CA/RDX复合材料超临界CO2间歇法微孔发泡研究

杨伟涛，应三九

(1.西安近代化学研究所，陕西 西安 710065；2.南京理工大学化工学院，江苏 南京 210094)



CA/RDX复合材料超临界CO2间歇法微孔发泡研究

杨伟涛1，应三九2

(1.西安近代化学研究所，陕西 西安 710065；2.南京理工大学化工学院，江苏 南京 210094)

摘要：以醋酸纤维素(CA)为黏结剂，黑索今(RDX)为含能组分，通过超临界二氧化碳(SC-CO2)间歇法发泡技术制备了微孔可燃复合材料CA/RDX。采用重量法研究了SC-CO2在该复合材料中的吸收与解吸收过程，同时利用Fickian扩散定律，研究了SC-CO2的扩散系数。采用扫描电子显微镜(SEM)研究了发泡温度和饱和压力对该材料内部形貌的影响规律。结果表明，CO2的吸收量与扩散系数随着饱和压力的提高而增加，在20MPa，40℃饱和条件下达到最大饱和量11.16%。RDX和CA界面降低了微孔异相成核能垒，因此增加了微孔成核速率，微孔产生于RDX和CA界面间的微裂纹并逐步长大成为完整的气泡；在微孔成核后，微孔尺寸随饱和温度与发泡温度差(ΔT)的增加而增大；由于CO2解吸收过程中扩散系数随饱和压力的增加而增加，因此饱和压力对泡孔形貌的影响不明显。

关键词：微孔可燃复合材料；醋酸纤维素；超临界CO2；SC-CO2；吸收解吸收行为；微孔发泡

Introduction

Cellulose acetate (CA) plays an important role in the field of regenerated fibers, and it can be produced from a few kinds of cheap and renewable materials, such as wood, cotton and recycled paper. CA is widely used for various applications, for instance, films, separation membrane, textiles, cigarette filters and coating. CA was also used in the ammunition field as an inert binder for crystalline explosives[1-2]. To date, it is a novel and effective idea to combine foamed CA binder-based composites with SC-CO2foaming technique. CA and crystalline explosives remain their known chemical properties and SC-CO2foaming grants the material porous structure and also new behavior. Foamed CA bonded composites have several advantages, such as light weight, heat insulation and high burning rate. This kind of foamed composites has great application potential in combustible cartridge cases and caseless ammunition[3-5].

SC-CO2is an inexpensive, safe and environmentally benign foaming agent[6-7]. Environment-friendly techniques can be used to prepare microcellular foams using SC-CO2. SC-CO2has many advantageous properties, such as a tuneable solvent power, the plasticization of glassy polymers (as a consequence of glass transition temperature depression) and enhanced diffusion rates. Especially, the use of CO2has increased greatly in the last few years due to the low critical conditions of CO2(31.1℃ and 7.1MPa). Meanwhile, the microcellular foaming process using SC-CO2allows an easy and complete separation from the polymer. The materials used to produce microcellular foams include polystyrene, polypropylene, polycarbonate, poly(methyl methacrylate) and so forth. In general, four steps are needed to prepare foamed materials: (1) sorption of SC-CO2;(2) cell nucleation;(3) cell growth;(4) stabilization of foam structures[8-9]. One of the key processes to control the porous structure is the nucleation and growth of gas cells dispersed throughout the polymer. The resulting structure depends on a balance between several mechanisms, such as diffusion of CO2in the polymer, the solubility and the plasticization effect, which depends intrinsically on the polymer and the saturation conditions.

But semi-crystalline polymers are considered difficult to foam due to lower gas solubility and the fact that the crystalline portions prevent the nucleation of bubbles[10]. As a result, the foamed structure is hard to achieve unless the objects were quenched from the melt so that the foamability of these materials was enhanced[11-12]. Also, the use of high temperature is not appropriate in fabrication of objects containing explosives.

Consequently, this work is focused on a two-step foaming process in which the CA bonded RDX composites are previously saturated under SC-CO2and foamed by a change of temperature. The microcellular processing technique employed in this study utilizes temperatures below the melting point such that the foaming process occurs while the CA is in a semi-crystalline state. Firstly, the solubility and diffusivity of CO2in the objects were investigated before foaming. Secondly, the foaming conditions of fabricating CA based combustible composites were explored and the inner foamed structures were investigated and discussed. Particular emphasis is given to the influence of the foaming temperature on the foam morphology.

1Experimental

1.1 Materials

The microcellular combustible objects were formulated based on RDX and CA. CA was supplied by Xi′an North Hui′an Chemical Industries Co., Ltd. Ethanol (AR, Nanjing Chemical Reagent Co., Ltd) and acetone (AR, Sinopharm Chemical Reagent Co., Ltd) were used as received. Average particle size of RDX is about 10 μm.

1.2Preparation of CA/RDX composites

CA (mass fraction of 35%) and RDX (mass fraction of 65%) particles were incorporated in a sigma blade mixer using acetone as solvent to obtain dough. The dough was extruded in a vertical hydraulic press. The extruded composites were dried until the volatile content was decreased to less than 0.5 percent.

1.3Microcellular foaming process

The solid composites were foamed by the two-step foaming process as shown in Figure 1. Sheet samples with 12.9mm×50mm×1mm were placed into the high pressure vessel. Then, the pressure was quenched. Lastly, the samples were heated for 1min.

Fig.1　Foaming equipments and foaming process using SC-CO2

1.4Sorption and Desorption Behaviour of SC-CO2in CA/RDX Sheet

The method of analyzing the sorption and desorption data, i.e. the gravimetric method, was described at length by previous investigators[13-14]. Figure 2 presents the schematic illustration of the sorption and desorption process and mass-loss analysis. For the purpose of this test, the specimens were pressurized in a pressure vessel wih CO2at saturation temperature,Tsat, and saturation pressure,Psat. After various sorption times (ts), the vessel was rapidly vented and the specimen was transferred to the electronic balance (Sartorius TE124S) under room temperature at atmospheric pressure for the desorption measurement. The first data as recorded at 30s. The weight of specimen (m) was recorded at an interval of 10s.

Fig.2　Schematic illustration of sorption and desorption process and mass-loss analysis

Since the ratio of thickness to length of specimen is very small, edge effects are negligible. A reasonable assumption of Fickian diffusion was applied and the governing equation was the basic mass transfer equation for a plane sheet system:

(1)

whereCis the concentration andDis the diffusivity. One-dimensional diffusion along the thicknessLof the planar sample and constant diffusivity were assumed in this study. With proper initial and boundary conditions and the application of the Laplace transform method, the following analytical solution was obtained:

(2)

whereMtis the amount of sorption into the planar specimen at timet, and can be expressed as the weight percentage of CO2per unit weight of specimen.M∞is equilibrium sorption amount.

The solution of Eq. (2) can be introduced to the desorption process after simplifying the equation for the desorption process:

Md/M∞=-4 (Dd/π)0.5(td0.5/L)

(3)

In the desorption process, the plot ofMd/M∞as a function of (t0.5/L) yields essentially a straight line with a slope of 4(Dd/π)0.5, which is readily solved forDd. Simultaneously, the amount of sorption of CO2into CA/RDX sheet,Ms, was estimated by extrapolating the measured desorption weight fraction,Md, to the zero desorption time.

Simplification of Eq. (2) by truncating at the first term in the summation yields the following form for the sorption process:

ln(1-Ms/M∞) =-(Dsπ2/L2)ts+ln(π2/8)

(4)

The sorption coefficient can be calculated by Eq.(4). The relationship of ln(1-Ms/M∞) andtsis linear, by whichDscan be obtained from the slope of the plot.

1.5Characterization of foam morphology

A QUANTA FEG 250 scanning electron microscope (SEM) was used to characterize the inner structures. The tested samples were immersed in liquid nitrogen and freeze-fractured.

2Results and Discussion

2.1Sorption course of SC-CO2into CA/RDX Sheet

The measured desorption mass fraction,Md, to the desorption time at 40℃ and 15MPa was shown in Figure 3. The linear relationship betweenMdandtd0.5indicates that the assumption of Fickian diffusion is acceptable. TheMsvalues at various sorption times,ts, are determined from the intercepts, and a sorption curve is then generated.

Fig. 3　The desorption curves of SC-CO2 in CA binder-based sample (Ts=40℃, Ps=15MPa)

Fig. 4　The dissolved SC-CO2 in CA binder-based sample (Ts=40℃, Ps=15MPa)

Fig. 5　ln(1-Ms/M∞)-ts/L2 curves and sorption coefficient of CA binder-based sample

Figure 4 shows the sorption curve where the sorption amountMswas plotted againstts. As we can see from Figure 4,Msincreased rapidly in the initial 1h, and increased slowly after 1h to equilibrium. An equilibrium uptake of 9.95% CO2is absorbed within 7h. Figure 5 is ln(1-Ms/M∞)-ts/L2from which the sorption diffusivity can be obtained. Because there are two sorption modes before and after 1h sorption time, two sorption coefficient were 0.2379mm2/h and 0.0187mm2/h, respectively. The decrease ofDsindicated that after CO2had penetrated the specimen, the diffusivity was weakened.The desorption data at different pressures are shown in Figure 6(a). The equilibrium sorption amountsM∞and diffusion coefficient at different pressures are shown in Figure 6(b). As the figures indicates, theM∞values and the desorption coefficient in Figure 6(b) increased with the increase of pressure. Although theM∞increased with pressures, the increasing ofDdresulted in more CO2escaping from the matrix during the period from the vessel to heating bath.

Fig.6　Desorption curves, M∞ and Dd for samples at different saturation pressures

Figure 7 presents desorption curves, sorption amounts and desorption coefficient at different temperatures. As Figure 7 indicates, the sorption amount decreased with the increase of saturation temperature, and in contrast, the desorption coefficient increased with temperature. It means that when we saturated the sample at high temperature, less CO2amount was obtained, and CO2more easily escaped from the matrix. For this reason, a high saturation temperature is usually not adopted in the two-step foaming process.

Fig.7　Desorption curves, M∞ and Dd for samples atdifferent saturation temperatures

2.2Characterization of inner structure

The objects containing 65% RDX were saturated at 40℃ and 15MPa for 10h, and then foamed at different foaming temperatures(Tf). Figure 8 shows the effect of temperature

on the inner structures. When samples were heated at a temperature under 100℃, there were microcracks around the RDX particles. When samples were heated at higher temperatures (110, 120, 130 and 140℃), microcracks expanded and developed into micropores. The pore size increased roughly with the increasing of foaming temperature.

Significantly, when pure CA saturated at 15MPa and 40℃ was foamed at 140℃, no cracks or pores were generated (also shown in Figure 8). RDX particles provide heterogeneous nucleation sites for the microfoaming process and decreased the foaming temperature matrix need to nucleation. According to the classical nucleation theory, lowering the interfacial energy and reducing the energy barrier can increase the nucleation rate[15]. At the same saturating pressure and temperature, the interfacial energy could be considered constant. The presence of an CA/RDX interface can sharply reduce the activation energy barrier to nucleation, thereby increasing the nucleation rate. This type of nucleation is defined as heterogeneous nucleation. Once the heterogeneous nucleation occurred, the driving force for cell growth is directly related to the ΔT(the increased temperature from saturation temperature to foaming temperature).

Unlike amorphous polymers, semi-crystalline polymers have sufficiently high mechanical properties (such as modulus and tensile strength) to maintain their structural integrity when the samples are heated. As a result, the materials do not flow rapidly enough, as compared to the diffusion of gas out of the polymer chains. The temperature of the semi-crystalline polymers must be raised above their melting points before the chains have enough mobility to accommodate growth. Hence, when samples was foamed under 100℃, the mobility of chains is not enough to ensure the nucleation developing to micropores. As a result, there are only microcracks around RDX particles.

Fig. 8　SEM images of specimens foamed at different foaming temperatures

The increasing saturation pressure causes an increase in the sorption amount of CO2, which affects the plasticization of the polymer binder. The plasticization is weakened with the diffusion of gas into pores. The pores stop growing due to the vitrification of polymer. Simultaneously, the pressure difference between inside and outside of the sample was also bigger when the saturation pressure is higher. Figure 9 shows the effect of saturation pressure on the morphology of foamed samples. As Figure 9 shows, it seems the size of micropores increased with pressure, resulting from the increased CO2uptake amount. But the effect of pressure is not as impressive as foaming temperatures. In the foaming process, the heterogeneous nucleation rate could be recognized as invariable[14], the increasing diffusion coefficient weakened the effect of saturation pressures on the CO2content at the moment when samples were dipped in oil bath, although the sorption amount when samples in vessel is higher at high pressures.

Fig. 9　SEM images of specimens foamed at various saturation pressures (Ts = 40℃, Tf = 110℃)

3Conclusions

(1) This study provided an available route, i.e. the two-step foaming process, to fabricate microcellular combustible objects which were composed of semi-crystalline polymer binders with a high melting point.

(2)The solubility and diffusivity of SC-CO2increased with increased pressures. The heterogeneous nucleation appeared around RDX particles and translated from cracks to pores at higher temperatures when the objects were still in solid-state.

(3) Heterogeneous nucleation reduced the activation energy, revealing an effective route to fabricate porous materials which are hard to foam.

References：

[1]Pillai A G S, Joshi M M, Barve A M, et al. Cellulose

acetate binder-based LOVA gun propellant for tank guns[J]. Defence Science Journal, 2013, 49(2): 141-149.

[2]Sanghavi R R, Kamale P J, Shaikh M A R, et al. HMX based enhanced energy LOVA gun propellant[J]. Journal of Hazardous Materials, 2007, 143(1/2): 532-534.

[3]Bhnlein-Mau J, Krber H. Technology of foamed propellants[J]. Propellants, Explosives, Pyrotechnics,2009, 34(3): 239-244.

[4]Bhnlein-Mau J, Eberhardt A, Fischer T S. Foamed Propellants[J]. Propellants, Explosives, Pyrotechnics, 2002, 27(3): 156-160.

[5]Yang W, Li Y, Ying S. Fabrication, thermoanalysis, and performance evaluation studies on RDX-based microcellular combustible objects[J]. Propellants, Explosives, Pyrotechnics, 2014, 39: 568-573.

[6]Ruiz J A R, Marc-Tallon J, Pedros M, et al. Micro and nano cellular amorphous polymers (PMMA, PS) in supercritical CO2assisted by nanostructured CO2-philic block copolymers - one step foaming process[J]. Journal of Supercritical Fluids, 2011, 58(1): 168-176.

[7]Ruiz J A R, Marc-Tallon J, Pedros M, et al. Two-step micro cellular foaming of amorphous polymers in supercritical CO2[J]. The Journal of Supercritical Fluids, 2011, 57(1): 87-94.

[8]Kazarian S G. Polymer processing with supercritical fluids[J]. Polymer Science Series Cc/C of Vysokomolekuliarnye Soedineniia, 2000, 42(1): 78-101.

[9]Goel S K, Beckman E J. Generation of microcellular polymeric foams using supercritical carbon dioxide. I: Effect of pressure and temperature on nucleation[J]. Polymer Engineering and Science, 1994,34(14): 1137-1147.

[10] Kumar V. Microcellular polymers: novel materials for the 21st century[J]. Cellular Polymers, 1993, 12(3): 207-223.

[11] Park C B, Cheung L K. A study of cell nucleation in the extrusion of polypropylene foams[J]. Polymer Engineering and Science, 1997, 37(1):1-10.

[12] Zheng W G, Lee Y H, Park C B. Use of nanoparticles for improving the foaming behaviors of linear PP[J]. Journal of Applied Polymer Science, 2010, 117(5): 2972-2979.

[13] Tang M, Du T B, Chen Y P. Sorption and diffusion of supercritical carbon dioxide in polycarbonate[J]. The Journal of Supercritical Fluids, 2004, 28(2): 207-218.

[14] Muth O, Hirth T, Vogel H. Investigation of sorption and diffusion of supercritical carbon dioxide into poly (vinyl chloride)[J]. The Journal of Supercritical Fluids, 2001, 19(3): 299-306.

[15]Colton, J S, Suh N P. Nucleation of microcellular foam: theory and practice[J]. Polymer Engineering and Science, 1987, 27(7): 500-503.

CLC number：TJ55;TB33Document Code：AArticle ID：1007-7812(2016)02-0022-05

Microcellular Foaming of CA/RDX Composites in a Batch Supercritical CO2Process

YANG Wei-tao1, YING San-jiu2

(1. Xi′an Modern Chemistry Research Institute, Xi′an 710065, China; 2. School of Chemical Engineering, Nanjing University of Science and Technology, Nanjing 210094, China)

Abstract:Microcellular combustible composites CA/RDX, based on cellulose acetate (CA) as binder and RDX as energetic component, were fabricated by the two-step microcellular foaming in supercritical CO2. The sorption and desorption behaviour of SC-CO2 in the composites were investigated by a weighing process, and the diffusivity was analyzed by Fickian diffusion. Scanning Electron Microscopy (SEM) was used to investigate the influence of foaming temperature and saturation pressure on the foamed structures. The results indicate that the CO2 sorption amount increases with the sorption pressure and the largest sorption amount reached to 11.16% at 20 MPa and 40℃, with a corresponding increase in sorption diffusivity. The presence of a RDX/CA interface can reduce the activation energy barrier to nucleation, thereby increasing the nucleation rate. The micropores generate from microcracks between RDX particles and the CA binder, and develope into viable bubbles. Once the nucleation has occurred, the driving force for cell growth is directly related to the increased temperature from saturation temperature to foaming temperature(ΔT). The pore size increases with ΔT. Simultaneously, the effect of saturation pressure is not evident as the foaming temperature, for the effect of pressure, was weakened by an increased diffusion coefficient.

Keywords:microcellular combustible composite; cellulose acetate; supercritical CO2 (SC-CO2); sorption and desorption behavior; microcellular foaming

DOI：10.14077/j.issn.1007-7812.2016.02.004

Received date:2015-11-11；Revised date:2015-11-19

Biography:YANG Wei-tao(1987- ), male, research field: application of energetic materials.E-mail:njyangweitao@163.com