. It showed that
Fig. 1. The component concentration in propellant, liner and insulation at different distance to liner interfaces.
253
40 50 60 70
cL /cP (%)
cI /cL (%)
NG
BTTN
AD
NG
BTTN
AD
24.7 21.9 20.4 11.3
22.7 19.0 18.9 19.7
281.8 228.6 292.9 272.7
8.7 9.0 13.1 5.6
6.5 7.6 11.9 10.0
35.5 36.7 31.4 32.1
concentrations of all migrated components were almost constant in propellant and insulation at distance to liner interface more than 1 mm, and those within 1 mm to liner interface were different from propellant and insulation. It revealed that migration appeared within 1 mm of the interfaces. The concentrations of migrated NG and BTTN in liner were less than those in propellant and more than those in insulation, but AD in liner was more than that in propellant and insulation. NG and BTTN were trace in insulation and much less than those in propellant, and AD was almost the same as in propellant although it was trace, too. It indicated that migration occurred during curing of the specimen and the migration of AD was different from that of NG and BTTN. To determine the difference of migration quantificationally, Table 1 listed the data of relative migration ratio of NG, BTTN and AD migrated from propellant to liner, and that of which migrated from liner to insulation after aging for one month at different temperatures. All the migration ratios of NG and BTTN were less than 25%, and migration ratios of AD were more than 200% from propellant to liner, and those from liner to insulation were less than those from propellant to liner. The migration ratios of AD were more than those of NG and BTTN at the same condition, and all the migration ratios decreased with temperature increasing from propellant to liner except AD at 60 ◦ C, and they were reverse from liner to insulation except those aging at 70 ◦ C and AD at 60 ◦ C. The results suggested that migration was changed with temperature and decomposition occurred at high temperature such as 70 ◦ C because NG decreased obviously at 70 ◦ C. It revealed that NG, BTTN migrated differently from AD, and liner could enrich AD and prevent NG and BTTN migrating from propellant to insulation in a way, just as the work we did previous [20]. Fig. 2 illustrates the changing tendency of contents of NG, BTTN and AD in propellant, liner, and insulation, and the bond strength of the propellant/liner/insulation specimen with time during aging at 70 ◦ C, respectively. It showed that all components present in propellant kept almost constant first, then decreased with time increasing at aging time after 3 × 106 s, that is to say, there were significant loss of plasticizer and stabilizer after 3 × 106 s of aging, while they increased first and decreased afterwards in liner and insulation at similar time. The maximal tensile strength ( m ) of the specimen remained almost constant at first period and decreased fast at the very time 5 × 106 s when components decreased in liner and insulation. The migration phenomenon occurs mainly due to the differential concentration of plasticizers and stabilizer between the additive part and the nonadditive part. All these components were added in propellant originally, and none was in liner, neither was in insulation, so they migrated from propellant to liner and insulation. It is well known that the maximal tensile strength is a kind of characterization for the linkage of a polymer network or a splice specimen, broken in polymer network or splice specimen linkage would result in sharp decrease of its maximal tensile strength. The relation between the maximal tensile strength and contents of components in liner and insulation suggested that migration appeared during the first period and decomposition occurred during the second period at 70 ◦ C.
254
Z.-p. Huang et al. / Journal of Hazardous Materials 229–230 (2012) 251–257
Fig. 2. The component concentration in propellant, liner and insulation, respectively during aging at 70 ◦ C and corresponding mechanic property of the specimen.
Fig. 3 illustrated chromatograms of propellant extractor aging at before and after 3 × 106 s. There are mainly four peaks before 3 × 106 s, and a new peak appears after 3 × 106 s. It proved the suggestion that decomposition occurred during the second period
70 ◦ C
Fig. 3. Chromatograms of propellant extractor before and after aging time 3 × 106 s.
at 70 ◦ C mentioned above. So migration was not the fatal factor to result in the invalidation of the specimen, while the chemical changes are the reason of their decrease in tensile strength. Kinetic parameters of migration could predict the tendency of migration of a mobile component, from which the mechanism of migration might be understood, so as some action might be taken to forbid the migration. The changing tendency of the specimen in contents of plasticizers and stabilizer aged at three other temperatures was similar as that aged at 70 ◦ C. The only difference was the migration rate. The higher the temperature, the faster the rate of migration is. From relation of contents of the components in propellant, liner and insulation with time at 70 ◦ C in Fig. 2 and those obtained at 40 ◦ C, 50 ◦ C and 60 ◦ C (not shown), the migration concentration versus time relations were obtained as c = k1 t + k2 t2 + b, and the constants k1 , k2 and b are listed in Table 2. It can be seen that k2 was much smaller than k1 , which indicated that the mechanism of migration of these compounds is zero order. Suceska et al. found that at a very early stage of evaporation of nitroglycerin from a double base rocket propellant, the evaporation can be described by the zero-order reaction model [15]. Though migration is not the same as evaporation, they are comparable because they are both physical changes in complex materials. The absolute value of k1 for NG is about 5 times that for BTTN and AD in the same condition. It corresponded to their molecular structure because NG and BTTN are homolog, and AD is aromatic compound, the molecular volume of NG is smaller than that of BTTN and AD, so the migrating rate of NG was faster than that of BTTN and AD. From above mentioned, the mechanism of migration of these compounds is zero order, so we calculated the apparent migration activation energy for the migration of NG, BTTN and AD by Arrhenius relation written as follows:
log10 [k1 ] = A −
Ea RT
(2)
4.43 4.76 4.75 3.92 – – – –5.41 0.40 0.56 1.22 2.31 0.015 0.014 0.013 0.015
−0.61 −12.3 2.86 5.43 0.82 −6.49 − −46.6 4.80 40.9 19.1 46.8 0.055 – 0.054 0.091
5.9 −0.087 5.6 −0.81 −0.39 −6.06 − −9.01 9.27 35.5 18.1 10.1 0.077 − 0.070 0.032
k1 (10−10 mol/(kg s))
255
–0.39 – –1.02 –15.0
– – – –
21.6 – 40.4 113
31.4 – 41.3 64.0
−0.64 −2.47 −4.89 −13.6
0.39 0.37 0.39 0.32
0.32 0.32 0.30 0.32
0.029 0.027 0.025 0.027
mol/(kg s))
−0.27 – −1.32 −14.0
k1 (10 b (mol/kg)
k2 (10
−16
mol/(kg s))
Liner
−10
−2.33 −5.08 −8.49 −27.3
−2.07 −5.54 −11.1 −29.2
40 50 60 70
40 50 60 70
BTTN
AD
–11.6 –24.1 –31.6 –38.7 40 50 60 70
k1 (10
NG
−10
T (◦ C)
Propellant
Fig. 4. The plots of lg [k1 ] against −1/RT of three components in propellant, liner and insulation.
Component
Table 2 The migration rate constants k1 , k2 and b of three components in propellant, liner and insulation at different temperatures.
2
mol/(kg s ))
b (mol/kg)
Insulation
k2 (10−17 mol/(kg s2 ))
b (10−3 mol/kg)
Z.-p. Huang et al. / Journal of Hazardous Materials 229–230 (2012) 251–257
where k1 is the migration rate constant at a given temperature, A is an empirically derived constant, Ea is the apparent activation energy for migration, R is the ideal gas constant, which is 8.3145 J mol−1 K−1 , T is the temperature of migration in Kelvin. The plots of log10 [k1 ] against −1/RT are illustrated in Fig. 4. It was found that the plots were good straight lines and the correlation coefficients were more than 95%. The apparent migration activation energy of three mobile components in propellant, liner and insulation were obtained by slopes of the lines and they are listed in Table 3. The Ea values were among 15 and 50 kJ/mol, which were much less than chemical energy, and almost the same as hydrogen bond energy, which was correspond to migration. The Ea of NG and BTTN were decreased in the direction of propellant to liner and insulation, and their difference was small. The Ea of AD in
256
Z.-p. Huang et al. / Journal of Hazardous Materials 229–230 (2012) 251–257
Table 3 The apparent migration activation energy of three materials in propellant, liner and insulation. Material
Ea (kJ/mol)
Propellant Liner Insulation
NG
BTTN
AD
20.9 19.9 17.4
30.5 29.7 28.9
36.4 46.1 20.5
liner was the largest, while that in insulation was the smallest, and their difference was higher than that of NG and BTTN. It suggested that the less the Ea, the faster and easier the migration, which was correspond to the migration rate constants. Diffusion coefficients of the migration components in propellant, liner and insulation may give important information for their migration [16–18]. The diffusion of a single substance from the interface into the isolation may be considered as ordinary diffusion into a semiinfinite slab and is described by Fick’s second law of diffusion. The equation assumes that there is only one diffusing component and that the diffusion coefficient is constant. The flux at the interface is given by Fick’s first law. The propellant is thought to be infinite for a thin layer of liner, so is for insulation. All components are assumed to be independent to migrate from propellant to liner and from liner to insulation. The migrating surface area is only on one side. So we can express the diffusion as following [18]:
D=
m
m0
2 i di 2
−1
(3)
t
Table 4 The average diffusion coefficients of migrating components in propellant, liner and insulation at 40 ◦ C.
Propellant Liner Insulation
has phenyl group was similar to that of liner and insulation which has long alkyl groups, and was different to that of propellant which has ether and urethane links. Further detail study about polarity of the materials will be done in our next work. It concluded that migration rate and energy were affected by the molecular volume of a mobile component and its diffusion property, and the amount of migration was resulted from the molecular polarity comparability of a mobile component to the based material. A schematic depiction of the specimen and the migration of NG, BTTN and AD is shown in Fig. 5. 4. Conclusions
where D is an average diffusion coefficient, m is the mass of the isolation sheet including the absorbed substance, m0 is the initial mass of the sheet, is the density of a diffused substance, i is the initial density of the sheet, di is the sheet thickness, t is the time when m is determined. Table 4 lists the average diffusion coefficient of the three components in propellant, liner and insulation. The average diffusion coefficients were in the range of 10−19 m2 s−1 to10−16 m2 s−1 , which was accordant to those determined by Grythe and Hansen [18]. The average diffusion coefficient of NG, BTTN were much higher than that of AD, and the value change tendency of all the components in the direction of propellant to liner and insulation was almost the same as Ea. The larger the diffusion coefficient is, the less the activation energy. It could be concluded that the migration activation energy was resulted from the diffusion property. From Table 1, the migration ratio of AD was much larger than that of NG and BTTN, which indicated that the final concentration was not resulted from migration rate, migration activation energy Ea or diffusion coefficient D. Why the observed behavior is the opposite of what is expected from the values of the diffusion coefficients and activation energies? It is because the migration activation energy and diffusion coefficient are kinetic parameters, which are affected by properties of the materials, temperature and time, and the migration ratio is a thermodynamic parameter, which is affected by properties of the materials and temperature, ignoring the time. Thus, it could be explained that the polarity of AD which
Material
Fig. 5. Schematic depiction of the specimen and the migration of NG, BTTN and AD.
¯ (10−18 m2 s−1 ) D NG
BTTN
AD
650 34.9 2.67
632 33.8 2.02
0.19 0.80 0.22
Migration appeared in the interfaces of NEPE propellant/HTPB liner/EPDM insulation, the smaller the apparent migration activation energy, the faster the migration rates, the larger the diffusion coefficient, and the less the amount of migration. The migration rate was affected by the molecular volume, and the amount of migration was resulted from the molecular polarity comparability. Acknowledgment We gratefully acknowledge financial support from the major Grant of the National Basic Research Program of China (973 Program 613142). References [1] M.S. Sureshkumar, C.M. Bhuvaneswari, S.D. Kakade, M. Gupta, Studies on the properties of EPDM-CSE blend containing HTPB for case-bonded solid rocket motor insulation, Polym. Adv. Technol. 19 (2008) 144–150. [2] A.M. Pang, J. Zheng, Prospect of the research and development of high energy solid propellant technology, J. Solid Rocket Techn. 27 (2004) 289–293. [3] W. Zhang, X. Fan, H. Wei, J. Li, Application of nitramines coated with nitrocellulose in minimum signature isocyanate–cured propellants, Propell. Explos. Pyrotech. 33 (2008) 279–285. [4] J.L. de la Fuente, O. Rodríguez, Dynamic mechanical study on the thermal aging of a hydroxyl-terminated polybutadiene-based energetic composite, J. Appl. Polym. Sci. 87 (2003) 2397–2405. [5] F. Zhao, S. Heng, R. Hu, H. Gao, F. Han, A study of kinetic behaviours of the effective centralite/stabilizer consumption reaction of propellants using a multi-temperature artificial accelerated ageing test, J. Hazard. Mater. 145 (2007) 45–50. [6] J.H. Yi, F.Q. Zhao, W.L. Hong, S.Y. Xu, R.Z. Hu, Z.Q. Chen, L.Y. Zhang, Effects of Bi-NTO complex on thermal behaviors, nonisothermal reaction kinetics and burning rates of NG/TEGDN/NC propellant, J. Hazard. Mater. 176 (2010) 257–261. [7] M.D. Judge, An investigation of composite propellant accelerated ageing mechanisms and kinetics, Propell. Explos. Pyrotech. 28 (2003) 114–119. [8] F.J. Wu, S. Peng, X.H. Chi, Study on microcosmic mechanics performance/structure of NPPE propellant/liner bonded interface, J. Solid Rocket Techn. 33 (1) (2010) 81–85. [9] J. Götz, Characterization of the structure in highly filled composite materials by means of MRI, Propell. Explos. Pyrotech. 27 (2002) 179–184. [10] F. Xu, N. Aravas, P. Sofronis, Constitutive modeling of solid propellant materials with evolving microstructural damage, J. Mech. Phys. Solid. 56 (2008) 2050–2073.
Z.-p. Huang et al. / Journal of Hazardous Materials 229–230 (2012) 251–257 [11] F.J. Wu, S. Peng, X.H. Chi, F.C. Zhong, XPS characterization of NEPE propellant/line bondline, J. Solid Rocket Techn. 32 (2009) 192–196. [12] S. Li, Y. Liu, X. Tuo, X. Wang, Mesoscale dynamic simulation on phase separation between plasticizer and binder in NEPE propellants, Polymer 49 (2008) 2775–2780. [13] H.L. Yin, Y. Wang, D.F. Li, Ingredient migration and their effect on NEPE propellant bonding system, J. Solid Rocket Techn. 32 (2009) 527–530. [14] A. Ünver, N. Dilsiz, M. Volkan, G. Akovalı, Investigation of acetyl ferrocene migration from hydroxyl-terminated polybutadiene based elastomers by means of ultraviolet–visible and atomic absorption spectroscopic techniques, J. Appl. Polym. Sci. 96 (2005) 1654–1661. [15] M. Suceska, S.M. Musanic, I.F. Houra, Kinetics and enthalpy of nitroglycerin evaporation from double base propellants by isothermal thermogravimetry, Thermochim. Acta 510 (2010) 9–16. [16] L. Gottlieb, S. Bar, Migration of plasticizer between bonded propellant interfaces, Propell. Explos. Pyrotech. 28 (2003) 12–17.
257
[17] J. Libardi, S.P. Ravagnan, A.M.F. Morais, A.R. Cardoso, Diffusion of plasticizer in a solid propellant based on hydroxyl-herminated polybutadiene, Polímeros 20 (2010) 241–245. [18] K.F. Grythe, F.K. Hansen, Diffusion, Rates and the role of diffusion in solid propellant rocket motor adhesion, J. Appl. Polym. Sci. 103 (2007) 1529–1538. [19] Z.P. Huang, J.J. Liu, S.L. Xu, J. Bai, X.G. Ma, Qualitative analysis of migration components in the interface of NEPE propellant, J. Solid Rocket Techn. 33 (2010) 41–44. [20] Z.P. Huang, L.M. Tan, Q.W. Cao, X.G. Ma, Quantitative analysis of migrating components in the interface of NEPE propellant/liner/insulation, Chin. J. Energ. Mater. 18 (2010) 330–334. [21] G.P. Sutton, Rocket Propulsion Elements: An Introduction to the Engineering of Rockets, 6th ed., Wiley, New York, 1992.