Growth of diamond in liquid metal at 1 atmosphere pressure

Abstract

Natural diamonds were (and are) formed (some, billions of years ago) in the Earth’s upper mantle in metallic melts in a temperature range of 900–1400°C and at pressures of 5–6 GPa1,2; indeed, diamond is thermodynamically stable under high pressure and high temperature (HPHT) conditions as per the phase diagram of carbon3. Scientists at General Electric invented and used a HPHT apparatus in 1955 to synthesize diamonds from melted iron sulfide at about 7 GPa and 1600°C4–6. There is an existing paradigm that diamond can be grown using liquid metals only at both high pressure (typically 5–6 GPa) and high temperature (typically 1300–1600°C) where it is the stable form of carbon7. Here, we describe the growth of diamond crystals and polycrystalline diamond films with no seed particles using liquid metal but at 1 atmosphere pressure, and at 1025°C, breaking this paradigm. Diamond grew at the interface of liquid metal composed of gallium, iron, nickel, and silicon and a graphite crucible, by catalytic activation of methane and diffusion of carbon atoms in the subsurface region of the liquid metal. Raman spectroscopy with 13C-labeling proves that methane introduced into the growth chamber is the carbon source for many of the regions of newly grown diamond. The new growth diamonds were studied by Raman spectroscopy, scanning and transmission electron microscopy, X-ray diffraction, and photoluminescence. Growth of (metastable) diamond in liquid metal at moderate temperature and 1 atm pressure opens many possibilities for further basic science studies and for the scaling of this type of growth.

Delivering C, B, and N Atoms into Liquid Metals in Which They Are Normally Insoluble

“…These ideas occurred to me while thinking about new methods to grow diamond (such as with or without diamond seeds in liquid metal; we have not yet tried the methods described below for growth of diamond but plan to in the future). It is my hope that what I describe is useful for many syntheses and for new science.…”

“…As an example of new science that might result from methods such as those outlined above, I mention recent modeling results obtained with AIMD (ab initio molecular dynamics) by colleagues at UNIST. In modeling, e.g., C atoms and Si atoms in liquid Ga (from a collaboration related to seedless diamond growth from a liquid metal containing majority component Ga now being studied in our laboratory), it has been learned that the two 2p electrons of a C atom strongly delocalize in the liquid Ga (as do not only the two 3p electrons of Si but also its 3s electrons). This was discovered from a study of the projected density of states of the “solvated” C and Si atoms in either a liquid Ga slab or bulk liquid Ga systems .…”