research papers\(\def\hfill{\hskip 5em}\def\hfil{\hskip 3em}\def\eqno#1{\hfil {#1}}\)

Journal logoSTRUCTURAL SCIENCE
CRYSTAL ENGINEERING
MATERIALS
ISSN: 2052-5206

The influence of deuteration on the crystal structure of hybrid halide perovskites: a temperature-dependent neutron diffraction study of FAPbBr3

CROSSMARK_Color_square_no_text.svg

aStructure and Dynamics of Energy Materials, Helmholtz-Zentrum Berlin für Materialien und Energie, Hahn-Meitner-Platz 1, Berlin, 14109, Germany, bInstitute of Chemistry, University of Potsdam, Karl-Liebknecht-Str. 24-25, Potsdam, 14476, Germany, and cInstitute of Geological Sciences, Freie Universität Berlin, Malteserstr. 74-100, Berlin, 12249, Germany
*Correspondence e-mail: alexandra.franz@helmholtz-berlin.de

Edited by R. Černý, University of Geneva, Switzerland (Received 18 December 2019; accepted 25 February 2020; online 20 March 2020)

This paper discusses the full structural solution of the hybrid perovskite formamidinium lead tribromide (FAPbBr3) and its temperature-dependent phase transitions in the range from 3 K to 300 K using neutron powder diffraction and synchrotron X-ray diffraction. Special emphasis is put on the influence of deuteration on formamidinium, its position in the unit cell and disordering in comparison to fully hydrogenated FAPbBr3. The temperature-dependent measurements show that deuteration critically influences the crystal structures, i.e. results in partially-ordered temperature-dependent structural modifications in which two symmetry-independent molecule positions with additional dislocation of the molecular centre atom and molecular angle inclinations are present.

1. Introduction

The aristotype of the perovskite-type crystal structure is defined as A[XII]B[VI]X3 stoichiometry with a corner-sharing network of octahedra in which the A cation is cuboctahedrally surrounded by oxygen atoms. The tetravalent B cation builds together with oxygen the aforementioned corner-sharing octahedral network. Hybrid perovskites show a corner-sharing network as well but the anion is a group 17 element: bromine, iodine or chlorine. Eponymous for hybrid perovskites is the replacement of the large inorganic A cation by an organic molecule e.g. methyl­ammonium [CH3NH3]+ (abbreviated as MA) or formamidinium [H2NCH=NH2]+ (FA), forming an organic–inorganic hybrid perovskite (see right-hand view in Fig. 1[link]). At 2.53 Å the radius of the formamidinium molecule (Kieslich et al., 2014[Kieslich, G., Sun, S. & Cheetham, A. K. (2014). Chem. Sci. 5, 4712-4715.]) is one-and-a-half times that of the inorganic A cation which results in an increase in the hybrid perovskite (∼6 Å) cubic lattice parameter compared to inorganic perovskite (∼4 Å).

[Figure 1]
Figure 1
Left: inorganic `classic' oxide perovskite as an example for ABX3 stoichiometry: visualization of the cubic aristotype with the A cation centred in the cuboctahedral void of corner-sharing BX6 octahedra. Right: hybrid perovskite structure visualization with formamidinium as the A cation and with X as a halide ion.

Different temperature-dependent crystal structures in perovskites can be visualized by the Bärnighausen family tree (Bärnighausen, 1980[Bärnighausen (1980). MATCH Commun. Math. Comput. Chem. 9, 139-175.]). This family tree describes in descending order and proceeding from the aristotype (space group [Pm{\bar 3}m]) crystallographic group–subgroup relations of the hettotype structures.

These derived space groups (with lower symmetry) show different kinds of distortions (see Megaw, 1973[Megaw, H. D. (1973). Crystal Structures: a Working Approach. London: Saunders.]). Symmetry lowering, resulting from tilting of the BX6 octahedra, is also discussed by Glazer (1972[Glazer, A. M. (1972). Acta Cryst. B28, 3384-3392.]), Woodward (1997[Woodward, P. M. (1997). Acta Cryst. B53, 32-43.]), Lufaso & Woodward (2001[Lufaso, M. W. & Woodward, P. M. (2001). Acta Cryst. B57, 725-738.]) and Bock & Müller (2002[Bock, O. & Müller, U. (2002). Z. Anorg. Allg. Chem. 628, 987-992.]). Commonly used is the three-letter-notation developed by Glazer in which the magnitude of tilting of the coordination octahedra around the [100], [010] and [001] directions of the perovskite unit cell relative to the Cartesian axes is specified. A superscript defines whether the adjacent layers rotate in the same (+) or in the opposite (−) direction.

Deuteration is often applied in the study of hydrogen-containing structures by neutron diffraction, as the very high incoherent scattering length of 1H results in high background noise (Sears, 2006[Sears, V. F. (2006). International Tables for Crystallography, Vol. C, Mathematical, Physical and Chemical Tables, edited by E. Prince, pp. 444-454. Dordrecht: Springer.]). Although it was assumed that the deuteration in most cases does not influence the crystal structure (Fisher & Helliwell, 2008[Fisher, S. J. & Helliwell, J. R. (2008). Acta Cryst. A64, 359-367.]), it has been reported by Shi et al. (2018[Shi, C., Zhang, X., Yu, C.-H., Yao, Y.-F. & Zhang, W. (2018). Nat. Commun. 9, 481.]) and Whitfield et al. (2016[Whitfield, P. S., Herron, N., Guise, W. E., Page, K., Cheng, Y. Q., Milas, I. & Crawford, M. K. (2016). Sci. Rep. 6, 35685.]) that small changes in bond length and angles can occur. Merz & Kupka (2015[Merz, K. & Kupka, A. (2015). Cryst. Growth Des. 15, 1553-1558.]) report on changes in the geometric arrangements of molecules in a crystal framework after the replacement of protium (H) by deuterium (D). Furthermore, Harwell et al. (2018[Harwell, J. R., Payne, J. L., Sajjad, M. T., Heutz, F. J. L., Dawson, D. M., Whitfield, P. S., Irvine, J. T. S., Samuel, I. D. W. & Carpenter, M. A. (2018). Phys. Rev. Mater. 2, 065404.]) report on changes in physical properties caused by deuteration of the FA molecule. Contradictory reports are given by Dunitz & Ibberson (2008[Dunitz, J. D. & Ibberson, R. M. (2008). Angew. Chem. Int. Ed. 47, 4208-4210.]) and Fortes & Capelli (2018[Fortes, A. D. & Capelli, S. C. (2018). Phys. Chem. Chem. Phys. 20, 16736-16742.]) who discuss the influence of deuteration on the temperature-dependent unit-cell volume. Since hydrogen and deuterium differ in mass and spin (Shi et al., 2018[Shi, C., Zhang, X., Yu, C.-H., Yao, Y.-F. & Zhang, W. (2018). Nat. Commun. 9, 481.]) and, furthermore, deuterium shows a smaller effective (van der Waals) radius (Dunitz & Ibberson, 2008[Dunitz, J. D. & Ibberson, R. M. (2008). Angew. Chem. Int. Ed. 47, 4208-4210.]), it is reasonable to assume an influence of the deuteration on the crystal structure of this hybrid perovskite.

In this paper we discuss the full structural solution for deuterated (D4) and hydrogenous HC(NH2)2PbBr3 (abbreviated as FAPbBr3) over the temperature range from 3 to 300 K and the impact of deuteration on the crystal structure.

2. Experimental procedure

2.1. Synthesis of FAPbBr3

Powder samples were synthesized using different routes [after Saidaminov et al. (2015[Saidaminov, M. I., Abdelhady, A. L., Maculan, G. & Bakr, O. M. (2015). Chem. Commun. 51, 17658-17661.]) and Baikie et al. (2013[Baikie, T., Fang, Y., Kadro, J. M., Schreyer, M., Wei, F., Mhaisalkar, S. G., Graetzel, M. & White, T. J. (2013). J. Mater. Chem. A, 1, 5628-5641.])] for the D4 and hydrogenous samples. Hydrogenous FAPbBr3 powder was synthesized from stoichiometric mixtures of FABr (99.99% from Ossila) and PbBr2 (98+%, extra pure, from Arcos Organics) in di­methyl­formamide (DMF, 99.8%, Roth) and homogenized overnight at room temperature, followed by evaporation of the solvent at 75°C, yielding 100% FAPbBr3.

2.2. Synthesis of (D4)-FAPbBr3

16.7 g (0.160 mol, 1 equiv.) of formamidinium acetate was dissolved in 25 ml milliQ water (18.2 MΩ) in a 100 ml two-neck flask equipped with reflux condenser, dropping funnel and magnetic stirring bar. The reaction mixture was cooled in an ice bath. Subsequently, 4.5 ml hydro­bromide acid (HBr in 47% water, 0.176 mol, 1.1 equiv.) in 10 ml milliQ water was added under constant stirring. Then, HBr was added to the reaction mixture. This mixture was heated to reflux for around one hour. Afterwards water and acetic acid were distilled in a rotary evaporator until a white powder occurred. Subsequently, another 4.5 ml of hydro­bromic acid in 10 ml milliQ water was added and distilled in the rotary evaporator until 19.87 g (99.35%) formamidinium bromide, as a white crystalline powder resulted.

In a 50 ml flask with stirring bar and a stopper, 8 g of formamidinium bromide was dissolved in ca 25 ml of deuterium oxide. Additionally 1–2 ml (0.064 mol) of deuterium bromide in D2O were added. After one hour of stirring, the deuterium oxide was removed by rotary evaporation. This procedure was repeated three times, yielding 7.95 g formamidinium bromide D4. Afterwards the dry product was used in the following steps without further purification. The final reaction step was similar to the aforementioned synthesis procedure for the hydrogenous FAPbBr3 only exchanging FABr with (D4)-FABr. The samples were stored under N2 to avoid potential degradation by oxygen and humidity.

2.3. Neutron powder diffraction and synchrotron X-ray powder diffraction

For a reliable detection of hydrogen/deuterium positions and to distinguish between C and N, neutron powder diffractograms were collected at the fine-resolution powder diffractometer E9 (FIREPOD) at the BER II neutron source at Helmholtz-Zentrum Berlin (HZB) (Franz & Hoser, 2017[Franz, A. & Hoser A. (2017). J. Large-Scale Res. Facil. 3, A103.]). A deuterated (D4) and a hydrogenated FAPbBr3 sample were placed in 6 mm diameter vanadium cans and measured in a temperature range of 3–300 K using a dry cryostat (TROK). A wavelength of λ = 1.7982 (1) Å (511-Ge) was applied.

X-ray powder diffraction data in the range of 5°< 2θ < 134° were collected for FAPbBr3 at the diffraction end station of KMC-2 beamline (BESSY II, HZB; Többens & Zander, 2016[Többens, D. S. & Zander, S. (2016). J. Large-Scale Res. Facil. 2, A49.]) using a radiation energy of 8048 eV [λ = 1.5406 (1) Å]. Additionally, for low-temperature experiments a modified Gifford–McMahon (GM) closed-cycle cryocooler, in-house label CCR-XRD, configured with a double Kapton cupola and helium exchange gas was used (HZB, 2018[HZB (2018). Temperature environments for KMC-2, https://www.helmholtz-berlin.de/user/experimental-infrastructures/sample-environment/se-at-bessy/Temp-environment-kmc2/.]). For an exact determination of the phase transition temperatures selected-region powder diffractograms (25°< 2θ < 35°) in the range from 20 to 285 K in 1 K steps were additionally taken.

2.4. Analysis

2.4.1. DFT calculations

As part of the structure determination process, in order to evaluate potential structure candidates, a number of Density Functional Theory (DFT) calculations were conducted on different fully ordered arrangements of the formamidinium cation in the ortho­rhombic structure, assuming space group Pnma and subgroups thereof. Calculations were performed from first principles with the program CRYSTAL14 (Dovesi et al., 2014[Dovesi, R., Orlando, R., Erba, A., Zicovich-Wilson, C. M., Civalleri, B., Casassa, S., Maschio, L., Ferrabone, M., De La Pierre, M., D'Arco, P., Noël, Y., Causà, M., Rérat, M. & Kirtman, B. (2014). Int. J. Quantum Chem. 114, 1287-1317.]) using 3D-periodic DFT with Gaussian basis sets and the PBE0 Hamiltonian (Adamo & Barone, 1999[Adamo, C. & Barone, V. (1999). J. Chem. Phys. 110, 6158-6170.]). Basis sets and other computational parameters were as used previously by us (Schuck et al., 2018[Schuck, G., Többens, D. M., Koch-Müller, M., Efthimiopoulos, I. & Schorr, S. (2018). J. Phys. Chem. C, 122, 5227-5237.]). The crystal structures were allowed to fully relax upon energy minimization. The molecule geometry from the optimization yielding the lowest energy was used as a rigid unit in both structure determination and refinement, and was kept unchanged in the final results (see Table S1).

2.4.2. Structure analysis and Rietveld refinement

The structure determination was performed by parallel tempering using FOX (Favre-Nicolin & Cerný, 2002[Favre-Nicolin, V. & Cerný, R. (2002). J. Appl. Cryst. 35, 734-743.]; https://objcryst.sourceforge.net). During this process, the positions of Pb, Br, FA were restrained to stay close to their sites within the perovskite structure. Geometry restraints of the molecule were treated as flexible using the software's default values for the initial search, followed by strict idealization and subsequent optimization as a rigid body. Structure solution calculations were performed independently for D4 and hydrogenated forms.

For tetragonal and orthorhombic perovskite structures, multiple attempts were conducted to improve the results using either symmetry reduction to a subgroup or multiple independent formamidinium molecules. None of these resulted in significant improvements.

Subsequent structure refinement applying the Rietveld method was carried out with the FullProf Suite (Rodríguez-Carvajal, 1993[Rodríguez-Carvajal, J. (1993). Physica B, 192, 55-69.]), using the rigid-body option of the software for the formamidinium cation. Compared with the structure determination software FOX, this allowed an enhanced modelling of the atomic displacement with individual isotropic and anisotropic displacement parameters, and TLS-displacement (translation-libration-screw rotation) of the molecule (Schomaker & Trueblood, 1968[Schomaker, V. & Trueblood, K. N. (1968). Acta Cryst. B24, 63-76.]). It was found, however, that anisotropic and TLS parameters correlated strongly with the molecular position, without resulting in significant improvements of the refinement. In our experience this is characteristic for structures with high disorder related to small deviations of the structure from a broken higher symmetry. Hence, only a simple displacement model was selected, with isotropic displacement parameters for all atoms and a uniform displacement parameter for all atoms of the formamidinium cation.

At temperatures of 10, 240 and 300 K, neutron diffraction data were collected for both (D4)-FAPbBr3 and FAPbBr3. It was neither possible to find a structure allowing a joint refinement of these data sets with a single set of atomic positions nor to describe the diffraction pattern of the deuterated sample with the structure parameters refined from the hydrogenated one and vice versa. The resulting structural differences are highly significant and very strong, indicating that the structural differences between (D4)-FAPbBr3 and FAPbBr3 discussed here are not negligible details.

3. Results

3.1. The cubic Pm[{\bar 3}]m structure at 300 K

The overall room-temperature structures of (D4)-FAPbBr3 and FAPbBr3 were found to be similar to studies performed by Schueller et al. (2018[Schueller, E. C., Laurita, G., Fabini, D. H., Stoumpos, C. C., Kanatzidis, M. G. & Seshadri, R. (2018). Inorg. Chem. 57, 695-701.]) and show the Glazer notation a0a0a0. No octahedral tilting is present.

In a refinement from laboratory X-ray powder diffraction data, Hanusch et al. (2014[Hanusch, F. C., Wiesenmayer, E., Mankel, E., Binek, A., Angloher, P., Fraunhofer, C., Giesbrecht, N., Feckl, J. M., Jaegermann, W., Johrendt, D., Bein, T. & Docampo, P. (2014). J. Phys. Chem. Lett. 5, 2791-2795.]) placed the central carbon of the FA molecule as the rotational centre at the 1b Wyckoff position at (½, ½, ½) in the centre of the cuboctahedral cage. Nitro­gen was placed twelvefold disordered at (y, y, ½) with y = 0.3363 (5), with both nitro­gen atoms in the molecule represented by the same Wyckoff position. This simplifying approach restricts the N—C—N angle to a reasonable value of 120°.

From single-crystal X-ray diffraction data, Govinda et al. (2018[Govinda, S., Kore, B. P., Swain, D., Hossain, A., De, C., Guru Row, T. N. & Sarma, D. D. (2018). J. Phys. Chem. C, 122, 13758-13766.]) refined this model by allowing both atoms to deviate from their high-symmetry site, with C at 6f (x, ½, ½) and N at 24l (x, y, ½). This refinement places the carbon 0.42 (1) Å away from the centre of the cage, but does not fundamentally change the model. In both cases, the high site symmetry [m\bar 3m] of 1b results in a sphere of disordered nitro­gen atoms around a carbon atom or a small sphere of disordered carbon atoms. Hydrogen atoms are not included in either model.

Our approach, using rigid-body modelling to restrict the number of free parameters, as a matter of principle places every atom of the molecule on a low symmetry, general Wyckoff site 48n (x, y, z). This of course results again in embedded spheres of carbon, nitro­gen and hydrogen. However, in contrast to previously published models the distribution of scattering power density within the spheres must be consistent with intramolecular geometry. This, together with the non-negligible scattering power of hydrogen, allows for a determination of the real positioning of the individual molecules underlying the disordered distribution.

Our results for FAPbBr3 are in good agreement with the above-mentioned results. Carbon is located almost in the middle of the cubic unit cell, at 0.49 Å from the centre, and very close to the positions found by Govinda et al. (2018[Govinda, S., Kore, B. P., Swain, D., Hossain, A., De, C., Guru Row, T. N. & Sarma, D. D. (2018). J. Phys. Chem. C, 122, 13758-13766.]). However, where the older models restrict the nitro­gen to a single sphere, we found the two independent nitro­gen atoms to occupy very different positions. One, 1.03 Å from the centre approximately at (y, y, ½), conforms to the position found in the previous studies. The other, 1.36 Å from the centre and approximately at (x, ½, ½), is a newly identified position. Both positions place nitro­gen at approximately 3.2 Å from the bromine anion, in positions to form hydrogen bonds. Overall, this structure is not significantly different from the published structures, but described by a more detailed model.

The situation is different for (D4)-FAPbBr3 (see Fig. 2[link]). In the deuterated compound the FA cation is found to be heavily decentred, with the central carbon atom at more than 2 Å distance from the 2b site and one of the nitro­gen atoms close to it. This displacement is accompanied by very high Debye–Waller factors in particular of the FA cation. Pb and Br also have higher Debye–Waller factors than in the hydrogenated compound. A possible reason for this high displacive disorder might be that the higher mass of the deuterium hinders the rotation of the cation and thus the breaking of the hydrogen bonds necessary for this rotation. The resulting stronger average hydrogen bond might pull the cation out of its average position in the centre of the cage. A problematic aspect of the observed structure is that it places one deuterium too close to Br, at a distance of only 1.73 Å. This, however, can be explained by geometric effects that make the distance between the average atom positions obtained by diffraction appear shorter than the real local distances. The low-temperature structure refinement shows that the displacement of Br from its high-symmetry position in the cubic structure can be more than 0.4 Å; this is in agreement with the r.m.s. displacement from the Debye–Waller factor. In addition the strong displacement of the cation in the direction lateral to the Br⋯D—N hydrogen bond also means that real local Br⋯D distances are much higher. In the following discussion of the low-temperature structures we will see that this motif of a highly displaced FA+ cation is present in all modifications of the deuterated compound.

[Figure 2]
Figure 2
Octahedral network of FAPbBr3 and (D4)-FAPbBr3 at 300 K (space group [Pm{\bar 3}m]). Displayed here is one FA position out of 48; left: FAPbBr3 unit cell with PbBr6 octahedra and the inclined FA molecule in the centre; Right: (D4)-FAPbBr3 with its FA molecule strongly dislocated from the unit-cell centre; the atoms are number and colour coded as follows: hydrogen – 1, yellow; deuterium – 2, red; nitro­gen – 3, light blue; carbon – 4, light grey, lead (octahedra) – grey; bromine – brown (edges of octahedra).

3.2. The tetragonal P4/mbm structure at 240 and 180K

The neutron data refinement of FAPbBr3 at 240 K and 180 K, and (D4)-FAPbBr3 at 240 K yield the tetragonal space group P4/mbm ([P{4 \over m}{2 \over b}{2 \over m}]) in which the octahedral network tilts in-plane around the c axis (a0a0c+) (see Fig. 3[link]). In this space group, the centre of the cage at site 2c has local symmetry mmm. As this symmetry is higher than the molecule symmetry of mm2, at least twofold disorder is inevitable. However, a significant tilt of the molecule breaks symmetry even further, resulting in a fourfold disorder with formamidinium on site 8j with site symmetry m. The placement of the molecule at this special site, where the mirror plane of the molecule coincides with the mirror plane of the space group, was not purported by the structural model, it resulted from the Rietveld refinement without significant deviations. While the preceding is valid regardless of deuteration state, the structures differ in the placement of the molecule. In the hydrogenated form, displacement from the centre of the cage is again small. This results in the formation of nearly symmetric, but weak N—H⋯Br hydrogen bonds with two opposing edges of the cage. In (D4)-FAPbBr3, the molecular centre atom carbon is strongly shifted out of the centre position. Consequently, only one N—D⋯Br hydrogen bond is formed, which is shorter. The position and orientation of the formamidinium cation in the hydrogenated form is very similar to the one underlying the disordered distribution in the cubic phase.

[Figure 3]
Figure 3
Octahedral network of FAPbBr3 at 180 and 240 K (left and middle) and (D4)-FAPbBr3 (right) shown along [001] (top) and [[10{\overline 1}]] (bottom); the crystal structures adopt space group P4/mbm with in-plane tilted octahedra around the c axis with the molecular distribution at 240 K and 180 K in [110]; in (D4)-FAPbBr3, the FA molecule is dislocated from its central position in the cuboctahedral void; the atoms are number and colour coded as follows: hydrogen – 1, yellow; deuterium – 2, red; nitro­gen – 3, light blue; carbon – 4, light grey, lead (octahedra) – grey; bromine – brown (edges of octahedra).

3.3. The orthorhombic Pnma structure at 140, 10 and 3 K

The measurements taken at the BERII neutron source were at 10 K for FAPbBr3, and at 140, 10 and 3 K for (D4)-FAPbBr3. Due to the extensively long measurement times for the hydrogenous sample (to reach satisfying neutron counts statistics of the powder pattern) only one measurement in the stability region of the orthorhombic phase could be performed.

The low-temperature phases (see Figs. 4[link], 5[link] and 6[link]) of FAPbBr3 adopt the orthorhombic crystal system with space group Pnma ([P{{{2_1}} \over n}{{{2_1}} \over m}{{{2_1}} \over a}]). The local symmetry at the centre of the cage is mb, a mirror plane perpendicular to b. This would allow for a fully ordered structure with the planar FA molecule inside the mirror plane. This is not realized. In the hydrogenated forms the planar FA molecule is oriented nearly exactly perpendicular to mb (see Fig. 5[link], left). This results in twofold disorder, with 0.5:0.5 partial occupation of symmetrically equivalent orientations. The placement of the FA cation close to the centre of the cage and its orientation are very similar to the positioning underlying the distribution in the tetragonal form of the compound.

[Figure 4]
Figure 4
Visualization of the (D4)-FAPbBr3 in which the refinement resulted in two symmetry-independent molecular positions: molecule 1 (left) is diagonally located in the Br-lozenge with the molecular centre atom C in the middle; molecule 2 (right) is strongly off-centred towards the PbBr6 network. The atoms are number and colour coded as follows: hydrogen – 1, yellow; deuterium – 2, red; nitro­gen – 3, light blue; carbon – 4, light grey, lead (octahedra) – grey; bromine – brown (edges of octahedra).
[Figure 5]
Figure 5
Visualization of octahedral networks of FAPbBr3 (left) and (D4)-FAPbBr3 (middle and right) at 10 K. (D4)-FAPbBr3 shows two symmetry-independent molecular positions in which the second molecule is strongly off-centred from the Br-lozenge centre. The atoms are number and colour coded as follows: hydrogen – 1, yellow; deuterium – 2, red; nitro­gen – 3, light blue; carbon – 4, light grey, lead (octahedra) – grey; bromine – brown (edges of octahedra).
[Figure 6]
Figure 6
(D4)-FAPbBr3 network at 3 K with visualization of the two symmetry-independent molecular positions and their different inclination angles. The atoms are number and colour coded as follows: hydrogen – 1, yellow; deuterium – 2, red; nitro­gen – 3, light blue; carbon – 4, light grey, lead (octahedra) – grey; bromine – brown (edges of octahedra).

The structures of deuterated (D4)-FAPbBr3 could not be described satisfactory with only one independent FA molecule. It was necessary to split it into two independent ones, resulting in an overall fourfold disorder (see Fig. 5[link], middle and right; Fig. 6[link]). The distribution of the cation over these two independent sites is stable, refining to occupation ratios of 0.58 (1): 0.42 (1) without any significant change over the whole temperature range (see Fig. 6[link]). The higher occupied position is once again similar to the one observed in the respective, D4 in this case, tetragonal form. The cation is in an orientation lateral to the mb plane, but shifted out of the centre of the cage and tilted, so that a single N—D⋯Br hydrogen bond can form. The cations at the lower occupied site, on the other hand, are shifted so far from the centre that one ND2 group extends into the lozenge spanned by bromine that forms the window between adjacent cages.

3.4. Temperature-dependent phase transitions

In the synchrotron diffraction pattern overview (see Fig. 7[link]) it can be clearly seen that the difference in the powder pattern is fairly small and the evidence-giving region is between 25° < 2θ < 35°. Thus, regions of interest scans were taken in 1 K steps and the phase transition temperatures could be determined. The 220/022 superlattice reflections show the orthorhombic–tetragonal phase transition at 157 K. The orthorhombic 221/122 superlattice reflections vanish at 264 K and as a consequence, mark the tetragonal–cubic phase transition (see Fig. 8[link]).

[Figure 7]
Figure 7
Overview of synchrotron powder diffraction data of FAPbBr3 taken from 20 K to 300 K in 30–60 K steps.
[Figure 8]
Figure 8
Synchrotron X-ray diffraction contour colour plot of a selected 2θ region, collected using a ramp rate of 1 K and showing the temperature-dependent phase transition of FAPbBr3 from Pnma to P4/mbm at 157 K (lower dashed black line) and from P4/mbm to [Pm{\bar 3}m] at 264 K (upper dashed black line).

Fig. 9[link] shows the temperature dependence of the lattice parameters of both hydrogenous FAPbBr3 and (D4)-FAPbBr3 as determined by the refinement of the neutron diffraction data. For a better comparability all lattice parameters are shown as pseudocubic (indexed as psc) with a, cpsc = [{\sqrt 2}ac] and bpsc = b/2. The orthorhombic, tetragonal and cubic lattice parameters are shown in Table S1. The thermal expansion of the pseudocubic lattice parameters of the hydrogenous sample follow – as expected – a clear trend up to 300 K. In contrast, the lattice parameter a of (D4)-FAPbBr3 shows an unusual behaviour shortly before the phase transition from Pnma to P4/mbm (Fig. 9[link], bottom view).

[Figure 9]
Figure 9
Evolution of the lattice parameters in terms of increasing temperature of hydrogenous (upper view) and deuterated (lower view) FAPbBr3. The inset in the bottom figure displays the zoomed region between 0 and 10 K. For a better comparability the pseudocubic lattice parameters are displayed. The error bars are smaller than the symbol sizes and the dashed lines are only a guide for the eye. The pseudocubic a lattice parameter (black squares) in (D4)-FAPbBr3 shows an unusual thermal behaviour shortly before the phase transition from Pnma to P4/mbm.

The connected PbBr6 octahedra are defined by the Pb—Br1—Pb and Pb—Br2—Pb angles (see Fig. 10[link], left-hand-side; the FA molecule is not displayed for clarity). With increasing temperature, these angles increase and converge to reach 180° at room temperature, leading to an arrangement of the octahedra as shown in Fig. 2[link] for FAPbBr3 and (D4)-FAPbBr3.

[Figure 10]
Figure 10
Comparison of the tilt angle Pb—Br2—Pb in FAPbBr3 and (D4)-FaPbBr3. The spatial extent caused by the two molecular positions leads to an increased tilting of the octahedral network. The dashed lines are only a guide for the eye.

Glazer notation of octahedral tilting in Pnma is ab+a with an out-of-phase tilting around the cubic a axes and an in-phase tilting around the cubic b axis ([001]cub = [101]orth, [010]cub = [010]orth and [100]cub = [10[\bar 1]]orth). The latter is shown in Fig. 10[link] (right) as Pb—Br2—Pb angle. In contrast to hydrogenous FAPbBr3, the tilt angle in (D4)-FAPbBr3 in the range from 3 to 240 K is lower and, furthermore, does not show a significant increase up to 240 K. The two symmetry-independent molecular sites which are present in all deuterated low-temperature modifications crystallizing in space group Pnma require more space in the voids between the octahedra. To compensate for this, the octahedral network is forced to tilt further. Since the FA molecule is located in the ac plane the lattice parameters are influenced by this as well.

Fig. 11[link] displays the mean unit-cell volumes of hydrogenous and deuterated FAPbBr3 from 3 to 300 K (calculated from the pseudocubic lattice parameters and chosen due to the enhanced comparability of the values). The graph shows the increased unit-cell volume of (D4)-FAPbBr3 due to deuteration. The unusual thermal behaviour of the a lattice parameters at 140 K (see Fig. 9[link]) is reflected in the value for the volume of (D4)-FAPbBr3 and is in the range of the value of the hydrogen-containing samples.

[Figure 11]
Figure 11
Comparison of mean unit-cell volumes of FAPbBr3 (black squares) and (D4)-FAPbBr3 (blue open circles). (D4)-FAPbBr3 shows the larger unit cell, compared to the FAPbBr3. The dashed lines are only a guide for the eye.

4. Summary

The strong influence of deuteration on the crystal structure of FAPbBr3 was demonstrated by a detailed neutron diffraction investigation over a wide temperature range. By deuteration of FAPbBr3 different partially-ordered crystal structures and increased lattice parameters have been observed. The deuterated FA molecule shows two symmetry-independent sites instead of one, leading to an increased tilt angle between the corner-sharing PbBr6 octahedra. (D4)-FAPbBr3 shows an additional molecular disordering in the Pnma region caused by a strong off-centre shift of the molecular position between the corner-sharing octahedra.

Supporting information


Computing details top

For all structures, program(s) used to solve structure: F.O.X. 1.9.6-#1295; program(s) used to refine structure: FULLPROF.2k 6.0.

(FAPbBr3_1H_10K) top
Crystal data top
CH5N2PbBr3c = 8.3604 (12) Å
Orthorhombic, PnmaV = 827.71 (19) Å3
Hall symbol: -P 2ac 2nZ = 4
a = 8.3685 (13) ÅConstant Wavelength Neutron Diffraction radiation
b = 11.8306 (9) ÅT = 10 K
Data collection top
Radiation source: nuclear reactor, E9-FIREPOD, BER II, HZB, Berlin, Germany2θmin = 4.445°, 2θmax = 134.420°, 2θstep = 0.075°
Refinement top
Rp = 1.37742 parameters
Rwp = 1.8440 restraints
Rexp = 0.961H-atom parameters constrained
RBragg = 14.512(Δ/σ)max = 0.05
1734 data points
Fractional atomic coordinates and isotropic or equivalent isotropic displacement parameters (Å2) top
xyzUiso*/UeqOcc. (<1)
P10.000000.000000.000000.0029 (12)*
Br10.040 (2)0.250000.008 (4)0.011 (4)*
Br20.771 (4)0.0164 (15)0.726 (3)0.044 (4)*
C10.485880.281230.01800.045 (2)*0.50000
H20.466590.367360.05710.045 (2)*0.50000
N30.595590.222590.09120.045 (2)*0.50000
H40.618800.141490.05960.045 (2)*0.50000
H50.658890.258540.18120.045 (2)*0.50000
N60.397910.243450.098910.045 (2)*0.50000
H70.410840.163430.140370.045 (2)*0.50000
H80.315060.294810.149410.045 (2)*0.50000
(FAPbBr3_D4_10K) top
Crystal data top
CHD4N2PbBr3c = 8.3736 (10) Å
Orthorhombic, PnmaV = 829.30 (12) Å3
Hall symbol: -P 2ac 2nZ = 4
a = 8.3712 (3) ÅConstant Wavelength Neutron Diffraction radiation
b = 11.8307 (9) ÅT = 10 K
Data collection top
Radiation source: nuclear reactor, E9-FIREPOD, BER II, HZB, Berlin, Germany2θmin = 4.435°, 2θmax = 134.411°, 2θstep = 0.075°
Refinement top
Rp = 1.79054 parameters
Rwp = 2.4460 restraints
Rexp = 1.049H-atom parameters constrained
RBragg = 11.889(Δ/σ)max = 0.05
1734 data points
Fractional atomic coordinates and isotropic or equivalent isotropic displacement parameters (Å2) top
xyzUiso*/UeqOcc. (<1)
Pb10.000000.000000.000000.016 (3)*
Br10.0072 (19)0.250000.005 (3)0.008 (4)*
Br20.773 (3)0.0192 (14)0.718 (3)0.052 (4)*
C1a0.474840.346710.004960.018 (3)*0.288 (4)
H2a0.418610.428410.019530.018 (3)*0.288 (4)
N3a0.605040.341840.08000.018 (3)*0.288 (4)
D4a0.662810.267550.09760.018 (3)*0.288 (4)
D5a0.649480.413470.12950.018 (3)*0.288 (4)
N6a0.407740.260100.073810.018 (3)*0.288 (4)
D7a0.455240.181550.064200.018 (3)*0.288 (4)
D8a0.306300.271280.138080.018 (3)*0.288 (4)
C1b0.374690.754720.160510.018 (3)*0.212 (4)
H2b0.427770.688410.231940.018 (3)*0.212 (4)
N3b0.247480.804450.216250.018 (3)*0.212 (4)
D4b0.192750.867140.154400.018 (3)*0.212 (4)
D5b0.202240.780080.323160.018 (3)*0.212 (4)
N6b0.442620.779410.024920.018 (3)*0.212 (4)
D7b0.398050.840810.04690.018 (3)*0.212 (4)
D8b0.541690.736570.00960.018 (3)*0.212 (4)
(FAPbBr3_D4_140K) top
Crystal data top
CHD4N2PbBr3c = 8.3939 (8) Å
Orthorhombic, PnmaV = 840.47 (15) Å3
Hall symbol: -P 2ac 2nZ = 4
a = 8.4303 (9) ÅConstant Wavelength Neutron Diffraction radiation
b = 11.8773 (12) ÅT = 140 K
Data collection top
Radiation source: nuclear reactor, E9-FIREPOD, BER II, HZB, Berlin, Germany2θmin = 4.434°, 2θmax = 134.410°, 2θstep = 0.075°
Refinement top
Rp = 1.62656 parameters
Rwp = 2.1750 restraints
Rexp = 1.251H-atom parameters constrained
RBragg = 11.688(Δ/σ)max = 0.05
1734 data points
Fractional atomic coordinates and isotropic or equivalent isotropic displacement parameters (Å2) top
xyzUiso*/UeqOcc. (<1)
Pb10.000000.000000.000000.019 (3)*
Br10.015 (2)0.250000.008 (3)0.012 (5)*
Br20.778 (3)0.0143 (19)0.722 (2)0.062 (6)*
C1a0.475950.340520.014050.042 (5)*0.293 (4)
H2a0.421890.422910.022630.042 (5)*0.293 (4)
N3a0.614340.333420.05440.042 (5)*0.293 (4)
D4a0.670550.258370.06560.042 (5)*0.293 (4)
D5a0.666840.404040.09710.042 (5)*0.293 (4)
N6a0.398200.255350.073020.042 (5)*0.293 (4)
D7a0.443170.176240.068320.042 (5)*0.293 (4)
D8a0.290900.268250.124460.042 (5)*0.293 (4)
C1b0.377220.745030.148210.042 (5)*0.207 (4)
H2b0.399480.660780.191880.042 (5)*0.207 (4)
N3b0.267660.803970.219160.042 (5)*0.207 (4)
D4b0.241780.883390.183280.042 (5)*0.207 (4)
D5b0.207070.769930.311740.042 (5)*0.207 (4)
N6b0.462050.780510.028560.042 (5)*0.207 (4)
D7b0.446290.858720.01720.042 (5)*0.207 (4)
D8b0.545230.729150.01970.042 (5)*0.207 (4)
(FAPbBr31H180K) top
Crystal data top
CH5N2PbBr3V = 422.58 (5) Å3
Tetragonal, P4/mbmZ = 2
Hall symbol: -P 4 2abConstant Wavelength Neutron Diffraction radiation
a = 8.4216 (4) ÅT = 180 K
c = 5.9582 (5) Å
Data collection top
Radiation source: nuclear reactor, E9-FIREPOD, BER II, HZB, Berlin, Germany2θmin = 4.435°, 2θmax = 134.410°, 2θstep = 0.075°
Refinement top
Rp = 1.23138 parameters
Rwp = 1.7060 restraints
Rexp = 1.064H-atom parameters constrained
RBragg = 14.037(Δ/σ)max = 0.05
1734 data points
Fractional atomic coordinates and isotropic or equivalent isotropic displacement parameters (Å2) top
xyzUiso*/UeqOcc. (<1)
Pb10.000000.000000.000000.0216 (19)*
Br10.000000.000000.500000.049 (7)*
Br20.2248 (11)0.7248 (11)0.000000.065 (4)*
C10.489320.01160.436290.062 (3)*0.25000
H20.462430.03850.262900.062 (3)*0.25000
N30.596880.095930.475990.062 (3)*0.25000
H40.626830.125880.635140.062 (3)*0.25000
H50.651660.150710.345740.062 (3)*0.25000
N60.411860.08910.590680.062 (3)*0.25000
H70.432190.06880.755790.062 (3)*0.25000
H80.329830.17110.545230.062 (3)*0.25000
(FAPbBr3D43K) top
Crystal data top
CHD4N2PbBr3c = 8.3655 (11) Å
Orthorhombic, PnmaV = 829.01 (17) Å3
Hall symbol: -P 2ac 2nZ = 4
a = 8.3783 (10) ÅConstant Wavelength Neutron Diffraction radiation
b = 11.8280 (13) ÅT = 3 K
Data collection top
Radiation source: nuclear reactor, E9-FIREPOD, BER II, HZB, Berlin, Germany2θmin = 4.455°, 2θmax = 134.730°, 2θstep = 0.075°
Refinement top
Rp = 1.99353 parameters
Rwp = 2.6850 restraints
Rexp = 1.394H-atom parameters constrained
RBragg = 12.001(Δ/σ)max = 0.05
1738 data points
Fractional atomic coordinates and isotropic or equivalent isotropic displacement parameters (Å2) top
xyzUiso*/UeqOcc. (<1)
Pb10.000000.000000.000000.012 (2)*
Br10.009 (2)0.250000.003 (4)0.008 (5)*
Br20.773 (3)0.0193 (15)0.718 (3)0.050 (5)*
C1a0.475170.345470.005890.016 (3)*0.285 (4)
H2a0.420210.427480.021810.016 (3)*0.285 (4)
N3a0.605220.340330.07920.016 (3)*0.285 (4)
D4a0.661820.265730.09800.016 (3)*0.285 (4)
D5a0.650710.412030.12760.016 (3)*0.285 (4)
N6a0.406790.258790.073350.016 (3)*0.285 (4)
D7a0.453070.179960.062440.016 (3)*0.285 (4)
D8a0.305580.270210.137820.016 (3)*0.285 (4)
C1b0.371130.754690.159290.016 (3)*0.215 (4)
H2b0.419770.685410.228290.016 (3)*0.215 (4)
N3b0.245790.806280.215730.016 (3)*0.215 (4)
D4b0.195190.871760.156130.016 (3)*0.215 (4)
D5b0.197860.780560.320910.016 (3)*0.215 (4)
N6b0.442160.780830.025750.016 (3)*0.215 (4)
D7b0.401770.844990.04370.016 (3)*0.215 (4)
D8b0.539430.736320.00950.016 (3)*0.215 (4)
(FAPbBr31H240K) top
Crystal data top
CH5N2PbBr3V = 426.83 (4) Å3
Tetragonal, P4/mbmZ = 2
Hall symbol: -P 4 2abConstant Wavelength Neutron Diffraction radiation
a = 8.4488 (3) ÅT = 240 K
c = 5.9795 (5) Å
Data collection top
Radiation source: nuclear reactor, E9-FIREPOD, BER II, HZB, Berlin, Germany2θmin = 4.318°, 2θmax = 141.642°, 2θstep = 0.075°
Refinement top
Rp = 1.09938 parameters
Rwp = 1.5170 restraints
Rexp = 0.645H-atom parameters constrained
RBragg = 12.321(Δ/σ)max = 0.05
1832 data points
Fractional atomic coordinates and isotropic or equivalent isotropic displacement parameters (Å2) top
xyzUiso*/UeqOcc. (<1)
Pb10.000000.000000.000000.0299 (15)*
Br10.000000.000000.500000.066 (7)*
Br20.2321 (12)0.7321 (12)0.000000.072 (4)*
C10.493210.00710.441070.072 (4)*0.25000
H20.455640.04470.276510.072 (4)*0.25000
N30.602070.101770.453090.072 (4)*0.25000
H40.641770.141460.602800.072 (4)*0.25000
H50.647980.147680.310480.072 (4)*0.25000
N60.426410.07390.613280.072 (4)*0.25000
H70.456970.04330.771320.072 (4)*0.25000
H80.342430.15790.589120.072 (4)*0.25000
(FAPbBr3D4-240K) top
Crystal data top
CHD4N2PbBr3V = 428.39 (8) Å3
Tetragonal, P4/mbmZ = 2
Hall symbol: -P 4 2abConstant Wavelength Neutron Diffraction radiation
a = 8.4652 (8) ÅT = 240 K
c = 5.9782 (8) Å
Data collection top
Radiation source: nuclear reactor, E9-FIREPOD, BER II, HZB, Berlin, Germany2θmin = 4.415°, 2θmax = 134.391°, 2θstep = 0.075°
Refinement top
Rp = 1.51340 parameters
Rwp = 2.1120 restraints
Rexp = 1.158H-atom parameters constrained
RBragg = 15.139(Δ/σ)max = 0.05
1734 data points
Fractional atomic coordinates and isotropic or equivalent isotropic displacement parameters (Å2) top
xyzUiso*/UeqOcc. (<1)
Pb10.000000.000000.000000.057 (5)*
Br10.000000.000000.500000.020 (5)*
Br20.2188 (9)0.7188 (9)0.000000.100 (6)*
C10.404600.09530.463330.255 (9)*0.25000
H20.348850.15110.320960.255 (9)*0.25000
N30.511620.011710.423920.255 (9)*0.25000
D40.567690.067780.550850.255 (9)*0.25000
D50.539440.039530.263730.255 (9)*0.25000
N60.359990.13990.662110.255 (9)*0.25000
D70.408180.09170.801420.255 (9)*0.25000
D80.275680.22420.678030.255 (9)*0.25000
(FAPbBr3_1H_300K) top
Crystal data top
CH5N2PbBr3V = 215.58 (1) Å3
Cubic, Pm3mZ = 1
Hall symbol: -P 4 2 3Constant Wavelength Neutron Diffraction radiation
a = 5.99609 (19) ÅT = 300 K
Data collection top
Radiation source: nuclear reactor, E9-FIREPOD, BER II, HZB, Berlin, Germany2θmin = 4.492°, 2θmax = 134.467°, 2θstep = 0.075°
Refinement top
Rp = 1.04339 parameters
Rwp = 1.3780 restraints
Rexp = 1.068H-atom parameters constrained
RBragg = 8.220(Δ/σ)max = 0.05
1734 data points
Fractional atomic coordinates and isotropic or equivalent isotropic displacement parameters (Å2) top
xyzUiso*/UeqOcc. (<1)
Pb10.000000.000000.000000.0415 (17)*
Br10.000000.000000.500000.082 (2)*
C10.509890.486840.579640.030 (7)*0.02083
H20.441290.403530.724340.030 (7)*0.02083
N30.378210.617390.466330.030 (7)*0.02083
H40.433470.699550.329250.030 (7)*0.02083
H50.217750.637380.515880.030 (7)*0.02083
N60.718250.449220.530710.030 (7)*0.02083
H70.791210.522700.397020.030 (7)*0.02083
H80.809210.344880.627870.030 (7)*0.02083
(FAPbBr3_D4_300K) top
Crystal data top
CHD4N2PbBr3V = 215.85 (2) Å3
Cubic, Pm3mZ = 1
Hall symbol: -P 4 2 3Constant Wavelength Neutron Diffraction radiation
a = 5.9986 (3) ÅT = 300 K
Data collection top
Radiation source: nuclear reactor, E9-FIREPOD, BER II, HZB, Berlin, Germany2θmin = 4.518°, 2θmax = 134.794°, 2θstep = 0.075°
Refinement top
Rp = 1.46639 parameters
Rwp = 1.9340 restraints
Rexp = 1.630H-atom parameters constrained
RBragg = 5.499(Δ/σ)max = 0.05
1738 data points
Fractional atomic coordinates and isotropic or equivalent isotropic displacement parameters (Å2) top
xyzUiso*/UeqOcc. (<1)
Pb10.000000.000000.000000.059 (2)*
Br10.000000.000000.500000.108 (3)*
C10.473230.271700.495290.201 (13)*0.02083
H20.362260.140360.440610.201 (13)*0.02083
N30.396520.474920.496110.201 (13)*0.02083
D40.493110.604510.545550.201 (13)*0.02083
D50.237580.504240.446540.201 (13)*0.02083
N60.674310.215580.555300.201 (13)*0.02083
D70.785320.331680.607900.201 (13)*0.02083
D80.720750.053140.549520.201 (13)*0.02083
 

Acknowledgements

The authors would like to acknowledge the use of the High Performance Computing Cluster DIRAC at the Helmholtz-Zentrum and are very grateful for the granted beam times at BESSYII, KMC-2 and BERII, E9 in Berlin.

References

First citationAdamo, C. & Barone, V. (1999). J. Chem. Phys. 110, 6158–6170.  Web of Science CrossRef CAS Google Scholar
First citationBärnighausen (1980). MATCH Commun. Math. Comput. Chem. 9, 139–175.  Google Scholar
First citationBaikie, T., Fang, Y., Kadro, J. M., Schreyer, M., Wei, F., Mhaisalkar, S. G., Graetzel, M. & White, T. J. (2013). J. Mater. Chem. A, 1, 5628–5641.  Web of Science CSD CrossRef CAS Google Scholar
First citationBock, O. & Müller, U. (2002). Z. Anorg. Allg. Chem. 628, 987–992.  Web of Science CrossRef CAS Google Scholar
First citationDovesi, R., Orlando, R., Erba, A., Zicovich-Wilson, C. M., Civalleri, B., Casassa, S., Maschio, L., Ferrabone, M., De La Pierre, M., D'Arco, P., Noël, Y., Causà, M., Rérat, M. & Kirtman, B. (2014). Int. J. Quantum Chem. 114, 1287–1317.  Web of Science CrossRef CAS Google Scholar
First citationDunitz, J. D. & Ibberson, R. M. (2008). Angew. Chem. Int. Ed. 47, 4208–4210.  Web of Science CrossRef CAS Google Scholar
First citationFavre-Nicolin, V. & Cerný, R. (2002). J. Appl. Cryst. 35, 734–743.  Web of Science CrossRef CAS IUCr Journals Google Scholar
First citationFisher, S. J. & Helliwell, J. R. (2008). Acta Cryst. A64, 359–367.  Web of Science CrossRef CAS IUCr Journals Google Scholar
First citationFortes, A. D. & Capelli, S. C. (2018). Phys. Chem. Chem. Phys. 20, 16736–16742.  CrossRef CAS PubMed Google Scholar
First citationFranz, A. & Hoser A. (2017). J. Large-Scale Res. Facil. 3, A103.  Google Scholar
First citationGlazer, A. M. (1972). Acta Cryst. B28, 3384–3392.  CrossRef CAS IUCr Journals Web of Science Google Scholar
First citationGovinda, S., Kore, B. P., Swain, D., Hossain, A., De, C., Guru Row, T. N. & Sarma, D. D. (2018). J. Phys. Chem. C, 122, 13758–13766.  CSD CrossRef CAS Google Scholar
First citationHanusch, F. C., Wiesenmayer, E., Mankel, E., Binek, A., Angloher, P., Fraunhofer, C., Giesbrecht, N., Feckl, J. M., Jaegermann, W., Johrendt, D., Bein, T. & Docampo, P. (2014). J. Phys. Chem. Lett. 5, 2791–2795.  CrossRef CAS PubMed Google Scholar
First citationHarwell, J. R., Payne, J. L., Sajjad, M. T., Heutz, F. J. L., Dawson, D. M., Whitfield, P. S., Irvine, J. T. S., Samuel, I. D. W. & Carpenter, M. A. (2018). Phys. Rev. Mater. 2, 065404.  CrossRef Google Scholar
First citationHZB (2018). Temperature environments for KMC-2, https://www.helmholtz-berlin.de/user/experimental-infrastructures/sample-environment/se-at-bessy/Temp-environment-kmc2/Google Scholar
First citationKieslich, G., Sun, S. & Cheetham, A. K. (2014). Chem. Sci. 5, 4712–4715.  CrossRef CAS Google Scholar
First citationLufaso, M. W. & Woodward, P. M. (2001). Acta Cryst. B57, 725–738.  Web of Science CrossRef CAS IUCr Journals Google Scholar
First citationMegaw, H. D. (1973). Crystal Structures: a Working Approach. London: Saunders.  Google Scholar
First citationMerz, K. & Kupka, A. (2015). Cryst. Growth Des. 15, 1553–1558.  Web of Science CrossRef CAS Google Scholar
First citationRodríguez-Carvajal, J. (1993). Physica B, 192, 55–69.  CrossRef Web of Science Google Scholar
First citationSaidaminov, M. I., Abdelhady, A. L., Maculan, G. & Bakr, O. M. (2015). Chem. Commun. 51, 17658–17661.  CrossRef CAS Google Scholar
First citationSchomaker, V. & Trueblood, K. N. (1968). Acta Cryst. B24, 63–76.  CrossRef CAS IUCr Journals Web of Science Google Scholar
First citationSchuck, G., Többens, D. M., Koch-Müller, M., Efthimiopoulos, I. & Schorr, S. (2018). J. Phys. Chem. C, 122, 5227–5237.  CrossRef CAS Google Scholar
First citationSchueller, E. C., Laurita, G., Fabini, D. H., Stoumpos, C. C., Kanatzidis, M. G. & Seshadri, R. (2018). Inorg. Chem. 57, 695–701.  CSD CrossRef CAS PubMed Google Scholar
First citationSears, V. F. (2006). International Tables for Crystallography, Vol. C, Mathematical, Physical and Chemical Tables, edited by E. Prince, pp. 444–454. Dordrecht: Springer.  Google Scholar
First citationShi, C., Zhang, X., Yu, C.-H., Yao, Y.-F. & Zhang, W. (2018). Nat. Commun. 9, 481.  CSD CrossRef PubMed Google Scholar
First citationTöbbens, D. S. & Zander, S. (2016). J. Large-Scale Res. Facil. 2, A49.  Google Scholar
First citationWhitfield, P. S., Herron, N., Guise, W. E., Page, K., Cheng, Y. Q., Milas, I. & Crawford, M. K. (2016). Sci. Rep. 6, 35685.  Web of Science CSD CrossRef PubMed Google Scholar
First citationWoodward, P. M. (1997). Acta Cryst. B53, 32–43.  CrossRef CAS Web of Science IUCr Journals 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.

Journal logoSTRUCTURAL SCIENCE
CRYSTAL ENGINEERING
MATERIALS
ISSN: 2052-5206
Follow Acta Cryst. B
Sign up for e-alerts
Follow Acta Cryst. on Twitter
Follow us on facebook
Sign up for RSS feeds