AGING OF EXPLOSIVE CRYSTALS (RDX) INVESTIGATED BY X-RAY DIFFRACTION

Transcription

1 1 AGING OF EXPLOSIVE CRYSTALS (RDX) INVESTIGATED BY X-RAY DIFFRACTION Michael Herrmann, Manfred A. Bohn Fraunhofer Institut für Chemische Technologie ICT, Pfinztal, Germany ABSTRACT Coarse and fine particles of standard and insensitive RDX (S-RDX and I-RDX) samples were aged up to 30 days and 90 C in air and argon, the aged and non-aged samples were investigated by means of X-ray diffraction and microstructure parameters of the crystalline powders were quantified. The investigations revealed different behavior of the RDX types. Whereas microstrain released in I-RDX, further crystal damage occurred in S-RDX during the aging treatment. INTRODUCTION AND AGING Reduced sensitivity (RS) variants of high explosives have created much interest for application in insensitive munitions as they offer the potential to reduce the susceptibility of munitions to sympathetic reactions and shock-based threats. When using such RS-variants the question arises, if a reduced sensitivity is affected by aging or if RS-qualities may even help preventing aging mechanisms. For instance Lochert et al. (2003) and Spyckerelle et al. (2008) nd significant changes of the shock sensitivity after aging of plastic-bonded explosive formulation PBXN-109 when RS-RDX (research department explosive, cyclotrimethylenetrinitramine) is included, which does not seem to be the case for standard RDX produced by the Bachmann process. Therefore Spyckerelle concluded aging must affect the RDX itself. On the analytical side a lot of work has been done at the crystal level with the aim to distinguish RS- from conventional qualities and to identify requirements for explosive particles which might explain or guarantee reduced sensitivity behavior of the corresponding plastic-bonded explosive PBX composition. X-ray diffraction tools were developed and refined at Fraunhofer ICT in order to quantify and assess internal crystal qualities of high explosives and even of coarse crystalline explosive powders embedded in a PBX (Herrmann et al., 2012). The

2 This document was presented at the Denver X-ray Conference (DXC) on Applications of X-ray Analysis. Sponsored by the International Centre for Diffraction Data (ICDD). This document is provided by ICDD in cooperation with the authors and presenters of the DXC for the express purpose of educating the scientific community. All copyrights for the document are retained by ICDD. Usage is restricted for the purposes of education and scientific research. DXC Website ICDD Website -

3 2 methods are applied in this work for the investigation of the aging behavior of a reduced sensitivity RDX compared to a standard quality of RDX. EXPERIMENTAL & EVALUATION Coarse and fine particles of standard and insensitive RDX samples (S-RDX and I-RDX) were aged up to 30 days and 90 C in air and argon. Details of the aging process and conditions are reported elsewhere (Bohn and Borne, 2013). The investigations of the microstructure of pristine and aged samples were performed on a Bragg-Brentano Diffractometer, D8 Advance from Bruker AXS, equipped with copper tube, 2.5 Soller collimators, 0.3mm divergence slit, anti-scatter screen, flip- opening. Table 1 summarizes samples and measurements. were monitored between 5 Additionally, coarse crystalline powders were measured using a different technique the scans or rocking curves (Herrmann et al. 2007). These scans were recorded on the same system, but equipped with a scintillation counter and secondary monochromator. Measurements were performed on the seven selected reflection conditions and ranges summarized in Table 2, with 0.01 step width and 4s/step counting time. The evaluation included phase identification and Rietveld analysis (whole pattern fit) using the data base of the International Centre for Diffraction Data ICDD and the tools of the program TOPAS from Bruker AXS (2008). It was based upon the crystal structure data of RDX reported by Choi and Prince (1972). Furthermore, the instrumental profile was calibrated through additional convolutions determined with the line and position standard reference material SRM 660a (2000) of the National Institute of Standards and Technology NIST. Thus, the evaluation yielded phase composition/purities, lattice parameters, mean crystal densities, crystallite size (Cry size L or L Vol-IB ) and/or the mean microstrain ( 0 ), the last two were deduced from the broadening of diffraction peak profiles. The rocking curves have been evaluated by pattern decomposition using the peak fit and Pearson VII analytical function provided by the program TOPAS from Bruker AXS. Peak width distributions and median peak widths X 50 deduced from peak fit data were used as a measure for internal crystal quality or relative microstrain of the RDX crystals (Herrmann et al. 2007).

4 3 Table 1: Samples and measurements. 2 scans air Ar 0d 15d 30d 15d 30d I-RDX, M3C 2 x 1 x 1 x 1 x 1 x I-RDX, Class 1 2 x 1 x 1 x 1 x 1 x S-RDX Typ I, Class 5 2 x 1 x 1 x 1 x 1 x S-RDX Typ I, Class 1 2 x 1 x 1 x 1 x 1 x scans air Ar 0d 15d 30d 15d 30d I-RDX, Class 1 1 x 1 x 1 x 1 x 1 x S-RDX Typ I, Class 1 1 x 1 x 1 x 1 x 1 x Table 2: Reflection conditions and ranges of rocking curves for RDX. hkl Position -scan 2 start stop (111) (210) (102) (131) (411), (132) (421), (223) (214), (133) RESULTS Sighting the scans revealed crystalline impurities of HMX (high-velocity military explosive, cyclotetramethylenetetranitramine) in S-RDX but not in I-RDX. Fig. 1 shows a relevant section of the diffraction patterns together with reference positions of RDX (blue markers) and HMX (green markers). The small peak at 20.6 identifies the -phase of HMX. Besides, the class 1 samples revealed strong intensity fluctuations and ragged profiles (Fig. 2), due to the poor orientation statistics of these coarse grained materials. Hence, only the the fine samples (S-RDX, Class 5 and I-RDX, M3C) were evaluated by means of Rietveld analysis. Fig. 2 shows, for instance, the Rietveld plot of fine I-RDX aged 15 days in air, where the green, red and grey patterns represent the measurement, the refined calculation and their differences, respectively. The Rietveld analysis of this pattern yielded 10µm mean crystallite size and a strain value of This means that no size broadening occurs in the RDX

5 4 samples but peak broadening due to micro strain, as 10µm is the maximum value of the size analysis Theta-scan Fig. 1: Diffraction patterns of S-RDX (black) and I-RDX (red) and markers of RDX (PDF ; blue) and HMX (PDF ; green) Fig. 2: Rietveld plot of 15 days aged I-RDX. Measured (green), calculated pattern (red) and difference (grey). Fig. 3 depicts a comparison of the behavior of microstrain in I-RDX and S-RDX with aging. The strain decreased continuously in I-RDX as function on aging, approaching a strain value of 0.13 after 30 days. No differences between aging in air and argon were found. This was different in S-RDX, where microstrain increased within the first 15 days but decreased

6 5 afterwards and finally also reached approximately 0.13 after 30 days. Higher deviations were found between aging in air and argon. Strain e0 Strain e I-RDX in air I-RDX in Ar 0d 15d 30d S-RDX in air S-RDX in Ar d 15d 30d Fig. 3: Comparison of microstrain of I-RDX (upper diagram) and S-RDX (lower diagram) plotted versus aging time. Fig. 4 depicts a section of a scan (beige), its pattern decomposition (red) and the difference (grey). The fitting procedures yielded 340 up to 640 resolvable peaks per sample, where each peak represents an individual crystallite. The cumulative numbers of peaks are plotted versus peak widths in Fig. 5. The curves differentiate I-RDX and S-RDX, since the curves of the S-RDX-samples are drawn to higher peak widths, which means to the right side of the diagram. For quantification of this effect the median peak widths of the curves in Fig. 5 were determined and plotted versus aging time at 90 C in Fig. 6. In this plot I-RDX and S-RDX show quite different behavior. Whereas the median peak widths of S-RDX start with significantly higher values, and underwent further broadening on aging the median peak widths of I-RDX even reduce slightly on aging. These changes are more pronounced in the second aging period between 15 and 30 days. Thus, the already remarkable difference of

7 6 median peak widths between I-RDX and S-RDX has more than doubled during the aging process. With the use of argon during the aging process, the median peak widths are shifted systematically to smaller values ,2 7,4 7,6 7,8 8 8,2 8,4 8,6 8,8 9 9,2 9,4 9,6 9,8 10 Fig. 4: Selected section of a scan and its pattern decomposition. Peak positions and measured pattern (beige), fitted pattern (red) and difference (grey). 1.0 Cumulative number of peaks 0.5 S-RDX 0d S-RDX 15d S-RDX 30d I-RDX 0d I-RDX 15d I-RDX 30d Peak width [ omega] Fig. 5: Cumulative number of scans plotted versus peak widths. The median peak widths are used as a measure for crystal quality in terms of microstrain.

8 7 Median peak width [ omega] I-RDX in argon I-RDX in air S-RDX in argon S-RDX in air Storage time [d] Fig. 6: Median peak widths of I-RDX and S-RDX. The curves reveal increased microstrain in S-RDX but release of microstrain in I-RDX on aging. DISCUSSIONS X-ray diffraction provides powerful tools for the investigation of internal microstructure parameters of energetic crystalline powders, particularly the rocking curve approach refined for the investigation of coarse material (class 1). Here the high number of evaluated peaks ensures a reliable statistical evaluation of peak broadening related to the microstructure. The investigations revealed remarkable higher microstrain in S-RDX crystals compared to I-RDX, which further increased during the aging process, whereas microstrain in I-RDX even released on aging. This means that a poor RDX crystal quality, which may include HMX impurities in the RDX lattice or in grain boundaries, undergoes further damage during aging whereas high quality RDX even improves due to thermal healing of defective areas in the crystals. This holds also for the fine powders but less pronounced and in case of S-RDX less conclusive. Aging in argon helped preventing or reducing crystal damage or even healing processes in the large crystals. The effect was not found in the fine fraction. However, when

9 8 applications in PBX are considered, the coarse crystals are assumed to dominate the sensitivity, and thus play the main role. Considering that microstrain at defects contributes to or initiates the creation of hot spots e. g. under shock loading, the use of I-RDX types in PBX not only reduces the original shock sensitivity of a PBX compared to the standard quality, but prevent RDX crystals from becoming more susceptible to unintended reactions under aging, which is not the case with S-RDX. REFERENCES Bohn M. A., Borne L. (2013) Investigation of the change in thermal and shock sensitivity by ageing of RDX charges bonded by HTPB-IPDI and GAP-N100 NDIA Insensitive Munitions and Energetic Materials Technology Symposium, San Diego, California, USA, Choi, C. S., Prince, E. ( Acta Crystallogr. B28, Herrmann, M., Kempa P. B., Doyle, S. (2007). Microstructure of energetic crystals grain by grain Z. Kristallogr. Suppl. 26, pp Herrmann, M., Kempa, P. B., Förter-Barth, U, Arnold, W., (2012)., Advances in X-ray Analysis, Vol. 55, Lochert, I. J., Franson, M. D., and Hamshere, B. L. (2003). Reduced Sensitivity RDX (RS-RDX) Part I: Literature Review and DSTO Evaluation DSTO Systems Sciences Laboratory, Edinburgh, Australia. Spyckerelle, C., Eck, G., Sjöberg, P., Amnéus, A.-M. (2008). Reduced Sensitivity RDX Obtained From Bachmann RDX Propellants Explos. Pyrotech. 33, No Bruker AXS (2008 Germany. National Institute of Standards & Technology (2000) Certificate Standard Reference Material 660a Gaithersburg, MD, USA. ICDD (2014), PDF (Database). International Centre for Diffraction Data, edited by Dr. Soorya Kabekkodu (Newtown Square, PA, USA).