Recent theoretical and experimental studies have produced several unusual and interesting results on dense lithium, the first metal in the periodic table. These include the deviation from simple metal behaviour, superconductivity at 17 K, and a metal to semiconductor transition 1-5 . Despite these efforts, at present there is no agreement on the location of the highpressure solid phases and melting curve of Li, and there is no clear picture of its phase diagram above 50 GPa (refs 4-7). Using powder and single-crystal high-pressure diffraction techniques, we have mapped out the lithium phase diagram up to 130 GPa over a wide temperature range between 77 and 300 K. Whereas the melting temperatures of materials usually rise under pressure, and even the lightest condensed gases, hydrogen and helium, melt at temperatures of the order of 10 3 K at 50 GPa (refs 8,9), we find that at these pressures lithium remains a liquid at temperatures as low as 190 K, by far the lowest melting temperature observed for any material at such pressure. We also find that in its solid state above 60 GPa, lithium adopts three novel and complex crystal structures not previously observed in any element. Estimates of the zeropoint energy suggest that quantum effects play a significant role in shaping the lithium phase diagram.The familiar properties and states of matter can be markedly modified by applying pressure and temperature. Besides those encountered in daily life (gas, liquid and solid), some exotic states, for example superfluids or superconductors, can be observed. Quantum effects, the energies of which are very small on an everyday scale, are responsible for the formation of these unusual forms of matter. To create any of these states, low temperatures are needed to decrease the energy of the system to the level where the quantum effects become dominant. Conversely, by applying pressure, and thereby bringing the atoms closer to each other, it is possible to increase the kinetic energy (that is, the zero-point energy) of the system. If the other energy terms that make up the total energy increase more slowly with pressure than the zero-point energy, it might be possible to reach a compression at which the quantum effects play the dominant role 10 . One of the obvious consequences of the zero-point energy being comparable to or in excess of differences in characteristic structural energies per atom would be melting of the solid under compression (cold melting) 10,11 . For light elements, such as hydrogen, melting influenced by the zero-point energy is expected to happen even at T = 0 (at compressions which are at present beyond the capabilities of experimental techniques), leading to a metallic liquid ground state with exotic properties 12 .Is it then possible to create a metallic liquid ground state in systems other than dense hydrogen? Most metallic elements with strong interatomic interactions are solids under normal conditions,

High-pressure high-temperature synchrotron diffraction measurements reveal a maximum on the melting curve of Na in the bcc phase at approximately 31 GPa and 1000 K and a steep decrease in melting temperature in its fcc phase. The results extend the melting curve by an order of magnitude up to 130 GPa. Above 103 GPa, Na crystallizes in a sequence of phases with complex structures with unusually low melting temperatures, reaching 300 K at 118 GPa, and an increased melting temperature is observed with further increases in pressure.

The long-unknown crystal structure of Bi-III has been solved. It comprises a body-centered-tetragonal (bct) "host" and a bct "guest" component made up of chains that lie in channels in the host; the guest is incommensurate with the host along the tetragonal c axis. Diffraction data for Sb-II reveal that it too can be fitted with the same composite structure. The structures of these two high-pressure phases of Bi and Sb are similar to those reported recently in the alkaline-earth metals Ba and Sr.

Recent advances in high-pressure experimental techniques have yielded high-quality x-ray diffraction data for the high-pressure phases of the group-15 elements Bi, Sb and As, and have made it possible to solve several longstanding problems in their structures. In particular, several complex incommensurate host-guest structures have been identified. This paper reviews the present state of knowledge of the structural transition sequences for these elements at high pressure and room temperature, including a summary of previous work, a detailed presentation of the new structures, and revised equations of state.