[(Ph)3PBr][Br7], [(Bz)(Ph)3P] 2[Br8], [(n -Bu)3MeN]2[Br 20], [C4MPyr]2[Br20], and [(Ph) 3PCl]2[Cl2I<s

Article abstract of DOI:10.1021/ic201291k

The five polyhalides [(Ph)₃PBr][Br₇], [(Bz)(Ph) ₃P]₂[Br₈], [(n-Bu)₃MeN] ₂[Br₂₀], [C₄MPyr]₂[Br₂₀] ([C₄MPyr] = N-butyl-N-meth

Full text of DOI:10.1021/ic201291k

ARTICLE
pubs.acs.org/IC
[(Ph)3PBr][Br7], [(Bz)(Ph)3P]2[Br8], [(n-Bu)3MeN]2[Br20],
[C4MPyr]2[Br20], and [(Ph)3PCl]2[Cl2I14]: Extending the Horizon of
Polyhalides via Synthesis in Ionic Liquids
Michael Wolff, Alexander Okrut, and Claus Feldmann*
Institut f€ur Anorganische Chemie, Karlsruhe Institute of Technology (KIT), Engesserstrasse 15, D-76131 Karlsruhe, Germany
S Supporting Information
b
ABSTRACT: The five polyhalides [(Ph)3PBr][Br7], [(Bz)(Ph)3-
P]2[Br8], [(n-Bu)3MeN]2[Br20], [C4MPyr]2[Br20] ([C4MPyr] =
N-butyl-N-methylpyrrolidinium), and [(Ph)3PCl]2[Cl2I14] were
prepared by the reaction of dibromine and iodine monochloride in
ionic liquids. The compounds [(Ph)3PBr][Br7] and [(Bz)(Ph)3-
P]2[Br8] contain discrete pyramidal [Br7]ꢀ and Z-shaped [Br8]2ꢀ
polybromide anions. [(n-Bu)3MeN]2[Br20] and [C4MPyr]2[Br20]
exhibit new infinite two- and three-dimensional polybromide
networks and contain the highest percentage of dibromine ever
observed in a compound. [(Ph)3PCl]2[Cl2I14] also consists of a
three-dimensional network and is the first example of an infinite
polyiodine chloride. All compounds were obtained from ionic liquids as the solvent that, on the one hand, guarantees for a high
stability against strongly oxidizing Br2 and ICl and that, on the other hand, reduces the high volatility of the molecular halogens.
’ INTRODUCTION
in extending the area of dye-sensitized thin-film solar cells (also
called Gr€atzel cells).19 Polyiodides and polybromides have also
been discussed in the context of high-power batteries or anti-
microbial filters and textiles.20,21 In addition, polybromides have
been evaluated as reactive species in Suzuki- or Heck-type
coupling reactions22,23 as well as in zincꢀbromine batteries.24
Thus, knowledge on the existence and stability of novel poly-
halides can contribute to all of these areas and may initiate novel
types of applications.
We now report on the novel compounds [(Ph)3PBr][Br7],
[(Bz)(Ph)3P]2[Br8], [(n-Bu)3MeN]2[Br20], and [C4MPyr]2[Br20]
containing the discrete anions [Br7]ꢀ and [Br8]2ꢀ as well as
infinite 2D and three-dimensional (3D) polybromide networks.
[(Ph)3PCl]2[Cl2I14], moreover, is obtained as the first example
of a polyiodine chloride 3D network. In addition, [(n-Bu)3MeN]2-
[Br20] and [C4MPyr]2[Br20] with nine molecules of dibromine
contain the highest percentage of elemental bromine ever ob-
served in a compound. All of the above polyhalides were obtained
via a comparably facile and straightforward strategy: synthesis in
ionic liquids.
The existence of polyhalides was first suggested by J€orgensen
in 1871.1 The salt [NH4][I3] was identified as the first example
and validated by crystal-structure determination later in 1935.2
Since then, polyhalides have been the subject of intensive
research efforts and have become standard chemistry textbook
knowledge during the last 80 years.3,4 The largest number of
known compounds is related to polyiodides, which have con-
siderable structural diversity.5 Their composition can be ratio-
nalized based on the general composition [Im+n] )
nꢀ [with m (I(0
and n (IꢀI)], ranging from [I3]ꢀ as the smallest representative,
through discrete polyiodide species (e.g., [I26]3ꢀ and [I29]3ꢀ) to
infinite-chain- and layer-type networks.5ꢀ8 Much less is known
regarding the lighter and more reactive halogens. Thus, poly-
bromides are limited to [Br3]ꢀ, linear [Br4]2ꢀ, Z-shaped [Br8]2ꢀ
and ring-type [Br10]2ꢀ 9ꢀ11
Moreover, the two compounds
.
[C5H6S4Br]+[(Br3ꢀ)(1/2Br2)] and [H4Tppz4+][(Brꢀ)2(Br42ꢀ)]
[H4Tppz = tetra(2-pyridyl)pyrazinium] contain infinite-chain-like
polybromide anions.12,13 The salt [TtddBr2]2+[(Brꢀ)2(Br2)3]
([TtddBr2]2+ = 4,5,9,10-tetrathiocino[1,2-b:5,6-b0]-1,3,6,8-tetra-
ethyldiimidazolyl-2,7-dibromodithionium) represents the only ex-
ample of an infinite two-dimensional (2D) polybromide network.14
With regard to chlorine and fluorine derivatives, only [Cl3]ꢀ,
V-shaped [(Cl3)(Cl2)]ꢀ, and [F3]ꢀ have been reliably identified
so far.15ꢀ17
’ EXPERIMENTAL METHODS
General Procedures. All synthetic work and sample manipulation
was performed using standard Schlenk techniques or gloveboxes. The
synthesis of the ionic liquids [C10MPyr]Br and [(n-Bu)3MeN][N(Tf)2]
Despite their role in the basic chemistry of halogens, poly-
halides are relevant for certain types of applications also. Here,
the redox system Iꢀꢀ[I3]ꢀ of classical analytical chemistry is the
most prominent.18 Moreover, polyiodides play an essential role
Received: August 2, 2011
Published: October 25, 2011
r
2011 American Chemical Society
11683
dx.doi.org/10.1021/ic201291k Inorg. Chem. 2011, 50, 11683–11694
|

Inorganic Chemistry
ARTICLE
Table 1. Crystallographic Data of Polyhalides (T = 200 K; λ(Mo Kα) = 71.073 pm)
[(Ph)3PBr][Br7]
[(Bz)(Ph)3P]2[Br8]
[(n-Bu)3MeN]2[Br20
]
[C4MPyr]2[Br20
]
[(Ph)3PCl]2[Cl2I14]
fw/g molꢀ1
space group, Z
a/pm
901.55
673.04
999.48
941.36
1221.47
P21/c, 4
1150.5(2)
1758.1(3)
1767.0(6)
90
P21, 2
P1, 2
C2/c, 8
P1, 2
1062.5(2)
1209.1(3)
1073.2(2)
90
904.5(2)
964.3(2)
1434.8(3)
103.77(3)
93.72(3)
93.73(3)
1208.7(4)
1.849
1634.0(3)
1024.2(2)
3315.4(7)
90
847.5(2)
1099.6(2)
1313.8(3)
88.77(3)
80.12(3)
75.85(2)
1169.3(4)
2.674
b/pm
c/pm
α/deg
β/deg
115.23(3)
90
96.67(3) °
90
123.02(2)
90
γ/deg
V/106 pm3
1247.2(4)
2.401
5511.3(19)
2.409
2996.8(13)
2.707
F
calc/g cmꢀ3
μ/cmꢀ1
12.93
6.736
14.543
17.127
7.487
R1/wR2 [I g 2σ(I)]
R1/wR2 (all data)
0.056/0.128
0.072/0.142
0.037/0.094
0.051/0.100
0.050/0.109
0.084/0.121
0.043/0.097
0.081/0.111
0.039/0.098
0.049/0.102
was carried out as described elsewhere.25,26 The ionic liquid [C4MPyr]OTf
was purchased from IOLITEC. All additional starting materials, including
Br2 (99.9%) and ICl (99.9%), were obtained from ABCR.
[C10MPyr]Br(400mg) and[C4MPyr][OTf] (380mg) (with[C10MPyr] =
N-decyl-N-methylpyrrolidinium; [C4MPyr] = N-butyl-N-methylpyrrolidi-
nium; [OTf] = triflate) at +100 °C, followed by the addition of excess
molecular dibromine (0.20 mL) at +50 °C, led to a light-red solution.
Cooling to room temperature resulted in the formation of cubelike, light-
red single crystals of [(Bz)(Ph)3P]2[Br8]. Subsequent to washing and
removal of the ionic liquid, the compound was obtained with a yield of
about 20%.
[(n-Bu)3MeN]2[Br20]. By the addition of an excess of molecular
dibromine (0.13 mL) to a solution of (2-bromophenyl)diphenylphosphine
(223 mg) in the ionic liquid [(n-Bu)3MeN][N(Tf)2] (0.10 mL) [with
[(n-Bu)3MeN]+ = (tributylmethyl)ammonium; [N(Tf)2] = bis(trifluoro-
methanesulfonyl)imide] and by subsequent heating to a temperature of
+50 °C, a light-red solution was obtained. Cooling to ꢀ15 °C resulted in
precipitation of the title compound as well as in partial crystallization of the
ionic liquid. Large quantities of deep-red single crystals were obtained by
slowly thawing this mixture to +5 °C. Note that the title compound was
again completely dissolved at a temperature of +10 °C. Subsequent to
washing and removal of the ionic liquid, the compound was obtained with a
yield of about 50%.
[C4MPyr]2[Br20]. By the addition of an excess of dibromine (0.33 mL)
to a eutectic, equimolar mixture of the ionic liquids [C10MPyr]Br (400 mg)
and [C4MPyr][OTf] (380 mg) (with C10MPyr = N-decyl-N-methylpyr-
rolidinium; C4MPyr = N-butyl-N-methylpyrrolidinium; [OTf] = triflate), a
light-red solution was obtained. Cooling from room temperature to ꢀ15 °C
resulted in precipitation of the title compound as well as in partial crystal-
lization of the ionic liquid. Deep-red, highly adhesive single crystals were
obtained in large quantities by slowly thawing this mixture to +5 °C.
Note that the title compound was again completely dissolved at a tem-
perature of +10 °C. Subsequent to washing and removal of the ionic
liquid, the compound was obtained with a yield of about 80%.
[(Ph)3PCl]2[Cl2I14]. By the addition of an excess of ICl (248 mg)
to a solution of triphenylphosphane (100 mg) and the ionic liquid
[C4MPyr][OTf] (0.10 mL) (with [C4MPyr] = N-butyl-N-methylpyr-
rolidinium; [OTf] = triflate), a dark-red solution was obtained. Cooling
from room temperature to +6 °C resulted in crystallization of the title
compound under the formation of a few deep-red, plate-shaped single
crystals. Subsequent to washing and removal of the ionic liquid, the
compound was obtained with a yield of about 10%.
All of the prepared polyhalides are extremely sensitive to moisture and
contact with air. With regard to the low melting point of the compounds,
suited single crystals for structure determination and refinement were
selected at about ꢀ20 °C under a cooled nitrogen flow and mounted in
perfluoropolyalkyl ether oil on top of a glass fiber. The purification of larger
quantities of polyhalides was, however, not straightforward. Because the title
compounds contain the same or a cation similar to that of the ionic liquid,
the solubility of both is largely similar in typical organic solvents. For this
reason, polyhalides are best separated from the ionic liquid on a glass frit
under cooled nitrogen by washing three to four times with small portions of
dried, ice-cooled acetonitrile or diethyl ether. Thereafter, the title com-
pounds can be dried in a flow of cooled nitrogen. Although large quantities
of all polyhalides are accessible in principle, complete separation of the
ionic liquid via the above washing procedure due to concurrent partial
dissolution of the polyhalides, nevertheless, reduces the final yield of
the pure compounds significantly. In summary, the solubility as well as the
temperature and moisture sensitivity of the polyhalides hampers the
separation of large quantities of the title compounds as required for
comprehensive analytical characterization. In the case of [(n-Bu)3MeN]2-
[Br20] and [C4MPyr]2[Br20], we were, however, successful in separating
sufficient quantities of the pure compounds (i.e., 50ꢀ100 mg).
Thermogravimetry (TG) and differential thermoanalysis were per-
formed with a STA409C device (Netzsch, Selb, Germany). The measure-
ments were performed in dried nitrogen. Crystals of [(n-Bu)3MeN]2[Br20]
and [C4MPyr]2[Br20] were selected under a flow of cool nitrogen (20 mg in
corundum crucibles) and heated thereafter from room temperature to
+600 °C at a rate of 10 °C minꢀ1. While [(n-Bu)3MeN]2[Br20] as well as
[C4MPyr]2[Br20] melt at +5 to +10 °C, both compounds were in the
liquid state when starting the TG analysis. Note, furthermore, that minor
amounts of black carbon remained inside the crucible.
[(Ph3)PBr][Br7]. Dissolving triphenylphosphine (343 mg) in an
eutectic, equimolar mixture of the ionic liquids [C10MPyr]Br (400 mg)
and [C4MPyr][OTf] (380 mg) (with [C10MPyr] = N-decyl-N-methyl-
pyrrolidinium; [C4MPyr] = N-butyl-N-methylpyrrolidinium; [OTf] =
triflate) at +100 °C, followed by the addition of an excess of molecular
dibromine (0.33 mL) at +50 °C, led to a light-red solution. Cooling this
solution to room temperature resulted in the formation of large quantities of
light-red, plate-shaped single crystals of [(Ph)3PBr][Br7]. Subsequent to
washing and removal of the ionic liquid, the compound was obtained with a
yield of about 40%.
X-ray Data Collection and Structure Solution. Data collec-
tion was performed at 200 K on an IPDS II image-plate diffractometer
(Stoe, Darmstadt) using graphite-monochromatized Mo Kα (71.073
pm) radiation and a low-temperature device. After data reduction via
X-RED (Stoe, Data Reduction Program, version 1.14, Darmstadt, 1999),
space group determination was carried out on the basis of systematic
[(Bz)(Ph)3P]2[Br8]. Dissolving benzyl(triphenyl)phosphonium bro-
mide (338 mg) in a eutectic, equimolar mixture of the ionic liquids
11684
dx.doi.org/10.1021/ic201291k |Inorg. Chem. 2011, 50, 11683–11694

Inorganic Chemistry
ARTICLE
absences via XPREP (Stoe, Data Reduction Program, version 1.14,
Darmstadt, 1999). Structure solution and refinement were performed by
direct methods and refined by full-matrix least squares on F2 with
SHELXTL (Bruker, Structure Solution and Refinement Package, version
5.1, Karlsruhe, 1998). Non-H atoms were refined with anisotropic
parameter of displacement. The position of the H atom could not be fixed
via Fourier refinement and was therefore modeled based on idealized
CꢀH bonds. X-SHAPE (Stoe, Crystal Optimisation for Numerical Absorp-
tion Correction, version 1.06, Darmstadt, 1999) was used to apply numerical
absorption correction based on crystal-shape optimization. Illustrations of
the crystal structures were prepared via DIAMOND (Crystal Impact,
Visuelles Informationssystem f€ur Kristallstrukturen, version 3.0d, Bonn,
2005). Details regarding crystal structure determination and refinement are
summarized in Table 1 and are deposited in the Supporting Information.
’ RESULTS AND DISCUSSION
Ionic-Liquid-Based Synthesis of Polyhalides. In view of the
synthesis of polyhalides [(Ph)3PBr][Br7], [(Bz)(Ph)3P]2[Br8],
[(n-Bu)3MeN]2[Br20], [C4MPyr]2[Br20], and [(Ph)3PCl]2-
[Cl2I14], the ionic liquids [C4MPyr][OTf], [C10MPyr]Br,
and [(n-Bu)3MeN][N(Tf)2] exhibit several beneficial features.
First, the ionic liquids are stable toward strongly oxidizing start-
ing materials such as elemental bromine and iodine monochlor-
ide. Second, the observed low vapor pressure of Br2 and ICl in the
applied ionic liquids is highly relevant. Qualitatively, this can be
visualized by the absence of the characteristic deep-brown (Br2)
and brownish-red (ICl) color above the liquid phase at moderate
temperatures (i.e., +10 to +30 °C). Finally, the combination of
[C10MPyr]Br and [C4MPyr][OTf] turned out to be exception-
ally advantageous.27 Herein, [C10MPyr]Br serves as a “bromide
donor”, whereas [C4MPyr][OTf] acts as a “liquifier” to establish
an eutectic mixture that is even liquid below room temperature
and thereby allows growing suited single crystals of the low-
melting halogen-rich compounds below room temperature (i.e.,
at ꢀ20 °C to +10).
Figure 1. Unit cell of [(Ph)3PBr][Br7] showing the unidirectional
orientation of the tripodal [Br7]ꢀ anions and (3 ꢁ 3) supercell along
the crystallographic (010) direction indicating a 3 + 1 coordination
(distorted tetrahedron, light red) around the central bromide anion
(Br1) resulting in a polybromide 2D network.
are uniformly oriented with their top along the crystallographic b
axis. This finding is in good agreement with the absence of a
center of inversion as well as with observation of a chiral structure
and a polar space group.
The successful preparation of novel polyhalides also points to
the role that ionic liquids may play regarding the synthesis of
highly reactive compounds in the future.28ꢀ30 To this concern,
the reaction of elemental metals and chalcogens in ionic liquids
has already led to interesting results. This includes, for instance, a
fascinating chlathrate-type germanium modification (i.e., 024-
Ge136),31 the observation of sulfur as a noncharged ligand (e.g.,
[(η1-S8)Cu(1,5,9-η3-S12)][Al(ORF)4])32 or promising layer-
type thermoelectrical materials (e.g., [Bi2Te2Br][AlCl4]).33 In
contrast, the chemistry of halogens in ionic liquids is dominated
by halogenidometalates exhibiting discrete or network-like build-
ing units.34ꢀ37 Reactions involving elemental halogens and poly-
halides, in contrast, have been sparsely performed. Thus, the
polyhalides [(Ph)3PBr][Br7], [(Bz)(Ph)3P]2[Br8], [(n-Bu)3-
MeN]2[Br20], [C4MPyr]2[Br20], and [(Ph)3PCl]2[Cl2I14] re-
present the first halogen-rich compounds prepared via an ionic-
liquid-based approach.
[(Ph)3PBr][Br7]. [(Ph)3PBr][Br7] was obtained as light-red,
plate-shaped crystals from an equimolar mixture of the ionic
liquids [C10MPyr]Br and [C4MPyr][OTf], containing dissolved
(Ph)3P and molecular dibromine in excess. According to crystal
structure analysis based on single crystals, the title compound
crystallizes monoclinically and surprisingly exhibits the polar,
noncentrosymmetric space group P21 (Table 1). Figure 1 dis-
plays the unit cell, containing [(Ph)3PBr]+ cations and pyramidal
[Br7]ꢀ anions. Note that all of the tripodal [Br7]ꢀ building units
At first sight, the BrꢀBr distances in [(Ph)3PBr][Br7] can be
categorized in two groups (Table 2). With 233.1ꢀ238.1 pm, the
shortest distances are observed for three Br2 units coordinated
to the Br atom at the top of the [Br7]ꢀ pyramid (Figure 2). These
BrꢀBr distances are significantly elongated in comparison to
the Br2 molecule in the solid state (227 pm),3,4 which points to
the influence of additional interactions. The atom representing
the top of the [Br7]ꢀ pyramid can be regarded as a bromide anion
and exhibits distances in a range of 287.4ꢀ290.7 pm to the three
Br2 units. Thus, the [Br7]ꢀ anion can be rationalized according to
the description [(Brꢀ)(Br2)3]. Although unknown as a polybro-
mide anion, such an arrangement is well-known for the polyiodides.5
For compounds containing [I7]ꢀ species, such as [Ph4P][I7],38
a
composition described by [(I3)ꢀ(I2)2] is, however, much more
common. The observed angles at the central Brꢀ (Br1) and Iꢀ (I1)
are finally very similar for [Br7]ꢀ (83.9ꢀ95.7°) as well as for [I7]ꢀ
(85.6ꢀ97.3°).
A more detailed consideration of the BrꢀBr distances shows
that extracting an isolated [Br7]ꢀ is too simplified (Figure 2). In fact,
each [Br7]ꢀ building unit is further connected to four additional
[Br7]ꢀ units via distances of 331.7 (Br1ꢀBr3) and 339.5 (Br3ꢀBr5)
pm to form an infinite layer-type arrangement. Moreover, a distance
of 349.9 pm (Br2ꢀBr8) between [Br7]ꢀ and the [(Ph)3PBr]+
11685
dx.doi.org/10.1021/ic201291k |Inorg. Chem. 2011, 50, 11683–11694

Inorganic Chemistry
ARTICLE
Table 2. Comparison of BrꢀBr Distances of the Polybromides Encountering Distances below the Doubled Van der Waals Radius
of Bromine (i.e., 370 pm)
Compound
Br0ꢀBr0/pm
Brꢀꢀ(Br2)/pm
(Br2)ꢀ(Br2)/pm
Br2 (solid) [3,4]
227
331 within layers (399 between layers)
[(Ph)3PBr][Br7]
[(Bz)(Ph)3P]2[Br8]
[(n-Bu)3MeN]2[Br20
233ꢀ238
231
287ꢀ291
250ꢀ310
294ꢀ314
291ꢀ316
332ꢀ350
360
]
231ꢀ233
229ꢀ234
327ꢀ344
325ꢀ358
[C4MPyr]2[Br20
]
Figure 3. Unit cell of [(Bz)(Ph)3P]2[Br8] with the Z-shaped [Br8]2ꢀ
anion.
the shorter first-order and the longer second-order BrꢀBr
distances is much less pronounced compared to that of
[(Ph)3PBr][Br7].
[(Bz)(Ph)3P]2[Br8]. [(Bz)(Ph)3P]2[Br8] was prepared simi-
larly to [(Ph)3PBr][Br7] by reacting excess bromine in an
equimolar, eutectic mixture of the ionic liquids [C10MPyr]Br
and [C4MPyr][OTf]. Instead of (Ph)3P, [(Bz)(Ph)3P]Br was
added here. As a result, light-red crystals of [(Bz)(Ph)3P]2[Br8]
were obtained that crystallize triclinically with space group P1
(Table 1). According to single-crystal structure analysis, the title
compound is constituted of [(Bz)(Ph)3P]+ cations and Z-shaped
[Br8]2ꢀ anions (Figure 3).
Figure 2. Intra- and intermolecular distances of the [Br7]ꢀ anion in
[(Ph)3PBr][Br7] (all distances in pm; thermal ellipsoids with 50%
probability of finding).
cation is observed. All of these BrꢀBr distances range below the
doubled van der Waals radius of the element (i.e., 370 pm).3,4 In
summary, the coordination around the central bromide anion
(Br1) can also assumed be as 3 + 1 (Figure 1). All four edges
of this tetrahedral arrangement show a different connectivity.
Hence, two edges of the distorted tetrahedra around Br1 are
directly interconnected via a single Br2 molecule (Br1ꢀBr2ꢀ
Br3 Br1, Br1 Br3ꢀBr2ꢀBr1), one edge involves two Br2
Although a Z-shaped [Br8]2ꢀ was already observed in the
compound [Q]2[Br8] (Q = quinuclidinium),5,10 a direct com-
parison of the two species, nevertheless, reveals significant dif-
ferences (Figure 4). First and most obvious, the angle Br2ꢀ
Br3ꢀBr4 with 88.0° in the title compound is much smaller com-
pared to [Q]2[Br8] (i.e., 107°). Consequently, the Z-like shape is
more compressed in [(Bz)(Ph)3P]2[Br8]. Second, the distances
Br1ꢀBr2ꢀBr3 with 251.8 and 249.8 pm in [(Bz)(Ph)3P]2[Br8]
are more equalized than those in [Q]2[Br8] (i.e., 243 and 266 pm).
Thus, the [Br8]2ꢀ anion in [(Bz)(Ph)3P]2[Br8] can be formulated
as [(Br3ꢀ)2(Br2)], whereas [Br8]2ꢀ in [Q]2[Br8] is better de-
scribed as [(Brꢀ)2(Br2)3] (Figure 4). In accordance with this view,
the BrꢀBr distance of the central bromine molecule for both species
only shows a minor difference and is slightly shorter in [(Bz)-
(Ph)3P]2[Br8] thanin[Q]2[Br8] (i.e., Br4ꢀBr4 with 230.8 vs 235.4
pm; Br4ꢀBr3 with 310.1 vs 317.1 pm). The different conforma-
tions of the [Br8]2ꢀ anions in [(Bz)(Ph)3P]2[Br8] and [Q]2[Br8],
finally, can be attributed to packing effects as well as to the shape and
volume of the respective cation. In addition, BrꢀH bridge bonding
has a significant influence. Thus, a remarkably short, and thereby
strong, H bridge bonding is observed for [Q]2[Br8] (i.e., 239 pm at
Br3).10 In contrast, just comparably weak interactions are observed
between [Br8]2ꢀ and [(Bz)(Ph)3P]+ in the title compound (276.7
and 278.4 pm at Br1; 287.8 pm at Br2).
3 3 3
3 3 3
molecules (Br1ꢀBr4ꢀBr5 Br3ꢀBr2ꢀBr1), and the final
3 3 3
edge (Br1ꢀBr6ꢀBr7) does not exhibit any additional BrꢀBr
interaction at all.
All distances in the [(Ph)3PBr]+ cation are in accordance with
the expectation. As discussed above, the cation is directly in-
volved in the polybromide network via long BrꢀBr distances.
Furthermore, long-ranging charge interaction between the cation
and the anionic polybromide naturally occurs. Finally, the short-
est Br HꢀC interaction is observed with 286.2 pm (to Br8 of
3 3 3
the [(Ph)3PBr]+ cation). With regard to what is discussed as a
HꢀBr bridge bonding (i.e., 240ꢀ280 pm),39ꢀ41 a significant
influence of such H bridge bonding can be excluded.
With regard to the alternative description of [(Ph)3PBr][Br7]
either to contain discrete [Br7]ꢀ anions or an infinite polybro-
mide 2D network, the first view seems much more viable in
summary. Thus, the distances inside the [Br7]ꢀ anion (233ꢀ291
pm) are significantly shorter compared to the longer distances to
interconnect [Br7]ꢀ (332ꢀ350 pm). This view is also coherent
with describing [(n-Bu)3MeN]2[Br20] and [C4MPyr]2[Br20] as
infinite 2D and 3D networks because the differentiation between
11686
dx.doi.org/10.1021/ic201291k |Inorg. Chem. 2011, 50, 11683–11694

Inorganic Chemistry
ARTICLE
Figure 4. Distances and angles of the [Br8]2ꢀ anion in [(Bz)(Ph)3P]2[Br8] (left) in comparison to [Q+]2[Br8] (right, Q+ = quinuclidinium)10
(all distances in pm; thermal ellipsoids with 50% probability).
As discussed for [(Ph)3PBr][Br7], second-order BrꢀBr inter-
actions (Br3ꢀBr3) can be discussed for [(Bz)(Ph)3P]2[Br8] and
indicate a tendency of the [Br8]2ꢀ anion to form infinite ∞1 [(Br3ꢀ)2-
(Br2)] chains. These second-order distances with 359.2 pm are,
however, even longer than those in [(Ph)3PBr][Br7]. Consequently,
extracting a discrete [Br8]2ꢀ anion is even more valuable for
[(Bz)(Ph)3P]2[Br8] also.
[(n-Bu)3MeN]2[Br20]. [(n-Bu)3MeN]2[Br20] was obtained by
reacting an excess of molecular dibromine with (PhBr)(Ph)2P
in [(n-Bu)3MeN][N(Tf)2]. While the deep-red crystals melt at
about +10 °C, crystal growth was performed by cooling to
ꢀ15 °C. Suitable single crystals for structure analysis were then
selected at about ꢀ20 °C. The formation of [(n-Bu)3MeN]2-
[Br20] can be rationalized to start with proceeding bromination
of (PhBr)(Ph)2P, resulting in a slow formation of Brꢀ anions. On
the basis of the alternative description [(n-Bu)3MeN]2[(Brꢀ)2-
Figure 5. Unit cell of [(n-Bu)3MeN]2[Br20] with the central bromide
anion (Br1) coordinated by five dibromine molecules (dark red with
(Br2)9], the title compound contains an exceptionally high
number of nine molecules of dibromine per formula unit.
bold line) to form distorted square-pyramidal [Br(Br2)5]ꢀ building
units (foreshadowed in light red) and the [(n-Bu)3MeN]+ cation
According to single-crystal structure analysis, [(n-Bu)3MeN]2-
[Br20] crystallizes monoclinically and, while considering all other
polyhalides presented here, with a relatively high space-group
symmetry (C2/c; Table 1). The compound consists of a central
bromide anion (Br1) that is coordinated by five dibromine mole-
cules (Figure 5). As a result, a distorted square-pyramidal co-
ordination with dibromine is obtained. Note that all central
bromide anions are equivalent by inversion symmetry. One of
the Br2 molecules (Br9ꢀBr9) serves as a direct linker between two
central bromide anions (Figure 6). The remaining four dibromine
molecules (Br2ꢀBr3, Br4ꢀBr5, Br6ꢀBr7, and Br8ꢀBr10) also
interlink the central bromide (Br1) but in each case involving a
second dibromine molecule (Br4ꢀBr5, Br2ꢀBr3, Br2ꢀBr3, and
Br9ꢀBr9) that is positioned almost perpendicular to the first. As a
consequence, the central Br1 is interconnected by one as well as by
two dibromine molecules (Figure 6).
between (disorder of Br5 and the [(n-Bu)3MeN]+ cation not shown
for clarity).
elongated in comparison to the pure element in the solid state
(227 pm).3,4 Next, the distances between dibromine molecules
and the central bromide anion range from 294.3 to 314.2 pm.
Finally, distances between 326.5 and 344.4 pm are observed
between different dibromine molecules. On the basis of these
distances, an infinite layer-type polybromide network is formed
that is oriented along the crystallographic a and b axes (Figure 7).
These polybromide layers are further interlinked to each other
via BrꢀBr interactions when including an even longer distance
between Br3 and Br5A (364.8 pm). This interaction is still with a
distance right below the doubled van der Waals radius of bromine
(370 pm)3,4 (Table 2 and Figure 7).
Three different types of BrꢀBr distances are observed in [(n-
Bu)3MeN]2[Br20] (Table 2). The shortest distances (230.9ꢀ
233.4 pm) represent the dibromine molecules, which are slightly
During structure refinement, a certain disorder was observed
for the Br5 atom and tackled by split-atom positions Br5A/Br5B
with 50% occupancy for each position (Figure 7). Such disorder
11687
dx.doi.org/10.1021/ic201291k |Inorg. Chem. 2011, 50, 11683–11694

Inorganic Chemistry
ARTICLE
of one or even more halide atoms has been observed for several
polyiodides also.5 Furthermore, the [NMe(n-Bu)3]+ cation
shows a certain disorder affecting the terminal C atoms of one
of the three butyl chains. Such disorder has also often been
observed for tetraalkylammonium cations and is here again
tackled by split-atom positions with an occupancy of 50%. The
relevant butyl chains of two cations in one position of finding
are exactly directed in front of each other, which is obviously
not possible. In fact, the bonding situation (i.e., H bridge
bonding) of the butyl chain as well as the required space in the
crystal lattice are more or less identical for both positions of
finding. Note that attempts to refine the crystal structure
by choosing the noncentrosymmetrical space group Cc still
involve structural disorder of the cation as well as of Br5, and
in addition lead to considerably larger residual values and
anisotropic parameters of thermal motion.
network are relevant to the overall bonding and contribute to the
stability of the polybromide 2D network.
At first sight, the [(n-Bu)3MeN]+ cation seems to act solely as
a kind of template within the polybromide network. A more
detailed view, however, indicates significant H bridge bonding
involving the disordered Br5 atom (Figure 8). With values of
241.1, 260.2, and 269.5 pm, the observed Br HꢀC distances
3 3 3
are in a range of what is discussed as a BrꢀH bridge bonding
(240ꢀ280 pm; Figure 8).39ꢀ41 In addition, a significantly longer
distance is observed between hydrogen and the central bromide
anion (Br1 H7A) with 311.4 pm. The latter rather points to a
3 3 3
dipoleꢀdipole interaction than to H bridge bonding. Taking the
Br1 H7A bonding situation as well as the resulting sterical
3 3 3
Figure 7. [(n-Bu)3MeN]2[Br20] with [(Brꢀ)3(Br2)4(Br2)1/2] and
[(Brꢀ)2(Br2)4] rings that are fused via edge-sharing (top). The resulting
polybromide 2D network exhibits layers along the crystallographic
(110) direction and a stacking along the c axis via longer BrꢀBr interac-
tions (i.e., 365 pm, dotted lines). The [(n-Bu)3MeN]+ cation and
disordered Br5 are omitted for clarity.
shielding due to the cation into account, the unusual square-
pyramidal [Br(Br2)5]ꢀ coordination around the central bromide
anion (Br1) can be coherently explained. In addition to H bridge
bonding, long-ranging Madelung-type charge interactions be-
tween the [(n-Bu)3MeN]+ cation and the anionic polybromide
Figure 6. Polybromide 2D network in [(n-Bu)3MeN]2[Br20], with the central bromide anion (Br1: light red) as the network node and the coordinating
dibromine molecules (Br2: dark red) as linkers (all distances in pm; all thermal ellipsoids with 50% probability).
11688
dx.doi.org/10.1021/ic201291k |Inorg. Chem. 2011, 50, 11683–11694

Inorganic Chemistry
ARTICLE
Figure 8. Shortest Br HꢀC distances between the bromine network and the [(n-Bu)3MeN]+ cation in [(n-Bu)3MeN]2[Br20] (all distances in pm;
3 3 3
all thermal ellipsoids with 50% probability).
The packing of the polybromide network in [(n-Bu)3MeN]2-
[Br20] can be described based on the distorted square-pyramidal
[(Brꢀ)(Br2)5] building unit (Figure 5). Such coordination is rare
nevertheless, be quantified in detail. Note that minor amounts of
black carbon remain inside the crucible subsequent to TG. The
overall good agreement between the experiment and calculation
finally evidences the phase purity of [(n-Bu)3MeN]2[Br20] and
its successful separation from the ionic liquid.
[C4MPyr]2[Br20]. A composition very comparable to that of
[(n-Bu)3MeN]2[Br20] was obtained with [C4MPyr]2[Br20].27 At
first glance, just the cation was exchanged and led to a second
compound, again containing a remarkable number of nine
molecules of dibromine ([C4MPyr]2[(Brꢀ)2(Br2)9]). Together,
[(n-Bu)3MeN]2[Br20] and [C4MPyr]2[Br20] contain the highest
percentage of dibromine ever observed and exhibit an even
higher X(0:XꢀI ratio as the iodine-richest polyiodide (Fc)3I29.5,6
In contrast to [(n-Bu)3MeN]2[Br20], [C4MPyr]2[Br20] is, how-
ever, with a significantly different structural arrangement and
connectivity of the Br atoms.
and has been sparsely observed for the polyiodides also.5,42
A
cross section of the layer-type polybromide illustrates the pack-
ing [(Brꢀ)(Br2)5] units and reveals an alternative description of
the polybromide network (Figure 7). Accordingly, the polybro-
mide layers are directed along the crystallographic (110) direc-
tion and stacked along the c axis with BrꢀBr interactions of 365
pm. Each layer is composed of [(Brꢀ)3(Br2)4(Br2)1/2] and
[(Brꢀ)2(Br2)4] rings that are fused via edge-sharing. In summary,
the BrꢀBr distances in [(n-Bu)3MeN]2[Br20] with 231ꢀ344 pm
are much less different from each other compared to [(Ph)3-
PBr][Br7] and [(Bz)(Ph)3P]2[Br8]. In contrast to the discrete
anions [Br7]ꢀ and [Br8]2ꢀ, this points to the presence of a
polybromide 2D network in [(n-Bu)3MeN]2[Br20].
Polybromides and dibromine-containing compounds, in general,
are rare to date. To this concern, the polybromide [TtddBr2]2+-
[(Brꢀ)2(Br2)3] and the bromidometalate [(C4H9)4N]2[Pt2Br10]-
(Br2)7 with six and seven molecules of dibromine per formula
unit, to the best of our knowledge, contain the highest percentage
of dibromine observed so far.14,43 Thus, with nine molecules of
dibromine, an even higher amount is found in [(n-Bu)3MeN]2-
[Br20]. Moreover, a comparison with polyiodides is intriguing.
Here, [Fc]3I29 ([Fc] = ferrocenium) is known as the iodine-
richest polyiodide and exhibits the highest halogen-to-halide
ratio (i.e., I(0:IꢀI = 26:3 = 8.67).5,6 Hence, [(n-Bu)3MeN]2-
[Br20] with a Br(0:BrꢀI ratio of 18:2 = 9.00 contains even more
of the elemental halogen than the iodine-richest polyiodide.
Because large quantities of [(n-Bu)3MeN]2[Br20] could be ob-
tained, the thermal stability of the compound was studied. Accord-
ing to TG, decomposition occurs with three steps (Figure 9): (a)
40ꢀ200 °C (ꢀ40.0%), (b) 200ꢀ220 °C (ꢀ24.0%), and (c)
220ꢀ340 °C (ꢀ30.6%). These steps can be rationalized based
on the following decomposition products: (a) 5Br2 (calcd
ꢀ40.0%), (b) 6HBr (calcd ꢀ24.3%), and (c) 2(C4H8Br)2-
(C4H9)(CH3)N (calcd ꢀ35.7%). Although several reactions
may partly occur in parallel (e.g., evaporation of dibromine,
radical bromination, evaporation of HBr, fragmentation of
the [(n-Bu)3MeN]+ cation), the thermal decomposition can,
[C4MPyr]2[Br20] was gained in yields of 80% by dissolving
elemental bromine in an equimolar mixture of the ionic liquids
[C10MPyr]Br and [C4MPyr]OTf. Note that the deep-red, highly
adhesive crystals melt at about +9 °C. [C4MPyr]2[Br20] crystal-
lizes triclinically with space group P1 (Table 1). Its central build-
ing unit is, similar to [(n-Bu)3MeN]2[Br20], a bromide anion
(Br1; Figure 10). Again, all central bromide anions are equivalent
by inversion symmetry. In contrast to [(n-Bu)3MeN]2[Br20], the
central bromide anion in [C4MPyr]2[Br20] is, however, coordi-
nated by six dibromine molecules. Four of these Br2 molecules
(Br2ꢀBr2, Br7ꢀBr8, Br8ꢀBr7, and Br9ꢀBr9) directly interlink
the bromide anion (Br1). The remaining two dibromine mol-
ecules (Br3ꢀBr4 and Br5ꢀBr6) also serve as linkers but in each
case involving a second dibromine molecule (Br5ꢀBr6 and
Br7ꢀBr8), positioned almost perpendicular to the first one.
Thus, Br1 is interconnected four times via two Br atoms, once via
three (Br1ꢀBr5ꢀBr6 Br7ꢀBr1) and once via four (Br1ꢀBr3ꢀ
3 3 3
Br4 Br6ꢀBr5ꢀBr1) Br atoms (Figure 10). In addition, a ninth
3 3 3
dibromine molecule (Br10ꢀBr10) is not coordinated to the central
bromide anion (Br1) but only to Br8.
Again, the BrꢀBr distances in [C4MPyr]2[Br20] can be
differentiated into three groups (Table 2 and Figure 10). The
shortest distances (229ꢀ234 pm) stem from dibromine mol-
ecules that are slightly longer than those observed for the pure
11689
dx.doi.org/10.1021/ic201291k |Inorg. Chem. 2011, 50, 11683–11694

Inorganic Chemistry
ARTICLE
Figure 9. TG of [(n-Bu)3MeN]2[Br20] (top) and [C4MPyr]2[Br20] (bottom) indicating three- and four-step weight losses at 40ꢀ340 and 60ꢀ560 °C.
element in the solid state (227 pm).3,4 Next, distances between
dibromine molecules and the central bromide anion range from
291 to 316 pm. Even longer distances between 325 and 358 pm
are finally observed between different dibromine molecules.
Again, the difference in length between these BrꢀBr distances
is much less distinctive compared to the clear gap between
shorter first-order and longer second-order distances in [(Ph)3-
PBr][Br7] and [(Bz)(Ph)3P]2[Br8]. Consequently, describing
(Figure 11). As a result, a 3∞[(Brꢀ)2(Br2)4(2Br2)2(Br2)] network
is obtained that is closely related to the CsCl type of structure.
This becomes obvious when reckoning the [C4MPyr]+ cation as
coordinated by eight [(Brꢀ)(Br2)4(2Br2)2]ꢀ building units and
vice versa (Figure 10).
For [C4MPyr]2[Br20], even the shortest Br HꢀC distance
3 3 3
(292 pm) ranges above typical values of BrꢀH bridge bonding
(240ꢀ280 pm).39ꢀ41 In contrast to [(n-Bu)3MeN]2[Br20], the
[C4MPyr]+ cation can, therefore, be considered as more or less
“naked”. In addition to its template-like function, long-ranging
Madelung-type charge interaction between [C4MPyr]+ and the
anionic network is naturely relevant to the overall stability of the
3D polybromide. According to TG (Figure 9), [C4MPyr]2[Br20]
decomposes with four steps up to a temperature of 550 °C, in-
cluding (a) 60ꢀ190 °C (ꢀ16.7%), (b) 190ꢀ290 °C (ꢀ28.1%),
(c) 290ꢀ420 °C (ꢀ23.7%), and (d) 420ꢀ550 °C (ꢀ25.1%).27
[C4MPyr]2[Br20] as a ∞[(Brꢀ)2(Br2)9] 3D network is much
3
more meaningful than any extraction of an isolated building unit.
Remarkably, the bromine network in [C4MPyr]2[Br20] is estab-
lished without involving any other element than bromine and
represents the first polybromide 3D network that is exclusively
established via interaction of bromine. Structurally, the bromine
network in [C4MPyr]2[Br20] can be rationalized based
on distorted, corner-sharing [(Brꢀ)(Br2)4(2Br2)2]ꢀ octahedra
11690
dx.doi.org/10.1021/ic201291k |Inorg. Chem. 2011, 50, 11683–11694

Inorganic Chemistry
ARTICLE
Figure 11. ∞[(Brꢀ)2(Br2)4(2Br2)2(Br2)] network in [C4MPyr]2-
[Br20] established by distorted corner-sharing [(Brꢀ)(Br2)4(2Br2)2]ꢀ
octahedra (foreshadowed in light red) with the central bromide anion
(Br1) as the network node that is interlinked by dibromine molecules
(dark red with bold line). The [C4MPyr]+ cation serves as a template
between layers of the polybromide 3D network.
3
Figure 10. [C4MPyr]2[Br20] with the central bromide anion (Br1, light
red) as the network node with coordinating dibromine molecules (Br2,
dark red) as linkers (all BrꢀBr distances in pm; all thermal ellipsoids
with 50% probability).
[(Ph)3PCl]2[Cl2I14]. While the synthesis in ionic liquids yielded
a variety of novel polybromide species, a straightforward exten-
sion is to replace Br2 by the homologous lighter and heavier
halogen, viz., iodine monochloride (ICl). Knowledge on poly-
iodine chlorides is rare to date and limited to small discrete
anions. Here, the linear trihalide with its isomers [ClꢀIꢀCl]ꢀ,
[IꢀIꢀCl]ꢀ, and [IꢀClꢀI]ꢀ has been frequently described.45ꢀ47
A crystal structure analysis of [ClꢀClꢀI]ꢀ is, however, still
lacking. Ab initio molecular orbital studies concerning the stability
of the mixed trihalide anions [X2Y]ꢀ in the gas phase and in
solution validate the isomers [XꢀYꢀX]ꢀ and [YꢀYꢀX]ꢀ to be
more stable than [YꢀXꢀX]ꢀ and [YꢀXꢀY]ꢀ if Y is heavier
than X.48,49 Moreover, [ClꢀIꢀIꢀCl]2ꢀ as a linear anion and
[ClꢀIꢀClꢀIꢀCl]ꢀ as a V-shaped anion were crystallographi-
cally characterized.50,51 A single infinite chainlike [I2Cl]ꢀ was
reported for [(PhenH)+]2[(I2Cl)ꢀ(ICl2)ꢀ] [PhenH = bis(1,10-
phenanthrolin-1-ium].46 Finally, the square-planar [ICl4]ꢀ has
been known for many years but is still part of a versatile
discussion addressing crystal structure, phase transitions, nuclear
quadrupole resonance spectroscopy, and measurements of the
heat capacity and dielectric properties.52ꢀ54
These values can be ascribed to a loss of (a) 2Br2 (calcd ꢀ17.0%),
(b) 7HBr (calcd ꢀ30.0%), (c) 2C3H6Br2 (calcd ꢀ21.4%), and (d)
C6H10NBr3/C6H11NBr2 (calcd ꢀ31.5%). The reliable quantifica-
tion as well as the clear differentiation between the thermal decom-
position of [(n-Bu)3MeN]2[Br20] and [C4MPyr]2[Br20] again
points to the phase purity of both bromine-rich compounds. In
this context, the different temperatures for complete decomposi-
tion (i.e., 340 °C for [(n-Bu)3MeN]2[Br20] and 560 °C for
[C4MPyr]2[Br20]) can be related to the different stabilities of the
respective cation. The slightly higher onset of the bromine
release (i.e., 40 °C for [(n-Bu)3MeN]2[Br20] and 60 °C for
[C4MPyr]2[Br20]), on the other hand, indicates a higher stability
of the polybromide 3D network compared to the 2D network.
Considering the melting point and vapor pressure of elemental
bromine, the infinite bromine-rich networks in [(n-Bu)3MeN]2-
[Br20] and [C4MPyr]2[Br20] turn out to be surprisingly stable.
While each single BrꢀBr bond is weak, this finding has to be
attributed to the manifold of BrꢀBr interactions that are almost
isotropically spread over the volume of the unit cell. Attempts to
further verify the BrꢀBr bonding situation via Raman spectroscopy
or quantum-chemical calculations have failed so far. In contrast to
very recent reports on Raman spectra of small, discrete polybromide
species (i.e., [Brn]ꢀ with n e 10),44 [(n-Bu)3MeN]2[Br20] and
[C4MPyr]2[Br20] exhibit just a broad, irregular-shaped absorption
between 300 and 100 cmꢀ1. This finding is to be expected in view of
the huge number of very similar BrꢀBr interactions inside the
infinite network. In addition, performing quantum-chemical calcula-
tions is definitely more than routine with regard to the restrictions of
the concrete compounds, viz., multielectron systems, low space-
group symmetry, infinite 3D networks, high polarizability of each Br
atom as well as the huge number of largely similar, comparably weak
halogenꢀhalogen interactions. Even for the long-known polyio-
dides, only a few quantum-chemical calculations are available by
now. Here, the structure of “small” polyiodides ([In]ꢀ with n e 10)
is still part of a controversial discussion.5ꢀ8,13,14 Note that the first
quantum-chemical calculations for “small” polybromides have been
reported quite recently.44
With a strategy that is very similar to the synthesis of poly-
bromides, [(Ph)3PCl]2[Cl2I14] is now obtained as the first poly-
iodine chloride 3D network. In concrete, excess iodine monochlor-
ide was reacted with (Ph3)P in [C4MPyr][OTf] as the ionic liquid
and resulted in deep-red, plate-shaped crystals exhibiting the
monoclinic space group P21/c (Table 1). According to the alter-
native description [(Ph)3PCl]2[(Clꢀ)2(I2)7], a central chloride
anion (Cl2) is here coordinated by five molecules of diiodine
(Figure 12). Four of these I2 molecules (I1ꢀI2, I3ꢀI4, I2ꢀI1, and
I4ꢀI3) directly interconnect the central chloride anions (Cl2) that
are crystallographically identical because of inversion symmetry. As a
result, a polyiodine chloride 2D network is formed. The remaining
disordered diiodine molecule (I5A/BꢀI6A/B) interlinks such a layer
to similar layers above and below. Interlinking to the next central Cl2
anion here involves an additional diiodine molecule, which is
arranged almost perpendicular to I5A/BꢀI6A/B. Besides the
square-pyramidal coordination with diiodine around the central
chloride anion, an additional diiodine molecule (I7ꢀI7) inter-
links the polyiodine chloride layers via I1 (I1 I7ꢀI7 I1).
3 3 3
3 3 3
11691
dx.doi.org/10.1021/ic201291k |Inorg. Chem. 2011, 50, 11683–11694

Inorganic Chemistry
ARTICLE
All ClꢀI distances with 301.2ꢀ309.3 pm are in a very similar
range. Extracting a discrete polyhalide anion is, therefore, not
meaningful. Thus, the central chloride anion (Cl2) can rather
be regarded as a network node that is interlinked by diiodine
molecules as a linker to form a 2D network. Herein, square-
pyramidal [Cl(I2/1)(I2/2)4]ꢀ building units are arranged via
corner-sharing to form layers along the crystallographic bc plane
(Figure 13). Noteworthy, the top of the [Cl(I2/1)(I2/2)4]ꢀ
pyramids alternates to [100] and [100] along the b axis, whereas
all pyramids are directed to [100] along the c axis. Comparing the
mean ClꢀI distance of 305.5 pm in [(Ph)3PCl]2[Cl2I14] to
polyhalides containing related ClꢀIꢀI building units is illustra-
tive. Thus, two groups of ClꢀI bonds become obvious. The first
group is represented by an isolated trihalide [ClꢀIꢀI]ꢀ exhibit-
ing a comparably short ClꢀI distance of 274.9 pm.47 In
[(PhenH)+]2[(I2Cl)ꢀ(ICl2)ꢀ], in contrast, the [ClꢀIꢀI]ꢀ
building units are interlinked by long-ranging contacts to form
infinite chains. This interlinking results in an attenuation of the
initial ClꢀI bond and significantly longer distances of 304 and
316 pm.46 Such a situation has been also observed for the anion
[ClꢀIꢀIꢀCl]2ꢀ with a ClꢀI distance of 307 pm.50 Here, strong
ClꢀH bonds are involved and result in a network via H bridge
bonding. Altogether, the ClꢀI bonding situation in [(Ph)3-
PCl]2[Cl2I14] is well in accordance with the above literature
data and points to the presence of an infinite polyiodine chloride
network.
2
As discussed above, the observed ∞[Cl(I2/1)(I2/2)4]ꢀ poly-
iodine chloride network is, furthermore, interconnected along
the a axis via two types of diiodine molecules (I5A/BꢀI6A/B and
I7ꢀI7) to form a 3D network. The layers are displaced by the
length of a ClꢀI bond (approximately 300 pm) along the c axis
(Figure 13). As a result, the diodine molecule (I5A/BꢀI6A/B) is
coordinated, on the one hand, to the central chloride anion (Cl2)
and, on the other hand, to I3 in layers positioned above and
below. The diodine molecule (I7ꢀI7) is not in contact to the
central chloride but only connected to I1 in one layer and I2
in layers above or below. The intermolecular IꢀI distances
(I1ꢀI7 = 359.4 pm) are quite strong compared to typical values
(i.e., 360ꢀ420 pm)5 and significantly shorter than the doubled
van der Waals radius of iodine (420 pm).3,4 Hence, the I1
3 3 3
I7ꢀI7 I1 interaction connects the 2∞[Cl(I2/1)(I2/2)4]ꢀ layers
3 3 3
along the a axis and plays a major role with regard to the
formation of the 3D network. Considerably longer distances
are observed in addition for I6A/B I3 (377.0 and 394.7 pm).
3 3 3
Note that the latter interaction is not essential with regard to the
polyiodine chloride 3D network.
With concern to the [P(Ph)3Cl]+ cation, again the question
arises whether any H bridge bonding occurs in the polyiodine
chloride network. With 294.4 pm, the shortest Cl H distance
3 3 3
is here observed inside the cation (Cl1 H6; Figure 14). All
Figure 12. 2 ꢁ 2 super cell of [P(Ph)3Cl]2[Cl2I14] with square-
pyramidal coordination (green) around the central chloride anion
([P(Ph)3Cl]+ cation omitted for clarity).
3 3 3
additional ClꢀH contacts are considerably longer than the sum
of the van der Waals radii (Cl = 175 pm; H = 120 pm). On the
Figure 13. [P(Ph)3Cl]2[Cl2I14] with the central chloride anion (Cl2, green) as the network node and coordinating diiodine molecules (I2, orange) as
linkers (all distances in pm; all thermal ellipsoids with 50% probability).
11692
dx.doi.org/10.1021/ic201291k |Inorg. Chem. 2011, 50, 11683–11694

Inorganic Chemistry
ARTICLE
Figure 14. Shortest Cl HꢀC distances between the polyiondine chloride network and the [PPh3Cl]+ cation in [P(Ph)3Cl]2[Cl2I14] (disorder of
3 3 3
I5A/I5B omitted for clarity; all distances in pm; all thermal ellipsoids with 50% probability).
other hand, the distance between Cl2 and H18 (310.6 pm) has to
be considered in more detail. Although this value is far above of
a H bridge bonding, the spatial vicinity between the central
chloride anion and the cation becomes obvious. On the basis
of this situation, the uncommon square-pyramidal coordination
of the [Cl(I2/1)(I2/2)4]ꢀ building unit can be rationalized simi-
larly to the case of [(n-Bu)3MeN]2[Br20]. All distances of the
[P(Ph)3Cl]+ cation, finally, are well in agreement with the
expectation.
highly interesting to verify the bonding situation. With regard
to infinite networks of the highly polarizable halogens and the
manifold of very similar, but weak interactions, such calcula-
tions will, however, be a challenge. With regard to application,
the newly discovered halogen-rich polybromide and polyio-
dine chloride networks could be interesting for saving storage
and handling the reactive halogens and interhalogens. More-
over, the compounds might be relevant for future high-power
batteries.
’ ASSOCIATED CONTENT
’ CONCLUSION
S
Supporting Information. X-ray crystallographic data in
The chemistry of the halogens is extended by the four novel
polybromides [(Ph)3PBr][Br7], [(Bz)(Ph)3P]2[Br8], [(n-Bu)3-
MeN]2[Br20], and [C4MPyr]2[Br20]. Hereof, [(n-Bu)3MeN]2-
[Br20] and [C4MPyr]2[Br20] represent infinite 2D and 3D
networks, containing the highest percentage of dibromine and
elemental halogen ever observed. [(Ph)3PCl]2[Cl2I14] consists
of an infinite halogen network also and represents the first example
of a polyiodine chloride 3D network. All compounds are accessible
by a surprisingly facile synthesis in ionic liquids. These solvents, on
the one hand, are stable against the strongly oxidizing starting
materials Br2 and ICl. On the other hand, ionic liquids reduce the
volatility of the elemental halogens and thereby allow a straightfor-
ward excess to halogen-rich compounds.
b
CIF format. This material is available free of charge via the
Internet at http://pubs.acs.org.
’ AUTHOR INFORMATION
Corresponding Author
*E-mail: claus.feldmann@kit.edu.
’ ACKNOWLEDGMENT
The authors are grateful to the Center for Functional Nano-
structures of the Deutsche Forschungsgemeinschaft at the KIT
for financial support.
Despite of almost 100 years of research on halogens and
polyhalides, the ionic-liquid-based approach allows one to extend
the horizon to a wide range of novel compounds, especially in view
of halogen-rich polyhalides. While just an exchange of the cation
results in new polybromide networks (i.e., [(n-Bu)3MeN]2[Br20],
[C4MPyr]2[Br20]), one may expect many more polyhalides and
halogen-rich compounds from the ionic-liquid-based approach.
Although each single halogenꢀhalogen interaction is weak, the
manifold of interactions that are largely isotropically arranged
inside the infinite networks can be regarded as an important
feature in view of the stability of the polybromides. Another
important issue is related to the long-ranging Madelung-type
charge interaction between the spacious cation and the anionic
polyhalide network. Quantum-chemical calculations could by
’ REFERENCES
(1) J€orgensen, S. M. J. Prakt. Chem. 1871, 3, 328–349.
(2) Mooney, R. C. L. Z. Kristallogr. 1935, 90, 143–150.
(3) Shriver, D.; Atkins, P. Inorganic Chemistry; Oxford Textbooks:
London, 2006.
(4) Hollemann, A. F.; Wiberg, N. Lehrbuch der Anorganischen
Chemie; de Gruyter: Berlin, 2007.
(5) Svenson, P. H.; Kloo, L. Chem. Rev. 2003, 103, 1649–1684.
(6) Tebbe, K. F.; Buchem, R. Angew. Chem. 1997, 36, 1345–1346.
Angew. Chem., Int. Ed. 1997, 36, 1345–1346.
(7) Tebbe, K. F.; Buchem, R. Eur. J. Inorg. Chem. 2000, 6, 1275–1282.
(8) Menon, S.; Rajasekharan, M. V. Inorg. Chem. 1997, 36, 4983–4987.
(9) Stromme, K. O. Acta Chem. Scand. 1959, 13, 2089–2100.
11693
dx.doi.org/10.1021/ic201291k |Inorg. Chem. 2011, 50, 11683–11694

Inorganic Chemistry
ARTICLE
(10) Robertson, K. N.; Bakshi, P. K.; Cameron, T. S.; Knop, O. Z.
Anorg. Allg. Chem. 1997, 623, 104–114.
(11) Cunningham, C. W.; Burns, G. R.; McKee, V. Inorg. Chim. Acta
1990, 167, 135–137.
(45) Aragoni, M. C.; Arca, M.; Devillanova, F. A.; Hursthouse, M. B.;
Huth, S. L.; Isaia, F.; Lippolis, V.; Mancini, A.; Verani, G. Eur. J. Inorg.
Chem. 2008, 3921–3928.
(46) Wang, Y. Q.; Wang, Z. M.; Liao, C. S.; Yan, C. H. Acta
(12) Bricklebank, N.; Skabara, P. J.; Hibbs, D. E.; Hurstouse, M. B.;
Malik, K. M. A. J. Chem. Soc., Dalton Trans. 1999, 3007–3013.
(13) Aragoni, M. C.; Arca, M.; Devillanova, F. A.; Hursthouse, M. B.;
Huth, S. L.; Isaia, F.; Lippolis, V.; Mancini, A.; Ogilvie, H. Inorg. Chem.
Commun. 2005, 8, 79–82.
Crystallogr., Sect. C 1999, 55, 1503–1506.
(47) Shilov, G. V.; Kazheva, O. N.; Dyachenko, O. A.; Chernovyants,
M. S.; Simonyan, S. S.; Goleva, V. E. Russ. J. Phys. Chem. 2002,
76, 1436–1443.
(48) Ogawa, Y.; Takahashi, O.; Kikuchi, O. J. Mol. Struct. 1998,
(14) Aragoni, M. C.; Arca, M.; Devillanova, F. A.; Isaia, F.; Lippolis,
V.; Mancini, A.; Pala, L.; Slawin, A. M. Z.; Woollins, J. D. Chem. Commun.
2003, 2226–2227.
429, 187–196.
(49) Ghassemzadeh, M.; Dehnicke, K.; Goesmann, H.; Fenske, D. Z.
Naturforsch. 1994, 49b, 602–608.
(15) Taraba, J.; Zak, Z. Inorg. Chem. 2003, 42, 3591–3594.
(16) Riedel, S.; K€ochner, T.; Wang, X.; Andrews, L. Inorg. Chem.
2010, 49, 7156–7164.
(50) Parigoridi, I. E.; Corban, G. J.; Hadjikakou, S. K.; Hadjiliadis,
N.; Kourkoumelis, N.; Kostakis, G.; Psycharis, V.; Raptopoulouc, C. P.;
Kubicki, M. Dalton Trans. 2008, 5159–5165.
(17) Braida, B.; Hiberty, P. C. J. Am. Chem. Soc. 2004, 126,
14890–14898.
(51) Parlow, A.; Hartl, H. Acta Crystallogr., Sect. B 1979, 35, 1930–1933.
(52) Yagi, Y.; Popov, A. I. J. Inorg. Nucl. Chem. 1967, 29, 2223–2230.
(53) Bateman, R. J.; Bateman, L. R. J. Am. Chem. Soc. 1972,
94, 1130–1134.
(54) Asaji, T.; Eda, K.; Fujimori, H.; Adachi, T.; Shibusawa, T.;
Oguni, M. J. Mol. Struct. 2007, 826, 24–28.
(18) Christian, G. D. Analytical Chemistry; Wiley: New York, 2003.
(19) Gr€atzel, M. Prog. Photovolt. 2006, 14, 429–442.
(20) Fabjan, C.;Drobits, J. InHandbook of Battery Materials; Besenhard,
J. O., Ed.; Wiley-VCH: Weinheim, Germany, 1999; pp 177ꢀ194.
(21) Khan, M. D.; Majibur, R.; Gotoh, Y.; Morikawa, H.; Miura, M.
Textile Res. J. 2009, 79, 1305–1311.
(22) Mora, M.; Jimenez-Sanchidrian, C.; Ruiz, J. R. Appl. Organomet.
Chem. 2008, 22, 122–127.
(23) Ceccon, A.; Crociani, L.; Santi, S.; Venzo, A.; Biffis, A.; Boccaletti,
G. Tetrahedron Lett. 2002, 43, 8475–8478.
(24) Kautek, W.; Conradi, A.; Sahre, M.; Fabjan, Ch.; Drobits, J.;
Bauer, G.; Schuster, P. J. Electrochem. Soc. 1999, 146, 3211–3216.
(25) Getsis, A.; Mudring, A. V. Z. Anorg. Allg. Chem. 2009, 635,
2214–2221.
(26) MacFarlane, D. R.; Meakin, P.; Sun, J.; Amini, N.; Forsyth, N.
J. Phys. Chem. B 1999, 103, 4164–4170.
(27) Wolff, M.; Meyer, J.; Feldmann, C. Angew. Chem., Int. Ed. 2011,
50, 4970–4973.
(28) Wasserscheid, P., Welton, T., Eds. Ionic Liquids in Synthesis;
Wiley-VCH: Weinheim, Germany, 2008.
(29) Krossing, I.; Slattery, J. M.; Daguenet, C.; Dyson, P. J. J. Am.
Chem. Soc. 2006, 128, 13427–13434.
(30) Giernoth, R. Angew. Chem. 2010, 122, 2896–2300. Angew.
Chem., Int. Ed. 2010, 49, 2834–2839.
(31) Guloy, A. M.; Ramlau, R.; Tang, Z.; Schnelle, W.; Braitinger,
M.; Grin, Y. Nature 2006, 443, 320–323.
(32) Santiso-Quinones, G.; Br€uckner, R.; Knapp, C.; Dionne, I.;
Passmore, J.; Krossing, I. Angew. Chem. 2009, 121, 1153–1157. Angew.
Chem., Int. Ed. 2009, 48, 1133–1137.
(33) Biswas, K.; Zhang, Q.; Chung, I.; Song, J. H.; Androulakis, J.;
Freeman, A. J.; Kanatzidis, M. G. J. Am. Chem. Soc. 2010, 32, 14760–14762.
(34) Xie, T.; Brockner, W.; Gjikaj, M. Z. Anorg. Allg. Chem. 2010,
636, 2633–2640.
(35) Kanatani, A.; Matsumoto, K.; Hagiwara, R. Eur. J. Inorg. Chem.
2010, 1049–1055.
(36) Deetlefs, M.; Hussey, C. L.; Mohammed, T. J.; Seddon, K. R.;
van den Berg, J. A.; Zora, J. A. Dalton Trans. 2006, 2334–2341.
(37) Babai, A.; Mudring, A. V. Inorg. Chem. 2005, 44, 8168–8169.
(38) Poli, G.; Gordon, J. C.; Khanna, R. K.; Fanwick, P. S. Inorg.
Chem. 1992, 31, 3165–3167.
(39) Ozeryanskii, V. A.; Pozharskii, A. F.; Bienko, A. J.; Sawka-
Dobrowolska, W. L.; Sobczyk, L. J. Phys. Chem. 2005, 109, 1637–1642.
(40) Tol, A. D.; Natu, A. D.; Puranik, V. G. Mol. Cryst. Liq. Cryst.
2007, 469, 69–77.
(41) Terao, H.; Gesing, T. M.; Ishihara, H.; Furukawa, Y.; Gowda,
B. T. Acta Crystallogr., Sect. E 2008, 65, m323.
(42) Horn, C.;Scudder, M.;Dance, I. Cryst. Eng. Comm. 2000, 9, 53–66.
(43) Berkei, M.; Bickley, J. F.; Heaton, B. T.; Steiner, A. Chem.
Commun. 2002, 2180–2181.
(44) Chen, X.; Rickard, M. A.; Hull, J. W.; Zheng, C.; Leugers, A.;
Simoncic, P. Inorg. Chem. 2010, 49, 8684–8689.
11694
dx.doi.org/10.1021/ic201291k |Inorg. Chem. 2011, 50, 11683–11694