Supramolecular chemistry of halogens: Complementary features of inorganic (M-X) and organic (C-X′) halogens applied to M-X...X′-C halogen bond formation

Article abstract of DOI:10.1021/ja0435182

Electronic differences between inorganic (M-X) and organic (C-X) halogens in conjunction with the anisotropic charge distribution associated with terminal halogens have been exploited in supramolecular synthesis based upon intermolecular M-X...X′-C halogen bonds. The synthesis and crystal structures of a family of compounds trans-[MCl₂(NC₅H ₄X-3)₂] (M = Pd(II), Pt(II); X = F, Cl, Br, I; NC ₅H₄X-3 = 3-halopyridine) are reported. With the exception of the fluoropyridine compounds, network structures propagated by M-Cl...X-C halogen bonds are adopted and involve all M-Cl and all C-X groups, M-Cl...X-C interactions show Cl...X separations shorter than van der Waals values, shorter distances being observed for heavier halogens (X). Geometries with near linear Cl...X-C angles (155-172°) and markedly bent M-Cl...X angles (92-137°) are consistently observed. DR calculations on the model dimers {trans-[MCl₂(NH₃)(NC ₅H₄X-3)]}2 show association through M-Cl...X-C (X ≠ F) interactions with geometries similar to experimental values. DFT calculations of the electrostatic potential distributions for the compounds trans-[PdCl₂(NC₅H₄X-3)₂] (X = F, Cl, Br, I) demonstrate the effectiveness of the strategy to activate C-X groups toward halogen bond formation by enhancing their electrophilicity, and explain the absence of M-Cl...F-C interactions. The M-Cl...X-C halogen bonds described here can be viewed unambiguously as nucleophile-electrophile interactions that involve an attractive electrostatic contribution. This contrasts with some types of halogen-halogen interactions previously described and suggests that M-Cl...X-C halogen bonds could provide a valuable new synthon for supramolecular chemists.

Full text of DOI:10.1021/ja0435182

Published on Web 03/31/2005
Supramolecular Chemistry of Halogens: Complementary
Features of Inorganic (M-X) and Organic (C-X′) Halogens
Applied to M-X‚‚‚X′-C Halogen Bond Formation
Fiorenzo Zordan,† Lee Brammer,*,† and Paul Sherwood‡
Contribution from the Department of Chemistry, UniVersity of Sheffield, Sheffield S3 7HF, U.K.,
and Computational Science and Engineering Department, CCLRC Daresbury Laboratory,
Daresbury, Warrington WA4 4AD, U.K.
Received October 26, 2004; E-mail: lee.brammer@sheffield.ac.uk
Abstract: Electronic differences between inorganic (M-X) and organic (C-X) halogens in conjunction with
the anisotropic charge distribution associated with terminal halogens have been exploited in supramolecular
synthesis based upon intermolecular M-X‚‚‚X′-C halogen bonds. The synthesis and crystal structures of
a family of compounds trans-[MCl2(NC5H4X-3)2] (M ) Pd(II), Pt(II); X ) F, Cl, Br, I; NC5H4X-3 )
3-halopyridine) are reported. With the exception of the fluoropyridine compounds, network structures
propagated by M-Cl‚‚‚X-C halogen bonds are adopted and involve all M-Cl and all C-X groups.
M-Cl‚‚‚X-C interactions show Cl‚‚‚X separations shorter than van der Waals values, shorter distances
being observed for heavier halogens (X). Geometries with near linear Cl‚‚‚X-C angles (155-172°) and
markedly bent M-Cl‚‚‚X angles (92-137°) are consistently observed. DFT calculations on the model dimers
{trans-[MCl2(NH3)(NC5H4X-3)]}2 show association through M-Cl‚‚‚X-C (X * F) interactions with geometries
similar to experimental values. DFT calculations of the electrostatic potential distributions for the compounds
trans-[PdCl2(NC5H4X-3)2] (X ) F, Cl, Br, I) demonstrate the effectiveness of the strategy to activate C-X
groups toward halogen bond formation by enhancing their electrophilicity, and explain the absence of
M-Cl‚‚‚F-C interactions. The M-Cl‚‚‚X-C halogen bonds described here can be viewed unambiguously
as nucleophile-electrophile interactions that involve an attractive electrostatic contribution. This contrasts
with some types of halogen-halogen interactions previously described and suggests that M-Cl‚‚‚X-C
halogen bonds could provide a valuable new synthon for supramolecular chemists.
Introduction
Some years ago Parthasarathy and colleagues examined the
nature of groups forming short contacts with organic halogens
(C-X) in the solid state and concluded that nucleophiles
approach approximately along the C-X bond axis, while
electrophiles approach the halogen in a roughly orthogonal
direction.4 They rationalized these observations in terms of
nucleophile interactions with the C-X group LUMO (its σ*
orbital) and electrophile interactions with the C-X group
HOMO, a predominantly p orbital-based lone pair (Scheme 1).
However, the situation is not quite as simple as this orbital model
alone suggests.
The interaction of nucleophiles with C-X groups has
subsequently been studied in detail by Allen and co-workers,5
who found a propensity in crystals for nitrogen and oxygen
atoms to form short contacts with carbon-bound halogens (X
) Cl, Br, I) at C-X‚‚‚O/N angles approaching 180°, consistent
with I. This is most pronounced for the heavier halogens.
Significantly, neither short contacts nor an orientational prefer-
ence is observed for interactions involving C-F groups.
Halogens have a ubiquitous presence in both inorganic and
organic chemistry, serving as monodentate or bridging ligands
for a wide variety of d-block, f-block, and main group metals
as well being common substituents in a large number of organic
compounds. Most frequently they lie at the periphery of
molecules. The resultant steric accessibility has the potential to
make halogenated compounds an attractive target for use in
supramolecular chemistry and crystal engineering wherein the
halogen atoms are directly involved in forming intermolecular
interactions. Indeed interest in packing arrangements of halo-
genated compounds goes back many years to what Schmidt
called the “chloro effect”, wherein the presence of chloro
substituents on aromatic compounds frequently resulted in
stacking arrangements with a resultant short (ca. 4 Å) crystal-
lographic axis.1,2 Chloroaromatic compounds have also been
intensely studied in the development of accurate anisotropic
atom-atom potentials for organic crystal structure prediction.3
† University of Sheffield.
‡ CCLRC Daresbury Laboratory.
(3) (a) Price, S. L.; Stone, A. J.; Lucas, J.; Rowland, R. S.; Thomleyt, A. E. J.
Am. Chem. Soc. 1994, 116, 4910. (b) Day, G. M.; Price, S. L. J. Am. Chem.
Soc. 2003, 125, 16434.
(4) Ramasubbu, N.; Parthasarathy, R.; Murray-Rust, P. J. Am. Chem. Soc. 1986,
108, 4308.
(5) Lommerse, J. P. M.; Stone, A. J.; Taylor, R.; Allen, F. H. J. Am. Chem.
Soc. 1996, 118, 3108.
(1) (a) Green, B. S.; Schmidt, G. M. J. Israel Chemical Society annual meeting
abstracts, 1971, 197. See also the discussion in ref 2, chapter 6, and further
references therein. (b) Cohen, M. D.; Schmidt G. M. J.; Sonntag, F. I. J.
Chem. Soc. 1964, 2000. (c) Schmidt, G. M. J. J. Chem. Soc. 1964, 2014.
(2) Desiraju, G. R. Crystal Engineering: The Design of Organic Solids;
Elsevier: Amsterdam, 1989.
9
10.1021/ja0435182 CCC: $30.25 © 2005 American Chemical Society
J. AM. CHEM. SOC. 2005, 127, 5979-5989
5979

A R T I C L E S
Zordan et al.
Scheme 1. Typical Interaction Geometries for Nucleophiles (Nu) I
and Electrophiles (El) II with Halocarbons (C-X) Rationalized in
Terms of HOMO-LUMO Interactions4
contrast, when bound to metal centers, halogens become much
stronger hydrogen bond acceptors8 and analogously form
stronger secondary interactions with other metal centers. This
contrast is illustrated by comparison of the electrostatic potential
associated with inorganic halogens (M-X) and with organic
halogen counterparts (C-X) shown in Figure 1. While both are
similarly anisotropic, the magnitude of the negative potential
is far greater for the inorganic halogens. Nevertheless, the
anisotropic potential provides the basis for directional interac-
tions that can be exploited in supramolecular assembly.
Scheme 2. Anisotropic Exchange Repulsion and Electrostatic
Interactions for Interactions of C-X Groups with Nucleophiles5
In the context of supramolecular synthesis, the properties of
halogens in organic molecules as electrophiles have been well
established for the heavier halogens.9 The resulting interactions
can be strong enough to influence the aggregation of organic
molecules in gas, liquid, and solid phase, and the term “halogen
bonding” has been introduced in order to establish an evident
analogy with hydrogen bonding.10 In addition, the hydrogen
bond acceptor properties of halogens bound to transition metals
n-
have led to the combination of halometalate ions, MXm (M
) metal, X ) Cl, Br), with cationic organic building blocks to
yield new hybrid ionic hydrogen-bonded materials, based on
supramolecular synthons such as N-H‚‚‚X-M, N-H‚‚‚X2M,
and N-H‚‚‚X3M.11
Complementary ab initio calculations using intermolecular
perturbation theory (IMPT) methods applied to a series of model
systems showed that approximately linear C-Cl‚‚‚O/N interac-
tions provide the lowest energy configuration. In all cases this
geometry is facilitated by an anisotropic exchange repulsion
contribution to the energy, which is at a minimum along the
C-Cl bond axis (Scheme 2, III). This is the origin of the
proposed anisotropic van der Waals radius for halogens6 and
provides an explanation for the packing of simple chlorohy-
drocarbons via C-Cl‚‚‚Cl-C short contacts,3a though other
opinions have been expressed on this matter.7 For the
CH3Cl‚‚‚OdCH2 model dimer5 the anisotropic electrostatic
potential around the halogen (see, Figure 1a) means that the
electrostatic interaction energy term is repulsive, but least
repulsive for a linear C-Cl‚‚‚O geometry (Scheme 2, IV),
thereby reinforcing the geometry favored by the exchange
repulsive term. In fact, in this example the overall interaction
is repulsive as the attractive charge transfer, polarization, and
dispersion terms are small in magnitude. However, analogous
calculations for the NC-CtC-Cl‚‚‚OdCH2 model dimer show
that while the anisotropic repulsion and electrostatic terms still
favor the linear C-Cl‚‚‚O geometry, the electron-withdrawing
alkyne and nitrile groups “activate” the C-Cl group, resulting
in an attractiVe electrostatic term, which combined with the
attractive second-order energy terms leads to attractive overall
C-Cl‚‚‚O interaction.
It is the exploitation of these differences between organic
and inorganic halogens that we seek to explore in developing a
new class of supramolecular synthons (M-X‚‚‚X′-C) for
applications in supramolecular chemistry and crystal engineer-
ing. Our strategy focuses on facilitating interactions between
inorganic halogens and suitably “activated” organic halogens
wherein an attractive electrostatic component may maximize
the efficacy of the anticipated directional interactions. Activation
of the organic halogen has been pursued through use of
halopyridines wherein the ring nitrogen atom exerts an electron-
withdrawing effect.12 The C-X groups can then be further
activated either by protonation of the pyridine nitrogen or its
coordination to a metal center (Scheme 3) or, as demonstrated
by Resnati and co-workers, by pyridine nitrogen methylation.13
The first approach (Scheme 3, i) has recently led to the synthesis
of hybrid ionic materials in which organic cations and halom-
etalate anions are linked via charge-assisted M-X‚‚‚X′-C halo-
gen bond synthons (X ) Cl, Br, I; X′ ) Cl, Br; M ) Co(II),
Cu(II)).14 Here we pursue the second approach (Scheme 3, ii)
and present the synthesis and characterization of a series of
molecular crystals of general formula trans-MCl2(NC5H4X-3)2
(M ) Pd(II), Pt(II); X ) F, Cl, Br, I; NC5H4X-3 ) 3-halopy-
ridine) designed in order to establish the applicability of the
M-Cl‚‚‚X-C interaction as an effective Lewis acid-Lewis
base pair synthon for supramolecular construction from neutral
We have previously noted the marked difference in hydrogen
bond acceptor capability between organic (C-X) and inorganic
(M-X) halogens that arises primarily from the far greater
polarity of the M-X bond.8 These studies confirm the prefer-
ence of electrophiles, in this case hydrogen bond donors (e.g.,
N-H), to interact with C-X groups with a geometry resembling
II.8a However, such interactions are weak not only for hydrogen
bond donors but where metal ions serve as the electrophiles. In
(9) (a) Corradi, E.; Meille, S. V.; Messina, M. T.; Metrangolo, P. Resnati, G.
Angew. Chem., Int. Ed. 2000, 39, 1782. (b) Metrangolo, P.; Resnati, G.
Chem. Eur. J. 2001, 7, 2511. (c) Bailey Walsh, R.; Padgett, C. W.;
Metrangolo, P.; Resnati, G.; Hanks, T. W.; Pennington, W. T. Cryst. Growth
Des. 2001, 1, 165. (d) Thallapally, P. K.; Desiraju, G. R.; Bagieu-Beucher,
M.; Masse, R.; Bourgogne, C.; Nicoud, J.-F. Chem. Commun. 2002, 1052.
(10) Legon, A. C. Angew. Chem., Int. Ed. 1999, 38, 2687.
(11) (a) Brammer, L.; Swearingen, J. K.; Bruton, E. A.; Sherwood, P. Proc.
Nat. Acad. Sci., U.S.A. 2002, 99, 4956. (b) Mareque Rivas, J. C.; Brammer,
L. Inorg. Chem. 1998, 37, 4756. (c) Lewis, G. R.; Orpen, A. G. J. Chem.
Soc., Chem. Commun. 1998, 1873.
(6) (a) Nyburg, S. C.; Wong-Ng, W. Proc. R. Soc. London 1979, A367, 29.
(b) Nyburg, S. C. Acta Crystallogr. 1979, A35, 641. (c) Nyburg, S. C.;
Faerman, C. H. Acta Crystallogr. 1985, B41, 274.
(12) Janiak, C. J. Chem. Soc., Dalton Trans. 2000, 3885.
(13) Logothetis, T. A.; Meyer, F.; Metrangolo, P.; Pilati, T.; Resnati, G. New J.
Chem. 2004, 28, 760.
(7) Desiraju, G. R.; Parthasarathy, R. J. Am. Chem. Soc. 1989, 111, 8725.
(8) (a) Brammer, L.; Bruton, E. A.; Sherwood, P. Cryst. Growth Des. 2001, 1,
277. (b) Aullo´n, G.; Bellamy, D.; Brammer, L.; Bruton, E. A.; Orpen, A.
G. Chem. Commun. 1998, 653.
(14) (a) Brammer, L.; M´ınguez Espallargas, G.; Adams, H. CrystEngComm 2003,
5, 343. (b) Willett, R. D.; Awwadi, F.; Butcher, R.; Haddad, S.; Twamley,
B. Cryst. Growth Des. 2003, 3, 301. (c) Freytag, M.; Jones, P. G. Chem.
Commun. 2000, 277.
9
5980 J. AM. CHEM. SOC. VOL. 127, NO. 16, 2005

Supramolecular Chemistry of Halogens
A R T I C L E S
Figure 1. Negative electrostatic potential distributions for (a) CH3Cl, (b) trans-PdCl(Me)(PH3)2, (c) CH3I, and (d) trans-PdI(Me)(PH3)2. Contour levels are
at 4 kcal/mol for the organic halogens (C-X) in (a) and (c) and at 10 kcal/mol for the inorganic halogens (M-X) in (b) and (d). (Figure adapted from ref
8a with permission of the American Chemical Society.)
Scheme 3. Approaches to “Activating” Organic Halogens (C-X) toward Interactions with Nucleophiles, Leading to Applications in Which
M-Cl‚‚‚X-C Halogen Bonds Can Be Used for Supramolecular Assembly
building blocks.15 The synthetic studies are supported by DFT
halopyridine, respectively, were prepared, the latter acidified using HCl.
The two solutions were layered in separate arms of the U-tube above
a basal layer of CH2Cl2. Both arms of the U-tube were then sealed
with Parafilm.
Method D (for synthesis of crystalline powders of compounds
5-8). Typically 40 mg of Na2PdCl4 was dissolved in 2 mL of MeOH/
CH2Cl2 (1:1) (solution I), and a stoichiometric amount of halopyridine
calculations on model systems and ab initio electrostatic
potential calculations on the trans-PdCl2(NC5H4X-3)2 series of
compounds. The resulting crystal structures provide compelling
evidence for the efficacy of the M-Cl‚‚‚X-C synthon (X *
F), which clearly defines extended 1D or 2D motifs in all cases.
was dissolved in 2 mL of MeOH previously acidified using 30 mg of
HCl (37% aqueous) (solution II). The two solutions I and II were kept
separated in a vial-in-vial system, and enough MeOH (ca. 20 mL) was
very slowly added into the larger vial in order to put the solutions in
contact. The process of diffusion was generally faster than that observed
with the U-tubes (method C), and within 2 weeks all the samples
prepared yielded very thin yellow needle-shaped crystals (same
morphology as those obtained from the U-tubes), not suitable for
analysis as single crystals but suitable for elemental analysis and powder
diffraction.
Synthesis of trans-[PtCl2(NC5H4F-3)2] (1). K2PtCl4 (30 mg, 0.072
mmol) in acidic aqueous solution (pH ≈ 4, 2 mL) was layered on
3-fluoropyridine (15 mg, 0.154 mmol) in CH2Cl2 (2 mL) using method
A. Yellow prismatic crystals of 1 (24.6 mg, yield 74.1%) resulted after
4 weeks. Anal. Calc: C, 26.08; H, 1.73; N, 6.08. Found: C, 25.91; H,
1.71; N, 5.87. The powder diffraction pattern (bulk sample) was
consistent with the pattern calculated from single-crystal data.
Synthesis of trans-[PtCl2(NC5H4Cl-3)2] (2). K2PtCl4 (30 mg, 0.072
mmol) in acidic aqueous solution (pH ≈ 4, 2 mL) was layered on
3-chloropyridine (16 mg, 0.140 mmol) in CH2Cl2 (2 mL) using method
A. Yellow prismatic crystals of 2 (34.8 mg, yield 98%) resulted after
1 week. Anal. Calc: C, 24.34; H, 1.62; N, 5.68. Found: C, 24.29; H,
1.58; N, 5.53. The powder diffraction pattern (bulk sample) was
consistent with the pattern calculated from single-crystal data.
Synthesis of trans-[PtCl2(NC5H4Br-3)2] (3a). K2PtCl4 (30 mg, 0.072
mmol) in acidic aqueous solution (pH ≈ 4, 4 mL) and 3-bromopyridine
(22 mg, 0.140 mmol) in CH2Cl2 solution (6 mL) were prepared. The
Experimental Section
General Procedures. All reagents (purchased from Aldrich or
Lancaster) and solvents were used as received. Single crystals of all
compounds were prepared by one of methods A-C. An alternative
method (D) was employed to prepare some samples as powders.
Elemental analyses were conducted by the Elemental Analysis Service,
Department of Chemistry, University of Sheffield. Powder diffraction
data were obtained using Cu KR radiation on a Bruker D8 diffractometer
in reflection mode. Thermal gravimetric analyses of 1 and 5‚MeOH
were conducted using a Perkin-Elmer Series 7 TGA instrument at a
scan speed of 1 °C/min.
General Procedures for Crystal Synthesis. Method A. Crystals
were obtained by diffusion between immiscible layers at room
temperature (20 °C). In small vials an acidified aqueous phase
containing the metal salt was layered upon a CH2Cl2 phase containing
the halopyridine. All the vials were sealed with a plastic lid during
crystal growth.
Method B. This method is a variant of method A in which a third
phase (toluene) was layered between the aqueous phase containing the
metal salt and an acidified MeOH phase containing the halopyridine
to reduce the rate of diffusion and thus mixing of the two components.
Method C. Crystals were obtained at room temperature (20 °C) by
a method based on slow diffusion in U-shaped tubes (internal diameter
5 mm). Methanolic solutions containing the metal salt and the
(15) Pennington has also reported neutral Cd-I‚‚‚I-I interactions: Bailey, R.
D.; Hook, L. L.; Pennington, W. T. Chem. Commun. 1998, 1181.
9
J. AM. CHEM. SOC. VOL. 127, NO. 16, 2005 5981

A R T I C L E S
Zordan et al.
aqueous solution (2 mL) was layered on the CH2Cl2 solution (2 mL)
using method A. Yellow needle-shaped crystals of 3a (40.8 mg, yield
97.3%) resulted after 4 weeks. Anal. Calc: C, 20.62; H, 1.37; N, 4.81.
Found: C, 20.44; H, 1.43; N, 4.67. The powder diffraction pattern (bulk
sample) was consistent with the pattern calculated from single-crystal
data.
(35 mg, 0.170 mmol), to which 20 mg of HCl (37% aqueous solution)
had been previously added, was diluted using MeOH (2 mL) (solution
B). Following method C, solution A (1 mL) and solution B (1 mL)
were respectively layered in separate arms of a U-tube above CH2Cl2
(2 mL). Thin yellow needle-shaped crystals of 8 were observed after 4
weeks. Thicker crystals (ca. 2 mg) were harvested after a further 4
weeks. A powder sample of 8 was prepared according to method D
(yield 39.3%). Anal. Calc: C, 20.43; H, 1.36; N, 4.76. Found: C, 20.41;
H, 1.32; N, 4.60. The powder diffraction pattern (bulk sample) was
consistent with the pattern calculated from single-crystal data.
Crystallography. X-ray data were collected on a Bruker SMART
1000 diffractometer using Mo KR radiation. Crystal structures were
solved and refined against all F2 values using the SHELXTL suite of
programs.16 A summary of the data collection and structure refinement
information is provided in Table 1. Data were corrected for absorption
using empirical methods (SADABS) based upon symmetry-equivalent
reflections combined with measurements at different azimuthal angles.17
Non-hydrogen atoms were refined anisotropically. Hydrogen atoms
were placed in calculated positions, refined using idealized geometries
(riding model), and assigned fixed isotropic displacement parameters.
For 5‚MeOH, the methanol molecule is disordered between two
positions related by a center of inversion, to each of which 50%
occupancy was assigned.
Synthesis of trans-[PtCl2(NC5H4Br-3)2] (3b). 3-Bromopyridine (40
mg, 0.253 mmol) in MeOH (2 mL), previously acidified using 30 mg
of HCl (37% aqueous), was layered on toluene (5 mL), which in turn
had been layered upon K2PtCl4 (10 mg, 0.024 mmol) in H2O (2 mL).
This procedure follows method B. Thin yellow needle-shaped crystals
of 3b were first observed after 2 weeks. Thicker crystals were harvested
after a further 5 weeks (12.3 mg, yield 88.0%). Anal. Calc: C, 20.62;
H, 1.37; N, 4.81. Found: C, 20.49; H, 1.39; N, 4.57. The powder
diffraction pattern (bulk sample) did not indicate a homogeneous phase
but could not simply be analyzed as a mixture of 3a and 3b.
Synthesis of trans-[PtCl2(NC5H4I-3)2] (4). K2PtCl4 (30 mg, 0.072
mmol) in acidic aqueous solution (pH ≈ 4, 2 mL) was layered on
3-iodopyridine (30 mg, 0.146 mmol) in CH2Cl2 (2 mL). Yellow laminar
crystals of 4 (46.9 mg, yield 96.3%) resulted after 1 week. Anal. Calc:
C, 17.75; H, 1.18; N, 4.14. Found: C, 17.48; H, 1.17; N, 3.80. The
powder diffraction pattern (bulk sample) was consistent with the pattern
calculated from single-crystal data.
Theoretical Calculations. Geometry optimizations for the four trans-
PdCl2(NH3)(NC5H4X-3) model complexes (X ) F, Cl, Br, I) and their
putative dimers were conducted using the GAMESS-UK18 package,
employing the B3LYP DFT functional.19 The basis set and pseudopo-
tentials used were LANL2DZ(dp);20 for the lighter elements C, H, N,
and F this was an all-electron Dunning DZP quality basis21 and for Cl,
Br, I, and Pd the effective core potentials (ECPs) of Hay and Wadt22
were used, and the associated basis sets were augmented with diffuse
and polarization function.23 The optimization for the dimer {PdCl2-
(NH3)(NC5H4I)}2 was initiated using a geometry resembling that
observed in the crystal structure of compound 8. Subsequent dimer
optimizations were initiated from the same geometry after replacing
the iodine atom with bromine, chlorine, and fluorine, respectively.
Electrostatic potentials for the PdCl2(NC5H4X-3)2 complexes
(X ) F, Cl, Br, I) were calculated at the B3LYP level using the
GAMESS-UK18 package, employing the effective core potentials
(ECPs) of Hay and Wadt,20,22 the LANL2DZ(dp)20,23 basis set for Pd,
and the Sadlej pVTZ (polarized valence triple-ú) basis sets20,24 on all
Synthesis of trans-[PdCl2(NC5H4F-3)2]‚MeOH (5‚MeOH).
Na2PdCl4 (25 mg, 0.084 mmol) was dissolved in MeOH (2 mL)
(solution A), and 3-fluoropyridine (17 mg, 0.175 mmol), to which 20
mg of HCl (37% aqueous solution) had been previously added, was
diluted using MeOH (2 mL) (solution B). Following method C, solution
A (1 mL) and solution B (1 mL) were respectively layered in separate
arms of a U-tube above CH2Cl2 (2 mL). Thick yellow needle-shaped
crystals of 5‚MeOH (<1 mg) resulted after 16 weeks. A powder sample
of 5‚MeOH was prepared according to method D (yield 54%). The
powder diffraction pattern (bulk sample) was consistent with the pattern
calculated from single-crystal data. Mass loss during combustion
analysis prevented accurate determination of composition by this
method. TGA analysis illustrates the problem in that a gradual loss of
mass occurs with an onset around 100 °C, culminating in a total of
70% mass loss by 275 °C. This contrasts with the unsolvated Pt
analogue (1), which is thermally stable up to 225 °C and then undergoes
a sharp loss of mass (ca. 25%) upon further heating.
Synthesis of trans-[PdCl2(NC5H4Cl-3)2] (6). Na2PdCl4 (10 mg,
0.034 mmol) was dissolved in MeOH (10 mL) (solution A), and
3-chloropyridine (20 mg, 0.176 mmol), to which 60 mg of HCl (37%
aqueous solution) had been previously added, was diluted using MeOH
(2 mL) (solution B). Following method C, solution A (1 mL) and
solution B (1 mL) were respectively layered in separate arms of a U-tube
above CH2Cl2 (2 mL). Thick yellow needle-shaped crystals of 6 (<1
mg) resulted after 8 weeks. A powder sample of 6 was prepared
according to method D (yield 71.6%). Anal. Calc: C, 29.68; H, 1.97;
N, 6.92. Found: C, 29.17; H, 1.84; N, 6.52. The powder diffraction
pattern (bulk sample) was consistent with the pattern calculated from
single-crystal data.
Synthesis of trans-[PdCl2(NC5H4Br-3)2] (7). Na2PdCl4 (25 mg,
0.084 mmol) was dissolved in MeOH (2 mL) (solution A), and
3-bromopyridine (25 mg, 0.158 mmol), to which 20 mg of HCl (37%
aqueous solution) had been previously added, was diluted using MeOH
(2 mL) (solution B). Following method C, solution A (1 mL) and
solution B (1 mL) were respectively layered in separate arms of a U-tube
above CH2Cl2 (2 mL). Thick yellow needle-shaped crystals of 7 (ca. 2
mg) resulted after 3 weeks. A powder sample of 7 was prepared
according to method D (yield 46.2%). Anal. Calc: C, 24.33; H, 1.62;
N, 5.68. Found: C, 24.27; H, 1.54; N, 5.39. The powder diffraction
pattern (bulk sample) was consistent with the pattern calculated from
single-crystal data.
(16) SHELXTL 5.1; Bruker Analytical X-Ray Instruments, Inc., 1998.
(17) (a) Sheldrick, G. M. SADABS, Empirical absorption correction program;
University of Go¨ttingen, 1995, based upon the method of Blessing.17b (b)
Blessing, R. H. Acta Crystallogr. 1995, A51, 33.
(18) GAMESS-UK is a package of ab initio programs written by Guest, M. F.;
van Lenthe, J. H.; Kendrick, J.; Schoffel, K.; Sherwood, P. with contribu-
tions from Amos, R. D.; Buenker, R. J.; van Dam, H. J. J.; Dupuis, M.;
Handy, N. C.; Hillier, I. H.; Knowles, P. J.; Bonacic-Koutecky, V.; von
Niessen, W.; Harrison, R. J.; Rendell, A. P.; Saunders: V. R.; Stone, A.
J.; de Vries, A. H. The package is derived from the original GAMESS
code due to Dupuis, M.; Spangler, D.; Wendoloski, J. NRCC Software
Catalog, Vol. 1, Program No. QG01 (GAMESS), 1980.
(19) (a) Stephens, P. J.; Devlin, F. J.; Chabalowski, C. F.; Frisch, M. J. J. Phys.
Chem. 1994, 98, 11623, and references therein. (b) GAMESS-UK uses the
VWN 3 (RPA-type) parametrization; for details on this issue see: Hertwig,
R. H.; Koch, W. Chem. Phys. Lett. 1997, 268, 345.
(20) All basis sets and ECP data were obtained from the Extensible Computa-
tional Chemistry Environment Basis Set Database, Version 02/25/04, as
developed and distributed by the Molecular Science Computing Facility,
Environmental and Molecular Sciences Laboratory, which is part of the
Pacific Northwest Laboratory, P.O. Box 999, Richland, WA 99352, and
funded by the U.S. Department of Energy. The Pacific Northwest
Laboratory is a multiprogram laboratory operated by Battelle Memorial
Institute for the U.S. Department of Energy under contract DE-AC06-
76RLO 1830. Contact David Feller or Karen Schuchardt for further
information.
(21) Dunning, T. H., Jr.; Hay, P. J. In Methods of Electronic Structure Theory,
Vol. 2; Schaefer, H. F., III. Ed.; Plenum Press: New York, 1977.
(22) (a) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 270. (b) Hay, P. J.;
Wadt, W. R. J. Chem. Phys. 1985, 82, 284. (c) Hay, P. J.; Wadt, W. R. J.
Chem. Phys. 1985, 82, 299.
(23) Check, C. E.; Faust, T. O.; Bailey, J. M.; Wright, B. J.; Gilbert, T. M.;
Sunderlin, L. S. J. Phys. Chem. A 2001, 105, 8111.
(24) Sadlej, A. J. Theor. Chim. Acta 1992, 81, 45, and references therein.
Synthesis of trans-[PdCl2(NC5H4I-3)2] (8). Na2PdCl4 (25 mg, 0.084
mmol) was dissolved in MeOH (2 mL) (solution A), and 3-iodopyridine
9
5982 J. AM. CHEM. SOC. VOL. 127, NO. 16, 2005

Supramolecular Chemistry of Halogens
A R T I C L E S
other atoms. Molecular geometries were based upon those in the crystal
structures of the corresponding molecules and the platinum analogues.
Average values for intra-ring distances and angles were used to give
the symmetrized ring geometries for all four compounds (see Supporting
Information). An average C-N-Pd-Cl torsion angle of 55° was used
for all calculations (cf. experimental values of 50-65°).
Results
Reaction of K2PtCl4 or Na2PdCl4 with the appropriate
3-halopyridine (NC5H4X-3) under acidic conditions permitted
preparation of the entire series of compounds trans-[MCl2-
(NC5H4X-3)2] (M ) Pt, Pd; X ) F, Cl, Br, I). Initial trials using
direct mixing of miscible solutions of the two reactants yielded
microcrystalline powders within minutes for the platinum
systems and within seconds for the palladium analogues. Thus,
slow mixing of solutions of the two reactants via immiscible
phases (method A) was necessary to prepare crystals of the
platinum compounds that were of suitable size for single-crystal
X-ray diffraction studies. Method C, using the U-tubes, permit-
ted the slowest mixing and proved necessary to prepare crystals
of the palladium compounds. Compounds 1-8 were character-
ized by low-temperature X-ray crystallography, with two
polymorphs being obtained for compound 3 under different
synthetic conditions. All compounds crystallize such that the
trans-[MCl2(NC5H4X-3)2] molecules adopt a crystallographi-
cally imposed centrosymmetric conformation. All compounds
except 5‚MeOH gave satisfactory elemental analyses. With the
exception of polymorph 3b, powder diffraction studies con-
firmed the presence of a single phase consistent with the single-
crystal structure determination for all compounds.
In all cases except that of the fluoropyridine systems, the
crystal structure contains a network of molecules linked via
M-Cl‚‚‚X-C interactions. Two network types were found,
namely, 1D tapes in which each molecule is linked to two
neighbors via pairs of M-Cl‚‚‚X-C interactions and 2D layers
in which each molecule is linked to four neighbors via individual
M-Cl‚‚‚X-C interactions.
Tape Structures of trans-[PtCl2(NC5H4Cl-3)2] (2), trans-
[PtCl2(NC5H4Br-3)2] (3b), and trans-[PdCl2(NC5H4I-3)2] (8).
The structures of 2 and 8 and the polymorph 3b comprise 1D
tape motifs propagated along the c-axis in which neighboring
molecules are centrosymmetrically linked via a pair of M-Cl‚
‚‚X-C interactions (2, X ) Cl, M ) Pt; 3b, X ) Br, M ) Pt;
8, X ) I, M ) Pd), as shown in Figures 2 and 3. In all cases
these are the only halogen-halogen interactions in the structure.
3b and 8 are isostructural, as is evident from unit cell parameters.
In 2 the Pt-Cl‚‚‚Cl-C interaction has a Cl‚‚‚Cl separation
slightly shorter than the sum of the van der Waals radii of the
two chlorine atoms and exhibits a large obtuse angle at the
organic chloride, but a much smaller angle at the metal-bound
chloride [(Pt)Cl‚‚‚Cl(C) 3.443(1) Å, Pt-Cl‚‚‚Cl 92.69(3)°,
C-Cl‚‚‚Cl 155.9(1)°]. The halogen-bonded ribbons further
aggregate in the three-dimensional structure via C-H‚‚‚Cl-Pt
hydrogen bonds [(Pt)Cl‚‚‚H(C) 2.864 Å, C-H‚‚‚Cl 128.0°,
Pt-Cl‚‚‚H 115.1° and (Pt)Cl‚‚‚H(C) 2.866 Å, C-H‚‚‚Cl 131.0°,
Pt-Cl‚‚‚H 76.4°] and via offset stacking interactions of cen-
trosymmetrically related halopyridine rings (Table 2 and Figure
12a, vide infra). Evidence for the supporting role of pairs of
C-H‚‚‚Cl-C interactions is also of note [(C)H‚‚‚Cl(C) 2.759
Å, C-H‚‚‚Cl 163.9°, C-Cl‚‚‚H ) 111.8° and (C)H‚‚‚Cl(C)
2.894 Å, C-H‚‚‚Cl 132.1°, C-Cl‚‚‚H ) 118.4°]. The shorter
1.
Table
9
J. AM. CHEM. SOC. VOL. 127, NO. 16, 2005 5983

A R T I C L E S
Zordan et al.
Table 2. Halogen-Halogen Interaction Geometries
(M)Cl‚‚‚X(C)
distance, Å
M
−
Cl‚‚‚
X
C
−
X
‚‚‚Cl
a
interaction
angle, deg
angle, deg (rCl +
rX), Åa RClX
2
Pt-Cl‚‚‚Cl-C
3.443(1)
92.69(3) 155.9(1)
3.50
0.98
0.95
0.93
0.95
0.89
1.03
0.98
0.91
3a Pt-Cl‚‚‚Br-C 3.429(3) 137.1(1) 165.7(3)
3.60
3.60
3.60
3.71
3.50
3.60
3.71
3.339(3) 126.1(1) 172.0(3)
3b Pt-Cl‚‚‚Br-C 3.431(2) 126.5(1) 166.6(2)
4
6
7
8
Pt-Cl‚‚‚I-C
3.306(2) 131.7(1) 170.0(2)
Pd-Cl‚‚‚Cl-C 3.596(4) 121.1(1) 159.3(3)
Pd-Cl‚‚‚Br-C 3.534(1) 121.69(3) 158.2(1)
Pd-Cl‚‚‚I-C
3.389(1) 121.98(3) 169.2(1)
a RClX ) d(Cl‚‚‚X)/rCl + rX, where rCl and rX are, respectively, the
van der Waals radii26 of Cl and X (following the definition of Lommerse
et al.5).
Figure 2. Views of 2 showing the tape network propagated via
Pt-Cl‚‚‚Cl-C interactions. Metal and chloride ligands are shown in red,
organic halogens in green, and all other atoms (N, C, H) in blue.
Figure 4. View of 3a showing the 2D network propagated via
Pt-Cl‚‚‚Br-C interactions. Colors are as in Figure 2.
C-H‚‚‚Cl 136.2° (138.6°), Pt-Cl‚‚‚H 101.0° (100.0°) for 3b;
(C)H‚‚‚Cl(Pd) 2.772, 2.860 Å, C-H‚‚‚Cl 140.9, 138.0°,
Pt-Cl‚‚‚H 104.3, 104.6° for 8] and via offset stacking interac-
tions of translationally related halopyridine rings (Table 2 and
Figure 12b, vide infra).
Layer Structures of trans-[PtCl2(NC5H4Br-3)2] (3a), trans-
[PdCl2(NC5H4Cl-3)2] (6), and trans-[PdCl2(NC5H4Br-3)2] (7).
The structures 6, 7, and polymorph 3a contain 2D motifs
propagated via M-Cl‚‚‚X-C interactions (6, X ) Cl, M )
Pd; 7, X ) Br, M ) Pd; 3a, X ) Br, M ) Pt) through which
each molecule interacts with four neighbors, as shown in Figures
4 and 5. Viewing each molecule as a 4-connected node, the 2D
network is of (4,4) topology. Again, in all three structures these
are the only halogen-halogen interactions present. Compounds
6 and 7 are isostructural, as is evident from unit cell parameters.
The structure of 3a is quite similar to the other pair, indeed
topologically identical, but has a lower space group symmetry,
which results in the presence of two independent molecules for
3a.25 A more undulating layer arrangement is found in 3a, where
the angle between the pyridine ring planes of neighboring
molecules reaches 59.8°, while the corresponding angle in 6
Figure 3. Views of 3b showing the tape network propagated via
Pt-Cl‚‚‚Br-C interactions. Compound 8 is isostructural with 3b, with
the tape propagated through Pd-Cl‚‚‚I-C interactions. Colors are as in
Figure 2.
of these interactions appears to directly reinforce the
Pt-Cl‚‚‚Cl-C interactions (Figure 13, vide infra).
In structures 3b and 8 the halogen-halogen interactions
show geometries that are qualitatively analogous to those in 2
[(Pt)Cl‚‚‚Br(C) 3.431(2) Å, Pt-Cl‚‚‚Br 126.5(1)°, C-Br‚‚‚Cl
166.6(2)° for 3b; (Pd)Cl‚‚‚I(C) 3.389(1) Å, Pd-Cl‚‚‚I 121.98-
(3)°, C-I‚‚‚Cl 169.2(1)° for 8]. The ribbons aggregate in sheets
via secondary C-H‚‚‚Cl-M interactions [(C)H‚‚‚Cl(Pt) 2.693
Å, C-H‚‚‚Cl 166.0°, Pt-Cl‚‚‚H 108.3° for 3b; (C)H‚‚‚Cl(Pd)
2.848 Å, C-H‚‚‚Cl 171.0°, Pt-Cl‚‚‚H 104.2° for 8], and these
sheets stack along the minor axis (c) of the unit cell via further
C-H‚‚‚Cl-M interactions [(C)H‚‚‚Cl(Pt) 2.871, (2.996) Å,
(25) Although Z′ ) 1 for 3a, there the asymmetric unit comprises two
independent half molecules, each situated at an inversion centre. For all
other structures reported herein, Z′ ) 0.5.
9
5984 J. AM. CHEM. SOC. VOL. 127, NO. 16, 2005

Supramolecular Chemistry of Halogens
A R T I C L E S
Figure 8. View of 5‚MeOH approximately along the a-axis showing
channel structure. Disordered methanol molecules, which lie in the channels,
are not shown. Colors are as in Figure 2.
Figure 5. View of
7 showing the 2D network propagated via
Pd-Cl‚‚‚Br-C interactions. Compound 6 is isostructural with 7, but the
network is propagated through Pd-Cl‚‚‚Cl-C interactions. Colors are as
in Figure 2.
Figure 9. DFT optimized geometry of the dimer of trans-[PdCl2(NH3)-
(NC5H4I)]. The bromo- and chloropyridine analogues adopt very similar
geometries. Colors are as in Figure 2.
C-Cl‚‚‚Cl 159.3(3)° for 6; (Pd)Cl‚‚‚Br(C) 3.534(1) Å,
Pd-Cl‚‚‚Br 121.69(3)°, C-Br‚‚‚Cl 158.2(1)° for 7]. In these
three structures, the layers are linked via offset stacking
interactions of translationally related halopyridine rings (Table
2, vide supra) and C-H‚‚‚Cl-M interactions [(C)H‚‚‚Cl(Pt)
2.683, 2.784 Å, C-H‚‚‚Cl 139.2°, 133.9°, Pt-Cl‚‚‚H ) 96.7,
96.3° for 3a; (C)H‚‚‚Cl(Pd) 2.783, 2.788 Å, C-H‚‚‚Cl 134.3°,
136.2°, Pd-Cl‚‚‚H ) 104.6°, 103.9° for 6; (C)H‚‚‚Cl(Pd) 2.793,
2.807 Å, C-H‚‚‚Cl 135.1°, 135.3°, Pd-Cl‚‚‚H ) 104.9°, 104.5°
for 7].
Figure 6. View of
Pt-Cl‚‚‚I-C interactions. Colors are as in Figure 2.
4 showing the 2D network propagated via
Layer Structure of trans-[PtCl2(NC5H4I-3)2] (4). Com-
pound 4 adopts a layer structure in which each molecule
is linked to four neighbors via Pt-Cl‚‚‚I-C interac-
tions [(Pt)Cl‚‚‚I(C) 3.306(2) Å, Pt-Cl‚‚‚I 131.7(1)°, C-I‚‚‚Cl
170.0(2)°] to give a 2D network of (4,4) topology. However,
the relative orientation differs from that of 3a, 6, and 7 in that
offset stacking interactions of translationally related halopyridine
rings occurs between neighboring molecules within the layer
rather than between layers (Figure 6; Table 2, vide supra). As
in the previous structures, the stacking interactions are supported
by C-H‚‚‚Cl-Pt hydrogen bonds [(C)H‚‚‚Cl(Pt) 2.701 Å,
C-H‚‚‚Cl 147.7°, Pt-Cl‚‚‚H ) 91.0°]. Stacking of the
layers is assisted by further C-H‚‚‚Cl-Pt hydrogen bonds
[(C)H‚‚‚Cl(Pt) 2.745 Å, C-H‚‚‚Cl 168.7°, Pt-Cl‚‚‚H 140.0°].
Structures of trans-[PtCl2(NC5H4F-3)2] (1) and trans-
[PdCl2(NC5H4F-3)2]‚MeOH (5‚MeOH). In the structures of
1 and 5‚MeOH halogen-halogen interactions are absent. In 1
molecules are linked primarily through C-H‚‚‚Cl-Pt hydrogen
bonds [(C)H‚‚‚Cl(Pt) 2.777, 2.784, 2.785 Å, C-H‚‚‚Cl 140.6°,
Figure 7. View of 1 showing C-H‚‚‚Cl-Pt hydrogen bonds (dashed lines)
and halopyridine ring stacking. Colors are as in Figure 2.
and 7 is 13.6° and 13.5°, respectively. The difference is
associated with a slightly different molecular conformation
adopted by 3a [C-N-M-Cl torsion angle 133.9°, 132.1° (3a),
121.1° (6), 121.2° (7)] and shorter halogen-halogen interactions
for 3a than for 6 and 7 [(Pt)Cl‚‚‚Br(C) 3.339(3), 3.429(3) Å,
Pt-Cl‚‚‚Br 126.1(1)°, 137.1(1)°, C-Br‚‚‚Cl 172.0(3)°, 165.7-
(3)° for 3a; (Pd)Cl‚‚‚Cl(C) 3.596(4) Å, Pd-Cl‚‚‚Cl 121.1(1)°,
9
J. AM. CHEM. SOC. VOL. 127, NO. 16, 2005 5985

A R T I C L E S
Zordan et al.
Figure 10. Calculated electrostatic potential for trans-[PdCl2(NC5H4X-3)2], (a) X ) I, (b) Br, (c) Cl, (d) F, shown in the plane of the pyridine rings. (Note
that the chloride ligands do not lie in this plane. Thus the negative potential associated with these ligands is substantially greater than appears in these
figures; e.g. see ref 8a.) Contour interval 2.5 kcal/mol. Colors: blue (most positive), red (most negative).
Figure 13. Role of C-H‚‚‚Cl-C interactions in the cross-linking of
Pt-Cl‚‚‚Cl-C bonded tapes in 2. Pt-Cl‚‚‚Cl-C interactions are also shown.
Colors are as in Figure 2.
2, vide supra). Long C-H‚‚‚F-C contacts (2.66 Å) are also
noted.
In 5‚MeOH the [PdCl2(NC5H4F-3)2] molecules assemble
into a layer arrangement via C-H‚‚‚Cl-Pd hydrogen bonds
Figure 11. Calculated electrostatic potential for trans-[PdCl2-
(NC5H4X-3)2] (X ) F, Cl, Br, and I, as indicated) plotted along the C-X
bond axis. In each case zero represents a distance from the halogen nuclear
position corresponding to the van der Waals radius26 of that halogen.
[(C)H‚‚‚Cl(Pd) 2.864 Å, C-H‚‚‚Cl 131.9°, Pd-Cl‚‚‚H 105°]
and weak C-H‚‚‚F-C interactions (2.56 Å) (Figure 8). The
sheets stack via offset stacking interactions involving transla-
tionally related halopyridine rings of neighboring molecules
(Table 2, vide supra) coupled with further C-H‚‚‚Cl-Pd
hydrogen bonds [(C)H‚‚‚Cl(Pd) 2.695, 2.770 Å, C-H‚‚‚Cl
139.9°, 135.2°, Pd-Cl‚‚‚H 152.4°, 102.4°]. The resulting
structure has channels running parallel to the a-axis which are
occupied by disordered methanol molecules based upon best
efforts to model the solvate crystallographically.
Theoretical Calculations on Dimers of the Model Com-
plexes trans-[PdCl2(NH3)(NC5H4X-3)] (X ) F, Cl, Br, I). To
further explore the fact that M-Cl‚‚‚X-C interactions are found
to be prevalent in the structures of trans-[MCl2(NC5H4X-3)2]
(X ) I, Br, Cl) but absent in trans-[MCl2(NC5H4F-3)2], we have
undertaken a series of DFT calculations to examine the
optimized geometries of putative dimers of the model complexes
trans-[PdCl2(NH3)(NC5H4X-3)] (X ) I, Br, Cl, F). The starting
dimer geometry used for subsequent optimizations was based
upon that observed between neighboring molecules in the tape
structure of 8 (and similar to that found in 2 and 3b). Optimized
dimer geometries (Figure 9) for the iodo, bromo, and chloro
derivatives showed good agreement with the observed geom-
etries within the tape structures of 8, 3b, and 2, respectively,
with Pd-Cl‚‚‚X-C interactions having approximately linear
C-X‚‚‚Cl angles and Pd-Cl‚‚‚X angles in the range 121-128°.
Average (Pd)Cl‚‚‚X(C) distances were calculated as 3.390 Å
(X ) I), 3.479 Å (X ) Br), and 3.429 Å (X ) Cl), all within
the sum of van der Waals radii for the pair of halogens in-
volved. By contrast, the trans-[PdCl2(NH3)(NC5H4F)] dimer is
Figure 12. Offset π-π stacking interactions found between 3-halopyridine
rings: (a) centrosymmetrically stacked; (b) translationally stacked. Colors
are as in Figure 2.
131.1°, 128.9°, Pt-Cl‚‚‚H 104.2°, 104.3°, 118.7°] and via offset
stacking interactions involving centrosymmetrically related
halopyridine rings of neighboring molecules (Figure 7; Table
9
5986 J. AM. CHEM. SOC. VOL. 127, NO. 16, 2005

Supramolecular Chemistry of Halogens
A R T I C L E S
much expanded from its starting geometry, with the shortest
(Pd)Cl‚‚‚F(C) separation being 4.622 Å, consistent with the
absence of such interactions in structures 1 and 5‚MeOH.
Electrostatic Potential Calculations on the Complexes
trans-[PdCl2(NC5H4X-3)2] (X ) F, Cl, Br, I). The electro-
static potential distribution was calculated for the series of
compounds trans-[PdCl2(NC5H4X-3)2] (X ) F, Cl, Br, I) using
averaged geometries close to those experimentally observed,
differing only in the C-X bond lengths between the set of four
compounds.
Clearly evident is the positive potential associated with the
organic halogens (X * F) (Figure 10). The potential decreases
in the sequence I > Br > Cl and becomes negative for fluorine,
as quantified along the direction of the C-X bond axis in Figure
11. This is consistent with the premise that placing the organic
halogen as a substituent on a coordinated pyridine ring provides
an enhancement of the electrophilicity of organic halogens
(Scheme 3) and highly pertinent to their propensity to form
M-Cl‚‚‚X-C (X * F) halogen bonds in the crystal structures
of the compounds reported herein.
M-Cl‚‚‚X-C interactions, the former as a Lewis base and
the latter as a Lewis acid. The attractive nature of the
M-Cl‚‚‚X-C interactions (X * F) can be assumed from the
fact that the electrostatic term will be attractive.5 This is evident
from the alignment of the region of electrostatic potential
minimum of the M-Cl group with the electropositive axial
region of the C-X groups. A repulsive electrostatic contribution
to putative M-Cl‚‚‚F-C interactions would result in a net
repulsive interaction and presumably is significant in accounting
for the absence of such interactions in the crystal structures of
1 and 5‚MeOH.
This chemical description of the M-Cl‚‚‚X-C interactions
is reinforced by the fact that normalized Cl‚‚‚X distances, RClX
,
decrease and the directionality of the interaction (linearity of
C-X‚‚‚Cl) increases with change of pyridine substituent X from
Cl to Br to I, consistent with the increased ability of the
C-X functional group to act as Lewis acid along the series of
halogens (Cl < Br < I). This decrease in Lewis acidity
(electrophilicity) arises from the diminishing positive electro-
static potential associated with the organic halogen (I > Br >
Cl; Figure 10a-c, Figure 11) and because of the analogous
progressive stabilization of the C-X σ* acceptor orbital
(Scheme 1, I). The trend in RClX values with X is also con-
sistent with our earlier observations involving charge-assisted
M-Cl‚‚‚X-C interactions.14a Thus, M-Cl‚‚‚X-C interactions
can be viewed as halogen bonds analogous to N‚‚‚X-C,
O‚‚‚X-C, and similar interactions and most probably stabilized
by similar attractive contributions (electrostatic and dispersion,
with small charge transfer and polarization components) coupled
with decreased repulsion for approaches along the C-X bond
axis.5 A full theoretical investigation of the nature of these
interactions is planned.
In this series of structures all halogens except fluorine
participate in M-Cl‚‚‚X-C halogen bonds. Furthermore, there
are no other halogen-halogen short contacts, though weak
hydrogen bonds of the type C-H‚‚‚Cl-M, and in some cases
C-H‚‚‚X-C, are observed (vide infra). This suggests that
M-Cl‚‚‚X-C halogen bonds can provide an effective synthon
for supramolecular synthesis. Indeed, to use the terminology
of Aakero¨y,28 the syntheses of 2, 3a, 3b, 4, 6, 7, and 8 provide
a 100% supramolecular yield based upon the M-Cl‚‚‚X-C
synthon (X ) I, Br, Cl), while structures of 1 and 5 (as
5‚MeOH) represent a 0% supramolecular yield for the putative
M-Cl‚‚‚F-C synthon. These synthons clearly merit further
investigation in the face of other competing synthons since even
some of the most widely used hydrogen-bonding synthons such
as carboxylic acid or amide dimers have quite low success rates
(33 and 8%, respectively) when considered in all crystal
structures, while these can rise to above 90% through choice
of appropriate molecular building blocks.30 However, their
success in these structures involving neutral building blocks
coupled with earlier evidence from analogous charge-assisted
interactions14,31 suggests that M-Cl‚‚‚X-C halogen bond
synthons could provide a highly valuable tool for supramolecular
assembly, particularly given the widespread accessibility of
halogenated inorganic and organic compounds.
Discussion
Supramolecular assembly typically involves a number of
different types of noncovalent interactions that facilitate mo-
lecular recognition and the assembly of supermolecules or
extended supramolecular structures. Discovering new interac-
tions and assessing the reliability of interactions and motifs in
terms of both geometric reproducibility and likelihood of
formation is a vital part of developing successful supramolec-
ular synthetic strategies. A successful strategy often takes
advantage of a correct assessment of the hierarchy of intermo-
lecular forces that dictate a given assembly. This concept has
been articulated in the context of hydrogen bonds by Etter27
and recently applied to some highly successful supramolecular
syntheses employing iso-nicotinamide by Aakero¨y and co-
workers.28 In the present study four classes of interactions have
been identified, namely, M-Cl‚‚‚X-C interactions, offset
π-stacking interactions between aromatic rings, C-H‚‚‚Cl-M
hydrogen bonds, and C-H‚‚‚X-C interactions. Their role in
the supramolecular assembly of the family of molecules is
discussed below.
Halogen Bonds. The most significant finding of this study
is that for all compounds, except the fluoropyridine systems in
1 and 5, their crystal structures contain networks within which
molecules are linked via M-Cl‚‚‚X-C interactions. The
geometries of these interactions are reliable and consistent (Table
2). Angles approaching linearity are observed at the organic
halide (C-X‚‚‚Cl), while much smaller obtuse angles are found
at the inorganic halide (M-Cl‚‚‚X).29 (M)Cl‚‚‚X(C) distances
are typically shorter than the sum of van der Waals radii for
the pair of halogens involved, i.e., RClX < 1.0.
These observations are strongly indicative that the inor-
ganic and organic halogens are serving different roles in the
(26) Bondi, A. J. J. Phys. Chem. 1964, 68, 441.
(27) Etter, M. C. Acc. Chem. Res. 1990, 23, 120.
(28) Aakero¨y, C. B.; Beatty, A. M.; Helfrich, B. A. J. Am. Chem. Soc. 2002,
124, 14425.
(29) This resembles the situation reported by Desiraju for C-X‚‚‚X′-C
interactions in which X * X′, wherein typically a near linear angle is
observed at the heavier (more electrophilic) halogen and a markedly bent
angle at the lighter (more nucleophilic) halogen. See: Pedireddi, V. R.;
Reddy, D. S.; Goud, B. S.; Craig, D. C.; Rae, A. D.; Desiraju, G. R. J.
Chem. Soc., Perkin Trans. 2 1994, 2353.
(30) Allen, F. H.; Motherwell, W. D. S.; Raithby, P. R.; Shields, G. R.; Taylor,
R. New J. Chem. 1999, 23, 25.
(31) Jones has reported a number of examples of C-X‚‚‚X- interactions: (a)
Freytag, M.; Jones, P. G. Ahrens, B.; Fischer, A. K. New J. Chem. 1999,
23, 1137. (b) Jones, P. G.; Vancea, F. CrystEngComm 2003, 303.
9
J. AM. CHEM. SOC. VOL. 127, NO. 16, 2005 5987

A R T I C L E S
Zordan et al.
Table 3. Inter-ring Geometries for Stacking Interactions
(viz., 125-140°) in all but three of the 21 interactions observed.
This suggests that C-H‚‚‚Cl-M hydrogen bonds, while im-
portant due to their shear numbers, probably play a supporting
role in the overall hierarchy of interactions given their readiness
to distort from optimum geometries.
interplanar
centroid-to-centroid
istance (Å)
offset angle
(deg)a
type of stacking
interactionb
distance (Å)
1
3.329
3.393
3.362
3.099
3.439
3.449
3.441
3.419
3.411
3.430
3.457
3.477
3.708
3.632
4.716
4.799
3.979
3.979
3.967
4.203
3.848
3.911
3.950
4.150
26.1
20.9
44.5
49.8
30.2
29.9
29.8
35.5
27.6
28.7
28.9
33.1
C
C
C
C
T
T
T
T
T
T
T
T
2
We finally turn to what is undoubtedly the weakest of the
four classes of noncovalent interactions identified in these crystal
structures, namely, the C-H‚‚‚X-C interactions. We have
previously shown that C-H‚‚‚X-C interactions show all the
geometric hallmarks of weak hydrogen bonds in terms of both
C-H‚‚‚X(C) and C-Cl‚‚‚H(C) angles,8a an assertion supported
in terms of C-H‚‚‚X(C) angle distribution by the earlier study
of Seddon and Aakero¨y,34 whereas Thallypally and Nangia
concluded35 that C-H‚‚‚X-C close contacts should generally
be considered as van der Waals interactions. In the present
family of structures there is some suggestion that dimerization
of neighboring molecules in certain directions is supported by
a cyclic R22(8) motif, most prominently illustrated in the
structure of 2 (Figure 13). Here geometries of the two
independent interactions are (C)H‚‚‚X(C) 2.759, 2.894 Å;
C-H‚‚‚X 163.9°, 132.1°; and C-Cl‚‚‚H 111.8°, 118.4°. The
former, which supports the Pt-Cl‚‚‚Cl-C interaction, has a
roughly linear geometry, makes an optimum angle at the
chlorine acceptor, and exhibits a (C)H‚‚‚X(C) distance some
0.19 Å shorter than the sum of hydrogen and chlorine van der
Waals radii. Tapes propagated by this cyclic R22(8) motif
involving either C-H‚‚‚Cl-C or C-H‚‚‚Br-C hydrogen bonds
have also been reported by Palmore and co-workers.36 Very
recently Desiraju and Howard have reignited the debate over
the potential importance of these types of weak interactions in
crystal engineering with a study of intramolecular O-H‚‚‚Cl-C
interactions.37
Polymorphism. Only one of the eight compounds studied
(trans-[PtCl2(NC5H4Br-3)2] (3)) was confirmed to be polymor-
phic. However, it is likely that polymorphism is more prevalent
at least in the nonfluorinated compounds given their similarity
in molecular shape and the fact that five structural arrangements
have been observed, viz., compound 2, isostructural compounds
3b and 8, compound 3a, isostructuctural compounds 6 and 7,
and compound 4. All molecular crystal structure arrangements
involve a compromise between efficient molecular packing38
and maximizing attractive over repulsive interactions. Thus, one
might anticipate that further exploration of crystallization
conditions may lead to alternative polymorphs for some of these
compounds that are consistent with the structural arrangements
already observed within this family.
3a
3b
4
5
6
7
8
a Angle between the centroid-centroid vector and the normal to one of
the pyridine ring planes (see ref 12). b C ) centrosymmetrically related
rings (see Figure 12b); T ) translationally related rings (see Figure 12a).
Stacking Interactions. Aromatic-aromatic or π-π stacking
interactions are well-known across the spectrum of molecules
containing aromatic rings. Such interactions can be rationalized
in terms of a set of rules proposed by Hunter and Sanders in
which the σ and π framework electrons are treated separately
wherein net attractive interactions arise where σ-π attractions
overcome π-π repulsions.32 Janiak has examined the stacking
interactions of pyridine and related ligands12 which are ap-
preciably less electron-rich than the prototypical benzene ring
and have a greater tendency to form offset parallel stacking
interactions which emphasize the σ-π attraction and benefit
from the electrostatic attraction between the permanent local
dipoles associated with such ligands. In all the structures
described in the present study, offset parallel stacking interac-
tions are present. Two distinct stacking geometries are found,
in which the stacked halopyridine rings from neighboring
molecules are related centrosymmetrically (Figure 12a) or by a
simple translation33 (Figure 12b). The former is noted by
Janiak to be a common configuration between pyridine lig-
ands,12 although the latter is more prevalent in the trans-
[PdCl2(NC5H4X-3)2] systems, occurring exclusively in com-
pounds 3-8. Geometries of the stacking interactions conform
well to typical values established by Janiak, with interplanar
separations all close to 3.4 Å and offset angles of 20-50° (Table
3). The prevalence of the translationally stacked interaction can
perhaps be attributed to its augmentation by a set of up to four
C-H‚‚‚Cl-M hydrogen bonds (Figure 12b) available for this
family of compounds.
C-H‚‚‚Cl-M and C-H‚‚‚X-C Interactions. That
C-H‚‚‚Cl-M interactions are viable hydrogen bonds, weaker,
but with similar though less sharply defined angular character-
istics as their stronger N-H‚‚‚Cl-M and O-H‚‚‚Cl-M
analogues, has been well established by a number of detailed
studies.8a,34,35 Thus, the occurrence of these hydrogen bonds
throughout these structures should be of no surprise. How-
ever, while the M-Cl‚‚‚H(C) acceptor angles lie overwhelm-
ingly within the preferred angular range (viz., 100-120°),
C-H‚‚‚Cl(M) angles are substantially displaced from linearity
Conclusions
This systematic study of the family of compounds trans-
[MCl2(NC5H4X-3)2] (M ) Pt, Pd; X ) I, Br, Cl, F) demonstrates
the importance of M-Cl‚‚‚X-C interactions where X * F. All
halogens participate in these interactions and in no other
halogen-halogen interactions, giving rise to 1D molecular tapes
or 2D layer structures. The roles of the inorganic halogen
(32) (a) Hunter, C. A.; Sanders, J. K. M. J. Am. Chem. Soc. 1990, 112, 5525.
(b) Hunter, C. A. Chem. Soc. ReV. 1994, 23, 101.
(36) (a) McBride, M. T.; Luo, T.-J. M.; Palmore, G. T. Cryst. Growth Des.
2001, 1, 39. (b) Luo, T.-J. M.; Palmore, G. T. Cryst. Growth Des. 2002, 2,
337.
(33) Note that the translation between rings is not normal to the ring planes but
rather along the centroid-centroid vector, which is inclined to the normal
of ring plane(s) at the offset angle (see Table 2).
(37) Banerjee, R.; Desiraju, G. R.; Mondal, R.; Howard, J. A. K. Chem. Eur. J.
2004, 10, 3373.
(34) Aakero¨y, C. B.; Evans, T. A.; Seddon, K. R.; Pa´linko´, I. New J. Chem.
1999, 23, 145.
(35) Thallypally, P. K.; Nangia, A. CrystEngComm 2001, 3, 114
(38) (a) Kitaigorodskii, A. I. Molecular Crystals and Molecules; Academic
Press: New York, 1973. (b) Pidcock, E.; Motherwell, W. D. S. Chem.
Commun. 2003, 3028.
9
5988 J. AM. CHEM. SOC. VOL. 127, NO. 16, 2005

Supramolecular Chemistry of Halogens
A R T I C L E S
(M-Cl) and the organic halogen (C-X) are distinct and
different, as evidenced by the C-X‚‚‚Cl and M-Cl‚‚‚X angles,
which consistently approach linearity at the organic halogen
while being markedly angular (ca. 120-130°) at the inorganic
chloride. The observed geometries are supported by DFT
calculations using model dimers and suggest an attractive
M-Cl‚‚‚X-C interaction within which the organic and inor-
ganic halogens function as electrophile and nucleophile, re-
spectively. This supposition is reinforced by the reduction in
interaction distance (RClX) as the pyridine substituent X de-
scends the halogen group and by the complete absence of
M-Cl‚‚‚F-C interactions. Thus, M-Cl‚‚‚X-C interactions
should clearly be classified as halogen bonds alongside
N‚‚‚X-C, O‚‚‚X-C, and other similar cases. The potential
efficacy of M-Cl‚‚‚X-C halogen bonds is highlighted by their
consistent presence and geometry in the presence of three other
identified categories of noncovalent interactions, namely, offset
π-π stacking, C-H‚‚‚Cl-M hydrogen bonds, and in some
cases C-H‚‚‚X-C hydrogen bonds, which collectively con-
tribute to the overall 3D structures observed.
Acknowledgment. L.B. and F.Z. are grateful to Harry Adams
(Sheffield) for helpful discussions regarding the crystal structure
determinations and the EPSRC (GR/R68733/01) for support of
this work.
Supporting Information Available: X-ray crystallographic
files in CIF format for compounds 1-8. Experimental and
calculated powder patterns for compounds 1-8. DFT optimized
geometries of dimers of trans-[PdCl2(NH3)(NC5H4X-3)]
model complexes (X ) I, Br, Cl, F). Geometries used in
DFT calculations of electrostatic potentials of trans-[PdCl2-
(NC5H4X-3)2] (X ) I, Br, Cl, F). This material is available free
of charge via the Internet at http://pubs.acs.org.
JA0435182
9
J. AM. CHEM. SOC. VOL. 127, NO. 16, 2005 5989