146 Z. Kristallogr. 225 (2010) 146–152 / DOI 10.1524/zkri.2010.1194 # by Oldenbourg Wissenschaftsverlag, München Evidence for the existence of a PbCO3-II phase from high pressure X-ray measurements Robert Minch*, I, Lars PetersI, 1 , Lars EhmII, III, Karsten KnorrIV, Oleg I. SiidraV, Vitali PrakapenkaVI, Przemyslaw DeraVI and Wulf DepmeierI I II III IV V VI Institut für Geowissenschaften, Christian-Albrechts-Universität zu Kiel, D-24118 Kiel, Germany Mineral Physics Institute, Stony Brook University, Stony Brook, NY 11794, USA National Synchrotron Light Source, Brookhaven National Laboratory, Upton, NY 11973, USA BrukerAXS GmbH, 76187 Karlsruhe, Germany Department of Crystallography, St. Petersburg State University, University Emb. 7/9, 199034 St. Petersburg, Russia GSECARS, University of Chicago, Chicago, IL 60637, USA Received June 16, 2009; accepted January 18, 2010 High pressure / X-ray diffraction / Phase transition / Lead carbonate Abstract. The high-pressure room temperature behavior of PbCO3 was investigated by angle-dispersive synchrotron powder diffraction up to 16.16(5) GPa. A phase transition to a high-pressure polymorph II of lead carbonate was observed at a pressure of approximately pc ¼ 8.7 GPa. Thereby, the symmetry is reduced from orthorhombic Pmcn to monoclinic P121/c1. The cell parameters at p ¼ 8.90(5) GPa are a ¼ 5.101(1)  A, b ¼ 8.303(3)  A,   c ¼ 5.602(2) A, b ¼ 89.51(4) , Z ¼ 4. The transformation is supposed to be of 2nd order. Introduction PbCO3 is one of the carbonates (CaCO3, BaCO3, SrCO3, and PbCO3) with aragonite-type structure at ambient conditions (De Villiers, 1971; Chevrier et al., 1992). According to the pressure homologue rule, stoichiometric compounds with larger cations are often considered as model structures for the behavior of phases with smaller cations at higher pressures (Liu and Basset, 1986). This implies, that all carbonates with aragonite structure should have similar phase diagrams, but would show phase transitions at different p–T conditions. High-pressure phase transitions of CaCO3 (aragonite), SrCO3, PbCO3 and BaCO3 were reported by various authors, e.g. Gillet et al., 1993; Lin and Liu, 1997a; Holl et al., 2000; Ono et al., 2005, 2005b; Ono, 2007. Lin and Liu (1997b) found that both, cerussite (PbCO3) and strontianite (SrCO3), show a post-aragonite transformation at 17 and 35 GPa, respectively. Catalli et al. (2005) observed a phase transition from an orthorhombic to a trigonal structure in PbCO3 at 15 GPa using IR spectroscopy. * Correspondence author (e-mail: robert@min.uni-kiel.de) 1 Present address: Institut für Kristallographie, RWTH Aachen University, 52056 Aachen, Germany Recently, it was proposed that PbCO3 undergoes an unexpected phase transition at about 8 GPa to a previously unknown phase II (Minch et al., 2010). The existence of PbCO3-II sets lead carbonate apart from the other aragonite-type carbonates. It was suggested previously that the structure of lead carbonate might not be “truly” aragonitetype (Durman et al., 1985). In contrast to other divalent elements (Ba, Ca, Sr), Pb has a 6s2 lone electron pair in its divalent state which sets it apart from the closed-shell earth-alkaline cations. The element lead might be regarded as a relativistic alkaline-earth metal (Mudring, 2006). The lone-pair occupies an inert orbital in the ligand sphere and thus can become stereochemically active (Siidra et al., 2008). The “stereochemically” active lone pair determines many of the physical properties of Pb(II) compounds, but its origin and nature have not yet been conclusively established (Stoltzfus et al., 2007). The studies carried out by Rieger and Mudring (2007) suggest that cation-anion interactions are the true driving forces responsible for lone pair distortions. To evidence the existence of the PbCO3-II phase and determine its structure, angle dispersive X-ray powder diffraction experiments were performed and their results are reported in this work. Experimental Extra pure PbCO3 was used for the experiments (99.999%, Alfa Aesar GmbH). High-pressure powder Xray diffraction patterns were taken up to 16.16(5) GPa at GSECARS beamline 13-ID-D at the Advanced Photon Source using the in situ high-pressure angle-dispersive Xray diffraction system. The beam size at the sample was 30  30 mm2. High-pressure patterns were collected at a wavelength of 0.3344  A using a mar165 CCD detector. The exposure time per image was about 10 s. Pressure was applied using a diamond anvil cell (DAC) with 300 mm diameter diamond culets. A Re gasket was preindented in the DAC to a thickness of about 40 mm 147 Evidence for the existence of a PbCO3-II with a hole of 100 mm in diameter serving as a sample chamber. The sample and three ruby chips (5 mm in size) for pressure calibration were loaded into the sample chamber. The pressure was determined by the ruby fluorescence method (Mao et al., 1978). Two DACs were used in the experiments. In the first cell argon was used as a pressure transmitting medium. A trace amount of Pt was mixed with the sample in an agate mortar, to act as an absorber for laser radiation for possible annealing of the sample, which, however, was not realized. Nitrogen was used as pressure transmitting medium in the second cell. Geometry parameters for the radial integration of the two-dimensional data and the sample to detector distance (202.604(1) mm) were determined from a CeO2 sample (Jephcoat et al., 1992) and the transformation into standard one-dimensional powder patterns was carried out using the FIT2D software (Hammersley et al., 1996). Lattice and structural parameters were derived using the Rietveld method (Rietveld, 1967) within the TOPAS suite of programs (Coelho, 2000). The background was described by a tenth-order polynomial and the peak profiles were modelled with pseudo-Voigt functions (Thompson et al., 1987). For PbCO3-II, it was necessary to add to this function an anisotropic broadening term to model the observed broadening with increasing pressure. In our analysis we found that the convolution with a suitable fourth-order spherical harmonic function reproduced this anisotropic broadening well. To partly compensate for the low scattering power of oxygen and carbon compared to lead restraints were introduced for C––O bond lengths (1.285  0.001  A) and for O––C––O angles (120  0.1 ). In addition, the z-parameters of the oxygen and carbon atoms were constrained to keep carbonate groups parallel to each other and perpendicular to the z-axis. The isotropic displacement parameters were constrained for all atoms to have identical values (McCusker et al., 1999). The strong preferred orientation of the gasket material Rhenium was accounted for by using a March-Dollase approach (Dollase, 1986). A second-order Birch-Murnaghan equation-of-state was used to determine the unit cell volume V0 at a pressure of 0 GPa and the bulk modulus B0 (Birch, 1978). Fig. 1. General projection of the crystal structure of PbCO3 along the c-axis at 0 GPa (left) (Chevrier et al., 1992) and 8.90(5) GPa (right). Note the antiparallel shift of neighbouring CO32 groups along b under compression. The combined effect of high compressibility along the c-axis and shift along b preserves the distance between the C––C atoms of the neighbouring CO32 groups. Fig. 2. Pressure dependence of diffraction patterns of PbCO3 from 0.30(5) GPa to 16.16(5) GPa at ambient temperature. The patterns are shifted with increasing pressure for clarity. Nitrogen was used as pressure transmitting medium in the experiments up to 11.2(2) GPa. The measurements at 9.50(5), 13.0(1) and 16.16(5) GPa were carried out using another DAC loaded with argon as pressure transmitting medium and Pt mixed with the sample for potential laser heating experiments. Tick marks denote reflection positions for Re (upper row), Pt (middle row) and Ar (lower row) and pertain only to measurements at 9.50(5), 13.0(1) and 16.16(5) GPa. Results At ambient conditions, lead carbonate is orthorhombic, space group Pmcn (standard setting: Pnma). The cell parameters reported by Chevrier et al., 1992, are a ¼ 5.179(1)  A, b ¼ 8.492(3)  A, c ¼ 6.141(2)  A, with Z ¼ 4. The lead atoms are nine-fold coordinated by oxygen and occupy Wyckoff position 4c. According to Chevrier et al., 1992, the carbonate groups are slightly aplanar and alternate along the c-axis. Carbon is at Wyckoff position 4c and oxygens are at 4c (O1) and 8d (O2) (Fig. 1). The evolution of the diffraction patterns as function of pressure up to 16 GPa at room temperature is presented in Fig. 2. Measurements up to 11.3(1) GPa were obtained using a DAC loaded with nitrogen as pressure transmitting medium. Three additional diffraction patterns taken at 9.5(2), 13.0(1) and 16.16(5) GPa were obtained using argon as pressure transmitting medium, and with the admix- 148 R. Minch, L. Peters, L. Ehm et al. Table 1. Refined unit cell parameters for PbCO3 as function of pressure. Standard deviations in parentheses were corrected by an estimated SCOR value of 3. p/GPa a/ A b/ A c/ A V/ A3 0 0.30(5) 5.179(1)a 8.492(3)a 6.141(2)a 270.082a 90a 5.173(1) 8.505(1) 6.111(1) 268.86(8) 90 4.50(5) 5.124(2) 8.428(2) 6.811(1) 250.89(8) 90 0.866 7.0(2) 7.40(5) 5.113(1) 5.113(3) 8.350(2) 8.350(3) 5.670(1) 5.669(1) 242.1(1) 242.6(1) 0.824 0.871 b/  c2 Pmcn 90 90 0.95 P121/c1 8.90(5) 5.101(1) Fig. 3. Observed (circles) and calculated (solid line) diffraction pattern for PbCO3 at 0.30(5) GPa. The difference and the tick marks for calculated reflection positions are plotted at the bottom. 8.303(3) 5.602(2) 237.3(1) 89.51(4) 0.756 9.5(2) 11.3(1) 5.07(1) 5.074(4) 8.23(2) 8.216(4) 5.56(1) 5.522(4) 231.7(8) 229.9(3) 88.11(9) 0.959 87.21(4) 1.105 13.0(1) 5.048(4) 8.127(8) 5.464(8) 223.7(4) 86.29(4) 0.673 a: Data taken from Chevrier et al., 1992 ture of a trace amount of Pt, as mentioned earlier. Figure 3 shows a plot of the structure refinement of lead carbonate at 0.30(5) GPa in space group Pmcn. All fits up to 7.40(5) GPa are of similar quality, thus compatible with the orthorhombic model. The fit to the experimental data at 8.90(5) GPa in the same space group looked reasonable only at a first glance, while a more critical examination  a  b Fig. 4. Observed (circles) and calculated (solid line) diffraction pattern for PbCO3 at 13.0(1) GPa in space group Pmcn (a) and. in space group P121/c1 (b) The differences are plotted at the bottom of the figures. Tick marks denote reflection positions for (rows from up to down) PbCO3, Ar, Re and Pt. showed small, but significant differences between observed and calculated patterns. From Raman spectroscopic data evidence for a phase transition to a phase II at about 8 GPa has been observed, thus between our data at 7.40(5) and 8.90(5) GPa (Minch et al. 2010). Therefore, the 8.90(5) GPa pattern was re-indexed using the TOPAS suite of programs (Coelho, 2000) and various lower symmetry models refined. The best agreement was found for the monoclinic space group P121/c1 (translationengleiche subgroup of Pmcn) with a monoclinic angle of b ¼ 89.51(4) . In order to demonstrate the symmetry breaking from orthorhombic to monoclinic, the results of structure refinements of the 13.0(1) GPa data in both space groups, Pmcn (a) and P121/c1 (b), are presented in Fig. 4. Table 1 lists the refined unit cell parameters at different pressures. All c2 values are in the order of magnitude around 1. Errors obtained by the refinement procedure were multiplied by an estimated SCOR value of 3 (Berar and Lelann, 1991), because of the well-known fact of systematically too small e.s.d.s from Rietveld refinements and image-plate data. Figure 5 shows the deviation of the monoclinic angle b from 90 as a function of pressure. The data were obtained by fitting all datasets from both experiments in space group P121/c1, down to the 0.30(5) GPa dataset. The critical pressure (pc) of the phase transition was estimated from the dependence of sin (b) vs. pressure to be 8.7 GPa (Fig. 5, lower inset). A linear fit of lg (90 -b) vs. lg (p–pc) is shown as an inset in Fig. 5. The slope of 0.58(9) is in good agreement with the assumption of a 2nd order phase transition, with (90 -b) as the order parameter in the frame of Landau theory. The symmetry change from Pmcn to P121/c1 via a second order phase transition is allowed by both, the Landau and the Lifshitz criteria (Stokes and Hatch, 1988). The symmetry relationship allows the development of a ferroelastic spontaneous strain. From the pressure dependence of the lattice parameters across the phase transition the components of the corresponding second rank tensor were calculated for the monoclinic setting as described by Ohashi et al., 1973. Only one component (e13 ¼ 0.5 ((a cos b/a0 sin b)  149 Evidence for the existence of a PbCO3-II Fig. 5. The deviation of the monoclinic angle b from 90 (90 -b) as function of pressure. Refinements have been carried out in space group P121/c1 for all measurements. The upper inset shows the plot lg (90 -b) vs. l g (p–pc ), from which a critical exponent g ¼ 0.58(9) for the order parameter Q  (p–pc) g can be deduced). The critical pressure Pc can be determined to be 8.7 GPa from the lower inset. If not indicated otherwise, the uncertainty corresponds to the size of the symbols used. (c cos b0/c0 sin b0)), where a, c and b are the high-pressure parameters and a0, c0 and b0 are extrapolated from the low pressure regime) contributes significantly to the scalar spontaneous strain es (Fig. 6). The changes of the other three (e11, e22, e33) are non-significant and around zero within experimental uncertainty. Refined atomic coordinates and bond lengths with angles for PbCO3-II in space group P121/c1 at 8.90(5) GPa with constrained z-parameters for oxygen and carbon atoms are given in Table 2 and 3, respectively. Table 2. Refined structural parameters for PbCO3-II from X-ray powder diffraction at 8.90(5) GPa/room temperature in space group P121/ c1 with constrained z-parameters for oxygen and carbon atoms. Bov is 0.7(2). Standard deviations in parentheses were corrected by an estimated SCOR value of 3. Atom x=a y=b z=c Pb 0.249(4) 0.074(1) 0.727(1) C 0.29(2) 0.72(1) 0.61 O(1) O(2.1) 0.27(2) 0.51(1) 0.56(1) 0.80(2) 0.61 0.61 O(2.2) 0.06(1) 0.81(1) 0.61 Fig. 6. Pressure evolution of the component e13 ¼ 0.5 ((a cos b/a0 sin b) – (c cos b0/c0 sin b0)) of the spontaneous strain. Using space group P121/c1 to fit experimental data collected at pressures above 13 GPa no longer yield results of satisfying quality. This points at a possible additional symmetry change in the pressure range between 13 and 16 GPa. According to Raman spectroscopic data (Minch et al., 2010), at pressures higher than 10 GPa, two highpressure phases (II, III) co-exist. Therefore, our evaluation of the dataset at 16.16(5) GPa, is most probably biased by an uncertain symmetry assignment and the results should be considered with due caution (see below). Hence, the cell parameters at this pressure are not shown in Table 1. The pressure dependence of the normalized lattice parameters a/a0, b/b0, c/c0 and the unit cell volume V/V0 is shown in Fig. 7. Note, the incompressibility of the a and b lattice parameters, in contrast with the continuously compressed c lattice parameter. In the limit of experimental resolution there is no significant discontinuity of neither the lattice parameters nor the volume, in full agreement with the 2nd order character of the phase transition; apparent jumps at about 9 GPa might well be due to the use of two different DACs (see above). Values of V0 ¼ 271(1)  A3 and B0 ¼ 50(1) GPa were obtained by a standard least squares fit. The linear axial compressibilities are ka ¼ 1 1 1 0.0017 GPa , kb ¼ 0.0023 GPa , kc ¼ 0.011 GPa . Table 3. Refined bond distances (in  A) and angles (in  ) for PbCO3-II at 8.90(5) GPa/room temperature in space group P121/c1 with constrained z-parameters for oxygen and carbon atoms. Standard deviations in parentheses were corrected by an estimated SCOR value of 3. Pb––O1 2.45(5) O1––C––O2.1 126(2) O2.2––Pb––O2.1 168(3) O1––Pb––O2.2 Pb––O1 Pb––O1 Pb––O2.1 Pb––O2.1 Pb––O2.1 Pb––O2.2 Pb––O2.2 Pb––O2.2 C––O1 C––O2.1 C––O2.2 2.63(6) 2.76(9) 2.45(9) 2.49(9) 2.69(9) 2.52(6) 2.65(6) 2.67(6) 1.26(6) 1.29(9) 1.35(8) O2.1––C––O2.2 O2.2––C––O1 O2.1––Pb––O2.1 O1––Pb––O2.1 O1––Pb––O2.1 O2.2––Pb––O1 O2.2––Pb––O2.1 O2.2––Pb––O2.1 O2.2––Pb––O2.2 O2.2––Pb––O1 O2.2––Pb––O2.1 O2.1––Pb––O1 116(3) 116(6) 73(1) 73(3) 52(3) 116(2) 106(3) 68(2) 120(2) 112(2) 110(2) 67(2) O1––Pb––O2.2 O1––Pb––O2.2 O1––Pb––O1 O1––Pb––O2.1 O1––Pb––O2.1 O2.2––Pb––O1 O2.2––Pb––O2.2 O2.2––Pb––O2.2 O2.2––Pb––O1 O2.2––Pb––O2.1 O2.2––Pb––O2.1 O2.1––Pb––O2.1 80(2) 94(3) 69(1) 147(2) 92(1) 145(2) 85(2) 67(1) 146(2) 67(2) 106(1) 86(2) O1––Pb––O1 O1––Pb––O2.2 O1––Pb––O2.2 O1––Pb––O1 O1––Pb––O2.1 O1––Pb––O2.1 O2.1––Pb––O1 O2.1––Pb––O2.2 O2.1––Pb––O1 O2.1––Pb––O2.2 O2.1––Pb––O2.2 O2.1––Pb––O2.1 70(2) 76(1) 74(2) 48(2) 140(2) 136(3) 112(3) 121(2) 116(3) 79(2) 51(3) 168(3) 119(2) 150 R. Minch, L. Peters, L. Ehm et al. -- Fig. 7. Evolution of the normalized lattice parameters a/a0 ( ), b/b0 (~), c/c0 (^) and the unit cell volume V/V0 (*) with pressure. The error bars for both, pressure and lattice parameters, correspond to the size of the symbols. The lines are curve fits of second-order BirchMurnaghan equation-of-states to the data. Results of two experiments are presented. Discussion There is a direct group-subgroup relationship between the space groups Pmcn and P121/c1. At ambient conditions three atoms are at special positions 4c (Pb, C, O1) and one (O2) is at general position 8d of space group Pmcn. In P121/c1, there is one non-equivalent site for the Pb and C atom and three sites for oxygen atoms (Table 1). The O2 oxygen site (8d) splits into two 4e general positions (designated O2.1 and O2.2 in Fig. 8). As can be seen from Fig. 8, the coordination of Pb atoms at ambient pressure is rather symmetric. The oxygen atoms are strongly bonded in the CO32 groups which leave only relatively weak bonds with the Pb2þ cation, thereby stabilizing a symmetric Pb2þ environment. The coordination number (9) of lead does not change across the phase transition. However, some Pb––O bonds become noticeably shorter (Fig. 8). Thus the coordination changes only slightly, with Pb––O bond distances varying from 2.45(5)  A to 2.76(9)  A. However, all changes of the bond lengths (Pb––O) are Fig. 8. Pb2þ cation coordination in the crystal structure of PbCO3 at 0 GPa (Chevrier et al., 1992) and 8.90(5) GPa. The O2 oxygen (designated as O2 at ambient pressure) site (8d) splits into two (designated as O2.1 and O2.2 atoms at 8.90(5) GPa) 4e general positions. The average errors at 0 and 8.9(5) GPa are 0.001 and 0.06  A, respectively. within 3s. The same observation is valid for the C––O bond lengths. They are varying from 1.26(8) to 1.35(8)  A. This is consistent within 3s with the reported C––O (1.2856(8)  A) bond length at ambient conditions (Chevrier et al., 1992). Hence, the changes reported here, if real, should be understood as a trend, rather than a statistically confirmed fact. The main structural change during the phase transition is, in fact, the combined effect of strong compression along the c-axis and the an anti-parallel displacement of the CO32 groups in the a-b plane (Fig. 1), such that the C––C distances are preserved during the transition (Fig. 9). This concerted rearrangement of CO32 groups leads to the reported slight change in the Pb coordination. An unconstrained refinement resulted in a slight tilting of the CO32 groups with respect to each other. However, as there are no published structures of carbonates with nonparallel carbonate groups, the structure of PbCO3-II was re-refined with constraints to keep the carbonate groups parallel as described in the experimental part. This did not change the overall R value, or any other significant parameter of the structure, at 8.90(5) GPa. Therefore, in Tables 2 and 3 we give the structural parameters resulting from this constrained refinement. Figure 9 shows the behavior of CO32 groups along the c-axis and evidences the displacive mechanism of the phase transition. Anti-parallel arrows in phase I symbolize the strong compression of PbCO3-I along the c-axis up to a point where the distance between the C-atoms of CO3-groups reaches a minimum of 2.87  A. The slightly canted anti-parallel arrows in PbCO3-II indicate that further compression of the c-axis results in a shear along the b-axis, thus preserving the minimum C––C-distance. One data point in the pressure range above 13 GPa apparently shows further shortening of the C––C distances. We believe this is an artefact due to the fact that we have refined the corresponding data in P121/c1, the space group of PbCO3-II. However, in Minch et al., 2010 evidence was Fig. 9. The evaluation of refined C––C distances with pressure. The dashed lines indicate the possible I–II phase boundary (pc) and the boundary between phase II and the two-phase region II þ III. For the meaning of antiparallel or canted antiparallel arrows in phase I and II, respectively, see text. The size of the symbols corresponds to errors in p. 151 Evidence for the existence of a PbCO3-II given for a two-phase region in this pressure range. Until now we have not been able to determine unambiguously the symmetry of PbCO3-III at 16.16(5) GPa with a combination of P121/c1 (for II) and either of two earlier proposed space groups for PbCO3-III (Catalli et al., 2005; Lin and Liu, 1997). As can be seen from Fig. 5, the pressure dependence of (90 -b) changes its behavior in the pressure range between 7–9 GPa, possibly indicating a deviation from a purely second order nature of the phase transition. Such a change in nature towards first-order behavior could be, for instance, due to coupling of the initial order parameter with a defect field in the vicinity of pc, when the correlation length diverges. The tentative phase diagram of PbCO3 based on Raman spectroscopic data shows a region of two coexisting polymorphs (I, II) in the 6.5–8 GPa pressure range (Minch et al., 2010), thus lending support to the idea of a slightly first order character of the phase transition. The V0 ¼ 271(1)  A3 obtained from the fits of a secondorder Birch-Murnaghan equation-of-state to the unit cell volume values is in good agreement with that published earlier by Chevrier et al., 1992 (V0 ¼ 270.082(6)  A3). The lattice compression is highly anisotropic, namely more than five times stronger along the c axis than along a or b. This behavior is found for other aragonite-type carbonates as well, e.g. for BaCO3, where the compression along c exceeds that along a and b even by a factor 10 (Holl et al., 2000). Note that in the aragonite-type structures “layers” of CO3-groups alternate with “layers” of cations along the c-axis, thus providing a clue to understanding the anisotropy. The pattern of phase transitions in PbCO3 differs definitively from those of other aragonite-type carbonates. In the case of witherite, BaCO3, mentioned above, a first-order transition at 7.2 GPa was observed by Holl et al., 2000 (Pmcn to P 31c). Santillán and Williams (2004) reported a similar post-aragonite phase transition in CaCO3 at 50 GPa. The space group of the high-pressure SrCO3 polymorph occurring at pressures above 10 GPa is P21212, as shown by Ono et al., 2005b. According to the pressure homologue rule, a post-aragonite phase transition in PbCO3 would be expected to occur at pressures somewhere between 10 and 50 GPa. The symmetry of PbCO3II (P121/c1) as reported here belongs to the same space group type as that of CaCO3-II. However, in the latter case, the transition starts from the calcite structure with space group R 3c. It is, however, possible that the observed PbCO3-II phase is a metastable transient phase like CaCO3-II or CaCO3-III rather than an equilibrium structure. From the data collected, this question cannot be answered unambiguously. From the fairly symmetric Pb2þ environment in PbCO3-II, we deduce that the lone electron pair does not significantly bias the high-pressure behavior of lead carbonate, at least not in the pressure range studied. To prove all phase boundaries in the phase diagram of PbCO3 (Minch et al., 2010), high pressure/temperature experiments would be needed. Acknowledgments. This research was supported by the Deutsche Forschungsgemeinschaft under project number KN 507/5-1 in the framework of the priority program: “Synthesis, ‘in situ’ characterization and quantum mechanical modeling of Earth Materials, oxides, carbides and nitrides at extremely high pressures and temperatures”. This work was performed at GeoSoilEnviroCARS (Sector 13), Advanced Photon Source (APS), Argonne National Laboratory. GeoSoilEnviroCARS is supported by the National Science Foundation – Earth Sciences (EAR-0622171) and Department of Energy – Geosciences (DE-FG02-94ER14466). Use of the Advanced Photon Source was supported by the U. S. Department of Energy, Office of Science, Office of Basic Energy Sciences, under Contract No. DEAC02-06CH11357. RF programme state contract # 02.740.11.0326. Special thanks go to anonymous reviewer for very helpful comments. References Berar, J.-F.; Lelann, P.: E.S.D.’s and Estimated probable error obtained in Rietveld refinements with local correlations. J. Appl. Crystallogr. 24 (1991) 1–5. Birch, F.: Finite strain isotherm and velocities for single-crystal and polycrystalline NaCl at high pressure and 300 K. J. Geophys. Res. 83 (1978) 1257. Catalli, K.; Santillan, J.; Williams, Q.: A high pressure infrared spectroscopic study of PbCO3-cerussite: constraints on the structure of the post-aragonite phase. Phys. Chem. Min. 32 (2005) 412-417. Chevrier, G.; Giester, G.; Heger, G.; Jarosch, D.; Wildner, M.; Zemann, J.: Neutron single-crystal refinement of cerussite, PbCO3, and comparison with other aragonite-type carbonates. Z. Kristallogr. 199 (1992) 67-74. Coelho, A. A.: TOPAS v2.0: General profile and structure analysis software for powder diffraction data. (2000). De Villiers, J.P.R.: Crystal structures of aragonite, strontianite, and witherite. Am. Mineral. 56 (1971) 758-761. Dollase, W.A.: Correction of intensities for preferred orientation in powder diffractometry: application of the March model. J. Appl. Crystallogr. 19 (1986) 267-272. Durman, R.; Jayasooriya U.A.; Kette S.F.: Is cerussite an Aragonite? Longitudinal Optical-Transverse Optical Splitting in the singlecrystal Raman spectra. J. Chem. Soc. Chem. Commun. (1985) 916-917. Gillet, P.; Biellmann, C.; Reynard, B.; McMillan, P.: Raman spectroscopic studies of carbonates part I: high-pressure and high-temperature behaviour of calcite, magnesite, dolomite and aragonite. Phys. Chem. Min. 20 (1993) 1-18. Hammersley, A.; Svensson S.; Hanfland M.; Fitch A.; Häusermann D.: Two-Dimensional Detector Software: From Real Detector to Idealised Image or Two-Theta Scan. High Pressure Res. 14 (1996) 235-248. Holl, C.; Smyth, J.; Laustsen, H.; Jacobsen, S.; Downs, R.: Compression of witherite up to 8 GPa and the crystal structure of BaCO3II. Phys. Chem. Min. 27 (2000) 467-473. Jephcoat, A.P.; Finger, L.W.; Cox, D.E.: High pressure, high resolution synchrotron x-ray powder diffraction with a position-sensitive detector. High Pressure Res. 8 (1992) 667-676. Lin, C.-C.; Liu, L.-G.: High pressure phase transformations in aragonite-type carbonates. Phys. Chem. Min. 24 (1997a) 977–987 Lin, C.-C.; Liu, L.-G.: Post-aragonite phase transitions in strontianite and cerussite: A high-pressure Raman spectroscopic study. J. Phys. Chem. Solids 58 (1997b) 977-987. Liu, L.; Basset, W.A.: Elemebts, Oxides and Silicates: High-Pressure Phases with implications for the Earth’s Interior. Oxford University Press, New York (1986). Mao, H.; Bell, P.; Shaner, J.; Steinberg, D.: Specific volume measurements of Cu, Mo, Pd, and Ag and calibration of the ruby R1 fluorescence pressure gauge from 0.06 to 1 Mbar. J. Appl. Phys. 49 (1978) 3276–3283. McCusker, L.B.; Von Dreele, R.B.; Cox, D.E.; Loueer, D.; Scardi, R.: Rietveld refinement guidelines. J. Appl. Cryst. 32 (1999) 36-50. Minch, R.; Dubrovinsky, L.; Kurnosov, A.; Ehm, L.; Knorr, K.; Depmeier, W.; Raman spectroscopic study of PbCO3 at high pressures and temperatures. Phys. Chem. Min. 37 (2010) 45–56. Mudring, A.-V.: Stereochemical activity of lone pairs in heavier main-group element compounds. In Inorganic Chemistry in Focus 152 III (Eds.: G. Meyer, D. Naumann, L. Wesemann), Wiley-VCH, (2006) 15-28. Ohashi, Y. and Burnham C. W.: Clinopyroxine lattice deformations: the roles of chemical substitution and temperature. Am. Mineral. 58 (1973) 843-849. Ono, S.; Kigegawa T.; Ohishi Y.; Tsuchiya J.: Post-aragonite phase transformation at CaCO3 at 40 GPa. Am. Mineral. 90 (2005) 667671. Ono, S.; Shirasaka, M.; Kikegawa, T.; Ohishi, Y.: A new high pressure phase of strontium carbonate. Phys. Chem. Min. 32 (2005b) 8-12. Ono, S.: New high pressure phases in BaCO3. Phys. Chem. Min. 34 (2007) 215-221. Rieger F.; Mudring A.-V.: Phase transition in Tl2TeO3: influence and origin of the thallium lone pair distortion. Inorg. Chem. 46 (2007) 446-452. R. Minch, L. Peters, L. Ehm et al. Rietveld H.: Line profiles of neutron powder diffraction peaks for structure refinements. Acta Crystallogr. 22 (1967) 151-152. Santillán, J.; Williams, Q.: A high pressure X-ray diffraction study of aragonite and the post-aragonite phase transition in CaCO3. Am. Mineral. 89 (2004) 1348-1352. Stokes, H.T., Hatch, D.M.: Isotropy subgroups of the 230 crystallographic space groups. World scientific publishing, Singapore 1988. Siidra, O.I.; Krivovichev S.V.; Filatov S.K.: Minerals and synthetic Pb (II) compounds with oxocentered tetrahedra: review and classification. Z. Kristallogr. 223 (2008) 114-125. Stoltzfus, M.W.; Woodward, P.M.; Seshadri, R.; Klepeis, J.-H.; Bursten B.: Structure and bonding in SnWO4, PbWO4, and BiVO4: lone pairs vs inert pairs. Inorg. Chem. 46 (2007) 3839-3850. Thompson, P.; Cox, D.E.; Hastings, J.B.: Rietveld refinement of Debye-Scherrer synchrotron X-ray data from Al2O3. J. Appl. Cryst. 20 (1987) 79-83.