Our opportunity to directly sample the deeper mantle is limited to the investigation of rocks and minerals delivered to the surface by magmas, such as kimberlites. Kimberlites bring to the surface mantle xenoliths formed at depths of up to 250 km, while the Earth's deepest sublithospheric minerals are found as inclusions in diamonds. During crystallisation, a small proportion (<5%) of diamonds captures tiny inclusions of other minerals. These minerals record information about the composition and processes occurring at depths in the range of 150–700 km and possibly even deeper. Less than 10% of all uncovered inclusions in diamonds belong to the sublithospheric mantle (Stachel & Harris, 2008), and, among these, majoritic garnet and ferropericlase are the dominant phases. Other minerals include broken-down davemaoite (former Ca–Si perovskite), bridgmanite (former Mg-perovskite), CF, NAL, and more rare inclusions. Recently unusual high-pressure phases including breyite, davemaoite, ice-VII, and ringwoodite have been discovered in inclusions in diamonds, making a breakthrough in our understanding of deep mantle petrology and composition.

Such discoveries would be impossible without new methods and analytical techniques, which have brought the possible size of investigated material from hundreds and tens of microns to just a few microns.

In this talk I am going to concentrate on new analytical advances and their applicability to diamond research. I will show examples of how the improvement of optical microscopy and Raman spectroscopy allows us to identify inclusions even a few hundred microns under the surface, how cutting-edge computer tomography allows us to identify multi-phase inclusions, and how synchrotron X-ray diffraction permits us to solve the structure of a 10-micron lamella within a complex multi-phase inclusion. I will conclude with how these new opportunities are helping to drive mantle research forward.

Stachel, T. & Harris, J. W. (2008). Ore Geology Reviews 34, 5-32.