Solid-state synthesis of mixed trihalides via reversible absorption of dihalogens by non porous onium salts

Article abstract of DOI:10.1039/c1ce05050h

1,6-Bis(trimethylammonium)hexane bis(trihalides) and mixed bis(trihalides) have been synthesized by treating the corresponding dihydrated halides with molecular dihalogens under gas-solid and solution conditions. Despite the starting halides being non-porous, the trihalide syntheses occur homogeneously, in quantitative yields, and reversibly. In all cases halogen bond prevails over hydrogen bond, dihalogens substitute for the hydration water of starting halide anions and trihalides are formed. The stability of the obtained trihalides is mainly due to cooperative halogen bond and cation templation effect. Hexamethonium halides are proven effective solids for the clathration and storage of molecular dihalogens. While the starting salts are not isostructural, all the formed trihalides and mixed trihalides are isostructural.

Full text of DOI:10.1039/c1ce05050h

View Online / Journal Homepage / Table of Contents for this issue
This paper is published as part of a CrystEngComm themed issue entitled:
Dynamic behaviour and reactivity in crystalline
solids
Guest Editors: Tomislav Friščić and Graeme Day
University of Cambridge, UK
Published in issue 13, 2011 of CrystEngComm
Image reproduced with permission of K. Užarević
Read an editorial article by Sir John Meurig Thomas, University of Cambridge:
Crystal engineering: origins, early adventures and some current trends
John Meurig Thomas
CrystEngComm, 2011, DOI: 10.1039/C1CE90016A
Other articles published in this issue include:
Drug-drug co-crystals: Temperature-dependent proton mobility in the molecular
complex of isoniazid with 4-aminosalicylic acid
Pawel Grobelny, Arijit Mukherjee and Gautam R. Desiraju
CrystEngComm, 2011, DOI: 10.1039/C0CE00842G
Are glycine cyclic dimers stable in aqueous solution?
Said Hamad and C. Richard A. Catlow
CrystEngComm, 2011, DOI: 10.1039/C0CE00877J
Solid-state synthesis of mixed trihalides via reversible absorption of dihalogens by
non porous onium salts
L. Meazza, J. Martí-Rujas, G. Terraneo, C. Castiglioni, A. Milani, T. Pilati, Pierangelo
Metrangolo and Giuseppe Resnati
CrystEngComm, 2011, DOI: 10.1039/C1CE05050H
Visit the CrystEngComm website for more cutting-edge crystal engineering research
www.rsc.org/crystengcomm

Dynamic Article Links <
C
CrystEngComm
Cite this: CrystEngComm, 2011, 13, 4427
www.rsc.org/crystengcomm
PAPER
Solid-state synthesis of mixed trihalides via reversible absorption of
dihalogens by non porous onium salts†
a
a
ab
ac
c
Lorenzo Meazza, Javier Mart ꢀı -Rujas, Giancarlo Terraneo,* Chiara Castiglioni, Alberto Milani,
d
ab
abd
Tullio Pilati, Pierangelo Metrangolo* and Giuseppe Resnati*
Received 13th January 2011, Accepted 5th May 2011
DOI: 10.1039/c1ce05050h
1
,6-Bis(trimethylammonium)hexane bis(trihalides) and mixed bis(trihalides) have been synthesized by
treating the corresponding dihydrated halides with molecular dihalogens under gas–solid and solution
conditions. Despite the starting halides being non-porous, the trihalide syntheses occur
homogeneously, in quantitative yields, and reversibly. In all cases halogen bond prevails over hydrogen
bond, dihalogens substitute for the hydration water of starting halide anions and trihalides are formed.
The stability of the obtained trihalides is mainly due to cooperative halogen bond and cation
templation effect. Hexamethonium halides are proven effective solids for the clathration and storage of
molecular dihalogens. While the starting salts are not isostructural, all the formed trihalides and mixed
trihalides are isostructural.
ꢀ
Introduction
(triiodide) anion I
ꢀ
3
3$(I
2
)
2
, and (iii) the bis(pentaiodide) anion
I5 3$(I ) .
2 4
The solid-state chemistry of organic porous solids has shown an
1
increase in research activity in recent years. Organic porous
Many reports concerning the peculiar structural and electronic
8
properties of polyhalides are available in the literature. In
networks and cages have revealed potential applications ranging
2
from guest inclusion and gas absorption to molecular trans-
4
port. Most of the above-mentioned properties are studied under
9
n
ꢀ
general, polyhalide chains X2m+n (X ¼ F, Cl, Br or I) give rise
3
to a particularly wide variety of structures that are assembled
ꢀ
ꢀ
ꢀ
ꢀ
starting from X , X
2
, and X
3
building blocks, where X and X
as the
3
external stimuli by exposing the porous solids to guest molecules
may be considered as the electron-donor moieties and X
10
2
(
gaseous or liquid) at different temperatures and pressures.
electron-acceptor. The smallest polyhalide species is the triha-
ꢀ
The dynamic properties upon external stimuli in nonporous
lide anion X
3
. Depending on the identity of the halogen atoms
ꢀ
5
crystalline organic solids have received far less attention, so far.
In recent studies, we have highlighted the ability of the nonpo-
rous a,u-bis(trimethylammonium)alkane diiodides to encapsu-
that form the X
ꢀ
3
species, these can be homo-atomic trihalides
ꢀ
X3 [(X1/X2–X3) ; X1 ¼ X2 ¼ X3 ¼ F, Cl, Br, I) or heter-
ꢀ
ꢀ
oatomic mixed trihalides X Y [(Y1/X2–X3) ; Y1, X2 ¼ X3 ¼
2
6
7
2
late diiodoperfluoroalkanes and I
through a size matching
can be
11
Cl, Br, I). Mixed trihalides X Y are less common than triha-
ꢀ
2
dynamic response in gas-solid reactions. Interestingly, I
2
ꢀ
lides X3 , usually as a consequence of their instability in solution.
absorbed in the crystal lattice of 1,6-bis(trimethylammonium)
hexane diiodide (3, hexamethonium iodide) to afford three
different type of adducts depending on the amount of absorbed
In particular, several homo- or heteroatomic aggregates have
been reported starting from iodine and bromine while examples
involving chlorine, or even more fluorine, are much less common,
especially in the solid state.
2
ꢀ
I
2
: (i) the metastable polyiodide dianion I
4
2
3$I , (ii) the bis
Mixed trihalides have been studied from both the theoretical
12
point of view, to understand the nature of their chemical bonds,
a
CNST-IIT@POLIMI, Via Pascoli 70/3, 20133 Milan, Italy
b
and the experimental point of view, for instance for their use as
reagents in electrophilic addition reactions to form double and
NFMLab, D.C.M.I.C. ‘‘Giulio Natta’’, Politecnico di Milano, Via L.
Mancinelli 7, 20131 Milan, Italy. E-mail: pierangelo.metrangolo@polimi.
it; Fax: +39 02 2399 3180; Tel: +39 02 2399 3041; giuseppe.resnati@
polimi.it.; +39 02 2399 3032
13
triple bonds. The broad range of possible interesting applica-
tions of trihalides and their mixed analogues in different fields
span from renewable energies (e.g. as new redox couples in Dye
c
D.C.M.I.C. ‘‘Giulio Natta’’, Politecnico di Milano, Piazza L. da Vinci 32,
2
0133 Milan, Italy
1
4–16
d
Sensitized Solar Cells (DSSCs) with improved properties),
to
ISTM-CNR, via Golgi 19, 20133 Milan, Italy
17
† Electronic supplementary information (ESI) available: Experimental
protocols, spectroscopic characterization and crystallographic data
conducting or magnetic materials and reagents in synthetic
18
chemistry.
2 2 2 2
(both single crystal and PXRD) for structures 1$(H O) , 2$(H O) , 1$
The chemical bonding in the trihalides and mixed trihalides
(
Br , 2$(I , and 2$(Br . CCDC reference numbers 797580–797583
2
)
2
2
)
2
2 2
)
19
can be easily rationalized in terms of the halogen bond (XB),
and 801987. For ESI and crystallographic data in CIF or other
electronic format see DOI: 10.1039/c1ce05050h
i.e. any non-covalent interactions involving the positive region
This journal is ª The Royal Society of Chemistry 2011
CrystEngComm, 2011, 13, 4427–4435 | 4427

(s-hole) of the electrostatic potential surface of halogen atoms.
In fact, the formation of trihalides can be understood by
The trihalide 1$(Br ) and the mixed trihalides 2$(I ) and 2$
2 2 2 2
(Br ) are stable at room temperature (r.t.) and ambient pressure
2 2
ꢀ
considering that the halide subunit X is the XB-acceptor (Lewis
but the dihalogen molecules can be removed by heating under
reduced pressure. Water is taken up from the air after the
halogen removal and the starting dihydrated hexamethonium
2
base, electron-donor) and the molecular halogen subunit X is
20
the XB-donor (Lewis acid, electron-acceptor). In terms of
supramolecular synthesis, the trihalide is formed when the halide
anion attractively interacts with the s-hole of molecular halogens
2 2 2 2
salts 1$(H O) and 2$(H O) are isolated.
The complete reversibility of the clathration process via gas-
solid reactions is proven and the dihydrated forms of hexame-
thonium bromide and chloride are shown to function as dynamic
(
n / s* donation) and a XB is formed on the extension of the
covalent bond of molecular halogens.
In the present work, moving from the ability of hexametho-
7
nium iodide 3 to clathrate I , we decided to establish if other 1,6-
2 2
nonporous solids for the absorption and storage of I and Br .
2
bis(trimethylammonium)hexane dihalides have a similar ability
to clathrate iodine and other halogens. We pursued the forma-
tion of trihalide and mixed trihalide species as a consequence of
the relevance of these anions in materials science in general and
Results and discussion
The general protocol used in these studies consisted in exposing
finely ground powders of 1$(H O) and 2$(H O) to vapours of
2
2
2
2
15–17
in DSSCs in particular.
The stabilization of unusual mixed
2 2 2
Br and I . The absorption of X and the formation of the new
trihalide species was designed by applying the principles of
supramolecular chemistry and crystal engineering. Specifically,
crystalline trihalide entities were monitored and studied by X-ray
powder diffraction (XRPD) and Raman spectroscopy. The
formation of the desired trihalides was proven by comparing the
experimental XRPD patterns with the simulated patterns from
the diffraction data obtained on single crystals from synthesis in
solution. All the reactions showed nearly quantitative yields and,
more importantly, were reversible (Scheme 1).
herein it will be described how the dihydrated form 1$(H
2 2
O) of
1
,6-bis(trimethylammonium)hexane dibromide (1) clathrates Br
2
via a gas-solid reaction and forms the salt 1,6-bis(trimethy-
lammonium)hexane bis(tribromide) 1$(Br ) . Similarly, the
dihydrated form 2$(H O) of hexamethonium dichloride (2)
2
2
2
2
clathrates I and Br and forms the bis(diiodochloride) and bis
2
2
(
(
dibromochloride) salts 2$(I2)2 and 2$(Br ) , respectively
2 2
2 2 2 2
Crystal structures of hexamethonium halides 1$(H O) and 2$(H O)
Scheme 1). Clearly, dihalogens substitute for hydration water in
the coordination sphere of the starting halide anions and triha-
lides are formed.
A search in the Cambridge Structure Database (CSD-version
5.31) revealed that the structures of both 1$(H O) and 2$(H O)
2
2
2
2
All the obtained trihalide complexes are isostructural. The
same products described above are obtained from reactions in
solution, which have been pursued in order to obtain samples
suitable for single crystal X-ray studies. This technique allowed
for the details of the trihalides structures to be established and
the different interactions responsible for the overall crystal
packing to be identified. The characterization of the trihalides
has been completed by Raman spectrometry studies corrobo-
rated by band assignment validated by high level theoretical
calculations.
2 2
were already reported (CSD codes: 1$(H O) , HMENAM; 2$
21
(H O) , HMTMAC). Unfortunately, these old crystallographic
2
2
data did not include hydrogens and the R-factors were quite
high. We decided to re-determine the crystal structures of 1$
(H O) and 2$(H O) in order to have detailed information on
2
2
2
2
the non-covalent interactions taking place in the crystal lattices.
Single crystals of 1$(H O) and 2$(H O) suitable for X-ray
2
2
2
2
analysis were obtained upon slow evaporation at r.t. of their
methanol solutions (see ESI). Despite the two organic salts
differing only by the halide anions, the crystal packings are not
isostructural (Fig. 1). The position of the halide anions with
respect to the cations is different in the two crystals, as is the
2 2
clathrated water connectivity. The bromide anions in 1$(H O)
present six Br/H–C hydrogen bonds below the sum of van der
Waals radii (SvdW) with three different molecules (at x, 1/2 ꢀ y,
ꢁ
1/2 + z and 1 ꢀ x, 1/2 + y, 1/2 ꢀ z); four contacts involve the
methyl groups, one the methylene a to the N atom and another
the methylene b. In 2$(H O) the chloride anions form five Cl/
2
2
H–C hydrogen bonds with four different molecules,
at ꢀ1 + x, y, z, and 1 ꢀ x, 1 ꢀ y, 1 ꢀ z with a methyl group, ꢀ1 +
x, ꢀ1 + y, z, and 1 ꢀ x, 1 ꢀ y, 2 ꢀ z with both a methyl and an
a methylene group. The differences between hydrogen bonds
involving the clathrated water molecules are more dramatic. In
both salts there are two quite strong X/H–O hydrogen bonds
with two different water molecules, the mean X/H distance
being about 78% of the SvdW in both structures. However, the
topology of the interactions is very different. Infinite chains
wherein the bromide anions and water molecules alternate are
2 2 2 2
present in 1$(H O) , while in 2$(H O) rhombic motifs are
Scheme 1 Synthesis of the complexes 1$(Br
2
)
2
, 2$(I
2
)
2
, and 2$(Br
2
O) .
2
)
2
via
formed where the vertexes are alternatively occupied by chloride
ions and water molecules (Fig. 1).
gas-solid reactions and their reversion to 1$(H
O) and 2$(H
2 2
2
4
428 | CrystEngComm, 2011, 13, 4427–4435
This journal is ª The Royal Society of Chemistry 2011

Fig. 2 (a) Gas-solid synthesis of 1$(Br
vapours of Br . (b) Experimental XRPD patterns of recrystallized
hexamethonium bromide dihydrate 1$(H O) . (c) Experimental XRPD
of hexamethonium tribromide 1$(Br from gas-solid reaction.
2 2 2 2
) by exposing solid 1$(H O) to
2
2
2
2 2
)
Fig. 1 Ball and stick view along the a-axis of the crystal structures of
$(H O) (top) and 2$(H O) (bottom). Only X/H–O hydrogen bonds
are reported (black dotted lines). Hexamethonium hydrogen atoms are
omitted for clarity. Colour code: C, dark gray; N, blue, Br, brown; Cl,
green; O, red; water hydrogen, light gray.
smaller than iodine, the unit cell of 1$(Br
than that of 3$(I
organized into alternating cation and anion columns and the
2
)
2
is slightly smaller
1
2
2
2
2
ꢀ
2
)
2
. Hexamethonium tribromide 1$Br
3
is
ꢀ
structure of the linear and asymmetric Br3 ion closely resembles
ꢀ
that of the I3 species in hexamethonium triiodide 3$(I2)2
ꢀ
2
3
(
Table 1). The Br3 anions show an offset stack along the c-axis
Synthesis and characterization of hexamethonium tribromide 1$(Br2)2
and form loosely interacting dimers thanks to type-I Br/Br
ꢀ
24
interactions (Fig. 3c,d). The interaction pattern of the Br
ꢀ
3
First we attempted the synthesis of the well-known Br
3
anion in
is
order to set up and to optimize the protocol for the successive
synthesis and Raman studies of mixed trihalides.
completed by five Br/H–C hydrogen bonds (below SvdW) that
ꢀ
the external bromine of Br
3
, not involved in type-I Br/Br
The compound 1$(H
talline powder was placed in a sealed jar and exposed to excess
2
O)
2
in the form of a white microcrys-
interactions, forms with the methyl hydrogens of three different
hexamethonium molecules.
22
Br
2
vapours at room temperature during 15 h. The white
The good agreement between the simulated XRPD pattern for
1$(Br ) obtained from solution crystallization and the experi-
mental XRPD pattern for the bulk polycrystalline sample from
powder gradually turned orange (Fig. 2a). The XRPD analysis of
the orange solid powder showed that the new obtained phase was
highly crystalline with no detectable peaks characteristic of
2
2
gas-solid reaction, proved that the structure of 1$(Br ) obtained
2
2
1
$(H O) being left (Fig. 2b,c).
2
from single-crystal X-ray diffraction data is fully representative
2
The same material was formed (XRPD and DSC analyses) on
crystallization from solution when excess Br dissolved in
methanol was slowly diffused into a methanol solution of
$(H O) . Precipitation produced orange single crystals suitable
of the bulk polycrystalline sample from gas-solid reaction. This
ꢂ
2
was also confirmed by the same melting point (ca. 125 C)
measure for samples from both solution and gas-solid reactions.
An important feature of solid a,u-bis(trimethylammonium)
alkanes dihalides is their ability to function as dynamically
porous materials showing a selective and reversible capture from
1
2
2
for single crystal X-ray diffraction. Single crystal X-ray analysis
revealed the formation of 1,6-bis(trimethylammonium)hexane
6
7
bis(tribromide) 1$(Br
hydration water of the starting hexamethonium bromide
2
)
2
. Bromine has substituted for the
2
the gas phase of diiodoperfluoroalkanes and I . Based on these
results, we decided to study the reversibility of the uptake of Br2
ꢀ
ꢀ
1
$(H O) and Br anions have been formed. Clearly, Br ions
2
by the solid hexamethonium bromide 1$(H O) .
2
3
2
2
ꢂ
The tribromide 1$(Br ) was heated up to ca. 80 C under
2 2
prefer to be halogen-bonded to molecular bromine than to be
hydrogen-bonded to water.
reduced pressure. Macroscopically, an evident colour change
from orange to white was immediately observed, suggesting the
The tribromide 1$(Br
methonium triiodide 3$(I
2
)
2
was nicely isostructural with hexa-
(Fig. 3a,b) and the unit cell dimen-
2
)
2
2 2 2
loss of Br and the reversion to 1$(H O) , water being uptaken
sions and angles are quite similar (Table 1). Bromine being
from the air during the sample handling after bromine removal.
This journal is ª The Royal Society of Chemistry 2011
CrystEngComm, 2011, 13, 4427–4435 | 4429

ꢀ
availability of stable complexes involving the I Cl ion might
2
open new perspectives in solid state halogenation reactions.
For this reason, we pursued the synthesis of a solid salt con-
ꢀ
taining the I
$(Br
The white powder of hexamethonium chloride 2$(H
exposed to excess I vapours in a sealed glass jar at r.t. for 48 h. A
dark red powder was produced; XRPD analysis showed that no
starting chloride 2$(H O) was present and a new crystalline
2
Cl anion by using the protocol optimized with
1
2 2
) .
2
2
O) was
2
2
2
phase was formed (Fig. 5a). The same material (as shown by m.p.
and XRPD) was obtained by crystallizing a 2 : 1 methanol
solution of I2 and 2$(H O) . Single crystal X-ray diffraction
2
2
analysis of a good quality light brown crystal from solution
showed that 1,6-bis(trimethylammonium)hexane bis(diiodo-
chloride) 2$(I
of the complex 2$(I )
2
)
2
had been formed. The simulated XRPD pattern
shows a good match between the experi-
2 2
mental diffraction pattern of the product obtained after the gas-
solid reaction and the simulated XRPD from single crystal data
(
Fig. 5b,c) and gas-solid and solution reactions show the same
ꢂ
melting point (ca. 155 C). Similar to 1$(Br
2
)
2
, the gas-solid
synthesis of 2$(I
Importantly, hexamethonium diiodochloride 2$(I ) is iso-
2 2
) occurs in quantitative yields.
2
2
7
structural with the tribromide and triiodide analogues 1$(Br )
2
2
and 3$(I ) , respectively. In fact, the crystal packing of 2$(I ) is
2
2
2 2
characterized by an alternation of cation and anion columns
Fig. 5d) and by the presence of couples of linear mixed trihalides
(
paired through type-I iodine/iodine interactions (Table 1)
between diiodochloride anions offset stacked along the c-axis.
A detailed analysis of the electron density in the single crystal
Fig. 3 Crystal structures of (a) hexamethonium triiodide 3$(I
2 2
) , and (b)
of the halogen bonded diiodochloride anion in 2$(I
2
)
2
, revealed
ꢀ
tribromide 1$(Br viewed along the c-axis. The alternation of cationic
2
2
)
2
that, as usual in mixed trihalides, the central position in the X Y
and anionic columns is shown. Hydrogen atoms are omitted for clarity.
Colour code: C, gray; N, blue; I, violet; Br, brown. (c) Detailed view of
anion and cation columns; the type-I Br/Br contacts connecting two
tribromide anions are dotted lines. (d) A couple of tribromide anions
connected by Br/Br contacts are shown along the c-axis.
ꢀ
anion, namely X2 in Y1/X2–X3 , is exclusively occupied by the
heavier halogen, i.e. iodine. The two external positions, namely
Y1 and X3, are occupied by chlorine and iodine atoms with
different probabilities. In particular, Y1, the external halogen
atom located closer to the trimethylammonium residue, is
populated for 95% by the chlorine atom and for 5% by the iodine
atom, while X3 has complementary occupancies. Chlorine is
smaller and more electronegative than iodine and thus tends to
form stronger hydrogen bonds and electrostatic interactions. The
populations described above optimize these intermolecular
interactions and chlorine atoms form seven hydrogen bonds with
XRPD analysis clearly showed that the reversion of 1$(Br
$(H O) was complete in 2 d (Fig. 4). Re-exposure of thus
formed 1$(H O) to Br vapours yielded quantitatively the tri-
bromide 1$(Br again. The reversibility of the Br uptake/
2 2
) into
1
2
2
2
2
2
2
)
2
2
release process was further confirmed by performing several
cycles, after which the hydrated salt 1$(H O) and the tribromide
2
2
the methyl groups of three hexamethonium cation molecules
ꢁ
1$(Br ) were recovered alternatively as proven by XRPD and
2 2
(
average distance dCl1/H4/H5 ¼ 3.010 A).
thermal analyses.
Similar to 1(Br ) and 3$(I ) , 2$(I ) is stable for months at r.t.
2
2
2 2
2 2
Finally, we monitored the stability of the obtained complex
and ambient pressure. Similarities among the three trihalides
2 2 2 2
1$(Br ) . A finely ground sample of the tribromide 1$(Br ) left in
extend also to the reversible nature of the dihalogen absorption.
ꢂ
air at r.t. for one month showed the same physical properties,
i.e. m.p. and XRPD pattern, of freshly prepared samples, thus
confirming that hexamethonium bromide functions as a perfect
molecular-storage system for bromine at room temperature and
ambient pressure.
In fact, when 2$(I
2
)
2
is heated up to 120 C under reduced
pressure for 2 d, 2$(H
demonstrated through XRPD. Here too, water is uptaken from
the air during the handling after the iodine removal.
2
2
O) is produced quantitatively, as
Synthesis and characterization of the hexamethonium
Synthesis and characterization of the mixed hexamethonium
dibromochloride 2$(Br2)2
diiodochloride 2$(I2)2
ꢀ
27
The Br Cl anion has been extensively studied in silico and in
2
ꢀ
The I
2
Cl anion is a mixed trihalide used, as its imidazolium salt,
solution, as its tetra n-butyl ammonium salt, for the electrophilic
28
25
in double and triple bonds halogenation. Reports about the
addition reaction to alkenes. However, there are (to now) no
ꢀ
ꢀ
26
Cl ion are pretty rare, whereas the
solid state structure of the I
2
crystal structures containing discrete Br
2
Cl anions in the CSD.
4
430 | CrystEngComm, 2011, 13, 4427–4435
This journal is ª The Royal Society of Chemistry 2011

Table 1 Structural data of trihalides and mixed trihalides in 1$(Br
2
)
2
, 2$(I
2
)
2
, 2$(Br
2
)
2
, and 3$(I
2
)
2
1$(Br
2
)
2
2$(I
2
)
2
2$(Br
2
)
2
2 2
3$(I )
X1/X2–X3
trihalide structure
a
a
a
a
Br1/Br2–Br3
Cl1/I2–I3
Cl1/Br2–Br3
I1/I2–I3
Space group
ꢁ
C 2/m
C 2/m
C 2/m
C 2/m
21.891(4)
7.6086(14)
7.8147(15)
90.00
a/A
ꢁ
21.450(4)
7.3232(10)
7.1474(12)
90.00
92.31(2)
90.00
1121.8(3)
2.647 (0.70)
2.463 (1.078)
179.60
3.581
152.72
22.3618(3)
7.5412(12)
7.1251(10)
90.00
91.388(12)
90.00
1201.1(3)
21.800(3)
7.2927(10)
6.8951(10)
90.00
91.743(12)
90.00
1095.7(3)
2.567 (0.69)
2.436 (1.067)
179.32
3.517
154.33
b/A
ꢁ
c/A
ꢂ
a/
ꢂ
b/
92.79(2)
90.00
ꢂ
g/
Unit cell/A
ꢁ
3
1300.1(4)
2.983 (0.72)
c
2.859 (1.072)
179.48
3.760
ꢁ
b
b
b
b
X1/X2 bond/A
ꢁ
X2–X3 bond/A
X1/X2–X3 angle/
X3/X3 bond/A
2.725 (0.72)
c
c
c
2.798 (1.049)
178.92
3.591
ꢂ
ꢁ
d
ꢂ
X2–X3/X3 angle /
153.72
151.14
a
b
Halogen atoms numbering corresponds to the numbering in CIF files available in Cambridge Structure Database. The number in brackets is
‘
normalized contact’, the ratio Nc ¼ Dij/(rvdWi + rPj), where Dij is the experimental distance between the atoms i and j, rvdWi is the van der
ꢁ
Waals radius of i (XB donor atom, 1.85, 1.98 A for Br and I, respectively), and rPj is the Pauling ionic radius for j (XB acceptor atom, 1.81, 1.95,
ꢁ
c
and 2.16 A for chloride, bromide, and iodide anions, respectively). The numbers in brackets indicate the percentage lengthening for the covalent
ꢀ
ꢁ
3 2 2 2
bond in X /X Y anion with respect to the covalent bond length in corresponding dihalogen molecule: 2.284 and 2.666 A for Br and I ,
d
ꢀ
respectively (value from www.webelements.com). One angle is given as C2v symmetry correlates two trihalides connected through the type-I
halogen/halogen interaction.
considerably lower affinity than iodide toward dihalogen
molecules.
In order to obtain a pure sample of the mixed trihalide
2$(Br
2
)
2
, finely ground hexamethonium chloride 2$(H
2 2
O) was
exposed to a stoichiometric amount of Br
2
vapours after the
29
standard gas-solid protocol. The XRPD spectrum of the
resulting crystalline material (Fig. 6a) revealed the quantitative
transformation of the starting compound 2$(H O) into a crys-
2
2
ꢀ
talline product free from Br3 anions, as indicated by the absence
ꢂ
of a peak around ca. 12.4 2q/ , and showing a very characteristic
ꢂ
peak, around ca. 8 2q/ , similar to a peak observed in the pattern
2 2
of 2$(I ) . Raman spectrometry studies, supported by band
assignment validated by high level theoretical calculations (see
ꢀ
onward), suggested the exclusive presence of the Br
the sample.
2
Cl anion in
To confirm this assignment through single crystal X-ray
analysis, we tried a solution synthesis of 2$(Br ) starting from
2
2
2$(H O) and stoichiometric Br2 and by using conditions
2 2
favouring a rapid crystal growth (to minimize equilibrium/
2 2
Fig. 4 (top) XRPD pattern of 1$(H O) simulated from single crystal X-
disproportionation reactions). Single crystal X-ray analysis of an
ꢀ
ray analyses. (bottom) Experimental XRPD pattern of the powder
ꢂ
2
orange plate unequivocally showed the presence of the Br Cl
obtained after heating 1$(Br
2
)
2
up to 80 C under reduced pressure for 2 d.
species wherein the central position in the triatomic anion is
occupied by bromine, the heavier halogen. Analysis of the elec-
ꢀ
ꢀ
The experimental free energy change for the equilibrium Br
ꢀ
2
Cl
ꢀ
tron density revealed that Br
ꢀ
3
ions were also present in the used
ꢀ
ꢀ
ꢀ1
ꢀ
+
Br Br
3
+ Cl in water is ꢀ1.8 kcal mol , namely Br
2
Cl is
crystal (Br
2
Cl /Br
3
ratio 7 : 1). The good agreement between the
and the experimental
thermodynamically unstable (as is the case for other mixed tri-
ꢀ
2 2
simulated XRPD pattern of 2$(Br )
XRPD spectrum obtained on the bulk polycrystalline sample
halides X
2
Y wherein Y is lighter than X) and tends to transform
ꢀ
into the Br3 anion through a series of equilibrium and dispro-
portionation reactions. The hexamethonium iodide templating
from the stoichiometric gas-solid reaction, confirmed the quan-
ꢀ
titative formation of the Br Cl ion under these conditions.
2
2
ꢀ
ꢀ
ability allowed the thermodynamic instability of I4 to be over-
7
come and we decided to test if the templating ability of the
Br Cl is isostructural with the other hexamethonium triha-
2
7
lides and mixed trihalides described or discussed in this paper
hexamethonium cation was strong enough to allow hexametho-
(Table 1) and similarities with 1$(Br
a molecular and supramolecular level. Chlorine is smaller than
bromine and the unit cell of 2$(Br is slightly smaller than that
2 2 2 2 2 2
) , 2$(I ) , 3$(I ) are both at
nium chloride 2$(H
ꢀ
2
O)
2
to overcome the thermodynamic insta-
bility of Br
2
Cl despite the chloride anion having
a
2 2
)
This journal is ª The Royal Society of Chemistry 2011
CrystEngComm, 2011, 13, 4427–4435 | 4431

Fig. 6 (a) The experimental XRPD pattern of 2$(Br
2
)
2
obtained from
the gas-solid reaction when stoichiometric Br
ꢀ
2
was used. (b) View of the
packing of the cations and Br
2
Cl anions approximately along the c-axis.
Colour code: C, gray; N, blue; Br, Brown; Cl, green.
After one month at room temperature and ambient pressure,
the XRPD pattern of 2$(Br ) remained unchanged proving that
2
2
ꢀ
the Br Cl stability is comparable in this crystal to the other
2
trihalides described in this paper.
ꢀ
ꢀ
ꢀ
Raman analysis of the Br3 , I Cl , and Br Cl ions
2
2
Raman spectroscopy is a powerful technique for the identifica-
30
tion of different polyhalide species. FT-Raman spectroscopy
measurements were carried out in order to corroborate the X-ray
diffraction (XRD) data from single crystals and powders on the
31
presence and arrangement of trihalides anions in the crystals.
Despite the apparent simplicity related to the description of
the vibrational properties of a triatomic molecule, the interpre-
tation of the Raman spectra of trihalides anions is not trivial in
this case, because of the effects of relevant intra- and intermo-
lecular interactions due to the clathration within the organic
crystal.
Fig. 5 Experimental XRPD pattern of (a) hexamethonium chloride
dihydrate 2$(H O) and (b) diiodochloride 2$(I obtained from the gas-
solid reaction. (c) XRPD pattern of diiodochloride 2$(I simulated from
single crystal X-ray analyses of a sample from solution synthesis. (d) View
On the basis of simple symmetry arguments, only one Raman
31
2
2
2
)
2
transition is observed in the two borderline cases of a perfectly
ꢀ
2
)
2
symmetric superanion (X
3
ꢀ
) and of a diatomic X
2
molecule
slightly perturbed by the X anions. In the first case the Raman
2 2
of the packing of 2$(I ) along the c-axis. Colour code: C, gray; N, blue; I,
+
line is assigned to the symmetric (S
bonds and in the latter one to the X–X stretching of the neutral
perturbed X diatomic molecule. On the other hand, the
g
) stretching of the two X–X
violet; Cl, green.
2
of 1$(Br ) . The stacking of the linear trihalide anions along the
2
appearance of two bands in the Raman spectrum can be
considered as an evidence of the existence of an asymmetric
2
c-axis is controlled by short type-1 bromine/bromine contacts.
As in 2$(I ) , this pairing of trihalide units locates chlorine atoms
ꢀ
supermolecule (i.e. the X3 asymmetric anion), which can be
2
2
at the two ends of the hexaatomic systems, namely conveniently
positioned to form three hydrogen bonds with the nearby and
positive methyl groups of three different hexamethonium
cations.
thought of as an intermediate case between the two structures
described above. Accordingly, any intermediate case is expected
to imply a relevant modulation of the (two) Raman active
transitions, both in frequencies and intensities. In addition, the
4
432 | CrystEngComm, 2011, 13, 4427–4435
This journal is ª The Royal Society of Chemistry 2011

peculiar molecular packing shown by the single crystal structure
studies implies strong intermolecular interactions between pairs
equilibrium molecular structure, which might strongly deviate
from that observed in the crystal. Since the molecular geometry is
the primary factor affecting the vibrational features, an unreal-
istic structure can prevent a correct simulation and interpretation
of the spectrum. Therefore, in cases (I) and (II), the calculations
of the Raman spectra have been carried out based on the atomic
positions obtained from X-ray analysis without any further
ꢀ
of trihalide anions (forming X3 dimers) and hydrogen bonds
between the anions and the organic moieties. Both these inter-
actions could affect the vibrational properties of the trihalides
anions.
In Fig. 7 the FT-Raman spectra of the three crystals 1$(Br
ꢀ
2
)
2
,
2$(I
2
)
2
, and 2$(Br
2
)
2
obtained via the gas-solid protocol are
optimization. In case (III) both the intramolecular and inter-
ꢀ
ꢀ1
reported in the frequency range 250–100 cm , where Raman
transitions associated to trihalides stretching vibrations are
observed. It is clear that the three spectra show different patterns
in frequencies and intensities, thus supporting the occurrence of
molecular geometries of Br
3
as well as the intermolecular
and the N atoms of the cations
ꢀ
distances and angles among Br
3
have been fixed to the values obtained by single crystal X-ray
diffraction, while the intramolecular parameters of the organic
cations have been re-optimized. The best description of the whole
pattern both in frequencies and relative Raman intensities can be
obtained with model (II), while the case of the isolated anion
[model (I)] is worst, as it gives relative intensities showing the
opposite trend with respect to the experimental ones. Interest-
ingly, the explicit introduction of hydrogen bonds [model (III)]
modifies the intensity ratio in the correct way, but largely over-
estimates the Raman activity of the higher frequency band.
Furthermore, it should be noted that the computed Raman
intensities of model (II) are dramatically enhanced with respect
to the model (I), showing an increase of the total Raman intensity
of about one order of magnitude (see Table SI-2 in ESI†). This
effect is stronger for the higher frequency band corresponding to
a normal mode mainly localized on the two bonds (of the two
anions) in closer contact, thus demonstrating and confirming
that the type-I halogen/halogen interaction is responsible for
a large polarization fluctuation of the whole dimer.
different molecular species in the three crystals. In particular, the
ꢀ
experimental spectrum of Br Cl proves that only traces
2
ꢀ
amounts of the Br
3
anion, if any, are present in 2$(Br
prepared under gas-solid conditions when stoichiometric
amounts of Br are used. In order to confirm the assignment of
2
)
2
batches
2
the observed bands as markers of the different trihalides identi-
fied by XRD analysis and to give a chemical/physical interpre-
tation of the Raman features, Density Functional Theory (DFT)
32
calculations have been carried out. A realistic simulation of
vibrational spectra requires that all the relevant intra- and
intermolecular interactions are taken into account properly. In
order to determine the relative importance of type-I halogen/
halogen interactions (forming hexaatomic adducts) relative to
the hydrogen bonds between the anions and the organic cation,
three different models have been chosen and carefully analyzed
ꢀ
ꢀ
in the case of Br
3
(Fig. 8): (I) an isolated Br
ꢀ
3
anion (Fig. 8, green
ꢀ
line); (II) a dimer of interacting Br
3
anions (namely a (Br
ꢀ
3
)
2
species; Fig. 8, red line); (III) a Br
3
anion interacting via
It is worth noting that the use of model (III) determines
a global intensification of the Raman activity with respect to
model (I), but the effect is relatively weak with respect to the
model (II). This result is particularly meaningful since it shows
that the vibrational features of these systems are mainly affected
by interactions between trihalides pairs, namely type-I halogen/
halogen contacts, while the hydrogen bond plays only a minor
role. For this reason, we can state that for a realistic prediction of
hydrogen bond with small models of the organic cations (Fig. 8,
blue line) In this case, a smaller basis set (6-311++G**) has been
33
used for the cations only.
It should be mentioned that, since in these models the effect of
the surrounding crystal field is not fully taken into account,
a geometry optimization of our model systems would come to an
ꢀ
ꢀ
ꢀ
Fig. 7 Black line: Experimental FT-Raman spectra of Br
ꢀ
3
, Br
2
Cl , and
Fig. 8 Comparison of the experimental FT-Raman spectrum of Br
anion (black) in 1$(Br
3
I
2
Cl in crystals 1$(Br
2
)
2
, 2$(Br
2
)
2
, and 2$(I
2
)
2
obtained via the gas-solid
2
)
2
crystals and DFT calculations on the three
protocol with excess Br
2
, stoichiometric Br
2
, and excess I , respectively.
2
different molecular models. In the case of model (II) (red) the Raman
intensity (normalized on a single trihalide molecule, see Table SI-2 of the
ESI†) has been further divided by a factor of 2.
Red lines: DFT computed Raman spectra of the same systems for dimers
of interacting anions (model (II), see text for details).
This journal is ª The Royal Society of Chemistry 2011
CrystEngComm, 2011, 13, 4427–4435 | 4433

the Raman pattern of the crystals under discussion, it is
mandatory to explicitly consider the anions dimers, while the
hydrogen bond can be neglected at least in a first approximation.
In Fig. 7 the DFT computed Raman spectra for dimers of
nature of the XB allows the process to be reverted at high
temperature and low pressure. The XB effectiveness in
substituting for hydrogen bonds in the coordination sphere of
halides is here directed by the templating ability of the hexame-
thonium cation. Considering the energetics of the process, the
halogen for water substitution in the coordination sphere of
halides may be quite general and its relevance extends to all cases
where water removal from the halides coordination sphere is
pursued.
ꢀ
ꢀ
ꢀ
3 2 2
Br , Br Cl , and I Cl anions, model (II), are compared with
the experimental FT-Raman spectra of the corresponding crys-
talline bulk samples. The agreement between experiments and
computations is very good and confirms that the observed
Raman bands can be taken as markers of the different trihalides
species occurring in the different crystals. Vibrational eigenvec-
tors analysis allowed us to state that the two Raman bands
Interestingly, the three crystalline complexes are isostructural
and this is even more remarkable if we consider how important
the nature of the halogen is in relation to the crystal packing, as
indicated by the non-isostructural packing of the starting
compounds 1$(H O) and 2$(H O) . The three trihalide deriva-
ꢀ
ꢀ
predicted for (Br3 ) and (Br Cl ) are assigned, respectively, to
2
2
2
the symmetric combination of stretching on symmetry-related
ꢀ
bonds of the two different anions. In particular, in Br Cl the
2
2
2
2
2
lower frequency band is associated to the symmetric Cl/Br
stretching and the higher frequency one to symmetric Br–Br
stretching. Also in this case, the vibration associated to the atoms
in closer contact (Br–Br bonds) showed the largest Raman
tives are isostructural also to the already described hexametho-
7
2 2
nium triiodide 3$(I ) .
These results confirm that the bis(trimethylammonium)alkane
dihalide salts, despite being nonporous in nature, function as
very effective dynamically porous materials trapping dihalogen
molecules from the gas phase and stabilizing trihalides and mixed
intensity (see details in ESI†).
ꢀ
The I
2
Cl case is peculiar since only one band is observed,
associated to a vibrational mode mainly due to the I–I stretching
of the dimer. Notice that this behaviour is rather surprising
because XRD data indicate the formation of a tightly bound
trihalides. Indeed, hexamethonium chloride 2$(H
ꢀ
2
O)
2
afforded
ꢀ
crystals containing the I Cl and Br Cl anion and their quality
2
2
and purity was good enough to allow for structural and Raman
characterization. The stabilization of these trihalides is due to
a synergistic combination of XB, hydrogen bond, and type-I
halogen/halogen contacts, as confirmed by X-Ray diffraction
and Raman analyses.
ꢀ
ꢀ
I2Cl superanion rather than a weakly interacting I /Cl
2
complex, for which one only Raman feature is indeed expected.
On the other hand, looking to the calculations (carried out at the
experimentally determined molecular geometry), we can find
another Raman active frequency (characterized by an almost
zero Raman activity) associated to the I–Cl stretching
The method we applied is general and can be extended to the
predictable preparation of other uncommon trihalides species.
This has the potential for enhancing significantly the structural
knowledge of trihalide species in the solid-state, as well as
providing a novel approach in the synthesis of new functional
organic solids.
(
symmetric combination of the two stretchings belonging to the
two anions). The fact that only one band occurs, is then
explained due to the intrinsically weak Raman activity of the I–
ꢀ
Cl bond in the I Cl superanion. This fact can be ascribed to the
2
quite polar nature of the I–Cl bonds: it is indeed generally
recognized that polar bonds often show stretching modes which
are practically silent in the Raman spectrum.
Acknowledgements
P.M., G.R., and G.T. thank Fondazione Cariplo (‘‘New-
Generation Fluorinated Materials as Smart Reporter Agents in
Conclusions
19
F MRI’’) and MIUR (‘‘Engineering of the Self-assembly of
In summary, we have applied the principles of supramolecular
chemistry and crystal engineering to control the gas-solid
synthesis of common trihalide and uncommon mixed trihalide
derivatives. Hexamethonium tribromide, diiodochloride, and
dibromochloride were synthesized by exposing crystals of dihy-
drated hexamethonium bromide and chloride to vapours of Br2
Molecular Functional Materials via Fluorous Interactions’’) for
financial support.
Notes and references
1 (a) G. Couderc and J. Hulliger, Chem. Soc. Rev., 2010, 39, 1545; (b)
S. Dalgarno, P. K. Thallapally, L. J. Barbour and J. L. Atwood,
Chem. Soc. Rev., 2007, 36, 236; (c) N. B. McKeown, J. Mater.
Chem., 2010, 20, 10588; (d) K. D. M. Harris, Supramol. Chem.,
2007, 19, 47; (e) J. A. Swift, A. M. Pivovar, A. M. Reynolds and
M. D. Ward, J. Am. Chem. Soc., 1998, 120, 5887.
2
and I . The trihalides were isolated in pure form if the stoichi-
ometry in the gas-solid reactions was conveniently chosen. The
gas-solid reactions occur homogeneously and are reversible if
suitable temperature and pressure are used. The mixed trihalides
ꢀ
ꢀ
2 (a) J. Rebek, Chem. Soc. Rev., 1996, 25, 255; (b) L. R. MacGillivray,
H. A. Spinney, J. L. Reid and J. A. Ripmeester, Chem. Commun.,
2 2
I Cl and Br Cl are hard to obtain as crystalline solids. The
obtained trihalides are perfectly stable at r.t. and ambient pres-
sure showing no detectable changes in the composition after one
month.
2
000, 517.
3 (a) A. P. C o^ te, A. I. Benin, N. W. Ockwig, M. O’Keeffe, A. J. Matzger
and O. M. Yaghi, Science, 2005, 310, 1166; (b) P. Sozzani, S. Bracco,
A. Comotti and R. Simonutti, Angew. Chem., Int. Ed., 2005, 44, 1816;
ꢀ
ꢀ
ꢀ
The experimental 0 K XB bond strength in Cl3 , Br3 , and I3
(
c) B. S. Ghanem, K. J. Msayib, N. B. McKeown, K. D. M. Harris,
ꢀ
134
are 99, 127, and 126 kJ mol
and it can thus be safely surmised
Z. Pan, P. M. Budd, A. Butler, J. Selbie, D. Book and A. Walton,
Chem. Commun., 2007, 67; (d) T. Tanaka, T. Tasaki and
Y. Aoyama, J. Am. Chem. Soc., 2002, 124, 12453; (e) J. L. Atwood,
L. J. Barbour and A. Jerga, Science, 2002, 296, 2367; (f) H. Kim,
Y. Kim, M. Yoon, S. Lim, S. M. Park, G. Seo and K. Kim, J. Am.
Chem. Soc., 2010, 132, 12200.
that in our systems the XB that bromide and chloride anions
form with I and Br are stronger than the hydrogen bonds with
2
2
water. At r.t. dihalogen molecules easily substitute for water in
the coordination sphere of the halide anions. The non-covalent
4
434 | CrystEngComm, 2011, 13, 4427–4435
This journal is ª The Royal Society of Chemistry 2011

4
(a) T. Meersmann, J. W. Logan, R. Simonutti, S. Caldarelli,
A. Comotti, P. Sozzani, L. G. Kaiser and A. Pines, J. Phys. Chem.
A, 2000, 104, 11665; (b) J. Mart ꢀı -Rujas, A. Desmedt,
K. D. M. Harris and F. Guillaume, J. Am. Chem. Soc., 2004, 126,
web/ins/2009-032-1-100
definition given here should be taken as provisional.
and
www.halogenbonding.eu).
The
20 (a) G. Cavallo, P. Metrangolo, T. Pilati, G. Resnati, M. Sansotera
and G. Terraneo, Chem. Soc. Rev., 2010, 39, 3772; (b)
P. Metrangolo and G. Resnati, Science, 2008, 321, 918; (c)
P. Metrangolo, T. Pilati, G. Terraneo, S. Biella and G. Resnati,
CrystEngComm, 2009, 11, 1187; (d) P. Metrangolo, F. Meyer,
T. Pilati, G. Resnati and G. Terraneo, Angew. Chem., Int. Ed.,
2008, 47, 6114; (e) K. Rissanen, CrystEngComm, 2008, 10, 1107; (f)
G. Cavallo, S. Biella, J. L u€ , P. Metrangolo, T. Pilati, G. Resnati
and G. Terraneo, J. Fluorine Chem., 2010, 131, 1165; (g) A. Abate,
S. Biella, G. Cavallo, F. Meyer, H. Neukirch, P. Metrangolo,
T. Pilati, G. Resnati and G. Terraneo, J. Fluorine Chem., 2009, 130,
1171; (h) A. Mele, P. Metrangolo, H. Neukirch, T. Pilati and
G. Resnati, J. Am. Chem. Soc., 2005, 127, 14972; (i) R. Liantonio,
P. Metrangolo, F. Meyer, T. Pilati, W. Navarrini and G. Resnati,
Chem. Commun., 2006, 1819.
1
1124; (c) J. Mart ꢀı -Rujas, K. D. M. Harris, A. Desmedt and
F. Guillaume, Mol. Cryst. Liq. Cryst., 2006, 456, 139–147; (d)
J. Mart ꢀı -Rujas, A. Desmedt, K. D. M. Harris and F. Guillaume,
J. Phys. Chem. B, 2006, 110, 10708; (e) J. Mart ꢀı -Rujas, A. Desmedt,
K. D. M. Harris and F. Guillaume, J. Phys. Chem. B, 2007, 111,
1
2339; (f) J. Mart ꢀı -Rujas, A. Desmedt, K. D. M. Harris and
F. Guillaume, J. Phys. Chem. C, 2009, 113, 736.
(a) J. L. Atwood, L. J. Barbour, A. Jerga and B. L. Schottel, Science,
2002, 298, 1000; (b) P. K. Thallapally, G. O. Lloyd, J. L. Atwood and
L. J. Barbour, Angew. Chem., Int. Ed., 2005, 44, 3848; (c) J. Tian,
P. K. Thallapally, S. J. Dalgarno and J. L. Atwood, J. Am. Chem.
Soc., 2009, 131, 13216.
P. Metrangolo, Y. Carcenac, M. Lahtinen, T. Pilati, K. Rissanen,
A. Vij and G. Resnati, Science, 2009, 323, 1461.
A. Abate, M. Brischetto, G. Cavallo, M. Lahtinen, P. Metrangolo,
T. Pilati, S. Radice, G. Resnati, K. Rissanen and G. Terraneo,
Chem. Commun., 2010, 46, 2724.
(a) E. H. Wiebenga, E. E. Havinga and K. H. Boswijk in Advances in
Inorganic Chemistry and Radiochemistry Academic, New York, 1961,
Vol. 3, p. 133; (b) I. V. Burykin, M. S. Chernov’yants and
N. V. Aleshina, Russ. Chem. Bull., 2007, 56, 1390; (c) A. J. Blake,
R. O. Gould, W.-S. Li, V. Lippolis, S. Parsons, C. Radek and
M. Schr o€ der, Angew. Chem., Int. Ed., 1998, 37, 293; (d) A. J. Blake,
F. Cristiani, F. A. De Villanova, A. Garau, L. M. Gilby,
R. O. Gould, F. Isaia, V. Lippolis, S. Parsons, C. Radek and
M. Schr o€ der, J. Chem. Soc., Dalton Trans., 1997, 1337.
A. I. Popov, in Halogen Chemistry, edited by V. Gutmann, Academic,
London, 1967, Vol. I, p. 255.
5
6
7
21 (a) HMENAM:K. Lonsdale, H. J. Milledge and L. M. Pant, Acta
Crystallogr., 1965, 19, 827; (b) HMTMAC:Y. Barrans and
C. Gardia, C.R. Acad. Sci., Ser. C (Chim), 1967, 264, 876.
22 After the 5 h of exposure of 1$(H
equilibrate with the atmosphere for 24 h. The same operation was
repeated 3 times. The polycrystalline compound 1$(H O) was
2 2 2
O) to Br gas, the jar was opened to
8
2
2
occasionally subjected to manual ‘‘stirring’’ to reduce the possibility
that different crystallites (e.g. those originally at the top and
bottom of the sample) would receive significantly different amounts
2
of Br gas exposure within the overall duration of the gas-solid
reaction.
23 L. Kloo, J. Rosdahl and P. H. Svensson, Eur. J. Inorg. Chem., 2002,
1203.
24 (a) G. R. Desiraju and R. Parthasarathy, J. Am. Chem. Soc., 1989,
111, 8725; (b) V. R. Pedireddi, D. S. Reddy, B. S. Goud,
D. C. Craing, A. D. Rae and G. R. Desiraju, J. Chem. Soc Perkin
Trans 2, 1994, 1, 2354.
25 C. Chiappe and D. Pieraccini, J. Org. Chem., 2004, 69, 6059.
26 (a) Y.-Q. Wang, Z.-M. Wang, C.-S. Liao and C.-H. Yan, Acta
Crystallogr., Sect. C: Cryst. Struct. Commun., 1999, 55, 1503; (b)
G. V. Shilov, O. N. Kazheva, O. A. D’yachenko,
M. S. Chernov’yants, S. S. Simonyan, V. E. Gol’eva and
A. I. Pyshchev, Zh. Fiz. Khim. (Russ.) (Russ. J. Phys. Chem),
2002, 76, 1436.
9
1
0 K.-F. Tebbe, in Homoatomic Rings, Chains and Macromolecules of
Main-Group Elements, edited by A. L. Rheingold, Elsevier,
Amsterdam, 1977, p. 551–606 and references therein.
1 (a) G. A. Landrum, N. Goldberg and R. Hoffmann, J. Chem. Soc.,
Dalton Trans., 1997, 3605; (b) E. Gershgoren, E. Gordon and
S. Ruhman, J. Chem. Phys., 1997, 106, 4806; (c) A. Sanov,
T. Sandford, L. J. Butler, J. Vala, R. Kosloff and W. C. Lineberger,
J. Phys. Chem. A, 1999, 103, 10244.
1
1
1
2 (a) G. C. Pimentel, J. Chem. Phys., 1951, 19, 446; (b) R. J. Hach and
R. E. Rundle, J. Am. Chem. Soc., 1951, 73, 4321; (c) G. A. Landrum,
N. Goldberg and R. Hoffmann, J. Chem. Soc., Dalton Trans., 1997,
27 Y. Ogawa, Y. Takahaski and O. Kikuchi, THEOCHEM, 1998, 429,
187 and references cited therein.
3
605 and references therein.
28 C. Chiappe, F. Del Moro and M. Raugi, Eur. J. Org. Chem., 2001,
3501.
3 (a) G. H. Schmid, D. G. Garratt in The Chemistry of Double-Bonded
Functional groups, edited by S. Patai, Wiley, New York, 1977, Suppl.
A, part 2, p. 725, and references cited therein; (b) P. B. D. De la Mare
and R. Bolton, in Electrophilic Addition to Unsaturated Systems, 2nd
ed., Elsevier, New York, 1982, p. 136.
2 2 2
29 White powder 2$(H O) was exposed to Br vapours. The colourless
crystals changed to orange within 1.5 d.
30 (a) P. Deplano, J. R. Ferraro, M. L. Mercuri and E. F. Trogu, Coord.
Chem. Rev., 1999, 188, 71; (b) H. Mittag, H. Stegemann, H. F u€ llbier
and G. Irmer, J. Raman Spectrosc., 1989, 20, 251; (c) M. C. Aragoni,
M. Arca, F. A. Devillanova, M. B. Hursthouse, S. L. Huth, F. Isaia,
V. Lippolis, A. Mancini, H. R. Ogilvie and G. Verani, J. Organomet.
Chem., 2005, 690, 1923; (d) G. C. Hayward and P. J. Hendra,
Spectrochim. Acta, 1967, 23A, 2309; (e) A. G. Maki and
R. Forneris, Spectrochim. Acta, 1967, 23A, 867; (f) W. B. Person,
G. R. Anderson, J. N. Fordemwalt, H. Stammreich and
R. Forneris, J. Chem. Phys., 1961, 35, 908.
31 FT-Raman spectra have been recorded with a Nicolet NXR 9650
Raman Spectrometer in a backscattering geometry on samples
prepared in capillary glass tube. The line at 1064 nm of a Nd:YVO4
laser was used as the excitation source; to prevent possible
degradations the beam diameter was kept defocused (1 mm
diameter) with a power measured on the samples of 260 mw. For
every spectrum 512 scans have been acquired and averaged.
14 Considerable efforts are being paid to assemble DSSCs in which the
ꢀ
ꢀ
usual electrolyte containing the couple I /I
Br /Br
redox potential than I /I
3
is replaced by the
couple has a more positive
and therefore, the mismatch between its
ꢀ
ꢀ
ꢀ
ꢀ
3
redox couple. The Br /Br
3
ꢀ
ꢀ
3
redox potential and that of the dye is reduced. This gap reduction
may overcome the energy loss observed during the dye regeneration
in DSSCs and in turn the DSSCs performances. For details see:
C. Teng, X. Yang, C. Yuan, C. Li, R. Chen, H. Tian, S. Li,
A. Hagfeldt and L. Sun, Org. Lett., 2009, 11, 5542.
5 (a) B. O’Regan and M. Gr €a tzel, Nature, 1991, 353, 737; (b)
M. Gr €a tzel, Inorg. Chem., 2005, 44, 6841; (c) A. Hagfeldt and
M. Gr €a tzel, Acc. Chem. Res., 2000, 33, 269.
1
1
1
6 R. Kawano and M. Watanabe, Chem. Commun., 2005, 2107.
7 (a) R. P. Shibaeva and E. B. Yagubskii, Chem. Rev., 2004, 104, 5347;
(b) T. J. Marks and D. W. Kalina, Extended Linear Chain Compounds,
edited by J. S. Miller, Plenum Press, New York, 1982, Vol. 1, p. 197.
8 (a) M. M u€ ller, M. Albrecht, V. Gossen, T. Peters, A. Hoffmann,
G. Raabe, A. Valkonen and K. Rissanen, Chem.–Eur. J., 2010, 16,
32 The hybrid exchange–correlation functional PBE1PBE has been
adopted and aug-cc-pVTZ basis set has been chosen for Cl and Br
atoms. In the case of I atoms, fully relativistic ECP28MDF effective
core potential with aug-cc-pVTZ basis set have been used. All these
1
1
2446; (b) M. D. Garc ꢀı a, J. Mart ꢀı -Rujas, P. Metrangolo,
C. Peinador, T. Pilati, G. Resnati and G. Terraneo,
CrystEngComm, 2011, DOI: 10.1039/c0ce00860e.
9 (a) P. Metrangolo, T. Pilati and G. Resnati, CrystEngComm, 2006, 8,
calculations have been carried out by means of Gaussian03 code [g03].
ꢀ
33 The models I, II and III have been initially tested on the Br
ꢀ
3
anion
ꢀ
1
2 2
and then have been exported to I Cl and Br Cl .
946; (b) The IUPAC Task Group required to propose a definition of
the halogen bond has not yet ended its activity. (see www.iupac.org/
34 K. E. Nizzi, C. A. Pommerening and L. S. Sunderlin, J. Phys. Chem.
A, 1998, 102, 7674.
This journal is ª The Royal Society of Chemistry 2011
CrystEngComm, 2011, 13, 4427–4435 | 4435