Figure 5: Structural data of types-I and -II GC as a function of pressure.

Because the interplanar spacings of GC are in broad agreement with the (002), (100)/(101) and (110)/(112) d-spacings of graphite, we use the d-spacing notation of graphite to represent the interlayer and intralayer distances of GC. (a) Interlayer distances analogous to (002) graphite. (b) Intralayer distances similar to (100)/(101) and (110)/(112) graphite. (c) Normalized interlayer distances. Open squares are data on type-I GC by ref. 11. (d) Normalized distances similar to graphite (100)/(101) and (110)/(112). Solid grey diamonds and dashed curves of graphite intralayer distances are calculated from ref. 59. From these data, volume compression of DMLG was evaluated. Note that the DMLG component of type-II GC appears slightly less compressible than type-I GC. (e) Interlayer distances derived from the first diffraction peak positions as a function of pressure (similar to graphite-(002)). The corresponding interlayer distances of graphite⁵⁹ are plotted as a dashed line for comparison. (f) Average first- (open circles) and second-neighbour (open diamonds) distances, derived from the G(r), as a function of pressure. The bond angles (solid squares) are derived from the ratio of the first- and second neighbour distances according to ref. 30. The precisions are less than 0.002 Å for bond lengths and 0.2° for bond angles. Errors in pressure are typically less than 0.15 GPa. Errors in Q position are estimated from the standard deviations in fitting the first two peaks in Fig. 4a, b and c. Errors in c, d, e are propagated from the uncertainties in Q. Errors in f are derived from the standard deviations in fitting the first two peaks in Fig. 4d. The current error bars (s.d.) are displayed at the lower right corner in (a–d) and plotted at the data in (e,f).