A Livermore software program called CHEETAH, an important, even indispensable tool for energetic materials researchers worldwide, was made more powerful in the summer of 1997 with the release of CHEETAH 2.0, an advanced version that simulates a wider variety of detonations. Derived from more than 40 years of experiments on high explosives at Lawrence Livermore and Los Alamos national laboratories, CHEETAH predicts the results from detonating a mixture of specified reactants. It operates by solving thermodynamic equations to predict detonation products and such properties as temperature, pressure, volume, and total energy released. The code is prized by synthesis chemists and other researchers because it allows them to vary the starting molecules and conditions to optimize the desired performance properties. One of the Laboratory's most popular computer codes, CHEETAH is used at more than 200 sites worldwide, including ones in England, Canada, Sweden, Switzerland, and France. Most sites are defense-related, although a few users, such as Japanese fireworks researchers, are in the civilian sector. In the U.S., the software has become the Department of Defense's preferred code for designing new explosives (Figure 1) and, to a lesser extent, propellants. (The Livermore work is supported under a Memorandum of Understanding between the Departments of Defense and Energy and is administered through the Office of the Secretary of Defense, Office of Munitions.) CHEETAH is also used by many defense contractors, such as Lockheed Martin and Thiokol, and by small detonation companies. CHEETAH was developed in 1993 by Livermore chemist Larry Fried and his colleagues at the Laboratory's High Explosives Applications Facility (HEAF) in an effort to update the long-standing TIGER thermochemical code. TIGER, in turn, was a derivative of the Laboratory's original RUBY code from the 1960s. The goal in creating CHEETAH, says Fried, was to make the use of thermochemical codes more attractive to high-explosive formulators through fast, yet scientifically rigorous codes, convenient user interface, and time-saving features such as a library of 200 starting reactants and 6,000 possible products.

Using CHEETAH CHEETAH's graphical user interface is designed for both Macintosh and Microsoft Windows or Windows NT operating systems. The interface sports three "views." The first view is the input window, from which most commands are entered. The second view is the main output file, in which text is displayed in black when CHEETAH is inactive and red when a calculation is being performed. The third view provides a concise summary of the calculations. Advanced Windows and Macintosh users can also access CHEETAH's command line interface. This command interface is the only form of input possible on systems running the UNIX operating system. The user first chooses the starting reactants by clicking on the reactant button represented by an Erlenmeyer flask icon. CHEETAH's database of starting reactants, which includes the most frequently used explosives and binders, saves the user the inconvenience of looking up thermodynamic constants for each reactant. Next, the user chooses one of three different kinds of calculations corresponding to a high-explosive detonation or to the firing of an artillery gun or a rocket. Runs are accomplished by clicking on the green-light icon. Alternatively, the user could employ the automatic formulator that adjusts the relative proportions of starting materials to match desired performance. For example, a shaped charge designed to penetrate armor needs to deliver its energy as quickly as possible, say, in 10 microseconds. By contrast, high explosives used in rock blasting must deliver energy more slowly, over tens of milliseconds. In this way, researchers can use CHEETAH to "test" new high-explosive formulations without resorting to actual small-scale tests. The States menu allows the user to specify the thermodynamic states that CHEETAH will calculate. For example, the user can determine the total energy of detonation or the energy of explosion at constant volume. Updates Expand Capabilities In 1996, the Livermore team released CHEETAH 1.40, which had extensive improvements to stability and user-friendliness and included advanced features to make its calculations more reliable over a wider range of material types and applications. Fried says that the most significant improvement in CHEETAH 2.0 is the addition of chemical kinetics, which should help greatly in treating nonideal (slow) detonation processes. The addition of a chemical kinetic framework, based upon modern Wood-Kirkwood detonation theory, allows for modeling of time-dependent phenomena such as partial detonation. Such nonideal phenomena are often poorly modeled by traditional Chapman-Jouguet thermodynamic theory, the basis for most of CHEETAH's calculations. Chapman-Jouguet theory assumes that thermodynamic equilibrium of the detonation products is reached instantaneously and that all products are consumed completely. In truth, actual situations may give different results because some components of the explosive react too late to drive the detonation front and because heat flows too slowly to bring all components into thermal equilibrium. Fried has found that with chemical kinetics, CHEETAH can predict the detonation velocity of slowly reacting materials such as PBXN-11 (a mixture of the explosive RDX and aluminum, ammonium perchlorate, and rubber binder, with a detonation velocity of 8 millimeters per microsecond) to within 0.2 millimeters per microsecond. A calculation ignoring kinetics is only accurate to within 2 millimeters, and thus, CHEETAH improves prediction of detonation velocity tenfold. CHEETAH and Weapons Stewardship The new chemical kinetics capability is very important for simulating the insensitive high explosives used in nuclear weapons. Simulating these materials has traditionally been difficult because they are much slower to react than classical high explosives. Yet, realistically modeling insensitive high explosives has acquired much greater importance in the current era of no nuclear testing and with the advent of the Department of Energy's Stockpile Stewardship and Management Program. "As the Laboratory's mission changes to nuclear weapon stewardship, we need to change our tools from those focusing on design to those looking at aging," says Fried. "High explosives change over time, and we need to know more about how those changes could affect their performance." Toward that end, a major effort was launched last year, in collaboration with Livermore computer scientists Steve Anderson and Shawn Dawson, to link CHEETAH to the extensive hydrodynamic codes of DOE's Accelerated Strategic Computing Initiative. The goal is to create more complete models of high-explosive detonations. User Satisfaction At Livermore, CHEETAH is being used to help guide the work of both synthesis and formulating chemists in a molecular design process similar to that found in pharmaceutical research (see June 1997 Science & Technology Review, pp. 4-13). It is part of the Laboratory's effort to provide more rigorous scientific structure for a field long dominated by intuition and trial and error. Fried reports high satisfaction among CHEETAH users. He notes that in the current era of constrained funding, the software can take the place of many actual experiments, thereby saving money by permitting the user to see the result of different formulations. And when cost savings are combined with safer and faster operation of experiments, there is even greater cause for user satisfaction. What's more, the program, now supplied on a CD-ROM, is free of charge to researchers in nonsensitive nations. There is also no charge for customer support in the form of e-mail dialogues with Fried. User problems and their resolutions are posted electronically in the hope of alerting users to commonly encountered problems. Fried and fellow researchers Clark Souers and Michael Howard continue to update and strengthen CHEETAH's capabilities. Fried is considering establishing a World Wide Web site so that a researcher lacking good computer resources could log onto one of Lawrence Livermore's smaller computer systems. A Web site could also provide a forum for discussions with users as well as provide ready electronic access to the 300-page manual. --Arnie Heller

Key Words: Accelerated Strategic Computing Initiative (ASCI), Chapman-Jouguet thermodynamic theory, CHEETAH, High Explosives Applications Facility (HEAF).

For further information contact Laurence Fried (510) 422-7796 (fried1@llnl.gov).