Compressed Arsenolite As4O6 and Its Helium Clathrate As4O6Â·2He

Article pubs.acs.org/crystal

Compressed Arsenolite As4O6 and Its Helium Clathrate As4O6·2He Piotr A. Guńka,*,† Kamil F. Dziubek,‡,§ Andrzej Gładysiak,§ Maciej Dranka,† Jacek Piechota,∥ Michael Hanfland,⊥ Andrzej Katrusiak,§ and Janusz Zachara*,† †

Faculty of Chemistry, Warsaw University of Technology, Noakowskiego 3, 00-664 Warszawa, Poland LENS, European Laboratory for Nonlinear Spectroscopy, Via Nello Carrara 1, 50019 Sesto Fiorentino, Italy § Faculty of Chemistry, Adam Mickiewicz University, Umultowska 89b, 61-614 Poznań, Poland ∥ Interdisciplinary Centre for Mathematical and Computational Modelling, University of Warsaw, Pawińskiego 5a, 02-106 Warszawa, Poland ⊥ ESRF, 6 rue Jules Horowitz, BP 220, F-38043 Grenoble Cedex, France ‡

S Supporting Information *

ABSTRACT: The crystal structure of arsenolite, the cubic polymorph of molecular arsenic(III) oxide, has been determined by single-crystal X-ray diffraction up to 30 GPa. The bulk of the crystal is monotonically compressed with no detectable anomalies to 60% of the initial volume at 30 GPa. The experimental As4O6 crystal compression exceeds that obtained by various theoretical models within the density functional theory framework. The As4O6 molecules are deformed toward a more tetrahedral shape. A surprising property of arsenolite immersed in helium has been revealed above 3 GPa; the As4O6· 2He clathrate is formed in the surface layer increasingly deeper with pressure. Interestingly, this is the first example of helium clathrate formed in situ with a solid molecular oxide and proof that helium may permeate even nonporous single crystals in highpressure diffraction studies. This indicates it is an important and general phenomenon that needs to be taken into account when conducting high-pressure diffraction studies in helium.

■

INTRODUCTION Arsenolite is a well-known mineral that, until recently, was generally used for various industrial, agricultural, chemical, and household applications because of its strong toxicity and exceptional properties. Polymorphism of arsenic(III) oxide has attracted attention for years. There are two minerals of As2O3: cubic arsenolite containing adamantoid As4O6 molecules packed in a three-dimensional (3D) diamond network1,2 and monoclinic claudetite I consisting of highly corrugated As2O3 layers.3−5 In the late 1940s, another monoclinic modification was discovered and called claudetite II because of its extended layers that resembled those of claudetite I.6,7 Well-known also is a glassy As2O3 modification. There have been a few studies of arsenic(III) oxide polymorphs under high pressure. The arsenolite−claudetite I equilibrium as a function of temperature and pressure was studied extensively in the 1960s. It was suggested that arsenolite undergoes a reversible transition to claudetite I at 110 ± 10 °C upon being subjected to shear stress.8 In another high-pressure (HP) study of arsenolite, claudetite I, and glassy As2O3, Soignard et al. compressed arsenolite under nonhydrostatic conditions up to 35 GPa and observed its amorphization above 20 GPa.9 Grzechnik studied arsenolite by Raman and infrared (IR) spectroscopies up to 16 GPa and suggested a phase transition at 6 GPa, lowering the cubic symmetry to tetragonal with two molecules per primitive © 2015 American Chemical Society

unit cell. He also noted that hydrostatic or quasihydrostatic pressure, as opposed to shear stress, does not induce the arsenolite−claudetite I transition.10 However, no structural information about the HP arsenolite and its compressibility was provided. Herein, we report the arsenolite structure determined by synchrotron single-crystal X-ray diffraction up to 30 GPa (ID09A beamline at the ESRF)11 and theoretically modeled by density functional theory (DFT) calculations that reveal a considerable band gap decrease. Above 3 GPa, we have also observed the appearance of additional weak diffraction spots, “ghost” reflections, which have been attributed to pressureinduced permeation of helium into the arsenolite host structure.

■

RESULTS AND DISCUSSION Crystal Structure and Diffraction Pattern. A homemade single crystal of arsenolite was studied in a diamond anvil cell at the ID09A beamline of the European Synchrotron Radiation Facility up to 30 GPa.11 At ambient pressure, arsenolite is a molecular crystal built of adamantoid As4O6 molecules (Figure 1) located at highly symmetric 43̅ m sites of cubic space group Received: March 20, 2015 Revised: June 15, 2015 Published: June 23, 2015 3740

DOI: 10.1021/acs.cgd.5b00390 Cryst. Growth Des. 2015, 15, 3740−3745

Crystal Growth & Design

Article

Figure 1. (a) Molecular structure of arsenolite with the As 4 tetrahedron colored green. (b) Pair of closest As4O6 molecules with As···O and closest As···As intermolecular separations marked with brown and light blue dashed lines, respectively. (c) (As4O6)4 cluster with As···As intermolecular distances marked with light blue dashed lines. (d) Arsenolite crystal structure with As4O6 molecules shown as a wireframe model. Structural voids represented by blue surfaces have been calculated in the Mercury13,14 program with a probe radius of 0.9 Å and a grid spacing of 0.5 Å.

Figure 3. Experimental intra- and intermolecular As···As distances (a) as well as As−O bond lengths and intermolecular As···O contacts (b) plotted as a function of pressure. Two kinds of intermolecular As···As distances indicated with squares and upward-pointing triangles correspond to the contacts in tetrahedral (As4O6)4 clusters (Figure 1c) and to the shortest intermolecular As···As distances (Figure 1b), respectively.

Fd3m ̅ . The bulk of the crystal is monotonically compressed over the whole pressure range of our study. No anomalies have been detected up to 30 GPa either in the lattice parameters (Figure 2) or in the intermolecular distances or molecular dimensions (Figures 3 and 4). This indicates that previously observed anomalies at 6 GPa in the Raman and IR spectra of mixed arsenolite and CsI powders10 were caused by the

Figure 4. O−As−O and As−O−As bond angles plotted as a function of pressure for both experimental and DFT-computed data.

nonhydrostatic effects in the sample rather than a phase transition in arsenolite. In our study, the He environment ensured perfectly hydrostatic conditions to 11.6 GPa, and pseudohydrostatic conditions above 11.6 GPa,12 whereas CsI is harder than arsenolite even under normal conditions. Pressure induces systematic distortions of the molecular geometry (Figures 3 and 4 and Figure S2 of the Supporting Information). Although no changes in the As−O bond length

Figure 2. Equation-of-state curves for experimental and DFTcomputed (p,V/Z) data. Filled and empty black circles denote experimental data measured for the compressed and decompressed sample, respectively. 3741



Article

can be noted within experimental error, As−O−As and O−As− O bond angles bend in the opposite sense, and consequently, the intramolecular As···As distances increase and O···O distances decrease (Figure 3 and Figures S2 and S3 of the Supporting Information). These changes, of ∼2% of the ambient-pressure values, are much smaller than the compression of intermolecular contacts. Oxygen atoms are pushed toward As···As intramolecular edges, which become longer as pressure increases and the molecule becomes more tetrahedral (Figure 1). This molecular shape change can be associated with the O···O packing interactions, previously described as a very important effect for silicate structures under high pressure.15 The intermolecular As···O distances are all well below the level of As and O van der Waals radii sum (3.37 Å),16 and they fall steadily with an increase in pressure (Figure 3b). The As··· As separation in the tetrahedral (As4O6)4 clusters decreases from 4.6090(5) Å at 0.1 MPa to 3.4268(7) Å at 30 GPa, passing the doubled As van der Waals radius (3.70 Å) at 15 GPa (Figures 1c and 3a and Figure S2 of the Supporting Information), which indicates a declining molecular character of the crystal toward a three-dimensional framework. The As··· As distance at 30 GPa equals 74% of the ambient-pressure value, while the analogous ratio for the As···O contact at 30 GPa is 79%, indicating that the compression of tetrahedral clusters is slightly more significant than the compression of pairs of closest As4O6 molecules (see Figure 1b,c). It is noteworthy that intra- and intermolecular As···As distances all converge to one value as pressure increases, resulting approximately in the fcc arrangement of closely packed As atoms (Figure S7 of the Supporting Information). Starting from 3.28(9) GPa, additional weak “ghost” reflections appear on the lower 2θ-angle side of the main reflections of the As4O6 sample (see Figure 5a and Figure S4 of the Supporting Information). The intensity of most of these “ghost” reflections increases with pressure, reaching 10−15% of the main reflections at the 29.83(5) GPa limit of our data (see Figure 5c). The “ghost” reflections can be indexed according to the face-centered cubic lattice with the parameter ag being ∼2% longer than that of the main As4O6 crystal (the subscript g indicates that this value has been derived from the “ghost” reflections). This ratio is somewhat smaller for the series of measurements with the compressed sample than for the decompression run (Figure 5b and Table S5 of the Supporting Information). After the possibilities of instrumental artifacts, spurious radiation, or multiple diffraction had been eliminated, the origin of the “ghost” reflections was associated with the formation of an inclusion compound of arsenolite with helium. Indeed, the structure of arsenolite comprises large octahedral voids at 0,0,0 (Figure 1d and Figure S5 of the Supporting Information), and helium, used as the pressure medium, is highly permeating. Several solid molecular compounds with light noble gas atoms within their structures were reported. It was shown that fullerene C60 is easily permeated with helium, already in the pressure range of approximately 0.1 GPa,17 and the HP structure of fullerene with neon atoms residing in the octahedral interstices was refined.18 Those experiments were performed on fullerene powders by means of high-pressure neutron diffraction. However, we were able to distinguish between reflections originating from the native and Hepermeated arsenolite crystals because of the single-crystal diffraction technique. To the best of our knowledge, such a nonequilibrated two-component system of a seemingly non-

Figure 5. (a) Arsenolite 4̅84̅ reflection profile extended to include the “ghost” reflection of the arsenolite−helium clathrate. The Bragg− “ghost” reflections pair recorded in the diffraction image is shown in the inset (the blue line corresponds to resolution d = 0.99 Å). (b) Lattice parameter a of arsenolite (black symbols) and ag of the lattice formed by the “ghost” reflections (red symbols). Upward-pointing triangles represent compression and downward-pointing triangles decompression. In the inset, the difference between the a and ag lattice parameters is plotted vs pressure. Black and red triangles correspond to the compressed and decompressed sample, respectively. (c) Intensities of seven selected “ghost” reflections plotted vs pressure. The legend includes the indices of reflections and their numbers in the peak list for the measurement at 11.08(5) GPa. The intensities plotted here were evaluated in the peak hunting procedure. 3742



Article

to 30 GPa, and this could be related to the fact that intermolecular contacts in senarmontite are shorter than in arsenolite at ambient pressure, as evidenced by the Sb···O and As···O contacts of 2.9018(9) and 3.041(2) Å, respectively.21 Interestingly, the geometry of the As4O6 molecule under 20.07(9) GPa is very similar to that of Sb4O6 at ambient pressure, as evidenced by the bond angles and bond valence values (see Table S6 of the Supporting Information). Thus, the crystal structure of senarmontite at a given pressure resembles that of arsenolite but under higher pressures. This suggests that arsenolite may undergo isostructural transitions above 30 GPa similar to those observed in senarmontite. Comparison of Crystal Structures between Experiment and Computations. We performed theoretical calculations of the arsenolite crystal structure under pressure utilizing various quantum mechanical models to compare their accuracy in the description of the system. We utilized DFT22,23 with two various basis sets: plane waves and localized Gaussian type atomic orbitals in the form of a pob-TZVP basis set.24 Two functionals with dispersion corrections treated with the Grimme method were studied, namely, the PBE-D2 functional from the GGA family of functionals25,26 and the hybrid B3LYPD* functional with the Grimme correction optimized for solid state calculations by Civalleri et al.27 A total of three models denoted PAW_PBE-D2, pob-TZVP_PBE-D2, and pobTZVP_B3LYP-D* were tested. The PAW calculations performed with the VASP code were conducted as constantvolume optimizations, and the external pressure was the result of the volume constraint; on the other hand, the pob-TZVP computations, effected in the CRYSTAL09 suite, were executed as full geometry optimizations under external pressure defined in the input. Interestingly, all of the models overestimate the As−O bond lengths by approximately 0.03−0.05 Å (cf. Figure S2 of the Supporting Information). The discrepancies for As··· O intermolecular distances are a little shorter, and the PBE-D2 model has been found to perform better than the B3LYP-D* model. The agreement between experimental and calculated As−O−As and O−As−O bond angles is remarkable, with most of the calculated values lying within one standard uncertainty (s.u.) of the experimental values. Equation of State. The third-order Birch−Murnaghan equation of state (EoS)28

porous inorganic host and its helium inclusion compound has not been reported until now. The “ghost” reflections provided unexpectedly rich structural information about the arsenolite−helium clathrate. The helium atom was located from the difference Fourier maps at 0,0,0 (Figure 6). The inclusion of He atoms in the refined model

Figure 6. Crystal structure of the As4O6·2He inclusion compound viewed along ∼[110]. As4O6 molecules are represented as a wireframe model whereas He atoms as light blue spheres. Dashed blue lines indicate the deformed octahedral surrounding of He by arsenic atoms (see Figure S5 of the Supporting Information for a closer view).

improved the fit between observed and calculated structure factors, while the relatively small values of the He atom displacement parameter testified to the full occupancy of this site and to the As4O6·2He stoichiometry. It is noteworthy that the octahedral cavities are interconnected by channels too narrow to allow the He atom to pass between the neighboring voids in the static structure (Figure 1d). However, the lattice and molecular vibrations are likely to make such transport possible, as was observed by Barbour and termed “porosity without pores”.19 Although the precise determination of the “ghost” reflections intensity was hampered by their diffusion and proximity of much stronger main reflections, the simple and highly symmetric structure of As4O6·2He allowed for the precise atomic position determination in this crystal. The closest As···He contact is 2.7456(18) Å at 7.71(8) GPa and 2.4517(14) Å at 29.83(5) GPa, whereas the closest O···He contacts are 2.698(11) and 2.444(4) Å, respectively. These distances are compatible with van der Waals radii of helium, arsenic, and oxygen when taking into account their considerable compression in this pressure range. The low intensity of ghost reflections stems from the fact that the helium clathrate is formed only on the surface of an arsenolite single crystal, but as the pressure increases, the intensity of “ghost” reflections also rises (Figure 5c), which indicates that more and more helium is pushed into the arsenolite structure and the thickness of the As4O6·2He surface layer increases. Recently, a high-pressure structural study of senarmontite− cubic antimony(III) oxide has been conducted. It undergoes two isostructural phase transitions at 3.5 and 10 GPa consisting in adamantane-like Sb4O6 molecules becoming a 3D covalent network. Cubic symmetry has been retained during the transitions, but significant changes in the bulk modulus have occurred.20 Similar effects were not observed for arsenolite up

p=

3B0 ⎡⎛ V0 ⎞7/3 ⎛ V0 ⎞5/3⎤ ⎢⎜ ⎟ − ⎜ ⎟ ⎥ × ⎝ V ⎠ ⎥⎦ 2 ⎢⎣⎝ V ⎠ ⎫ ⎧ ⎤⎪ ⎡⎛ V ⎞2/3 3 ⎨1 + (B′0 − 1)⎢⎜ 0 ⎟ − 1⎥⎬ ⎪ ⎢⎣⎝ V ⎠ 4 ⎦⎥⎪ ⎩ ⎭ ⎪

where V0 is volume at zero pressure, B0 is the bulk modulus, and B′0 is bulk modulus first derivative with respect to pressure, has been fitted to the experimental as well as DFT-calculated (p,V/Z) set of points using weights for both pressure and volume s.u.’s. The resulting EoSs parameters are listed in Table 1, and the EOS curves are plotted together with experimental and computed points in Figure 2. The agreement between the experimental and PAW_PBED2 data is extraordinary. It is noteworthy that these calculations were performed in the P1 space group, and the analysis of the optimized geometries indicates that they all exhibit the symmetry of the cubic Fd3̅m space group, which supports our experimental result that no lowering of symmetry occurs in arsenolite upon compression. Surprisingly, the agreement is 3743



Article

Table 1. Third-Order Birch−Murnaghan EoS Parameters for the Experimental and Computational Set of (p,V/Z) Points experiment PAW_PBE-D2 pob-TZVP_PBE-D2 pob-TZVP_B3LYPD* a

V0/Za (Å3)

B0 (GPa)

B′0

168.7(6) 169.1(4) 162.8(14) 172.0(18)

11.8(5) 10.0(5) 21(2) 18(2)

8.6(2) 10.5(4) 7.0(5) 6.4(5)

V0/Z is the molecular volume at zero pressure.

worst for the pob-TZVP_B3LYP-D* model even though the B3LYP functional is used extensively for computations of molecules and the dispersion correction has been optimized for molecular crystals. The application of the Gaussian basis set leads to significant overestimation of the bulk modulus and also to a significant increase in its standard uncertainty. The crystal structure of arsenolite helium clathrate was modeled using the PAW_PBE-D2 model that was found to perform best for arsenolite. The agreement between experimental and DFT EoS curves is very good (Table S7 and Figure S6 of the Supporting Information), which further confirms our structural model of the crystalline phase giving rise to the “ghost” reflections. Band Structure. The band structure and density of states (DOS) have been computed to see if any interesting changes occur in the arsenolite’s electronic structure upon compression, given that As···As separations fall below the sum of As van der Waals radii as pressure increases (see Figure 3). The PAW method with a plane wave basis set was utilized with a PBE exchange correlation functional and Grimme’s D2 correction for dispersion. Computations have shown that arsenolite is an insulator at 0.06 GPa with a band gap of 4.2 eV that drops to 2.7 eV, making cubic As2O3 semiconducting at 27.8 GPa (Figure 7). Electronic bands tend to be more diffuse at high pressure, but no significant changes in the As and O partial DOS relative to one another have occurred with the increase in pressure.

■

Figure 7. Band structure and density of states (DOS) in arsenolite crystal structures at 0.06 GPa (top) and 27.8 GPa (bottom). Green and red lines denote As and O partial DOS, respectively. X stands for (0.5, 0.5, 0), Γ for (0, 0, 0), and Λ for (0.5, 0.5, 0.5) points in the reciprocal lattice.

CONCLUSIONS The compression of the arsenolite crystal structure between 0 and 30 GPa is monotonic. It has been shown that arsenolite does not undergo any phase transition at 6 GPa, lowering its symmetry,10 and it does not undergo amorphization above 20 GPa.9 In the smoothly compressed structure, the As4O6 molecules are slightly distorted toward a tetrahedral shape, with the As atoms displaced outside and O pushed inside relative to the ambient-pressure structure. The arsenolite structure is exceptionally stable up to 30 GPa at least, when the crystal volume is compressed to 60% of the ambient pressure value and the intermolecular distances fall well below the van der Waals distances. The experimentally measured compression is consistent but slightly stronger than any of the quantum mechanical predictions. None of the applied DFT models pointed to a phase transition either. It has been found that the plane wave basis set and PBE functional augmented with the Grimme’s dispersion correction (PBE-D2) yields the best predictions and the application of the pob-TZVP Gaussian basis set leads to the overestimation of the arsenolite bulk modulus. The computations of arsenolite band structure indicate that its band gap decreases from 4.2 eV at ambient pressure to 2.7 eV at 27.8 GPa. Surprisingly, this study has revealed the formation of As4O6·2He clathrate above 3 GPa, and its crystal structure has been determined. As4O6·2He is

formed on the surface of the arsenolite sample, and the He permeation depth increases with pressure upon compression. The clathrate decomposes upon decompression. Such a gradual dynamic process of He permeation has not been observed before for a nonporous molecular inorganic host to the best of our knowledge, possibly because similar studies were performed mainly on powdered samples of very fine grains. Our results show that helium may permeate even nonporous single crystals in high-pressure diffraction studies and affect the diffraction pattern of studied materials. This effect should be taken into account when surprising features arise in the derived structural models, given that, unlike in this experiment, “ghost” reflections may overlap with main reflections that can be more diffuse than the ones observed for arsenolite. Further investigations in different pressure-transmitting media as well as research of the helium penetration time dependence are planned to gain deeper insight into the phenomenon. 3744



■

Article

(20) Pereira, A. L. J.; Gracia, L.; Santamaría-Pérez, D.; Vilaplana, R.; Manjón, F. J.; Errandonea, D.; Nalin, M.; Beltrán, A. Phys. Rev. B 2012, 85, 174108. (21) Whitten, A. E.; Dittrich, B.; Spackman, M. A.; Turner, P.; Brown, T. C. Dalton Trans. 2004, No. No. 1, 23−29. (22) Hohenberg, P.; Kohn, W. Phys. Rev. 1964, 136, B864−B871. (23) Kohn, W.; Sham, L. J. Phys. Rev. 1965, 140, A1133−A1138. (24) Peintinger, M. F.; Oliveira, D. V.; Bredow, T. J. Comput. Chem. 2013, 34, 451−459. (25) Perdew, J. P.; Burke, K.; Ernzerhof, M. Phys. Rev. Lett. 1996, 77, 3865−3868. (26) Perdew, J. P.; Burke, K.; Ernzerhof, M. Phys. Rev. Lett. 1997, 78, 1396. (27) Civalleri, B.; Zicovich-Wilson, C. M.; Valenzano, L.; Ugliengo, P. CrystEngComm 2008, 10, 405−410. (28) Birch, F. Phys. Rev. 1947, 71, 809−824.

ASSOCIATED CONTENT

S Supporting Information *

Experimental and computational details, experimental and DFT-computed structural parameters, and cif files with experimental structures. The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.cgd.5b00390.

■

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected]. *E-mail: [email protected]. Funding

This work has been supported by the National Science Centre, Grant DEC-2012/05/N/ST5/00283. K.F.D. gratefully acknowledges the Polish Ministry of Science and Higher Education for financial support through the “Mobilnośc ́ Plus” program. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The VASP calculations were conducted in the Interdisciplinary Centre for Mathematical and Computational Modelling (Computational Grant G28-3), whereas the CRYSTAL09 computations were performed using resources provided by the Wroclaw Centre for Networking and Supercomputing (http://wcss.pl) (Grant 260). We acknowledge the European Synchrotron Radiation Facility for provision of synchrotron radiation facilities.

■

REFERENCES

(1) Bozorth, R. M. J. Am. Chem. Soc. 1923, 45, 1621−1627. (2) Pertlik, F. Czech. J. Phys. 1978, 28, 170−176. (3) Becker, K. A.; Plieth, K.; Stranski, I. N. Z. Anorg. Allg. Chem. 1951, 266, 293−301. (4) Frueh, A. J. Am. Mineral. 1951, 36, 833−850. (5) Pertlik, F. Monatsh. Chem. 1978, 109, 277−282. (6) Kürbs, E.; Plieth, K.; Stranski, I. N. Z. Anorg. Chem. 1949, 258, 238−246. (7) Pertlik, F. Monatsh. Chem. 1975, 106, 755−762. (8) White, W. B.; Dachille, F.; Roy, R. Z. Kristallogr. 1967, 125, 450− 458. (9) Soignard, E.; Amin, S. A.; Mei, Q.; Benmore, C. J.; Yarger, J. L. Phys. Rev. B 2008, 77, 144113. (10) Grzechnik, A. J. Solid State Chem. 1999, 144, 416−422. (11) Merlini, M.; Hanfland, M. High Pressure Res. 2013, 33, 511−522. (12) Takemura, K. High Pressure Res. 2007, 27, 465−472. (13) Macrae, C. F.; Edgington, P. R.; McCabe, P.; Pidcock, E.; Shields, G. P.; Taylor, R.; Towler, M.; van de Streek, J. J. Appl. Crystallogr. 2006, 39, 453−457. (14) Macrae, C. F.; Bruno, I. J.; Chisholm, J. A.; Edgington, P. R.; McCabe, P.; Pidcock, E.; Rodriguez-Monge, L.; Taylor, R.; van de Streek, J.; Wood, P. A. J. Appl. Crystallogr. 2008, 41, 466−470. (15) Prewitt, C. T.; Downs, R. T. In Ultrahigh-pressure mineralogy: Physics and chemistry of the earth’s deep interior; Reviews in Mineralogy; Mineralogical Society of America: Washington, DC, 1998; pp 283− 317. (16) Bondi, A. J. Phys. Chem. 1964, 68, 441−451. (17) Schirber, J. E.; Kwei, G. H.; Jorgensen, J. D.; Hitterman, R. L.; Morosin, B. Phys. Rev. B 1995, 51, 12014−12017. (18) Morosin, B.; Jorgensen, J. D.; Short, S.; Kwei, G. H.; Schirber, J. E. Phys. Rev. B 1996, 53, 1675−1678. (19) Barbour, L. J. Chem. Commun. 2006, No. No. 11, 1163−1168. 3745


Compressed Arsenolite As4O6 and Its Helium Clathrate As4O6Â·2He

Recommend Documents