THE only solid carbon dioxide most of us know is the dry ice we use for cooling or in Halloween witch's brew. But researchers have known one other form of it for some time and recently isolated several more in the laboratory. One of the new-found forms is nearly as hard as diamond, and indications are that it will conduct heat just as well, too. Basic research in Livermore's mission-related work has led to yet another surprising and important scientific discovery.
Carbon dioxide (CO2) is a simple and nearly inert molecule. It is abundant in the atmosphere of Earth and other planets and is also a major byproduct of high-explosive detonations. Understanding how it and the other byproducts of high-explosive detonations—water, carbon, and nitrogen—behave at detonation pressures and temperatures is important for creating accurate models for codes that simulate nuclear weapons performance. Studying the optical, mechanical, and energetic properties of the individual reaction products at high pressures and temperatures provides important experimental data for improving Livermore's models to predict the performance of high explosives.
The split second of a high-explosive detonation may produce temperatures as high as 5,500 kelvin and pressures up to half a million times that of Earth's atmosphere. Chemists and material scientists have suggested a number of molecular phases for CO2 at various high pressures and temperatures, but their structure and stability have not been fully characterized. That is changing, however.

In experiments using Livermore's diamond anvil cell to slowly increase pressures up to millions of Earth atmospheres, a team led by physicist Choong-Shik Yoo has verified two forms of solid CO2 never before seen in the laboratory, known as CO2-IV and CO2-V. The team also has experiments under way on a third form, CO2-II, which has weaker chemical bonds. All three forms are very different from one another as well as from the two previously known molecular phases, CO2-I (dry ice) and CO2-III.
The phase diagram for carbon dioxide shows the pressures and temperatures at which changes in phase occur for CO2 based on results to date. All five phases are solid in one form or another, but they differ greatly in their molecular configuration and crystal structure and, more importantly, in the nature of their chemical bonding and the strength of intermolecular interactions. The intermolecular interactions change from weak quadrupolar interactions in the linear molecular phases (I and III), to relatively strong dipolar interactions in the bent-molecular phase IV, and eventually to strong covalent bonds in the polymeric phase V. (Results of the most recently discovered phase, CO2-II, are just beginning to be understood.)
The evolving intermolecular interactions are in line with recent theoretical and experimental results that suggest that many simple molecules will become polymeric and even metallic at high pressures and temperatures. These findings have important implications for the chemistry of high-explosive detonations, which entail similar temperatures and pressures.
Carbon dioxide crystallizes into CO2-I with a pressure of 1.5 gigapascals at room temperature. Increased pressure on CO2-I produces CO2-III, which becomes a highly strained, high-strength solid above 20 to 30 gigapascals. CO2-III is fairly stable at room temperature to about 70 gigapascals.
The Livermore team turned up both the pressure and temperature to isolate CO2-V, the polymeric solid, using the synchrotron facility at Grenoble, France. They found that CO2-V has extremely low compressibility, similar to that of cubic boron nitride. This suggests that the carbon dioxide polymer could be used as a superhard material just as cubic boron nitride and diamond are. What is especially interesting about CO2-V is that it shows nonlinear optical behavior. The frequency of light that passes through it doubles, which may lead to a new class of generating materials for high-power lasers.
More recently, Yoo's team discovered CO2-IV at pressures between 12 and 30 gigapascals. Its bent molecule is the precursor to polymeric CO2-V. Also, CO2-IV has proved to be optically nonlinear.

Crystallizing a Liquid
To create CO2-V, the team placed a small amount of condensed, liquid CO2 between the two tips of the diamond anvil cell and squeezed it to pressures of about 40 gigapascals, or 400,000 times the air pressure at Earth's surface. At the same time, they used a neodymium-doped yttrium-lithium-fluoride (Nd:YLF) laser to indirectly heat the CO2 to 1,800 kelvins.
When the compressed sample was heated, the team was surprised to see a visible emission of green light. The incoming laser light at the infrared wavelength of 1,054 nanometers had been frequency doubled to 527-nanometer light, which is green. The green light being emitted is the second harmonic of the laser light used to heat the sample.
Because CO2 absorbs the Nd:YLF light poorly, the team heated the sample indirectly by scattering micrometer-sized ruby chips in it and heating the chips with the laser. In other experiments, they used a platinum foil or a rhenium gasket for heating. The experiment was repeated 20 times and yielded the same results each time, regardless of the heating material used. "We wanted to verify that the frequency doubling we saw was not the result of contamination from the heating materials," says Yoo.
The sample, reduced to just a few micrograms, had become an extended solid phase of CO2. X-ray diffraction, Raman spectroscopy, and other forms of analysis showed that the carbon and oxygen atoms had rearranged themselves in chainlike patterns connected by single carbon-oxygen bonds, a structure similar to that of a high-temperature modification of quartz.
The team found CO2-IV by laser-heating CO2-III at pressures between 12 and 30 gigapascals to 1,000 kelvins and quenching the sample to 300 kelvins. They found that with increasing pressure, the interactions weakened. At about 80 gigapascals, the quenched sample collapsed into an amorphous solid. If laser heating was continued with increasing pressure, the CO2-IV transformed into polymeric CO2-V above 30 gigapascals.

Stabilizing CO2-V
If this new, very hard CO2-V can be stabilized at ambient temperatures and pressures, it will have many uses. New classes of high explosives, nonlinear optical materials with high thermal and mechanical stability, high-strength glass, and superhard materials for tools are all candidates. Crystals that can double the frequency of laser light from infrared to green would be valuable for Livermore's inertial confinement fusion energy program.
So far, once CO2-V has been created at high pressures and temperatures, it retains its structure at room temperature but only at pressures above 1 gigapascal. Below that pressure it collapses and is no longer a polymer. Nearly half a century ago, when scientists were trying to produce the first synthetic diamonds, the same problem arose. Diamond could be synthesized only at high temperatures and pressures until scientists learned its growth mechanism. Then they were able to synthesize it at lower pressures and temperatures using catalysis. Now diamond is routinely manufactured at ambient pressure with very little heat. Says Yoo, "We'll be looking for something like that with this new carbon dioxide."
Other applications of the team's research are truly out of this world: the experimental temperatures and pressures used to create CO2-V are comparable to those inside giant gas planets like Uranus and Neptune. Who knows what lurks in their interiors?
—Katie Walter

Key Words: carbon dioxide, diamond anvil cell.

For more information contact Choong-Shik Yoo (925) 422-5848 (

Back to October 2000 // Science & Technology Review 2000 // Science & Technology Review // LLNL Homepage