Synthesis of FeN4 at 180 GPa and its crystal structure from a submicron-sized grain

Bykov, M.; Khandarkhaeva, S.; Fedotenko, T.; Sedmak, P.; Dubrovinskaia, N.; Dubrovinsky, L.

doi:10.1107/S2056989018012161

research communications

CRYSTALLOGRAPHIC
COMMUNICATIONS

ISSN: 2056-9890

Volume 74| Part 10| October 2018| Pages 1392-1395

https://doi.org/10.1107/S2056989018012161

Open

access

Synthesis of FeN₄ at 180 GPa and its crystal structure from a submicron-sized grain

Maxim Bykov,^a ^* Saiana Khandarkhaeva,^a Timofey Fedotenko,^b Pavel Sedmak,^c Natalia Dubrovinskaia ^b and Leonid Dubrovinsky ^a

^aBayerisches Geoinstitut, University of Bayreuth, 95440 Bayreuth, Germany, ^bMaterial Physics and Technology at Extreme Conditions, Laboratory of Crystallography, University of Bayreuth, 95440 Bayreuth, Germany, and ^cEuropean Synchrotron Radiation Facility, BP 220, 38043 Grenoble Cedex, France
^*Correspondence e-mail: maks.byk@gmail.com

Edited by M. Weil, Vienna University of Technology, Austria (Received 7 August 2018; accepted 28 August 2018; online 7 September 2018)

Iron tetranitride, FeN₄, was synthesized from the elements in a laser-heated diamond anvil cell at 180 (5) GPa and 2700 (200) K. Its crystal structure was determined based on single-crystal X-ray diffraction data collected from a submicron-sized grain at the synchrotron beamline ID11 of ESRF. The compound crystallizes in the triclinic space group P $[\overline{1}]$ . In the asymmetric unit, the Fe atom occupies an inversion centre (Wyckoff position 1d), while two N atoms occupy general positions (2i). The structure is made up from edge-sharing [FeN₆] octahedra forming chains along [100] and being interconnected through N—N bridges. N atoms form catena-poly[tetraz-1-ene-1,4-diyl] anions [–N=N—N—N–]_∞²⁻ running along [001]. In comparison with the previously reported structure of FeN₄ at 135 GPa [Bykov et al. (2018). Nat. Commun. 9, 2756], the crystal structure of FeN₄ at 180 GPa is similar but the structural model is significantly improved in terms of the precision of the bond lengths and angles.

Keywords: polynitrides; iron tetranitride; high-pressure single-crystal X-ray diffraction; crystal structure.

CCDC reference: 1864279

1. Chemical context

Polynitrogen compounds have attracted great interest because of their potential applications as high-energy-density materials. Recently, a variety of nitrogen-rich compounds containing polymeric and oligomeric nitrogen chains, N₅ or N₆ rings, or even more complex networks have been predicted to be stable at high pressures (Steele & Oleynik, 2016 , 2017 ; Zhang et al., 2017 ; Xia et al., 2018 ). Predicted lithium and caesium pentazolates LiN₅ and CsN₅ were successfully synthesized at high-pressure conditions via the reaction between a metal or metal azide and nitrogen (Laniel et al., 2018 ; Steele et al., 2017 ). Recently, Bykov and co-workers synthesized two compounds containing polymeric nitrogen chains, viz. an inclusion compound ReN₈·N₂ (Bykov et al., 2018b ) and iron tetranitride, FeN₄ (Bykov et al. 2018a ) via the reaction between Fe or Re and nitrogen in a laser-heated diamond anvil cell (DAC). The crystal structures of these compounds were studied at pressures up to 135 GPa by means of single-crystal X-ray diffraction (SCXRD).

The higher the pressures, the more challenging are synthesis and diffraction studies in DACs, even at dedicated high-pressure stations at the 3rd generation synchrotron facilities where the X-ray beam can be focused down to 2–3 µm. First of all, at pressures exceeding 150 GPa, the size of the sample is of only about 10 µm or less, and single-crystalline grains of the reaction product(s) are often of submicron size, which results in a drastic worsening of the signal-to-noise ratio in SCXRD. Additionally, the contribution of parasitic diffraction from the gasket material increases with pressure because the sample chamber becomes smaller upon compression. Submicron focusing of the X-ray beam, which is possible on some synchrotron beamlines, can provide suitable conditions to collect SCXRD data at multi-megabar pressures. Here we report the synthesis of FeN₄ from the elements at a pressure of about 180 GPa and provide the structure refinement for FeN₄ against SCXRD data at this pressure. The X-ray beam focusing down to 0.3×0.3 µm² at the synchrotron beamline ID11 (ESRF, Grenoble, France) allowed us to collect SCXRD data from an FeN₄ grain with linear dimensions of about 0.5 µm.

2. Structural commentary

The crystal structure (Fig. 1a,b) and the unit-cell volume (Fig. 2) of FeN₄ at 180 GPa are in a good agreement with the structural model for this compound at 135 GPa and its equation of state as reported by Bykov et al. (2018a). Despite the increased pressure, as a result of the application of the submicron beam focusing, the quality of the SCXRD data collected at 180 GPa turned out to be much better. Thus, the quality of the structure refinement of FeN₄ based on the 180 GPa data set is significantly improved in comparison with that for the 135 GPa data set. This is evident from a comparison of such important refinement indicators such as the data-to-parameter ratio (7.1 vs 4.8), Δρ_max/Δρ_min (0.76/−0.56 vs 0.98/−1.09 e Å⁻³) and R₁[I>2σ(I)] (0.040 vs 0.064). Furthermore, the precision of the bond lengths and angles is significantly improved (Table 1).

Table 1
Selected bond lengths for FeN₄ at 135 GPa (Bykov et al., 2018a) and at 180 GPa (this study)

	135 GPa	180 GPa
Fe—N1	1.73 (2)	1.707 (10)
Fe—N1ⁱ	1.73 (2)	1.707 (10)
Fe—N2ⁱⁱ	1.81 (3)	1.783 (14)
Fe—N2ⁱⁱⁱ	1.81 (3)	1.783 (14)
Fe—N2^iv	1.78 (3)	1.763 (6)
Fe—N2^v	1.78 (3)	1.763 (6)
N1—N1^vi	1.29 (5)	1.277 (14)
N1—N2	1.30 (3)	1.298 (8)
N2—N2^vii	1.43 (4)	1.37 (3)
Symmetry codes: (i) −x + 1, −y, −z; (ii) x + 1, y, z + 1; (iii) −x, −y, −z − 1; (iv) −x + 1, −y, −z − 1; (v) x, y, z + 1; (vi) −x, −y − 1, −z − 1; (vii) −x, −y − 1, −z − 2.

Figure 1
The crystal structure of FeN₄ at 180 GPa. (a) and (b) Projections of the crystal structure along [100] and [001], respectively. (c) A fragment of the crystal structure showing the coordination of Fe atoms. [Symmetry codes: (i) x, y, 1 + z; (ii) 1 − x, −y, −z; (iii) −x, 1 − y, −z; (iv) 1 + x, 1 + y, 2 + z; (v) −x, −1 − y, −1 − z; (vi) 1 + x, 1 + y, 1 + z; (vii) 1 − x, −y, 1 − z; (viii) 1 + x, 1 + y, z.] (d) The crystal structure of FeN₄ in polyhedral representation.

Figure 2
Pressure-dependence of the unit-cell volume of FeN₄. Blue points and the equation of state (blue line) are taken without modification from Bykov et al. (2018a

) [V_94.5 = 29.94 (4) Å³, K_94.5 = 603 (22) GPa, K′_94.5 = 4.0 (fixed)]. Red point – current study.

The Fe1 atom occupies an inversion centre of space-group type P $[\overline{1}]$ (Wyckoff position 1d), while the two nitrogen atoms N1 and N2 occupy general positions (2i). The iron atom is coordinated by six nitrogen atoms, forming a distorted octahedron. [FeN₆] octahedra share opposite edges, thus forming infinite chains along [100]. These chains are interconnected through N—N bridges as shown in Fig. 1d. The covalently bonded nitrogen atoms form infinite zigzag chains running along [001] (Fig. 1a–c). The N1 atom has a trigonal–planar coordination, while N2 is tetrahedrally coordinated, suggesting sp² and sp³ hybridization, respectively. In agreement with the study of Bykov et al. (2018a), the N—N distances increase in the following order d(N1—N1) < d(N1—N2) < d(N2—N2) (Table 1), supporting the conclusion that the N1—N1 bond is a double-bond, while N1—N2 and N2—N2 bonds are single bonds. Therefore, the nitrogen atoms form catena-poly[tetraz-1-ene-1,4-diyl] anions [–N=N—N—N–]_∞^2–.

The key parameters for the synthesis of polynitrides are pressure–temperature conditions and the choice of metal and/or nitrogen precursors. High temperatures and pressures are required to overcome the kinetic barrier for breaking the triple N≡N bond, to increase the chemical potential of nitrogen and to stabilize the reaction products (Sun et al., 2017 ). It is known that increasing pressure allows compounds with higher nitrogen content to be obtained, e.g. for the Fe—N system Fe_xN (x = 2–8) can be synthesized at ambient and low pressures (Ertl et al. 1979 ), FeN at 12 GPa (Clark et al., 2017 ), FeN₂ at 60 GPa, and FeN₄ at 106 GPa (Bykov et al., 2018a). Interestingly, at a given pressure, different metals stabilize different types of nitrogen networks. For example, ReN₈·N₂ synthesized at 106 GPa contains polydiazene chains [–N=N–]_∞ (Bykov et al., 2018b), whereas alkali metals form pentazolate salts at even lower pressures (Laniel et al., 2018; Steele et al., 2017), i.e. the type of metal, the variety of its oxidation states, and its ionic radius play an important role in the chemistry of the nitrogen network. The current study shows that FeN₄ can be synthesized in a broad pressure range from 106 to 180 GPa. Such an extended stability range for this compound may be related to the favourable sixfold coordination of Fe. On one hand, it perfectly matches the 18 e⁻ rule (Bykov et al., 2018a), and on the other hand, for the Fe—N system coordination number 6 is geometrically preferable. Further systematic studies of various metal polynitrides will allow empirical rules for the design of novel materials at different pressure and temperature conditions to be formulated.

3. Synthesis and crystallization

A piece of iron powder (Sigma Aldrich, 99.99%) was loaded inside a sample chamber of a BX90-type diamond anvil cell equipped with double-bevelled Boehler–Almax type diamonds (culet diameter 40 µm). Nitrogen was used as a pressure-transmitting medium and as a reagent for the synthesis. The sample was compressed up to 180 GPa and laser-heated from both sides up to 2700 (200) K. The pressure was determined using the equation of state (EoS) of hcp-iron. As there are several equations of state of iron in the literature (Table 2), for a given unit-cell volume of iron V_Fe = 15.171 (5) Å³ one can get slightly different pressures in the range 173.5 to 187.5 GPa with an average of 179.8(5.2) GPa. Taking into account this uncertainty in the pressure determination, we accepted the rounded value of 180 GPa.

Table 2
Pressures for FeN₄ synthesis based on different 3rd order Birch–Murnaghan EoS's of hcp-Fe reported in the literature [V_Fe = 15.171 (5) Å³/unit cell]

Reference	V₀ (Å³)	K (GPa)	K′	Pressure (GPa)
Dewaele et al. (2006 )	22.468 (24)	165 (fixed)	4.97 (4)	173.5(2.2)
Fei et al. (2016 )	22.428 (fixed)	172.7(1.4)	4.79 (5)	174.1(1.4)
Sakai et al. (2014 )	22.18 (20)	179.6(2.2)	4.91 (12)	174.9(2.1)
Mao et al. (1990 )	22.35 (3)	164.8(3.6)	5.33 (9)	179.8(4.3)
Yamazaki et al. (2012 )	22.15 (5)	202 (7)	4.5 (2)	181.0(5.6)
Dubrovinsky et al. (2000 )	22.35 (3)	155.6(3.5)	5.81 (6)	183.7(4.8)
Garai et al. (2011 )	22.33 (3)	164 (2)	5.52 (5)	183.9(2.5)
Boehler et al. (2008 )	22.46 (4)	160 (6)	5.6 (2)	187.5(8.2)

In order to locate the FeN₄ grain in the sample chamber we used the following strategy: we collected 27 × 27 = 729 still images with the exposure time of 6 s. Before taking the next image, either the horizontal or vertical motor was moved by 0.5 µm, allowing a 13 × 13 µm² X-ray diffraction map of the sample chamber to be built up (Fig. 3). The images were then analyzed with XDI software (Hrubiak, 2017 ).

Figure 3
X-ray diffraction imaging of the sample chamber at 180 GPa. The colour intensity is proportional to the intensity of the following reflections: the (100) reflection of Re for the green region; the (101) reflection of Fe for the orange region; the sum of the (101), (1 $[\overline{1}]$ 1), (0 0 2), and ( $[\overline{1}]$ $[\overline{1}]$ 2) reflections of FeN₄ for the blue region.

4. Refinement

Crystal data, data collection details and structure refinement details are summarized in Table 3. We have used the same non-reduced unit-cell setting and the structure model of FeN₄ at 135 GPa (Bykov et al., 2018a) for refinement of the crystal structure of FeN₄ at 180 GPa. As a result of the limited angular range caused by the laser-heated DAC and the very small crystal size, the resolution of the data set was not sufficient to refine the atoms with anisotropic displacement parameters. Hence they were refined with isotropic displacement parameters only.

Table 3
Experimental details

Crystal data
Chemical formula	FeN₄
M_r	111.89
Crystal system, space group	Triclinic, P $[\overline{1}]$
Temperature (K)	293
a, b, c (Å)	2.4473 (10), 3.4688 (14), 3.5144 (13)
α, β, γ (°)	105.22 (4), 110.60 (4), 91.39 (3)
V (Å³)	26.72 (2)
Z	1
Radiation type	Synchrotron, λ = 0.30996 Å
μ (mm⁻¹)	1.33
Crystal size (mm)	0.0005 × 0.0005 × 0.0005

Data collection
Diffractometer	ID11 @ ESRF
Absorption correction	Multi-scan (ABSPACK; Oxford Diffraction, 2005 $[Oxford Diffraction (2005). ABSPACK. Oxford Diffraction Ltd, Abingdon, England.]$ )
T_min, T_max	0.967, 1.000
No. of measured, independent and observed [I > 2σ(I)] reflections	117, 71, 70
R_int	0.020
(sin θ/λ)_max (Å⁻¹)	0.901

Refinement
R[F² > 2σ(F²)], wR(F²), S	0.040, 0.082, 1.18
No. of reflections	71
No. of parameters	10
Δρ_max, Δρ_min (e Å⁻³)	0.76, −0.56
Computer programs: CrysAlis PRO (Rigaku OD, 2018 $[Rigaku OD (2018). CrysAlis PRO. Rigaku Oxford Diffraction, Yarnton, UK.]$ ), SHELXT (Sheldrick, 2015a ), SHELXL2014 (Sheldrick, 2015b ) and OLEX2 (Dolomanov et al., 2009 ).

Supporting information

CCDC reference: 1864279

Crystal structure: contains datablock I. DOI: https://doi.org/10.1107/S2056989018012161/wm5459sup1.cif

Structure factors: contains datablock I. DOI: https://doi.org/10.1107/S2056989018012161/wm5459Isup2.hkl

checkCIF report

Computing details top

Data collection: CrysAlis PRO (Rigaku OD, 2018); cell refinement: CrysAlis PRO (Rigaku OD, 2018); data reduction: CrysAlis PRO (Rigaku OD, 2018); program(s) used to solve structure: SHELXT (Sheldrick, 2015a); program(s) used to refine structure: SHELXL2014 (Sheldrick, 2015b); molecular graphics: OLEX2 (Dolomanov et al., 2009); software used to prepare material for publication: OLEX2 (Dolomanov et al., 2009).

Iron tetranitride top

Crystal data top

FeN₄	Z = 1
M_r = 111.89	F(000) = 54
Triclinic, P1	D_x = 6.953 Mg m⁻³
a = 2.4473 (10) Å	Synchrotron radiation, λ = 0.30996 Å
b = 3.4688 (14) Å	Cell parameters from 68 reflections
c = 3.5144 (13) Å	θ = 2.8–16.1°
α = 105.22 (4)°	µ = 1.33 mm⁻¹
β = 110.60 (4)°	T = 293 K
γ = 91.39 (3)°	Irregular, black
V = 26.72 (2) Å³	0.001 × 0.001 × 0.001 mm

Data collection top

ID11 @ ESRF diffractometer	71 independent reflections
Radiation source: synchrotron	70 reflections with I > 2σ(I)
Synchrotron monochromator	R_int = 0.020
ω scans	θ_max = 16.2°, θ_min = 2.8°
Absorption correction: multi-scan (ABSPACK; Oxford Diffraction, 2005)	h = −3→3
T_min = 0.967, T_max = 1.000	k = −5→4
117 measured reflections	l = −6→5

Refinement top

Refinement on F²	0 restraints
Least-squares matrix: full	Primary atom site location: dual
R[F² > 2σ(F²)] = 0.040	w = 1/[σ²(F_o²) + (0.0282P)² + 0.3122P] where P = (F_o² + 2F_c²)/3
wR(F²) = 0.082	(Δ/σ)_max < 0.001
S = 1.18	Δρ_max = 0.76 e Å⁻³
71 reflections	Δρ_min = −0.56 e Å⁻³
10 parameters

Special details top

Geometry. All esds (except the esd in the dihedral angle between two l.s. planes) are estimated using the full covariance matrix. The cell esds are taken into account individually in the estimation of esds in distances, angles and torsion angles; correlations between esds in cell parameters are only used when they are defined by crystal symmetry. An approximate (isotropic) treatment of cell esds is used for estimating esds involving l.s. planes.

Fractional atomic coordinates and isotropic or equivalent isotropic displacement parameters (Å²) top

	x	y	z	U_iso*/U_eq
Fe	0.5000	0.0000	0.0000	0.0072 (4)*
N1	0.163 (4)	−0.346 (4)	−0.485 (2)	0.0066 (10)*
N2	0.065 (3)	−0.309 (4)	−0.861 (2)	0.0068 (10)*

Geometric parameters (Å, º) top

Fe—Feⁱ	2.4473 (10)	Fe—N2^vii	1.763 (6)
Fe—Feⁱⁱ	2.4473 (10)	N1—N1^viii	1.277 (14)
Fe—N1	1.707 (10)	N1—N2	1.298 (8)
Fe—N1ⁱⁱⁱ	1.707 (10)	N2—Fe^ix	1.763 (6)
Fe—N2^iv	1.783 (14)	N2—Fe^x	1.783 (14)
Fe—N2^v	1.783 (14)	N2—N2^xi	1.37 (3)
Fe—N2^vi	1.763 (6)

Feⁱ—Fe—Feⁱⁱ	180.0	N2^vii—Fe—Feⁱⁱ	46.7 (5)
N1—Fe—Feⁱ	96.6 (4)	N2^iv—Fe—Feⁱⁱ	133.98 (18)
N1ⁱⁱⁱ—Fe—Feⁱⁱ	96.6 (4)	N2^iv—Fe—Feⁱ	46.02 (18)
N1—Fe—Feⁱⁱ	83.4 (4)	N2^vii—Fe—N2^v	92.7 (5)
N1ⁱⁱⁱ—Fe—Feⁱ	83.4 (4)	N2^vi—Fe—N2^v	87.3 (5)
N1ⁱⁱⁱ—Fe—N1	180.0	N2^v—Fe—N2^iv	180.0 (7)
N1—Fe—N2^vii	81.3 (4)	N2^vii—Fe—N2^iv	87.3 (5)
N1ⁱⁱⁱ—Fe—N2^vii	98.7 (4)	N2^vi—Fe—N2^iv	92.7 (5)
N1ⁱⁱⁱ—Fe—N2^vi	81.3 (4)	N2^vi—Fe—N2^vii	180.0
N1—Fe—N2^vi	98.7 (4)	N1^viii—N1—Fe	118.9 (9)
N1—Fe—N2^iv	90.5 (5)	N1^viii—N1—N2	109.6 (9)
N1—Fe—N2^v	89.5 (5)	N2—N1—Fe	129.4 (9)
N1ⁱⁱⁱ—Fe—N2^v	90.5 (5)	Fe^ix—N2—Fe^x	87.3 (5)
N1ⁱⁱⁱ—Fe—N2^iv	89.5 (5)	N1—N2—Fe^x	110.9 (12)
N2^vi—Fe—Feⁱ	46.7 (5)	N1—N2—Fe^ix	126.9 (5)
N2^vi—Fe—Feⁱⁱ	133.3 (5)	N1—N2—N2^xi	107.6 (12)
N2^vii—Fe—Feⁱ	133.3 (5)	N2^xi—N2—Fe^x	113.1 (6)
N2^v—Fe—Feⁱⁱ	46.02 (18)	N2^xi—N2—Fe^ix	109.9 (9)
N2^v—Fe—Feⁱ	133.98 (18)

Symmetry codes: (i) x+1, y, z; (ii) x−1, y, z; (iii) −x+1, −y, −z; (iv) x+1, y, z+1; (v) −x, −y, −z−1; (vi) −x+1, −y, −z−1; (vii) x, y, z+1; (viii) −x, −y−1, −z−1; (ix) x, y, z−1; (x) x−1, y, z−1; (xi) −x, −y−1, −z−2.

Selected bond lengths for FeN₄ at 135 GPa (Bykov et al., 2018a) and at 180 GPa (this study) top

	135 GPa	180 GPa
Fe—N1	1.73 (2)	1.707 (10)
Fe—N1ⁱ	1.73 (2)	1.707 (10)
Fe—N2ⁱⁱ	1.81 (3)	1.783 (14)
Fe—N2ⁱⁱⁱ	1.81 (3)	1.783 (14)
Fe—N2^iv	1.78 (3)	1.763 (6)
Fe—N2^v	1.78 (3)	1.763 (6)
N1—N1^vi	1.29 (5)	1.277 (14)
N1—N2	1.30 (3)	1.298 (8)
N2—N2^vii	1.43 (4)	1.37 (3)

Symmetry codes: (i) -x+1, -y, -z; (ii) x+1, y, z+1; (iii) -x, -y, -z-1; (iv) -x+1, -y, -z-1; (v) x, y, z+1; (vi) -x, -y-1, -z-1; (vii) -x, -y-1, -z-2.

Pressures for FeN₄ synthesis based on different 3rd order Birch–Murnaghan EoS's of hcp-Fe reported in the literature [V_Fe = 15.171 (5) Å³/unit cell] top

Reference	V₀ (Å³)	K (GPa)	K'	Pressure (GPa)
Dewaele et al. (2006)	22.468 (24)	165 (fixed)	4.97 (4)	173.5(2.2)
Fei et al. (2016)	22.428 (fixed)	172.7(1.4)	4.79 (5)	174.1(1.4)
Sakai et al. (2014)	22.18 (20)	179.6(2.2)	4.91 (12)	174.9(2.1)
Mao et al. (1990)	22.35 (3)	164.8(3.6)	5.33 (9)	179.8(4.3)
Yamazaki et al. (2012)	22.15 (5)	202 (7)	4.5 (2)	181.0(5.6)
Dubrovinsky et al. (2000)	22.35 (3)	155.6(3.5)	5.81 (6)	183.7(4.8)
Garai et al. (2011)	22.33 (3)	164 (2)	5.52 (5)	183.9(2.5)
Boehler et al. (2008)	22.46 (4)	160 (6)	5.6 (2)	187.5(8.2)

Acknowledgements

The diffraction experiments were performed on beamline ID11 at the European Synchrotron Radiation Facility (ESRF), Grenoble, France.

References

Boehler, R., Santamaría-Pérez, D., Errandonea, D. & Mezouar, M. (2008). J. Phys. Conf. Ser. 121, 022018. CrossRef Google Scholar
Bykov, M., Bykova, E., Aprilis, G., Glazyrin, K., Koemets, E., Chuvashova, I., Kupenko, I., McCammon, C., Mezouar, M., Prakapenka, V., Liermann, H.-P., Tasnádi, F., Ponomareva, A. V., Abrikosov, I. A., Dubrovinskaia, N. & Dubrovinsky, L. (2018a). Nat. Commun. 9, 2756. Web of Science CrossRef Google Scholar
Bykov, M., Bykova, E., Koemets, E., Fedotenko, T., Aprilis, G., Glazyrin, K., Liermann, H.-P., Ponomareva, A. V., Tidholm, J., Tasnádi, F., Abrikosov, I. A., Dubrovinskaia, N. & Dubrovinsky, L. (2018b). Angew. Chem. Int. Ed. 57, 9048–9053. Web of Science CrossRef Google Scholar
Clark, W. P., Steinberg, S., Dronskowski, R., McCammon, C., Kupenko, I., Bykov, M., Dubrovinsky, L., Akselrud, L. G., Schwarz, U. & Niewa, R. (2017). Angew. Chem. Int. Ed. 56, 7302–7306. Web of Science CrossRef Google Scholar
Dewaele, A., Loubeyre, P., Occelli, F., Mezouar, M., Dorogokupets, P. I. & Torrent, M. (2006). Phys. Rev. Lett. 97, 215504. Web of Science CrossRef Google Scholar
Dolomanov, O. V., Bourhis, L. J., Gildea, R. J., Howard, J. A. K. & Puschmann, H. (2009). J. Appl. Cryst. 42, 339–341. Web of Science CrossRef CAS IUCr Journals Google Scholar
Dubrovinsky, L., Saxena, S., Tutti, F., Rekhi, S. & LeBehan, T. (2000). Phys. Rev. Lett. 84, 1720–1723. Web of Science CrossRef Google Scholar
Ertl, G., Huber, M. & Thiele, N. (1979). Z. Naturforsch. Teil A, 34, 30–39. Google Scholar
Fei, Y., Murphy, C., Shibazaki, Y., Shahar, A. & Huang, H. (2016). Geophys. Res. Lett. 43, 6837–6843. Web of Science CrossRef Google Scholar
Garai, J., Chen, J. & Telekes, G. (2011). Am. Mineral. 96, 828–832. Web of Science CrossRef Google Scholar
Hrubiak, R. (2017). XDI. High Pressure Collaborative Access Team, Geophysical Laboratory, Carnegie Institution of Washington, Argonne, Illinois 60439, USA. Google Scholar
Laniel, D., Weck, G., Gaiffe, G., Garbarino, G. & Loubeyre, P. (2018). J. Phys. Chem. Lett. 9, 1600–1604. Web of Science CrossRef Google Scholar
Mao, H. K., Wu, Y., Chen, L. C., Shu, J. F. & Jephcoat, A. P. (1990). J. Geophys. Res. 95, 21737. CrossRef Web of Science Google Scholar
Oxford Diffraction (2005). ABSPACK. Oxford Diffraction Ltd, Abingdon, England. Google Scholar
Rigaku OD (2018). CrysAlis PRO. Rigaku Oxford Diffraction, Yarnton, UK. Google Scholar
Sakai, T., Takahashi, S., Nishitani, N., Mashino, I., Ohtani, E. & Hirao, N. (2014). Phys. Earth Planet. Inter. 228, 114–126. Web of Science CrossRef Google Scholar
Sheldrick, G. M. (2015a). Acta Cryst. A71, 3–8. Web of Science CrossRef IUCr Journals Google Scholar
Sheldrick, G. M. (2015b). Acta Cryst. C71, 3–8. Web of Science CrossRef IUCr Journals Google Scholar
Steele, B. A. & Oleynik, I. I. (2016). Chem. Phys. Lett. 643, 21–26. Web of Science CrossRef Google Scholar
Steele, B. A. & Oleynik, I. I. (2017). J. Phys. Chem. A, 121, 8955–8961. Web of Science CrossRef Google Scholar
Steele, B. A., Stavrou, E., Crowhurst, J. C., Zaug, J. M., Prakapenka, V. B. & Oleynik, I. I. (2017). Chem. Mater. 29, 735–741. Web of Science CrossRef Google Scholar
Sun, W., Holder, A., Orvañanos, B., Arca, E., Zakutayev, A., Lany, S. & Ceder, G. (2017). Chem. Mater. 29, 6936–6946. Web of Science CrossRef Google Scholar
Xia, K., Gao, H., Liu, C., Yuan, J., Sun, J., Wang, H.-T. & Xing, D. (2018). Sci. Bull. 63, 817–824. Web of Science CrossRef Google Scholar
Yamazaki, D., Ito, E., Yoshino, T., Yoneda, A., Guo, X., Zhang, B., Sun, W., Shimojuku, A., Tsujino, N., Kunimoto, T., Higo, Y. & Funakoshi, K. (2012). Geophys. Res. Lett. 39, l20308. Web of Science CrossRef Google Scholar
Zhang, S., Zhao, Z., Liu, L. & Yang, G. (2017). J. Power Sources, 365, 155–161. Web of Science CrossRef Google Scholar

This is an open-access article distributed under the terms of the Creative Commons Attribution (CC-BY) Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original authors and source are cited.