Isentropic compression loading of octahydro-1,3,5,7-tetranitro-1,3,5,7-tetrazocine(HMX) and the pressure-induced phase transition at 27 GPaD. E. Hare, J. W. Forbes, D. B. Reisman, and J. J. Dick Citation: Applied Physics Letters 85, 949 (2004); doi: 10.1063/1.1771464 View online: http://dx.doi.org/10.1063/1.1771464 View Table of Contents: http://scitation.aip.org/content/aip/journal/apl/85/6?ver=pdfcov Published by the AIP Publishing Articles you may be interested in Isothermal equations of state of beta octahydro-1,3,5,7-tetranitro-1,3,5,7-tetrazocine at high temperatures J. Appl. Phys. 97, 053513 (2005); 10.1063/1.1856227 On the nucleation mechanism of the β-δ phase transition in the energetic nitramine octahydro-1,3,5,7-tetranitro-1,3,5,7-tetrazocine J. Chem. Phys. 121, 5550 (2004); 10.1063/1.1782491 Temperature-dependent shear viscosity coefficient of octahydro-1,3,5,7-tetranitro-1,3,5,7-tetrazocine (HMX): Amolecular dynamics simulation study J. Chem. Phys. 112, 7203 (2000); 10.1063/1.481285 An estimate of the linear strain rate dependence of octahydro-1,3,5,7-tetranitro-1,3,5,7-tetrazocine J. Appl. Phys. 86, 6717 (1999); 10.1063/1.371722 Equation of state, phase transition, decomposition of β-HMX (octahydro-1,3,5,7-tetranitro-1,3,5,7-tetrazocine) athigh pressures J. Chem. Phys. 111, 10229 (1999); 10.1063/1.480341

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:

131.193.242.21 On: Thu, 27 Nov 2014 21:13:54

Isentropic compression loading of octahydro-1,3,5,7-tetranitro-1,3,5,7-tetrazocine (HMX) and the pressure-induced phase transition at 27 GPa

D. E. Hare,a) J. W. Forbes,b) and D. B. ReismanLawrence Livermore National Laboratory, Livermore, California 94551

J. J. DickLos Alamos National Laboratory, Los Alamos, New Mexico 87545

(Received 5 March 2004; accepted 19 May 2004)

The 27 GPa pressure-induced epsilon–phi phase transition in octahydro-1,3,5,7-tetranitro-l,3,5,7-tetrazocine(HMX ) is explored using the isentropic compression experiment(ICE) techniqueat the Sandia National LaboratoriesZ-machine facility. Our data indicate that this phase transitionis sluggish and if it does occur to any extent under the time scaless200–500 nsd and strain ratess53105d typical of ICE loading conditions, the amount of conversion is small. ©2004 AmericanInstitute of Physics. [DOI: 10.1063/1.1771464]

Octahydro-l,3,5,7-tetranitro-l,3,5,7-tetrazocine(HMX ) isthe main reactive ingredient of many high-performance highexplosive formulations. There are two phase transitions thathave been found along the room-temperature isotherm.1 Theone at 12 GPa is fromb to « phase and is described as beingconformational and martensitic. It apparently occurs withoutan abrupt volume change. The other one(« to f) is discon-tinuous, occurs at 27 GPa, and is accompanied by a 4% vol-ume decrease.

The presence of phase transitions may have importantconsequences for the detailed understanding of detonation inHMX and its formulations. For example, the peak pressureof the detonation wave will depend on whether the transitionoccurs rapidly according to Zeldovich–Von Neumann–Doering theory. The question is whether shock-loading con-ditions give sufficient time for the nucleation and growth ofa phase before the reaction starts to chemically alter andconsume the HMX. Like shock loading, Isentropic compres-sion experiment(ICE) loading also has the ability to dynami-cally probe phase transitions at a high strain rate and yet ithas the potential to achieve significantly higher pressureswithout provoking reaction in the sample.2

There are numerous excellent references on the ICEtechnique so this will not be described in detail here.3,4 Verybriefly, an enormous electrical current pulse is used to drivea compressive ramp wave in a conductive aluminum sub-strate via the Lorentz force. This ramp wave(duration ap-proximately 200 ns) propagates from the substrate into HMXsamples of carefully quantified thickness, ranging from about400 to 600mm in thickness. The HMX samples5 were of twocrystal orientations(010) and (110) and were tamped usingoptically transparent windows[Poly(methymethacrylate)(PMMA), LiFs100d, and NaCls100d were all used in thisparticular experiment]. The velocity of the respective HMX/window interface was measured with VISAR(Velocity Inter-ferometer System for any Reflector) interferometry equip-ment from VALYN VISAR (Albuquerque, NM). Peakinterface velocities on the order of 2.5 km/s were obtainedin the case of NaCl windows and 2.1 km/s in the case of LiF

windows, indicating that the peak stress achieved in HMXwas on the order of 40 GPa and 49 GPa, respectively. Usingthe current history, the VISAR data, and detailed thicknessmeasurements of aluminum substrate and HMX samples, theVISAR data are simulated using the Trac-II magnetohydro-dynamic code.

In order to accurately assess the occurrence or absenceof the phase transition, it is important to be able to model it.An HMX loading curve based on the quotidian equation ofstate(QEOS) fit of the isotherm data is used.6,7 Although the12 GPa (b to «) transition will probably not show up inVISAR records due to its lack of volume dependency, the27 GPa transition should show up if it occurs. Our use of theQEOS form assumes that the transition goes to completioninstantaneously.

Comparison of simulation to experiment for thes110dorientation of HMX using LiFs100d windows shows no ob-vious indications of a phase transition for this orientation. Onthe other hand, in Fig. 1, we see that the data for the thickersample of thes010d HMX orientation shows some hint of

a)Electronic mail: hare2@llnl.govb)Present address: University of Maryland, Dept. of Mechanical Engineer-

ing, College Park, MD 20742.FIG. 1. s010d HMX with NaCl window, VISAR data vs Trac-II simulationusing QEOS for HMX.

APPLIED PHYSICS LETTERS VOLUME 85, NUMBER 6 9 AUGUST 2004

0003-6951/2004/85(6)/949/3/$20.00 © 2004 American Institute of Physics949 This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:

131.193.242.21 On: Thu, 27 Nov 2014 21:13:54

anomaly at about 2.1 km/s, but it is not nearly at the levelindicated in the simulation, suggesting that the transitionmay be sluggish and very incomplete on the time scale of theexperiment. One complication is that we used NaCl windowsfor our s010d HMX data. NaCl is known to have a phasetransition at 26 GPa,8 close to the HMX phase transition.

To further investigate the possible influence of NaCl, wededuced an accurate pressure drive for simulations using theHayes backwards integration technique9,10 on VISAR datafrom a reference interface consisting of LiFs100d windowdirectly on the aluminum substrate. The Al/LiF deducedpressure drive above was used in Trac-II to simulate theVISAR data from a NaCls100d window directly on the sub-strate. These simulations use a simple linear NaCl Hugoniotequation-of-state lacking any phase transition information.The Al/NaCl simulation and Al/NaCl data displayed in Fig.2 disagree at about 2.1 km/s. Figure 3 tells the same story inanother way: Here, the Al/NaCl VISAR data were backwardintegrated to obtain a pressure drive and then compared withthe more reliable aforementioned pressure drive deducedfrom the phase-transition-free Al/LiF VISAR data. Lackinga phase transition in NaCl, the two drives should be identicalwithin experimental and computational uncertainties. Thediscrepancies illustrate that the phase transition in the NaClwindow can obscure accurate observations regarding a27 GPa phase transition in HMX.

It should be pointed out that the lack of evidence for aphase transition in thes110d data does not negate the possi-bility of its occurrence in thes010d or other orientations asthere are examples where the crystal orientation relative towave propagation direction is believed to be an importantfactor in inducing phase transformation, as in KCl,11 graphiteto diamond,12 and CdS.13

Finally, we derived a best-fit linear Hugoniot for unre-acted HMX based on our HMXs110d data using Trac-IIsimulations. The constant gamma/volume assumption wasused with aG0 of 1.20 based on thermodynamic data fromliterature. The Hugoniot given byUs=3.45 km/s+1.90*UPfit our data well in the range up to interface velocities of

1.7 km/s. By contrast, a Hugoniot computed by a least-squares fit of existing shock compression data from LosAlamos10 for both solvent pressed HMX (density1.891 g/cc) and single crystal HMX(density 1.900 g/cc)gives Us=3.31 km/s+1.55*UP. Further work is needed tosatisfactorily obtain an accurate unreacted Hugoniot at highpressures.

In summary, both thes110d and s010d orientations ofHMX have been taken up to at least 40 GPa using the ICEtechnique. If the 27 GPa« to f phase transition in HMXdoes indeed occur, it is sluggish. Thes010d orientation needsto be re-examined using LiF windows, which do not have theadditional complication of having their own phase transitionas the NaCl does.

The authors would like to acknowledge the outstandingtechnical assistance of Allen J. Elsholz and Frank Garcia ofLawrence Livermore National Laboratory as well as thetechnical staff of theZ-machine facility at Sandia NationalLaboratories. They also thank K. S. Vandersall and L. G.Green for their expert lapping of the HMX samples. Theauthors are grateful to Dennis B. Hayes for helpful discus-sions. Y. M. Gupta made us aware of relevant works onshock-induced phase transitions. The authors thank R. J.Simpson and P. T. Springer of LLNL for critical funding andsupport.

1C-H Yoo and H. Cynn, J. Chem. Phys.111, 10229(1999).2D. E. Hare, D. B. Reisman, F. Garcia, L. G. Green, J. W. Forbes, M. D.Furnish, Clint Hall, and R. J. Hickman inShock Compression of Con-densed Matter-2003, edited by M. D. Furnish( AIP, Melville, N.Y. 2004).

3D. B. Reisman, A. Toor, R. C. Cauble, C. A. Hall, J. R. Asay, M. D.Knudson, and M. D. Furnish, J. Appl. Phys.89, 1625(2001).

4C. A. Hall, J. R. Asay, M. D. Knudson, W. A. Stygar, R. B. Spielman, T.D. Pointon, D. B. Reisman, A. Toor, and R. C. Cauble, Rev. Sci. Instrum.72, 3587(2001).

5H. H. Cady and L. C. Smith, Studies on the polymorphs of HMX, LosAlamos Scientific Laboratory report, Los Alamos, NM, LAMS-2652(TID-4500, 1961).

6R. M. More, K. H. Warren, D. A. Young, and G. B. Zimmerman, Phys.Fluids 31, 3059(1988).

7D. A. Young and E. M. Covey, J. Appl. Phys.78, 3748(1995).

FIG. 2. Simulation of the Al/NaCl VISAR interface, vs the actual datarecord. The phase-transitionless linear Hugoniot equation-of-state was usedto simulate the NaCl.

FIG. 3. Comparison of the pressure drives created by the backward integra-tion of the Al/NaCl and Al/LiF VISAR records. The linear Hugoniot wasused for the NaCl.

950 Appl. Phys. Lett., Vol. 85, No. 6, 9 August 2004 Hare et al.

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:

131.193.242.21 On: Thu, 27 Nov 2014 21:13:54

8J. Wackerle and H. L. Stacy inShock Compression of Condensed Matter–1989, edited by S. C. Schmidt, J. N. Johnson, and L. W. Davidson(Elsevier Science, New York, 1990).

9D. B. Hayes, Sandia National Laboratories Report Albuquerque, NM,SAND200I-1440(2001).

10LASL Shock Hugoniot Data, edited by S. P. Marsh( University of Cali-fornia Press, Berkeley, Calif., 1980), p. 335, 595–596.

11D. B. Hayes, J. Appl. Phys.45, 1208(1974).12D. J. Erskine and W. J. Nellis, Nature(London) 349, 317 (1991).13M. D. Knudson and Y. M. Gupta, J. Appl. Phys.91, 9561(2002).

Appl. Phys. Lett., Vol. 85, No. 6, 9 August 2004 Hare et al. 951

This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to IP:

131.193.242.21 On: Thu, 27 Nov 2014 21:13:54