Perovskite-related hybrid noble metal iodides: Formamidinium platinum iodide [(FA)2PtIVI6] and mixed-valence methylammonium gold iodide [(MA)2AuIAuIIII6]

Article abstract of DOI:10.1016/j.ica.2017.04.060

Noble metal halide compounds are an exciting class of materials which display unique crystal chemistry with temperature and pressure. Here we report two new compounds, single-valence formamidinium platinum iodide [(FA)₂Pt^IVI₆] and mixed-valence methylammonium gold iodide [(MA)₂Au^IAu^IIII₆]. Structural changes between 300 K and 100 K have been monitored. The compounds crystallize in two different perovskite-related structure classes, with the added complexity of a disordered, “tumbling”, organic cation at 300 K. This tumbling is significantly reduced at 100 K. (FA)₂Pt^IVI₆ undergoes in-phase octahedral tilting similar to related vacancy-ordered perovskite materials. At 100 K, (MA)₂Au^IAu^IIII₆ undergoes a transition which facilitates stronger Au–I covalent bonding as made evident by a contraction of the unit cell and distances between Au and I atoms in the ab plane.

Full text of DOI:10.1016/j.ica.2017.04.060

Accepted Manuscript
Research paper
Perovskite-Related Hybrid Noble Metal Iodides: Formamidinium Platinum Io-
dide [(FA)2Pt IV I6] and Mixed-Valence Methylammonium Gold Iodide
[(MA)2Au I Au III I6]
Hayden A. Evans, Emily C. Schueller, Sara R. Smock, Guang Wu, Ram
Seshadri, Fred Wudl
PII:
S0020-1693(17)30392-4
http://dx.doi.org/10.1016/j.ica.2017.04.060
ICA 17565
DOI:
Reference:
Inorganica Chimica Acta
To appear in:
Accepted Date:
30 April 2017
Please cite this article as: H.A. Evans, E.C. Schueller, S.R. Smock, G. Wu, R. Seshadri, F. Wudl, Perovskite-Related
Hybrid Noble Metal Iodides: Formamidinium Platinum Iodide [(FA)2PtIV I6] and Mixed-Valence Methylammonium
Gold Iodide [(MA)2AuI AuIII I6], Inorganica Chimica Acta (2017), doi: http://dx.doi.org/10.1016/j.ica.2017.04.060
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Perovskite-Related Hybrid Noble Metal Iodides:
Formamidinium Platinum Iodide [(FA)2PtIV I6] and
Mixed-Valence Methylammonium Gold Iodide
[(MA)2AuIAuIIII6]
Hayden A. Evans,∗,†,‡ Emily C. Schueller,‡ Sara R. Smock,† Guang Wu,†
Ram Seshadri,∗,‡ and Fred Wudl∗,¶
†Department of Chemistry and Biochemistry, University of California
Santa Barbara, California 93106, United States
‡Materials Research Laboratory, University of California
Santa Barbara, California 93106, United States
¶Materials Department, University of California
Santa Barbara, California 93106, United States
E-mail: hevans@mrl.ucsb.edu; seshadri@mrl.ucsb.edu; wudl@chem.ucsb.edu
1

Abstract
Noble metal halide compounds are an exciting class of materials which display
unique crystal chemistry with temperature and pressure. Here we report two new
compounds, Single-valence formamidinium platinum iodide [(FA)2PtIV I6] and mixed-
valence methylammonium gold iodide [(MA)2AuIAuIIII6]. Structural changes be-
tween 300 K and 100 K have been monitored. The compounds crystallize in two dif-
ferent perovskite-related structure classes, with the added complexity of a disordered,
“tumbling”, organic cation at 300 K. This tumbling is significantly reduced at 100 K.
(FA)2PtIV I6 undergoes in-phase octahedral tilting similar to related vacancy-ordered
perovskite materials. At 100 K, (MA)2AuIAuIIII6 undergoes a transition which facili-
tates stronger Au–I covalent bonding as made evident by a contraction of the unit cell
and distances between Au and I atoms in the ab plane.
1. Introduction
Noble metal chemistry is exceedingly impactful and relevant, from catalysis to biomedical
research.1 These elements are interesting in that they can provide selective reactivity as
well as low activation barriers for oxidation and reduction. Furthermore, noble metal salts
and oxides frequently display unique electrical properties owing to the propensity of the
noble metal to occur in more than one oxidation state (mixed-valence). Mixed-valence
platinum and gold complexes have been studied as one dimensional metallic conductors.2
Though the class of mixed-valence alkali and hybrid platinum halide materials have
been studied before,3 the mixed-valence alkali gold halide complexes A2[AuIX2][AuIIIX4]
(A = K, Rb, Cs; X = Cl, Br, I), referred to as A2Au2X6 compounds hereafter, have received
more attention. These materials were first prepared and described crystallographically in
19794 as chloride and bromide derivatives5 and later as iodides.6 These materials crystal-
lize with both linear AuII2− and square-plane AuIIII−4 subunits, or alternatively described,
as a distorted perovskite structure of elongated and compressed AuI6 octahedra that stack
2

alternatively along the [001] and [110] directions. Under pressure, these materials un-
dergo a phase transition to a cubic perovskite structure (stabilizing the Au(II) ion), where
the structure of Cs2Au2Cl6 above 12.5 GPa and at room temperature undergoes a transition
7
¯
from the tetragonal I4/mmm space group to the cubic Pm3m, and Cs2Au2I6 undergoes
a transition from I4/mmm to either Immm or Ibmm.8 As pressure is increased, these
AuII6 and AuIIII6 octahedra are driven closer to equivalency, and when equivalent, the
electronic structure of Au nears the d9 valence state. Just beneath the phase transition
pressure, strong electron-phonon interactions are due to the mixed-valence effect of the
Au (which is understood to be fluctuating rapidly), whereas just above the transition, the
strong electron-phonon interactions arise from the Jahn-Teller effect.9 This phenomenon
is undoubtedly of scientific interest, especially for studying the interactions of itinerant
electrons and the implications of high pressure on conductivity.
Other hybrid organic-inorganic gold iodides have been prepared previously with uni-
valent Au,10 as well as a more recent report of two mixed-valence gold iodide compounds
made with the counter cations (NH3(CH2)7NH3)2+ and (NH3(CH2)8NH3)2+ which stabi-
lize two dimensional mixed-valence Au-I layers similar to the alkali gold iodides.11 These
compounds are exciting as potential superconductors, as well as potential members of
the flourishing hybrid perovskite and perovskite related optoelectronic field.12 In this re-
port we describe the high and low temperature structures of two hybrid compounds:
a vacancy-ordered double perovskite formamidinium platinum iodide [(FA)2PtI6] and a
mixed-valence methylammonium gold iodide [(MA)2Au2I6]. We describe the structural
changes that occur with temperature, which are relevant to future high pressure studies
of these materials, as well as to related hybrid perovskite optoelectronic material research
that is focused on the dynamics of the small organic cations and their implication on device
performance.
3

2. Experimental
Formamidinium iodide (CH(NH2)2I) was prepared by combining formamidine acetate with
hydroiodic acid. Solid formamidine acetate (1.00 eq, 99%, Sigma-Aldrich) was added to
room temperature hydroiodic acid (2.0 eq., 57% wt/wt in aqueous solution, Spectrum.)
After stirring for 2 hours, the excess solvent was removed using a rotary evaporator. The
residue was washed with boiling toluene to remove any triazine contaminant, recrystal-
lized from ethanol, vacuum filtered and dried to give white needle habit crystals. The
crystals were placed in a vial wrapped in aluminum foil and stored in a glove box.
Methylammonium iodide (CH3NH3I) was prepared following an acid-base reaction of
hydroiodic acid with methylamine. A solution of CH3NH2 (1.00 eq. 33% wt/wt solution
in absolute ethanol, Spectrum) was added to a 0 ◦C solution of hydroiodic acid (1.05 eq.,
57% wt/wt in aqueous solution, Spectrum). After stirring for 2 hours at 0 ◦C, the excess
solvent was removed using a rotary evaporator. The residue was recrystallized twice from
a hot ethanol/water mixture, washed with ether, and dried under vacuum overnight to
give white, plate habit crystals. The crystals were placed in a vial wrapped in aluminum
foil and stored in a desiccator.
The (FA)2PtI6 was prepared by dissolving 100 mg (0.376 mg) of PtCl2 (Strem, 99.9)
in 3.0 g (13 mol HI) fresh 57% wt/wt hydroiodic acid (Spectrum). The hydroiodic acid
used contained free iodine due to absence of stabilizer. After dissolution, 84.21 mg (0.752
mmol) of CH(NH2)2I was added, prompting formation of a dark precipitate. The solution
was brought to a boil, held for 10 minutes, and slowly cooled. Dark cubes of (FA)2PtI6
were washed with ether, isolated via vacuum filtration, and stored in a glove box. Pt(II) is
oxidized to Pt(IV) by the free iodine present in the stabilizer-free hydroiodic acid used in
the synthetic procedure.
The (MA)2Au2I6 was prepared by dissolving 100 mg (0.254 mmol) HAuCl4 (Strem,
99.9985%) in 3.0 g (13 mol HI) 57% wt/wt hydroiodic acid (Spectrum). The hydroiodic
acid used contained free iodine due to absence of stabilizer. After dissolution, 56.86 mg
4

(0.5078 mol) CH3NH3I solid was added, prompting a dark precipitate to form. The solu-
tion was brought to a boil and slowly cooled. Dark colored, plates of (MA)2Au2I6 form,
which were isolated via filtration. (MA)2AuI6 is very hygroscopic and soluble in most sol-
vents such as ethanol and ether. Single crystals were kept in the hydroiodic acid mother
liquor and removed just prior to the diffraction data collection. Crystals of (MA)2Au2I6
were flushed with nitrogen during the diffraction experiments.
Single crystal X-ray diffraction data of the title compounds were collected on a Bruker
KAPPA APEX II diffractometer equipped with an APEX II CCD detector using a TRIUMPH
˚
monochromator with a Mo Kα X-ray source (λ = 0.71073 A). The crystals were mounted
on a cryoloop under Paratone-N oil and kept under nitrogen. Absorption correction of the
data was carried out using the multiscan method SADABS.13 Subsequent calculations were
carried out using SHELXTL.14 Structure determination was done using intrinsic methods.
All hydrogen atom positions were omitted. Structure solution, refinement, and creation of
publication data was performed using SHELXTL. Crystal structures were prepared using
the VESTA software suite.15
3. Results and Discussion
3.1 Crystallographic analysis
Figure 1 shows the 300 K and 100 K structures for (FA)2PtI6. Figure 1(a) depicts the 300 K
structure viewed down the c-axis, and illustrates the isolated PtI−6 2 octahedra and disor-
dered formamidinium cations. This structure is the same as K2PtCl6 where Pt forms a
face-centered cubic lattice with formamidinium ions in the tetrahedral cavities (in place of
the potassium ions) corresponding to the Fm3m cubic space group.16 The formamidinium
¯
moiety has a reliably placed C atom but disordered N atoms. However, the disorder ap-
pears to have partial preferential orientation, or “clicking”, as is depicted in Figure1(a).
These orientations provide the greatest level of hydrogen bonding between the amine
5

Figure 1: Crystal structure of (FA)2PtI6. (a) 300 K structure viewed down the c-axis, (b)
100 K, viewed down the c-axis emphasizing octahedral titling and (c) 100 K structure,
CCDC submission numbers, RT = 1513695, LT = 1513696.
6

Figure 2: Positive Fourier difference map of 300 K (FA)2PtI6 structure with 95% probability
−3
˚
ellipsoids. The F(max)(eA ) was 1.1059 for this data set, and is displayed with a max
level of 0.75. (a) Electron density surrounding carbon atom. (b) Full 300 K structure
viewed down the c-axis.
7

hydrogens and nearest neighbor I atoms. Figure1(b) shows the tetragonal P4/mnc struc-
ture at 100 K, where the PtI−6 2 octahedra tilt in phase in the ab plane, as is commonly
seen in related perovskite and metal vacancy-ordered perovskite structures. The tilting of
the vacancy-ordered perovskite, as described by Glazer tilting nomenclature,17 is a0a0c+,
meaning that the octahedra rotate in phase equally in the ab plane, and not at all along the
c-axis. Figure 1(c) illustrates the structure of (FA)2PtI6 at 100 K, where the carbon atom
is reliably placed. However, in both room and low temperature structures, there was diffi-
culty assigning N density with certainty. To further our understanding of these structures,
we determined Fourier difference maps to describe the positive electron density surround-
ing the formamidinium carbon.
Figure 3: Fourier difference map of the 100 K (FA)2PtI6 structure with 95% probability
−3
˚
ellipsoids. F(max)(eA ) was 2.597 for this data set, and is displayed with a max level
of 0.97. (a) Full structure. (b) Electron density surrounding carbon. (c) Electron density
¯
along the 110, 100, and 110 planes. These were chosen to cut the formamidinium carbon
atom into 60 degree segments. The two areas of strong density in the center of each plane
are the the C atoms, and were purposefully placed for ease of comparison.
8

In Figure 2(a), it is clear that rings of electron density exist in the form of a tetrahedron,
with smaller areas of electron density are observed further out. These densities suggest
that on average, the nitrogens of the formamidinium cation are found primarily in one of
the four lobes, with the possibility of the formamidinium cation orienting in six different
directions. This more predictable behavior is expected when compared to a smaller or-
ganic cation methylammonium, which is known to shift and tumble to a greater degree,
as we describe later for the (MA)2Au2I6 compound. The electron density surrounding the
formamidinium carbon in the low temperature structure, on the other hand, is less easily
described. Figure 3(a) displays the fourier difference of the full structure, Figure 3(b) the
density surrounding the carbon, and Figure 3(c) the electron density along crystallographic
planes of interest that intersect the formamidinium carbon atom. The data suggests that
the formamidinium cation is statically disordered both in position and orientation as the
PtI6 octahedra tilt, preventing translational symmetry of the N atoms to be reasonably
discerned. In Figure 3(b), the oval shaped C ellipsoid and the density surrounding the car-
bon, leads us to believe that the formamidnium cation can off center substantially without
preference during cooling, presenting many possible orientations for the formamidinium
cation on average. The electron density maps along the planes of interest agree with this
description, as there are many areas of weak electron density which could be attributed
¯
to N density. However, as can be inferred from the 110 plane, there is observable elec-
tron density in the shape of the formamidinium cation (an elongated X shape), which may
be a more probable location for the formamidinium cation. The 110 plane suggests that
though the carbon and one nitrogen may reside in the same plane, one of the formami-
dinium nitrogens may not. These many degrees of freedom of the formamidinium cation
prevent reliable crystallographic description at low temperature, but when compared to
the alkali cation containing vacancy-ordered double perovskites, the polar formamidinium
cation provides potentially accessible properties like ferroelectricity, if the material were
aligned with an electric field. This warrants future temperature and electric field depen-
9

dent studies. Furthermore, we expect pressure-induced structural changes to be another
area of exploration, similar to work done for comparable vacancy-ordered palladium io-
dide double perovskites.18
Figure 4: Crystal structure of [(MA)2AuIAuIIII6]. (a) 300 K, emphasizing that at room
temperature, the structure is best described as discrete subunits of AuI−2 and AuI4−. An
additional view of the disordered methylammonium cation is also included
˚
. The bond length between Au and I is upper bound at 3.3 A. (b) 100 K, viewed down the
a-axis emphasizing the now slight tilting of the Au–I octahedra and square planes. The
˚
upper bound bond length remains 3.3 A, indicating that there is increased extended Au–I
covalent bonding in the ab plane. CCDC deposition numbers, RT = 1513697, LT =
1513698.
Figure 4 depicts the crystal structure of (MA)2Au2I6 at 300 K and 100 K. Figure 4(a) and
(b) show the tetragonal I4/mmm room temperature structure, and the low-temperature
P43212 structure viewed down the a-axis, respectively. The structure of (MA)2Au2I6 at
10

room temperature contains discrete subunits of AuI−2 and AuI4− in addition to consider-
ably disordered methylammonium cations. Much like the modeling that has described
methylammonium in methylammonium lead iodide (MAPbI3),19 it is clear that the cation
is tumbling within its cavity, changing orientation rapidly as it hydrogen bonds between
one of the twelve nearest neighbor iodide atoms. This is possible due to the low energy
barrier for the cation to shift.20 Interestingly, the structure of (MA)2Au2I6 at low tempera-
ture, Figure 4(b), undergoes a distortion which forms buckled Au-I sheets. This suggests
that the structure at low temperatures has increased covalent bonding between the Au of
the AuI−2 subunits and the neighboring I of the AuI−4 moieties. Prior reports have described
that the bonding between gold and iodine is analogous to polyiodides due to similar elec-
tronegativities of Au and I (2.54 and 2.66, respectively),21 with the AuI2 subunit directly
comparable with the linear I−3 anion. If we apply the same criteria for determining covalent
˚
bonding, where the I–I bonding distance is upper bound at 3.3 A, then at 100 K sheets of
Au–I square planes and octahedra form infinitely in the ab plane, as is depicted in Figure
4(b). In regards to the methylammonium cation at low temperature, the structure was
solved with a translational symmetry describing the orientation of the methylammonium
cation but with understandable difficulty in distinguishing between the nitrogen or carbon
atoms of the methylammonium. This is due to low Z contrast. It could be quite sim-
ilar to methylammonium lead iodide where orientation of the methylammonium cation
alternates consistently but this is hard to discern.22
4. Conclusion
The structures of the vacancy-ordered double perovskite (FA)2PtI6 and the mixed-valence
(MA)2Au2I6 are reported at 300 K and 100 K. At 300 K, the structure of (FA)2PtI6 can be
¯
described with the cubic space group Fm3m, and at 100 K with the tetragonal space group
P4/mnc. This reduced symmetry is caused by in-phase tilting of the discrete Pt–I octa-
11

Table 1: Crystallographic information for title compounds
Empirical formula (MA)2Au2I6
Diffractometer
Monochrometer
(MA)2Au2I6
(FA)2PtIV I6
(FA)2PtIV I6
Bruker KAPPA APEX II
TRIUMPH
mode
Omega scan
300
Temperature (K) 300
Crystal habit, color
Crystal system Tetragonal
Space group I4/mmm
100
Plate, gold
100
Cube, black
Tetragonal
P43212
Cubic
¯
Tetragonal
P4/mnc
(No. 128)
674.65
Fm3m
(No. 139)
Volume (A ) 895.8(2)
(No. 96)
1720.1(2)
8.3513(4)
8.3513(4)
24.663(2)
4
(No. 225)
1401.8(2)
11.192(5)
11.192(5)
11.192(5)
4
3
˚
˚
a (A) 8.490(7)
7.6428(5)
7.6428(5)
11.6659(7)
2
˚
b (A) 8.490(7)
˚
c (A) 12.428(9)
Z
2
ρ (g mol−1) 1213.42
Dens. (g cm−3) 4.521
Abs. (mm−1) 26.669
F000 1028
1213.42
4.686
27.776
2032
2.575
1046.63
4.959
23.193
1784
3.153
to 30.533
1046.63
4.906
23.870
872
3.187
to 25.242
2.906
2θ range
to 27.028
to 27.039
Reflections
2018(299)
(Unique)
10354 (1887)
0.0529
1032 (144)
0.0390
687 (394)
0.0155
Rint 0.0551
Final R1, wRR
0.0396, 0.1139 0.0469, 0.0870 0.0507, 0.1074 0.0266, 0.0530
[I > 2σ(I)]
R1, wRR, (all data) 0.0433, 0.1159 0.0469, 0.0900 0.0507, 0.1074 0.0269, 0.0531
2.068
and −1.742
GOF 1.246
4.457
and −2.101
1.074
1.082
and −5.908
1.291
1.806
and −2.955
1.348
−3
˚
∂F (eA )
12

hedra. This tilting is seen in the related and well-studied K2PtCl6 compound, as well as
other similar vacancy-ordered double perovskite compounds. The (FA)2PtI6 material at
300 K has a predictably disordered formamidinium cation. At 100 K, the formamidinium
cation presents static disorder, preventing accurate assignment of the formamidinium N
atoms. Due to the polar nature of the formamidinium cation, future temperature and
electric field-dependent studies are of interest to examine induced polarization of the ma-
terial. The (MA)2Au2I6 material crystallizes in the tetragonal space group of I4/mmm at
room temperature, and in the tetragonal space group P43212 at low-temperature. This
compound contains discrete AuI−2 and AuI4− subunits at room temperature, but upon cool-
ing, an increased Au–I covalent bonding pattern occurs in the ab plane, producing buckled
Au–I sheets. Due to the similar electronegativities of Au and I, this material at 100 K can
be viewed as an extended polyiodide analogue. Additionally, like the related alkali metal
compounds, we expect (MA)2Au2I6 to display interesting crystal chemistry under pressure,
forming a cubic perovskite and stabilizing the electronically interesting Au(II) ion.
Acknowledgments
This work was supported by the U.S. Department of Energy, Office of Science, Basic Energy
Sciences under award number DE-SC-0012541. The research involved the use of shared
experimental facilities of the Materials Research Laboratory at UCSB supported by the
MRSEC Program of the National Science Foundation under Award No. DMR 1121053. We
would like to thank Dr. Geneva Laurita for fruitful discussions.
References
(1) a) M. C. Gimeno, The Chemistry of Gold, Wiley-VCH, Weinheim, 2009, 1–63; b) F. R. Hartley,
The Chemistry of Platinum and Palladium, Wiley, New York, 1973 ; c) L. R. Kelland, N. P. Far-
13

rell, Platinum-Based Drugs in Cancer Therapy, Humana Press: Totawa, 2000; d) M. Albrecht,
G. Martin van Koten, Angew. Chem. Int. Ed. 2001, 40, 3750–3781.
(2) a) F. Mehran, B. A. Scott, Phys. Rev. Lett. 1973, 31, 1347–1350; b) J. M. Williams, Adv. Inorg.
Chem., 1983, 26, 235–268 and references therin; c) K. Uemura, K. Fukui, K. Yamasaki, K.
Matsumoto, M. Ebihara, Inorg. Chem. 2010, 49, 7323–7330; d) C. H. Hendon, A. Walsh, N.
Akiyama, Y. Konno, T. Kajiwara, T. Ito, H. Kitagawa, K. Saki, Nat. Commun. 2016, 7, 1–6.
(3) a) G. Thiele, D. Wagner, Z. Anorg. Allg. Chem. 1978, 446, 126–130 ; b) G. Thiele, C. Mrozek,
D. Kammerer, K. Wittmann, Z. Naturforsch. 1983, 38b, 905–910; c) G. Thiele, H. W. Rotter, W.
Ba¨chle, Z. Anorg. Allg. Chem. 1989, 572, 55–62; d) W. Baechle, H. W. Rotter, G. Thiele, R. J. H.
Clark, Inorg. Chim. Acta. 1992, 191, 121–129; e) G. Thiele, H. W. Rotter, W. Ba¨chle, Z. Anorg.
Allg. Chem. 1994, 620, 1271–1277.
(4) a) J. Stra¨hle, J. Gelinek, J. Ko¨lmel, Z. Anorg. Allg. Chem. 1979, 41, 241–260; b) J. Stra¨hle, J.
Gelinek, J. Ko¨lmel, A. M. Nemecek, Z. Naturforsch. B Chem. Sci. 1979, 34, 1047–1052.
(5) a) N. Matsushita, H. Kitagawa, N. Kojima, Acta Crystallogr. Sect. C-Cryst. Struct. Commun.
1997, 53, 663666; b) S. C. Riggs, M. C. Shapiro, F. Corredor, T. H. Geballe, I. R. Fisher, G. T.
McCandless, J. Y. Chan, J. Cryst. Growth 2012, 355, 13–16.
(6) a) H. Tanino, K. Takahashi, Solid State Commun. 1986, 59, 825–827; b) H. Kitagawa, N.
Kojima, T. Nakajima, J. Chem. Soc. Dalton. Trans, 1991, 11, 3121–3125; c) N. Kojima, A.
Tanaka, H. Sakai, Y, Maeda, Nucl. Instr. Method. Phys. Res., 1993, B76, 366–367; d) K. Ikeda,
N. Kojima, Y. Ono, Y. Kobayashi, M. Seto, X. J. Liu, Y. Moritomo, Hyperfine Interact. 2004, 154,
1–4.
(7) N. Matsushita, H. Ahsbahs, S. S. Hafner, N. Kojima, Rev. High Pressure Sci. Technol. 1998, 7,
329–331.
(8) S. Wang, S. Hirai, M. C. Shapiro, S. C. Riggs, T. H. Geballe, W. L. Mao, I. R. Fisher, Phys. Rev.
B. 2013, 87, 54104-1–54104-6.
(9) H. Kitagawa, N. Kojima, H. Takahashi, N. Mori, Synth. Metals 1993, 56, 1726–1729
14

(10) a) P. Braunstein, A. Mueller, H. Boegge, Inorg. Chem. 1986, 25, 2104–2106; b) J. R. Ferraro,
M. A. Beno, R. J. Thorn, H. H. Wang, K. S. Webb, J. M. Williams, J. Phys. Chem. Solids 1986,
47, 301–308; c) Z. Tang, A. P. Litvinchuk, H.-G. Lee, A. M. Guloy, Inorg. Chem. 1998, 37,
4752–4753; d) S. D. Gupta, J. Am. Chem. Soc. 1914, 36, 747–751.
(11) J. Guan, Z. Tang, A. M. Guloy, Chem. Commun. 1999, 18, 1833–1834.
(12) a) B. Saparov, D. B. Mitzi, Chem. Rev. 2016, 116, 4558–4596; b) Z. Xu, D. B. Mitzi, Inorg.
Chem. Commun. 2003, 42, 6589–6591; c) D. B. Mitzi, Inorg. Chem. 2000, 39, 6107–6113; d)
A. Kojima, K. Teshima, Y. Shirai, T. Miyasaka, J. Am. Chem. Soc. 2009, 131, e) G. Kieslich, S.
Sun, A. K. Cheetham, Chem. Sci. 2015, 6, 3430–3433; f) G. Kieslich, S. Sun, A. K. Cheetham,
Chem. Sci. 2014, 5, 4712–4715; g) A. J. Lehner, D. H. Fabini, H. A. Evans, C. A. H´ebert, S.
R. Smock, J. Hu, H. Wang, J. W. Zwanziger, M. L. Chabinyc, R. Seshadri, Chem. Mater. 2015,
27, 7137–7148; h) D. H. Fabini, T. Hogan, H. A. Evans, C. C. Stoumpos, M. G. Kanatzidis, R.
Seshadri, J. Phys. Chem. Lett. 2016, 7, 376–381; i) H. A. Evans, A. J. Lehner, J. G. Labram, D.
H. Fabini, O. Barreda, S. R. Smock, G. Wu, M. L. Chabinyc, R. Seshadri, F. Wudl, Chem. Mater.
2016, 28, 3607–3611; j) D. H. Fabini, G. Laurita, J. S. Bechtel, C. C. Stoumpos, H. A. Evans,
A. Van der Ven, M. G. Kanatzidis, R. Seshadri, J. Am. Chem. Soc. 2016, 138, 11820–11832; k)
D. H. Fabini, J. G. Labram, A. J. Lehner, J. S. Bechtel, H. A. Evans, A. Van der Ven, F. Wudl, M.
L. Chabinyc, R. Seshadri, Inorg. Chem., 2017, 56, 11–25
(13) SADABS; Sheldrich, G. M. University of Gottingen: Germany, 2005.
(14) SHELXTL PC, Version 6.12; Bruker AXS Inc.: Madison, WI, 2005.
(15) K. Momma, F. Izumi, Commission on Crystallogr. Comput. 2006, 7, 106–119.
(16) a) P. J. Ewing, L. Pauling, Z. Kristallogr. 1928, 68, 223230 b) R. J. Williams, D. R. Dillin, W.
O. Milligan, Acta Crystallogr. Sect. B-Struct. Sci. 1973, 29, 1369–1372.
(17) A. Glazer, Acta Crystallogr. Sect. B-Struct. Sci. 1972, 28, 3384–3392
(18) B. Schu¨pp, P. Heines, A. Savin, H.-L Keller, Inorg. Chem. 2000, 39, 732–735
15

(19) A. M. A. Leguy, J. M. Frost, A. P. McMahon, V. G. Sakai, W. Kockelmann, C. Law, X. Li, F.
Foglia, A. Walsh, B. C. Oregan, Nat. Commun. 2015, 6, 1–10.
(20) C. Quarti, G. Grancini, E. Mosconi, P. Bruno, J. M. Ball, M. M. Lee, H. J. Snaith, A. Petrozza,
F. D. Angelis, J. Phys. Chem. Lett. 2014, 5, 279–284.
(21) a) P. H. Svensson, L. Kloo, Chem. Rev. 2003, 103, 16491684; b) N. N. Greenwood, A. Earn-
shaw, Chemistry of the Elements, Pergamon Press, New York, 1990, Chapter 28 c) S. Madhu,
H. A. Evans, V. V. T. Doan-Nguyen, J. G. Labram, G. Wu, M. L. Chabinyc, R. Seshadri, F. Wudl,
Angew. Chem. Int. Ed. 2016, 55, 8032–8035.
(22) M. T. Weller, O. J. Weber, P. F. Henry, A. M. Di Pumpo, T. C. Hansen, Chem. Commun. 2015,
51, 4180–4183.
16