Arene-mercury complexes stabilized aluminum and gallium chloride: Synthesis and structural characterization

Article abstract of DOI:10.1021/ja011003u

Reaction of HgCl₂ with 2 equiv of MCl3 in an aromatic solvent yields Hg(arene)₂(MCl₄)₂ where, arene = C₆H₅Me, M = Al (1), Ga (2); arene = C₆H₅Et, M = Al (3) and Ga (4); o-C₆H₄Me₂, M = Al (5), Ga (6); C₆H₃-1,2,3-Me₃, M = Al (7) and Ga (8). The solid-state structures of compounds 1-5 and 7 have been determined by X-ray crystallography. In the solid state, compounds 1-4 and 7 exist as neutral complexes in which two arenes are bound to the mercury, and the MCl₃ groups are bound through bridging chlorides to the mercury; compound 5 exists as a cation-anion pair [Hg(o-C₆H₄Me₂)2(A1C14)][AlCl₄]. However, in solution compounds 1-8 all exist as neutral complexes. The structures of Hg(arene)₂(AlCl₄)₂ and [Hg(arene)₂(AlCl₄)]⁺ have been determined by DFT calculations {B3LYP level} to facilitate the assignment of the ¹³C CPMAS NMR spectra and are in good agreement with the X-ray diffraction structures of compounds 1 and 5. Reaction of HgCl₂ with MCl₃ in benzene, m-xylene, and p-xylene results in the formation of liquid clathrates whose spectroscopic characterization is consistent with ionic structures, [Hg(arene)₂(MCl₄)][MCl₄]. The calculated energy difference between Hg(C₆H₅Me)₂(AlCl₄)₂ and [Hg(C₆H₅Me)2(A1C14)] [AlCl₄] is discussed with respect to the structure of compound 5 in the solid state versus solution state and the proposed speciation in the liquid clathrates.

Full text of DOI:10.1021/ja011003u

J. Am. Chem. Soc. 2001, 123, 11219-11228
11219
Arene-Mercury Complexes Stabilized Aluminum and Gallium
Chloride: Synthesis and Structural Characterization
Alexander S. Borovik,1a Simon G. Bott,1b and Andrew R. Barron*,1a
Contribution from the Department of Chemistry and Center for Nanoscale Science and Technology,
Rice UniVersity, Houston, Texas 77005, and Department of Chemistry, UniVersity of Houston,
Houston, Texas 77204
ReceiVed April 20, 2001
Abstract: Reaction of HgCl2 with 2 equiv of MCl3 in an aromatic solvent yields Hg(arene)2(MCl4)2 where,
arene ) C6H5Me, M ) Al (1), Ga (2); arene ) C6H5Et, M ) Al (3) and Ga (4); o-C6H4Me2, M ) Al (5), Ga
(6); C6H3-1,2,3-Me3, M ) Al (7) and Ga (8). The solid-state structures of compounds 1-5 and 7 have been
determined by X-ray crystallography. In the solid state, compounds 1-4 and 7 exist as neutral complexes in
which two arenes are bound to the mercury, and the MCl3 groups are bound through bridging chlorides to the
mercury; compound 5 exists as a cation-anion pair [Hg(o-C6H4Me2)2(AlCl4)][AlCl4]. However, in solution
compounds 1-8 all exist as neutral complexes. The structures of Hg(arene)2(AlCl4)2 and [Hg(arene)2(AlCl4)]+
have been determined by DFT calculations {B3LYP level} to facilitate the assignment of the 13C CPMAS
NMR spectra and are in good agreement with the X-ray diffraction structures of compounds 1 and 5. Reaction
of HgCl2 with MCl3 in benzene, m-xylene, and p-xylene results in the formation of liquid clathrates whose
spectroscopic characterization is consistent with ionic structures, [Hg(arene)2(MCl4)][MCl4]. The calculated
energy difference between Hg(C6H5Me)2(AlCl4)2 and [Hg(C6H5Me)2(AlCl4)][AlCl4] is discussed with respect
to the structure of compound 5 in the solid state versus solution state and the proposed speciation in the liquid
clathrates.
Introduction
Group 13 halides are well-known as catalysts for the Friedel-
Crafts alkylation and acylation of aromatic hydrocarbons. The
highly Lewis acidic group 13 halide activates the alkyl or acyl
halide, via either complexation or ionization (e.g., eq 1),2 by
placing a positive charge on the â-substituent (Figure 1a).
RCl + AlCl3 h RCl‚‚‚AlCl3 h [R]+[AlCl4]-
(1)
The increase of positive charge on the â-substituent is a general
effect of the coordination of aluminum Lewis acids to both
organic (Figure 1b) and inorganic (Figure 1c) carbonyls, for
example, activation of ketones to alkylation and reduction3 and
activation of transition metal carbonyl ligands toward the methyl
migration.4 We have recently observed that the coordination of
an alcohol to an aluminum Lewis acid results in an increase of
the pKa of the alcohol O-H by at least 7 units.5 Although
* To whom correspondence should be addressed. Web site: http://
www.rice.edu/barron.
(1) (a) Department of Chemistry, Rice University. (b) Department of
Chemistry, University of Houston.
Figure 1. Schematic representation of the “activation” of organic and
inorganic substrates by aluminum Lewis acids.
(2) (a) Whitmore, F. C. J. Am. Chem. Soc. 1932, 54, 3274. (b) Olah, G.
A.; Kuhn, S. J.; Flood, S. H. J. Am. Chem. Soc. 1962, 84, 1688. (c) Olah,
G. A.; Kobayashi, S.; Tashiro, M. J. Am. Chem. Soc. 1972, 94, 7448.
(3) (a) Power, M. B.; Nash, J. R.; Healy, M. D.; Barron, A. R.
Organometallics 1992, 11, 1830. (b) Power, M. B.; Bott, S. G.; Atwood, J.
L.; Barron, A. R. J. Am. Chem. Soc. 1990, 112, 3446. (c) Power, M. B.;
Bott, S. G.; Clark, D. L.; Atwood, J. L.; Barron, A. R. Organometallics
1990, 9, 3086 and references therein.
(4) (a) Butts, S. B.; Holt, E. M.; Strauss, S. H.; Alcock, N. W.; Stimson,
R. E.; Shriver, D. F. J. Am. Chem. Soc. 1979, 101, 5864. (b) Butts, S. Beda;
Strauss, S. H.; Holt, E. M.; Stimson, R. E.; Alcock, N. W.; Shriver, D. F.
J. Am. Chem. Soc. 1980, 102, 5093.
seemingly unrelated, this activation may also be considered as
an example of placing an increased positive charge on the
â-substituent, the alcohol hydrogen (Figure 1d), hence increasing
its electrophilicity and making it more susceptible to reaction
with nucleophiles such as aluminum alkyls.
There has been an increased interest in the development of
new Lewis acidic compounds as catalysts and cocatalysts,
especially with regard to olefin polymerization. Most of these
efforts have focused on using electron-withdrawing substituents,
low coordination numbers, or multiple centers.6 As an alternative
approach, applying the concept of increasing the Lewis acidity
(5) McMahon, C. N.; Bott, S. G.; Barron, A. R. J. Chem. Soc., Dalton
Trans. 1997, 3129.
10.1021/ja011003u CCC: $20.00 © 2001 American Chemical Society
Published on Web 10/20/2001

11220 J. Am. Chem. Soc., Vol. 123, No. 45, 2001
BoroVik et al.
Table 1. Selected Bond Lengths (Å) and Angles (deg) in Hg(arene)2(MCl4)2
Al
C6H5Me
(1)
Ga
C6H5Me
(2)a
Al
C6H5Et
(3)
Ga
C6H5Et
(4)
Al
o-C6H4Me2
(5)
Al
C6H3-1,2,3-Me3
(7)
M
arene
Hg-C
2.32(1)
2.72(1)
2.677(2)
2.176(4)
2.102(5)-2.118(3)
81.1(1)
2.349(9)
2.71(1)
2.652(2)
2.239(2)
2.144(3)-2.166(3)
82.4(1)
2.30(1)
2.72(2)
2.648(4)
2.184(5)
2.080(6)-2.169(5)
84.5(6)
2.33(1)
2.75(2)
2.634(4)
2.230(4)
2.128(4)-2.155(4)
85.6(2)
2.27(2), 2.40(2)
2.64(2), 2.74(2)
2.761(3), 2.768(3)
2.166(4), 2.156(5)
2.09(1)-2.169(5)
75.2(1)
2.405(9), 2.44(1)
2.45(1), 2.46(1)
2.661(2), 2.758(3)
2.181(4), 2.178(4)
2.103(5)-2.124(4)
89.90(8)
Hg-Cl
M-Clbr
M-Clter
Cl-Hg-Cl
C-Hg-C
Hg-Cl-M
135.1(5)
110.2(2)
131.3(6)
110.6(1)
129(1)
112.3(2)
126(1)
111.4(2)
125.5(6)
90.8(1), 91.2(1)
140.0(4)
106.1(1), 107.9(1)
a Borovik, A. S.; Bott, S. G.; Barron, A. R. Angew. Chem., Int. Ed. 2000, 39, 4117.
of metals through their “activation” by another Lewis acid
(Figure 1e) has drawn our interest. In this regard we note that
aluminum halides have been previously employed as activators
for transition metals through a similar complexation (see Figure
1f).7
and co-workers with a highly electronegative trifluoroacetate
ligand.16 Based on the possibility that group 13 halide Lewis
acids could “activate” other weaker Lewis acids, we have
investigated the effect of AlCl3 and GaCl3 on the stability of
Hg‚‚‚arene complexes.
Although the Lewis acid behavior of mercuracarboranes has
been extensively studied by Hawthorne and co-workers,8 the
Lewis acidic nature of group 12 halides, in particular those of
mercury, has been studied much less than that of the group 13
halides.9 However, the chemistry of mercury(II) salts with
aromatic hydrocarbons is well developed regarding electrophilic
attack on aromatic compounds (aromatic mercuration), and Olah
et al. have shown that Hg-arene complexes are involved as
intermediates.10 Kochi and co-workers have shown that the
activation of the arene is related to a charge-transfer transition
in the π-complex.11,12 Crabtree and co-workers have proposed
a π-complex as a key intermediate in a variety of photochemical
C-C bond forming reactions,13 while the characterization of a
series of Hg(I)-arene complexes has been reported.14 Mercury-
(II)-arene complexes are well-established as important inter-
mediates, however, simple complexes have only been charac-
terized spectroscopically.15 The only structural characterization
of a nonsolvate mercury(II) complex was reported by Kochi
Results and Discussion
The reaction of HgCl2 with 2 mol equiv of MCl3 (M ) Al,
Ga) in a substituted aromatic solvent (C6H6-xRx) yields a colored
solution (yellow to orange, depending on the arene), from which
crystalline material may be obtained in moderate to high yield
(eq 2) for C6H5Me, M ) Al (1), Ga (2);17 C6H5Et, M ) Al (3),
Ga (4); o-C6H4Me2, M ) Al (5), Ga (6); C6H3-1,2,3-Me3, M )
Al (7), Ga (8).
HgCl2 + MCl3 f Hg(arene)2(MCl4)2
(2)
In contrast, reaction of HgCl2 with 2 mol equiv of AlCl3 in
benzene, m-xylene, p-xylene yields liquid clathrates, see below.
Compounds 1-8 are stable when exposed to O2, CO, and
CO2, but decompose upon irradiation by ambient light, exposure
to moisture, or being subjected to chlorinated or coordinating
solvents (e.g., CHCl3, THF, Et2O and MeCN). The solubility
of each compound in its respective solvent (i.e., compound 1
in toluene) suggests a simple Lewis acid-base complex of the
metal halides rather than a cation/anion pair, vide infra. This is
supported by solution conductivity measurements that indicate
neutral compounds in solution for compounds 1-8. Although
elemental analysis is consistent with the given formulas and
the EI mass spectrum confirms the presence of the arene (see
(6) See, for example: (a) Radzewich, C. E.; Guzei, I. A.; Jordan, R. F.
J. Am. Chem. Soc. 1999, 121, 8673. (b) Korolev, A. V.; Guzei, I. A.; Jordan,
R. F. J. Am. Chem. Soc. 1999, 121, 11606. (c) Dagorne, S.; Guzei, I. A.;
Coles, M. P.; Jordan, R. F. J. Am. Chem. Soc. 2000, 122, 274. (d) Munoz-
Hernandez, M.; Keizer, T. S.; Parkin, S.; Patrick, B.; Atwood, D. A.
Organometallics 2000, 19, 4416. (e) Atwood, D. A.; Jegier, J. J. Chem.
Soc., Chem. Commun. 1996, 1507. (f) Nelson, S. G.; Kim, B.; Peelen, T.
J. J. Am. Chem. Soc. 2000, 122, 9318. (g) Wuest, J. D., Acc. Chem. Res.
1999, 32, 81, and references therein.
(7) (a) Le N.; Jean P.; Youinou, M. T.; Osborn, J. A. Organometallics
1992, 11, 2413. (b) Le N.; Jean P.; Osborn, J. A. Organometallics 1991,
10, 1546.
(8) See, for example: (a) Hawthorne, M. F.; Zheng, Z. Acc. Chem. Res.
1997, 30, 267. (b) Hawthorne, M. F.; Yang, X.; Zheng, Z. Pure Appl. Chem.
1994, 66, 245. (c) Lee, H.; Diaz, M.; Knobler, C. B.; Hawthorne, M. F.
Angew. Chem., Int. Ed. 2000, 39, 776. (d) Lee, H.; Diaz, M.; Hawthorne,
M. F. Tetrahedron Lett. 1999, 40, 7651. (e) Badr, I. H. A.; Diaz, M.;
Hawthorne, M. F.; Bachas, L. G. Anal. Chem. 1999, 71, 1371 and references
therein.
(9) (a) The Chemistry of Mercury; McAuliffe, C. A., Ed.; Macmillan:
Toronto, 1977. (b) Dean, P. A. W. Prog. Inorg. Chem. 1978, 24, 109. (c)
Tschinkl, M.; Schier, A.; Riede, J.; Gabba¨ı, F. P. Organometallics 1999,
18, 2040. (d) Saito, S.; Zhang, J.; Koizumi, T. J. Org. Chem. 1998, 63,
6029. (e) Zhuang, R.; Mueller, A. H. E. Macromolecules 1995, 28, 8043.
(f) Persson, I.; Sandstroem, M.; Goggin, P. L. Inorg. Chim. Acta 1987,
129, 183. (g) Wuest, J. D.; Zacharie, B. Organometallics 1985, 4, 410 and
references therein.
Experimental Section), the solution H and 13C NMR do not
1
allow for structural determination due to the H/D exchange with
other aromatic solvents, for example, C6D6.17,18 However, the
solid-state structures of compounds 1-5 and 7 have been
determined by X-ray crystallography.
X-ray Crystallography. The molecular structures of com-
pounds 1, 3, 5, and 7 are shown in Figures 2-5, respectively.
Compounds 217 and 4 are isomorphous with their aluminum
analogues.19 Selected bond lengths and angles for compounds
1-5 and 7 are given in Table 1. The solid-state structures of
Hg(arene)2(MCl4)2 appear to fall into two general categories:
(16) W. Lau, J. C. Huffman, and J. K. Kochi, J. Am. Chem. Soc. 1982,
104, 5515.
(10) See, for example: Olah, G. A.; Yu, S. H.; Parker, D. G. J. Org.
Chem. 1976, 41, 1983 and references therein.
(11) Lau, W.; Kochi, J. K. J. Am. Chem. Soc. 1986, 108, 6720.
(12) Fukuzumi, S.; Kochi, J. K. J. Org. Chem. 1981, 46, 4116.
(13) Fowley, L. A.; Less, J. C., Jr.; Crabtree, R. H.; Siegbahn, P. E. M.
J. Organomet. Chem. 1995, 505, 57.
(14) Frank, W.; Dincher, B., Z. Naturforsch. 1987, 42b, 828.
(15) (a) Damude, L. C.; Dean, P. A. W. J. Organomet. Chem. 1979,
181, 1. (b) Damude, L. C.; Dean, P. A. W.; Sefcik, M. D.; Schaefer, J. J.
Organomet. Chem. 1982, 226, 105.
(17) We have previously reported the isolation and structural character-
ization of [Hg(C6H5Me)2(GaCl4)2]: Borovik, A. S.; Bott, S. G.; Barron, A.
R. Angew. Chem., Int. Ed. 2000, 39, 4117.
(18) Borovik, A. S.; Bott, S. G.; Barron, A. R. ACS National Meeting,
American Chemical Society: Washington, DC, August, 2000.
(19) Although the molecular structures of compounds 6 and 8 have not
been determined by X-ray crystallography due to twinning issues, the
observation of a “double cell” as compared to their aluminum analogues
(along with their spectroscopic characterization) suggests that compounds
6 and 8 are isostructural to compounds 5 and 7, respectively.

Arene-Mercury Complexes
J. Am. Chem. Soc., Vol. 123, No. 45, 2001 11221
Figure 2. Molecular structure of Hg(C6H5Me)2(AlCl4)2 (1). Thermal
ellipsoids shown at the 30% level, and hydrogen atoms are omitted for
clarity.
Figure 5. Molecular structure of Hg(C6H3-1,2,3-Me2)2(AlCl4)2 (7).
Thermal ellipsoids shown at the 30% level, and hydrogen atoms are
omitted for clarity.
2, 3, and 5). In contrast, the o-xylenes in compound 5 are in a
meso orientation (see Figure 4).
As may be seen from Table 1, each arene in compounds 1-5
is bound in a highly asymmetric η2 manner with the Hg-C
bond para to the arene’s methyl substituent being shorter
(0.34-0.42 Å) than the bond in the meta position. The calcu-
lated structures for the neutral complexes, Hg(C6H6)2(AlCl4)2
and Hg(C6H5Me)2(AlCl4)2, and the presence of a single 13C
CPMAS NMR resonance, per arene ring, assignable to Hg-C
for compounds 1-6 are consistent with η1 rather than η2
coordination of the arene (see below). The coordination of the
two C6H3-1,2,3-Me3 ligands in compound 7 appears to be closer
to η2 coordination, see Table 1. This is confirmed by 13C
CPMAS NMR spectroscopy, see below. It is noteworthy that
the crystal structure of compound 7 is the only one of those
discussed herein with arene‚‚‚arene π-stacking; the intermo-
lecular distance of 3.31 Å is less than the sum of the van der
Waal radii of the arene rings (3.7 Å).20
Figure 3. Molecular structure of Hg(C6H5Et)2(AlCl4)2 (3). Thermal
ellipsoids shown at the 20% level, and hydrogen atoms are omitted for
clarity.
It should be noted that the Hg-C distances in compounds
1-5 are on either side of the values reported by Kochi and
co-workers for [Hg2(µ-O2CCF3)4(η2-C6Me6)2] (2.56 and 2.58
Å)12 and are significantly shorter than those observed for the
intramolecular Hg‚‚‚arene interactions (ca. 3.2 Å).21 Further-
more, the Hg-C distances in compounds 1-5 are shorter than
those observed for the Hg(I)-arene complex, [Hg2(C6M6)]-
[AlCl4]2,14 [2.43(3) and 2.41(3) Å] in which the arene is a
stronger π-donor. It is also worth noting that a typical Hg-C
σ-bond is 2.1-2.2 Å22 in length which is only slightly shorter
than the shortest Hg-C interactions in the compounds discussed
herein, see Table 1. Although the Kochi complex was the first
crystallographically characterized mercury-arene π-complex,
several examples had been spectroscopically characterized,23,24
and Tsunoda and Gabba¨ı have recently reported the structural
characterization of a benzene “solvate” supramolecule that
contains a µ6-η2:η2:η2:η2:η2:η2 benzene sandwiched between six
[Hg3(C6F4)3] units.25
Figure 4. Molecular structure of the [Hg(C6H4Me2)2(AlCl4)] cation
(5). Thermal ellipsoids shown at the 30% level, and hydrogen atoms
are omitted for clarity.
(20) Handbook of Chemistry and Physics, 60th ed.; CRC Press: Boca
Raton, Florida, 1980; p D-194.
(21) (a) Lampe, P. A.; Moore, P. Inorg. Chim. Acta 1979, 36, 27. (b)
Cathy, A. J.; Chaichit, N.; Gatehouse, B. M. Acta Crystallogr., Sect. B 1980,
36, 786.
(22) Kamenar, B.; Penavic, M. Inorg. Chim. Acta 1972, 6, 191.
(23) Eliezer, I.; Avinur, P. J. Chem. Phys. 1971, 55, 2300.
(24) Vezzosi, I. M.; Peyronel, G.; Zanoli, A. F. Inorg. Chim. Acta 1974,
8, 229.
neutral (C6H5Me, C6H5Et, and C6H3-1,2,3-Me3) and ionic (o-
C6H4Me2). Irrespective of the overall charge of the mercury
complex, the two arene ligands are π-bound to the mercury
center (see Figures 2-5). The substituted arene ligands are
oriented in a rac manner in compounds 1-4 and 7 (cf., Figures
(25) Tsunoda, M.; Gabba¨ı, F. P. J. Am. Chem. Soc. 2000, 122, 8335.

11222 J. Am. Chem. Soc., Vol. 123, No. 45, 2001
BoroVik et al.
Scheme 1. Schematic Representation of the Potential
Interconversion from Neutral [Hg(arene)2(AlCl4)2] (cf, 1 and
3) to Cationic [Hg(arene)2(AlCl4)2]+ (cf., 5) via an
Asymmetrical Complex (cf, 7)
Figure 6. View of the close cation/anion interaction in [Hg(C6H4-
Me2)2(AlCl4)][AlCl4]. Thermal ellipsoids shown at the 30% level, and
hydrogen atoms are omitted for clarity.
The X-ray structure of compound 5 shows it to be a cation/
anion pair in the solid state, that is, [Hg(C6H4Me2)2(AlCl4)]-
[AlCl4], in which the coordinated AlCl4 moiety is bound λ2 to
the mercury (Figure 4), rather than λ1 as observed in compounds
1-4 and 7. Despite the presence of a single chelating [AlCl4]-
anion, the geometry and bond distances about mercury in
compound 5 are similar to those in [Hg(C6H5R)2(MCl4)2] (see
Table 1). A weak interaction (along the crystallographic a-axis)
between the cation and anion in adjacent chains (Hg‚‚‚Cl )
3.29 Å) is close to the sum of the van der Waal radii, and the
geometry about the mercury is distorted due to this weak
interaction, see Figure 6.
A consideration of the orientation of the MCl3 unit (Figures
2, 3, and 5) shows that the longer terminal M-Cl is oriented
toward the mercury. The resulting Hg‚‚‚Cl distances (3.56-
3.83 Å) are outside what would ordinarily be considered a
bonding interaction; however, the effect on the M-Cl distance
and the orientation of the MCl3 unit suggest that there is an
electrostatic interaction between the chloride and the mercury.
While this long-range interaction is symmetrical in compounds
1-4 due to their C2 axis symmetry, a distinct asymmetry is
observed for compound 7 [Hg(1)‚‚‚Cl(11) ) 3.57 Å, Hg(1)‚‚‚
Cl(23) ) 3.68 Å]. Upon a comparison of the Hg‚‚‚Cl-Al
interactions in compounds 1, 7, and 5 we can propose that these
structures are part of a continuum between neutral (covalent)
and ionic structures (Scheme 1). This observation offers the
question: is the interconVersion of neutral and ionic structures
facile?
Table 2. Experimental and Calculated (DFT) 13C NMR Spectra
calcd shift
δ (ppm)
exptl shift
δ (ppm)
compound
atoma
Hg(C6H5Me)2(AlCl4)2
C(1)
C(2)
C(3)
C(4)
C(5)
C(6)
C(11)
C(11)
C(12)
C(13)
C(14)
C(15)
C(16)
C(111)
C(112)
155.7
132.5
135.6
104.1
138.0
132.4
25.5
156.8
146.7
138.9
102.4
130.3
134.1
24.7
159.4
136.6
140.1
100.7
143.1
136.6
24.0
158.4
147.7
139.0
105.5
135.6
139.0
22.4
(1)
[Hg(C6H4Me2)2(AlCl4)]+
(5)
23.5
22.4
a See Figures 2 and 4 for atom numbering scheme.
for compounds 1 and 5. It is interesting to note that the number
of aromatic/aliphatic resonances correlates well with the crystal-
lographic symmetry. For example, the toluene molecules in
compound 1 are related by crystallographic C2 symmetry, and
the 13C CPMAS NMR spectrum shows a single CH3 resonance
and four CH resonances (one of which is due to two overlapping
resonances). In contrast, the 13C CPMAS NMR spectrum for
compound 7 shows two sets of resonances for magnetically
distinct C6H3-1,2,3-Me3 ligands (see Figure 7a) consistent with
the two crystallographically independent ligands per mercury
center (see Figure 5). A dipolar dephasing experiment was
performed (total dephasing delay ) 50 ms) to identify the
methyl-substituted aromatic carbon signals. In the case of
compound 7, six of the original aromatic resonances are
suppressed as a result of the dipolar dephasing experiment (see
Figure 7b).
NMR and UV-Visible Spectroscopy. As noted above we
1
are unable to obtain solution H and 13C NMR spectra for
compounds 1-8 due to a facile H/D exchange. Dissolution of
compounds 1-8 in C6D6 results in H/D exchange and the
formation of C6D5H and the appropriate substituted arene, that
is, C6D6-xMex.26 Unfortunately, Hg(arene)2(MCl4)2 compounds
are insoluble in fluorinated solvents (i.e., Freon), and adverse
reactions occur in CS2 or chlorinated solvents.
Solid-state 13C CPMAS NMR spectra have been obtained for
compounds 1-8, see Experimental Section. Peak assignments
were obtained by a combination of dipolar dephasing experi-
ments27 and comparison with calculated (DFT) chemical shifts
(see Experimental Section). Table 2 gives a comparative
example of the calculated and experimental 13C NMR shifts
A common observation for the 13C CPMAS spectra of
compounds 1-6 is that all but one of the aromatic C-H signals
are unusually deshielded and the other is exceptionally shielded
(δ 100-106 ppm). A similar shielding effect was observed for
(26) This exchange reaction is explored in detail elsewhere.
(27) Alemany, L. B.; Grant, D. M.; Alger, T. D.; Pugmire, R. J. J. Am.
Chem. Soc., 1983, 105, 6697.

Arene-Mercury Complexes
J. Am. Chem. Soc., Vol. 123, No. 45, 2001 11223
a
Table 3. UV-Visible Spectra of Hg(arene)2(MCl4)2
M
arene
λ (nm)
ꢀ (mol-1cm-1
)
notes
Al
Al
Al
C6H6
C6H5Me
C6H5Et
285
325
288
297
337
331
310
301
315
279
305
289
298
334
332
301
320
2000
4500
1300
1300
4300
6500
7000
2500
2000
5000
7000
5300
5600
6500
4000
2400
2300
b
c
Al
Al
Al
Al
o-C6H4Me2
m-C6H4Me2
p-C6H4Me2
b
b
c
C6H3-1,2,3-Me3
Ga
Ga
Ga
C6H6
C6H5Me
C6H5Et
b
c
Ga
Ga
Ga
o-C6H4Me2
p-C6H4Me2
C6H3-1,2,3-Me3
b
c
Figure 7. 13C CPMAS NMR spectra of Hg(C6H3-1,2,3-Me3)2(AlCl4)2
(7) showing the presence of magnetically inequivalent arene ligands
(a). The identity of the methyl-substituted aromatic carbon signals is
enabled by a dipolar dephasing experiment (b) in which the aromatic
CH resonances are suppressed.
a [Hg(O2CCF3)4(C6H6-n)Men)2], λ ) 288-315 nm. b Upper layer of
liquid clathrate, see text. c Two well resolved peaks.
compounds 5 and 7 may be predicted to be isostructural. Third,
13C CPMAS NMR may be used as a structural probe for these
complexes.
We have been unable to obtain satisfactory 199Hg solution
NMR for any of the compounds and we were able to obtain a
solid-state 199Hg MAS NMR (35.84 MHz) spectra only for Hg-
(C6H5Me)2(GaCl4)2 (2).17 However, it is interesting that the
observed chemical shift (δ -1970) is downfield from that
reported for HgCl2 (δ -1497 ppm)28 and closer to that of Hg2+
aqueous salts (δ -2253 to -2361 ppm)29 than simple Lewis
base complexes, for example, [HgCl2{P(nBu)3}2] (δ -404
ppm).30 Thus, 199Hg NMR spectroscopy is consistent with a
highly electropositive mercury center.31
The solid-state 27Al MAS NMR spectrum of compound 5
shows a single resonance with two maxima,32 while the solution
27Al NMR spectrum shows a single complex resonance at 105
ppm. Similar complex spectra are observed for compounds 1
Figure 8. Plot of the crystallographic Hg-C bond distance (with esd’s)
versus the 13C CPMAS NMR chemical shift for the aromatic carbons
with the closest Hg‚‚‚C contacts in Hg(arene)2(MCl4)2.
and 7, see Experimental Section. The chemical shifts of the 27
-
Al NMR resonances are similar to those previously reported
for the [AlCl4]- anion.30
the C6HMe5 and 1,2,4,5-C6H2Me4 complexes of Hg(SbF6)2.15b
Unfortunately, no detectable J(199Hg-13C) multiplets are evident
that might help identify the signals of the carbons coordinated
to the mercury. However, based on DFT calculations (Table
2), the most shielded aromatic C-H signal in each spectrum
can be assigned to the carbon bound to mercury. The presence
of a single Hg-C resonance for compounds 1-6 is consistent
with η1 rather than η2 coordination of the arene. In contrast,
the 13C CPMAS spectra of compounds 7 and 8 show four
resonances associated with Hg‚‚‚C interactions (e.g., see Figure
7a). These resonances are in the range 112-125 ppm, signifi-
cantly removed from the range observed for compounds 1-6.
This difference may be rationalized by a consideration of the
relative Hg‚‚‚C distances in the X-ray structure of compound 7
in comparison to those in compounds 1-5. Figure 8 shows a
plot of Hg‚‚‚C distance versus the 13C CPMAS chemical shifts
for the two closest arene carbon atoms in compounds 1-5 and
7. Clearly, a correlation between Hg‚‚‚C distance versus the
13C chemical shift exists. Three observations may be made from
the data in Figure 8. First, compounds 1-5 are indeed best
considered to have η1 coordination of the arene in the solid
state, whereas coordination in compound 7 is best described as
η2. Second, given the similarity of the 13C CPMAS NMR
chemical shift for compounds 5 and 6, as well as compounds 7
and 8, the solid-state structures of the gallium analogues of
The UV-visible spectra of compounds 1-8 (as well as those
of the upper layer of the liquid clathrates described below) are
given in Table 3. It is interesting to note that the UV-visible
spectrum of compound 5 follows the trend for compounds 1,
3, and 7, suggesting that in solution, compound 5 exists as a
neutral complex in o-xylene solution. Compounds 3, 4, 7, and
8 all show two well-resolved peaks with similar molar absorp-
tivities; however, it is unclear whether this is due to the lowering
of symmetry or the presence of two isomers in solution. A
comparison of the UV-visible spectra for the gallium com-
pounds with their aluminum analogues shows that the absorp-
tions are generally similar. On the basis of DFT calculations
the UV-visible absorption is found to be due to an areneπ f
Hgs charge transfer, see below.
(28) Sens, M. A.; Wilson, N. K.; Ellis, P. D.; Odom, J. D. J. Magn.
Reson. 1974, 15, 191.
(29) (a) Maciel, G. E.; Borzo, M. J. Magn. Reson. 1973, 10, 388. (b)
Kru¨ger, H.; Lutz, O.; Nolle, A.; Schwenk, A. Z. Physik. 1975, A273, 325.
(30) Kidd, R. G.; Goodfellow, R. J. In NMR and the Periodic Table;
Harris, R. K., Mann, B. E., Eds.; Academic Press: New York, 1978; Chapter
8, p 195.
(31) Petrosyan, V. S.; Reutov, O. A. J. Organomet. Chem. 1974, 76,
123.
(32) (a) Alemany, L. B. Appl. Magn. Reson. 1993, 4, 179. (b) Smith,
M. E.; van Eck, E. R. H. Prog. Nucl. Magn. Reson. Spectrosc. 1999, 34,
159.

11224 J. Am. Chem. Soc., Vol. 123, No. 45, 2001
BoroVik et al.
Table 4. Calculated Bond Lengths (Å) and Angles (deg) in Hg(arene)2(MCl4)2 and [Hg(arene)2(AlCl4)]+
Hg(arene)2(MCl4)2
[Hg(arene)2(AlCl4)]+
calcd
C6H5Me
calcd
exptl
2.32(1)
2.72(1)
2.677(2)
2.176(4)
2.102(5)-2.118(3)
81.1(1)
exptl
C6H6
2.402
2.900
2.463
2.393
C6H5Me
C6H6
o-C6H4Me2
Hg-C
2.382
2.863
2.471
2.371
2.782
2.466
2.364
2.181
85.24
110.06
92.49
2.349
2.812
2.554
2.353
2.184
84.61
111.75
91.32
2.496, 2.561
2.351
2.741
2.349, 2.365
2.186
83.58
2.27(2), 2.40(2)
2.64(2), 2.74(2)
2.761(3), 2.768(3)
2.166(4), 2.156(5)
2.09(1)-2.169(5)
75.2(1)
Hg‚‚‚C
Hg-Cl
Al-Clbr
Al-Clter
Cl-Hg-Cl
C-Hg-C
Hg-Cl-M
2.383
2.200-2.245
95.94
2.204-2.243
102.20
112.51
117.85
109.37
119.43
135.1(5)
110.2(2)
108.99
91.93, 93.21
125.5(6)
90.8(1), 91.2(1)
Figure 9. Calculated structures for (a) Hg(C6H5Me)2(AlCl4)2, (b) [Hg(C6H5Me)2(AlCl4)]+ and (c) Hg(C6H6)2Cl2.
The absorptions for Hg(arene)2(AlCl4)2 are all at higher
energies compared to their gallium analogues, suggesting a
stronger Hg‚‚‚arene interaction in the former, in line with
expected differences in Lewis acidity between GaCl3 and AlCl3.
Finally, it should be noted, that while the absorption for [Hg2(µ-
O2CCF3)4(arene)2] is transient,12,13 the absorptions for Hg-
(arene)2(MCl4)2 are sustained indefinitely.
delocalized component across the AlCl3 units. It is clear from
the DFT calculations that the Hg‚‚‚arene interaction involves
two π orbitals on each arene (I and II). Such a combination
agrees with the asymmetric bonding of the arene to mercury,
see above.
DFT Calculations. To better understand the relationship
between the neutral and ionic structures as well as to assign
the 13C NMR spectra, DFT calculations were performed at the
B3LYP level using the 6-31G** basis set for C and H and
Stuttgart RLC ECP basis set for Hg, Cl and Al. Calculations
were performed on Hg(C6H6)2(AlCl4)2, Hg(C6H5Me)2(AlCl4)2,
[Hg(C6H6)2(AlCl4)]+, [Hg(C6H5Me)2(AlCl4)]+, and [Hg(o-
C6H4Me2)2(AlCl4)]+. The optimized calculated structural pa-
rameters for each model compound are given in Table 4 along
with the appropriate crystallographic data. Exemplary calculated
structures for Hg(C6H5Me)2(AlCl4)2 and [Hg(C6H5Me)2(AlCl4)]+
are shown in Figure 9a and b, respectively.
On the basis of the calculated energy [339 kJ‚mol-1 (C6H6)
and 330 kJ‚mol-1 (C6H5Me)] for the dissociation of one
[AlCl4]- group from Hg(arene)2(AlCl4)2 (eq 3), the cation-
anion interaction is clearly very strong.
Hg(arene)2(AlCl4)2 f [Hg(arene)2(AlCl4)]+ + [AlCl4]-
(3)
The calculated structures of Hg(arene)2(AlCl4)2 correspond
to a true energy minimum, having all positive vibrational
frequencies. In contrast, in the case of [Hg(o-C6H5Me)2(AlCl4)]+
one imaginary frequency was calculated for the minimized
structure indicating a first-order saddle point. The 3D potential
energy surface of [Hg(o-C6H4Me2)2(AlCl4)]+ is very shallow,
for example, the energy difference between Cs and C1 sym-
It is interesting that the dissociation of AlCl3 from Hg(C6H6)2-
(AlCl4)2 (eq 4) is more favored (112 kJ‚mol-1) than the
dissociation of [AlCl4]- (eq 3).
Hg(C6H6)2(AlCl4)2 f HgCl(C6H6)2(AlCl4) + AlCl3 (4)
We have no evidence for the dissociation of AlCl3 (or GaCl3)
in solution, suggesting that the formation in the solid state of
the ionic structure (i.e., the o-xylene derivatives) is due to a
large lattice stabilization energy. Furthermore, the formation of
ionic species in solution (see below) must be moderated by
strong ion solvation.
As noted above, the UV-visible spectra of compounds 1-8
consist of an absorption for Hg(arene)2(MCl4)2 and are all
between 279 and 337 nm, resulting in a characteristic yellow
to orange color. Single-excitation configuration interaction (CIS)
calculations were performed on the neutral complexes Hg-
(C6H6)2(AlCl4)2, Hg(C6H5Me)2(AlCl4)2, and [Hg(o-C6H4Me2)2-
(AlCl4)]+. Three singlet excited states were calculated for each
of the complexes. Although the calculated excitation energies
are about 6-17% higher than the experimentally determined
metries is only 0.71 kJ‚mol-1
.
As may be seen from Table 4, the overall structures for both
neutral and cationic structures are reasonably reproduced at the
present level of theory. However, the Hg-C distances are calcu-
lated to be slightly longer than those observed in the crystal
structures, while the Hg-Cl distances are underestimated. De-
spite these differences, the calculated 13C NMR shifts are very
close to those observed in the solid-state 13C CPMAS NMR,
see Table 2.
The Hg‚‚‚arene interaction is best described as being strongly
ionic in character with small contributions from the s and p
orbitals on mercury. In addition, there appears to be no d
character in the bonding. As may be seen from Figure 10, the
bonding orbital surfaces for Hg(C6H5Me)2(AlCl4)2 have a strong

Arene-Mercury Complexes
J. Am. Chem. Soc., Vol. 123, No. 45, 2001 11225
Figure 10. Calculated surfaces for the Hg‚‚‚arene interaction in Hg(C6H5Me)2(AlCl4)2.
Figure 11. Calculated LUMO (a) and HOMO (b) surfaces of Hg(C6H5Me)2(AlCl4)2.
values,33 their trend is undoubtedly correct. All of the calculated
excited states have similar excitation energies, and each occurs
between the HOMO (or a filled molecular orbital close in
energy) and the LUMO. As may be seen from Figure 11b, the
HOMO surface for Hg(C6H5Me)2(AlCl4)2 is essentially a
combination of aromatic π orbitals. The other two energetically
similar filled orbitals are also aromatic π in character. The major
contribution of the LUMO (Figure 11a) comes from the Hg 6s
orbital. Thus, the observed UV-visible spectra are due to an
areneπ f Hgs charge transfer.
Liquid Clathrates. As noted above, the reaction of HgCl2
with MCl3 in benzene, m-C6H4Me2, p-C6H4Me2 yields liquid
clathrates, for example, Figure 12. Clathrate formation is con-
centration-dependent. Thus, if HgCl2 and AlCl3 are reacted in
benzene at a concentration below 0.01 M (Hg) a homogeneous
solution is formed; above this threshold value, the clathrate
Figure 12. Liquid chlathrate formed from the reaction of HgCl2 with
AlCl3 in p-xylene showing the presence of two colored layers associated
with neutral (upper) and ionic (lower) complexes.
forms. The threshold for m-C6H4Me2 and p-C6H4Me2 is 0.16
and 0.05 M, respectively.
(33) The overstimation of excitation energies has been previously reported
for a number of simple molecules, see: (a) Foresman, J. B.; Frisch, A. E.
Exploring Chemistry with Electronic Structue Methods, 2nd ed.; Gaussian:
Pittsburgh, PA, 1996; Chapter 9. (b) Muguruma, C.; Koga, N.; Hatanaka,
Y.; El-Sayed, I.; Mikami, M.; Tanaka, M. J. Phys. Chem. 2000, 104, 4928.
The upper layer of the clathrate in each case is pale yellow
in color, while the lower layers are bright orange, see Figure
12. The layers can be separated by decanting the upper layer.
The homogeneous solutions formed below the threshold con-

11226 J. Am. Chem. Soc., Vol. 123, No. 45, 2001
BoroVik et al.
centrations are spectroscopically identical to the upper layer of
the clathrate. Conductivity measurements on the individual
layers indicate that the species in the upper layer are neutral,
while those in the lower layer are 1:1 electrolytes. The arene:
Hg ratio of the clathrates was determined to be ca. 8.5 for C6H6
and m-C6H4Me2, and ca. 7.5 for p-C6H4Me2. The formation of
the clathrates is summarized by the equilibrium in eq 5.
characterized spectroscopically.15 Unfortunately, the Hg‚‚‚arene
interaction in HgCl2(arene)2 is not sufficiently robust to allow
for crystallization and X-ray structural characterization. A
measure of the effectiveness of MCl3 in increasing the Lewis
acidity of the mercury, and thus enhancing the coordination of
arenes, may be obtained from (a) UV-visible spectroscopy and
(b) DFT calculations.
The UV spectrum of HgCl2 dissolved in toluene has been
reported to consist of an absorption at 274 nm.15 The spectral
bands observed for compounds 1 (325 nm) and 2 (305 nm) are
at a lower energy than that of the HgCl2 complex, indicative of
a greater decrease in the π-π* energy in the arene ring for the
mixed metal complexes.
Hg(arene)2(AlCl4)2 + n arene h
[Hg(arene)2(AlCl4)][AlCl4]‚n(arene) (5)
The concentrations of the upper layer of the benzene and
p-xylene clathrates are insufficient to allow for NMR charac-
terization. In contrast, the m-xylene clathrate exhibits a 27Al
NMR spectrum identical to that observed for the isolable neutral
compounds. The concentrations of the upper layer for all the
clathrates (and the homogeneous solutions) are suitable for UV
visible spectroscopy and are shown in Table 2. In each case,
the spectra are similar to those of the isolated compounds, Hg-
(arene)2(MCl4)2.
DFT calculations were performed on Hg(C6H6)2Cl2 at the
B3LYP level using the 6-31G** basis set for C and H and
Stuttgart RLC ECP basis set for Hg and Cl; the calculated
structure is shown in Figure 9c. The Hg-Cl distances in Hg-
(C6H6)2Cl2 (2.311 Å) are significantly shorter than in the
calculated structure of Hg(C6H6)2(AlCl4)2 (2.463 Å) and ap-
proach the values observed in HgCl2 by X-ray crystallography
[2.283(9) Å].35 This is expected in comparing a terminal versus
bridging chloride ligand. Also, the Cl-Hg-Cl angle calculated
for Hg(C6H6)2Cl2 (152.3°) is closer to linear (as in HgCl2) than
the equivalent angles in Hg(arene)2(AlCl4)2 [experimental )
75.2(1)-89.90(8)°, calculated ) 95.94-102.20°]. More impor-
tantly, the closest Hg‚‚‚C interactions in Hg(C6H6)2Cl2 (2.813
Å) are significantly larger than those calculated for Hg(C6H6)2-
(AlCl4)2 (2.402 Å). Thus, the presence of AlCl3 decreases the
Hg‚‚‚C distance by ca. 0.4 Å; a significant activation. Further-
more, it is worth noting that the Hg‚‚‚C interaction in Hg(C6H6)2-
Cl2 is comparable to the second closest distance in Hg(C6H6)2-
(AlCl4)2, which, based upon 13C spectroscopy, is very weak.
The 27Al NMR spectra of the lower clathrate layers show
single broad resonances consistent with AlCl4-. The 13C NMR
spectra of the clathrates shows a single set of resonances for
the arene, suggesting that there is a dynamic equilibria between
coordinated and “free” arene. The 1H NMR the CH3 resonances
(for m- and p-xylene) appear as a single sharp signal, but the
aromatic CH signals are very broad, suggesting a second slower
exchange process. This may be accounted for by a degenerate
H/H exchange of the aromatic CH groups, related to the H/D
exchange observed in deuterated solvents.
The term “liquid clathrates” is generally used to designate
nonstoichiometric liquid inclusion compounds which form upon
the interaction of aromatic molecules with certain ionic moi-
eties.34 It is generally believed that liquid clathrates are formed
when the parent compound possesses the following conditions:
the substance must have a relatively low lattice energy; the
substance must be capable of exhibiting a very strong cation-
anion interaction; association into tight ion pairs or other units
must be prevented. The lattice energy for Hg(arene)2(MCl4)2
should be low, given the lack of any unusual intermolecular
distances in the solid state, and thus meets the first requirement.
Based upon the calculated energy (ca. 330 kJ‚mol-1) for the
Conclusions
We have demonstrated that group 13 Lewis acids may be
used to “activate” other Lewis acidic complexes. In this regard,
stable Hg‚‚‚arene complexes have been prepared by the reaction
of HgCl2 with 2 mol equiv of MCl3 (M ) Al, Ga) in an aromatic
solvent.36 For C6H5Me, C6H5Et, o-C6H4Me2, and C6H3-1,2,3-
Me3 a neutral complex Hg(arene)2(MCl4)2 is formed in solution
and retained in the solid state except with o-C6H4Me2, for which
an ionic structure is observed. In contrast, reaction of HgCl2
with MCl3 in benzene, m-C6H4Me2, p-C6H4Me2 yields liquid
clathrates. We propose that the upper layer of each clathrate
contains a dilute solution of the neutral compound, Hg(arene)2-
(MCl4)2, while the lower layer is consistent with a 1:1 electrolyte
system, that is, [Hg(arene)2(MCl4)][MCl4]. On the basis of the
high calculated energy for dissociation of [AlCl4]- from the
neutral complex we propose that the formation of ionic
compounds is energetically moderated by the formation of the
clathrates. It is unclear at this time what factors control the
formation of a neutral complex versus a clathrate, and why only
the o-C6H4Me2 derivative exists as a cation/anion pair in the
solid state. We are continuing our studies on these mercury-
arene complexes, in particular, their application as H/D exchange
catalysts.
-
dissociation of one AlCl4 group from Hg(arene)2(AlCl4)2 (eq
3), the cation-anion interaction is clearly strong. From con-
ductivity measurements which confirm a 1:1 electrolyte, we may
presume that the aromatic molecules strongly solvate the ions
precluding ion pair formation.
We note that the clathrates reported herein are distinct from
the more usual class of clathrates. In a traditional clathrate the
upper layer is essentially solvent, while the lower “clathrate”
layer contains the ionic salts and solvent. In our clathrates, the
upper layer contains a neutral component “Hg(arene)2(MCl4)2”,
while the lower contains the ionic component “[Hg(arene)2-
(MCl4)][MCl4]”. The presence of mercury arene complexes in
both layers is clearly seen in the photograph shown in Figure
12.
The Hg‚‚‚arene interaction in Hg(arene)2(MCl4)2 and Hg-
(arene)2(MCl4)][MCl4] is significantly stronger than for either
of the constituent halides, that is, HgCl2 or MCl3. Thus, the
group 13 halides appear to activate the mercury toward the
Activation of HgCl2 toward Arene Binding by MCl3. As
noted in the Introduction, Hg‚‚‚arene complexes have been
(34) (a) Atwood, J. L.; Atwood, J. D. In Inorganic Compounds with
Unusual Properties; King, R. B., Ed.; Advances in Chemistry, Vol. 150;
American Chemical Society, Washington, DC, 1976; Chapter 11. (b)
Atwood, J. L. In Recent DeVelopments in Separation Science; Li, N. N.,
Ed.; CRC Press: Cleveland, 1977; Vol. 3, pp 195-209. (c) Atwood, J. L.
In Inclusion Compounds; Atwood, J. L., Davies, J. E. D., MacNicol, D.
D., Eds.; Academic Press: London, 1984; Vol. 1, pp 375-405.
(35) Subramanian, V.; Seff, K. Acta Crystallogr. 1980, B36, 2132.
(36) We note that stable silver arene complexes have been prepared by
the Lewis acid abstraction of fluoride from AgF, see: Hatop, H.; Roesky,
H. W.; Labahn, T.; Roepken, C.; Sheldrick, G. M.; Bhattacharjee, M.
Organometallics 1998, 17, 4326.

Arene-Mercury Complexes
J. Am. Chem. Soc., Vol. 123, No. 45, 2001 11227
coordination of the arenes. Based upon X-ray crystallography,
13C CPMAS NMR spectroscopy, and DFT calculations the
Hg‚‚‚arene interaction is found to range from predominantly
η1 coordination to close to η2 coordination. At this time we
are unclear as to the factors that control the mode of coordina-
tion.
Experimental Section
NMR spectra were obtained on Bruker AM-250 and Avance 200,
400, and 500 spectrometers. Chemical shifts are reported relative to
internal solvent resonances. 13C and 199Hg MAS spectra were obtained
at 50.32 and 35.84 MHz respectively, using Bruker Avance 200
spectrometer. A 7 mm zirconium dioxide rotor was used for all spectra,
with the spin rates up to 7 kHz. 199Hg spectra were recorded with direct
polarization (4 µs 28° rf pulses). Centerband signals were located by
varying the spinning rate. A 20.53 ms FID was acquired with high
level proton decoupling and a 50 s relaxation delay without decoupling.
A total of 1024 scans were required to get acceptable spectra. The FID
was processed with 70 Hz of line broadening. Chemical shift was
referenced using 0.5 M solution of HgCl2 in 75% EtOH with 25% of
D2O (δHg ) -1497 ppm).37 Mass spectra were obtained on a Finnigan
MAT 95 mass spectrometer operating with an electron beam energy
of 70 eV for EI mass spectra. UV-visible spectral data were recorded
on a Varian Cary 4 spectrometer and are given in Table 3. Microanaly-
ses were performed by Oneida Research Services, Inc., Whitesboro,
NY. Unfortunately, the extreme air sensitivity of several compounds
resulted in highly variable analysis results. The synthesis of Hg(C6H5-
Me)2(GaCl4)2 (2) was reported previously.17 Solvents and all arenes
were distilled and degassed prior to use.
Hg(C6H5Me)2(AlCl4)2 (1). Toluene (25 mL) was added to the solid
mixture of anhydrous HgCl2 (1.00 g, 3.68 mmol), and AlCl3 (0.984 g,
7.37 mmol). The resulting yellow solution was heated at approximately
100 °C while being vigorously stirred to allow all HgCl2 to dissolve.
In about 15 min, heating was stopped, and the reaction flask was
wrapped in aluminum foil to prevent the decomposition reaction caused
by bright light. Yellow crystals grew within 1 h at room temperature.
Yield: 90%. Mp 95 °C. 13C CPMAS NMR (50.32 MHz): δ 159.4
(1C, CMe), 143.1 (1C, m-CH), 140.1 (1C, m-CH), 136.6 (2C, o-CH),
100.7 (1C, Hg‚‚‚CH), 24.0 (1C, CH3). 27Al NMR (52.15 MHz, toluene
with C6D6 as an external reference): δ 105 (W1/2 ) 780 Hz). 27Al MAS
NMR (52.15 MHz): δ 118 and 86 (W1/2 ) 3370 Hz).32
Hg(C6H5Et)2(AlCl4)2 (3). Prepared in a manner similar to that for
compound 1, but using ethylbenzene (25 mL), anhydrous HgCl2 (1.00
g, 3.68 mmol), and AlCl3 (0.984 g, 7.37 mmol) to yield yellow crystals.
Yield: 80%. Mp 74 °C. 13C CPMAS NMR (50.32 MHz): δ 163.2
(1C, CCH2), 140.2 (1C, m-CH), 137.1 (1C, m-CH), 135.9 (o-CH), 103.7,
(1C, Hg‚‚‚CH), 30.8 (1C, CH2), 15.4 (1C, CH3).
Hg(C6H5Et)2(GaCl4)2 (4). Prepared in a manner similar to that for
compound 1, but using ethylbenzene (25 mL), anhydrous HgCl2 (1 g,
3.68 mmol), and GaCl3 (1.298 g, 7.37 mmol). Yellow crystals were
grown within 1 h at room temperature. Yield: 78%. Mp 59 °C. 13C
CPMAS NMR (50.32 MHz): δ 162.8 (1C, CCH2), 139.9 (1C, m-CH),
136.2 (3C, m-CH and o-CH), 105.0 (1C, Hg‚‚‚CH), 30.9 (1C, CH2),
16.2 (1C, CH3).
[Hg(o-C6H4Me2)2(AlCl4)][AlCl4] (5). Prepared in a manner similar
to that for compound 1, but using o-xylene (25 mL), anhydrous HgCl2
(1.00 g, 3.68 mmol), and AlCl3 (0.984 g, 7.37 mmol) to yield dark-
yellow crystals. Yield: 80%. Mp 99 °C. 13C CPMAS NMR (50.32
MHz): δ 158.4 (1C, CCH3), 147.7 (1C, CCH3), 139.0 (2C, o-CH),
135.6 (1C, m-CH), 105.5 (1C, Hg‚‚‚CH), 22.4 (2C, CH3). 27Al MAS
NMR (52.15 MHz): δ 89 and 82 (W1/2 ) 870 Hz). 27Al NMR (52.15
MHz, in o-xylene, C6D6 as an external lock solvent): δ 105 (W1/2
630 Hz).
)
[Hg(o-C6H4Me2)2(GaCl4)][GaCl4] (6). Prepared in a manner similar
to that for compound 1, but using o-xylene (25 mL), anhydrous HgCl2
(1.00 g, 3.68 mmol), and GaCl3 (1.298 g, 7.37 mmol), to yield dark-
yellow crystals. Yield: 85%. Mp 77 °C. 13C CPMAS NMR (50.32
(37) NMR and the Periodic Table; Harris, R. K., Mann, B. E., Eds.;
Academic Press: New York 1978; p 268

11228 J. Am. Chem. Soc., Vol. 123, No. 45, 2001
BoroVik et al.
MHz): δ 159.0 (1C, CCH3), 146.3 (1C, CCH3), 139.3 (3C, o-CH and
m-CH), 102.9 (1C, Hg‚‚‚CH), 22.5 (1C, CH3).
and Cs symmetries were imposed on neutral and cationic molecules,
respectively. Vibrational frequencies were then evaluated for benzene
complexes to verify the existence of the true potential minimum and
to determine zero-point energies. 13C NMR chemical shifts for Hg-
(C6H5Me)2(AlCl4)2 and [Hg(o-C6H4Me2)2(AlCl4)]+ complexes were
calculated at the same level of theory. Vertical excitation energies and
corresponding oscillator strengths for Hg(C6H6)2(AlCl4)2, Hg(C6H5Me)2-
(AlCl4)2, and [Hg(o-C6H4Me2)2(AlCl4)]+ were calculated by the single-
excitation configuration interaction (CIS)40 method at the ground-state
stationary points of the B3LYP level.
Crystallographic Studies. Data for compounds 1, 2-5, and 7 were
collected on a Bruker CCD SMART system, equipped with graphite
monochromated Mo KR radiation (λ ) 0.71073 Å) and corrected for
Lorentz and polarization effects. The structures were solved using the
direct methods program XS41 and difference Fourier maps and refined
by using full matrix least-squares methods. All non-hydrogen atoms
(except the ethyl groups in compounds 3 and 4) were refined with
anisotropic thermal parameters. Hydrogen atoms were introduced in
calculated positions and allowed to ride on the attached carbon atoms
[d(C-H) ) 0.95 Å]. Refinement of positional and anisotropic thermal
parameters led to convergence (see Table 5).
Hg(C6H3-1,2,3-Me3)2(AlCl4)2 (7). Prepared in a manner similar to
that for compound 1, but using 1,2,3-trimethylbenzene (20 mL),
anhydrous HgCl2 (0.50 g, 1.84 mmol), and AlCl3 (0.491 g, 3.68 mmol).
Yellow crystals were grown over a few days at -19 °C. Yield: 70%.
Mp 94 °C. 13C CPMAS NMR (50.32 MHz): δ 153.1 (3C, br, CCH3),
148.1 (3C, br, CCH3), 136.3 (1C, o-CH), 135.0 (1C, o-CH), 125.1 (1C,
Hg‚‚‚CH), 117.2 (2C, Hg‚‚‚CH), 112.8 (1C, Hg‚‚‚CH), 23.8 (2C, CH3),
21.6 (1C, CH3), 20.4 (1C, CH3), 16.9 (1C, CH3), 15.5 (1C, CH3). 27Al
MAS NMR (52.15 MHz): δ 90 and 76 (W1/2 ) 2040 Hz).32 27Al NMR
(52.15 MHz, in 1,2,3-trimethylbenzene, C6D6 as an external lock
solvent): δ 105 (W1/2 ) 2040 Hz).
Hg(C6H3-1,2,3-Me3)2(GaCl4)2 (8). Prepared in a manner similar to
that for compound 1, but using 1,2,3-trimethylbenzene (20 mL),
anhydrous HgCl2 (0.5 g, 1.84 mmol), and GaCl3 (0.65 g, 3.68 mmol).
Yellow crystals were grown over a few days at -19 °C. Yield: 80%.
Mp 83 °C. 13C CPMAS NMR (50.32 MHz): δ 152.9 (1C, CCH3),
152.2 (1C, CCH3), 150.8 (2C, CCH3), 147.5 (1C, CCH3), 146.4 (1C,
CCH3), 136.0 (1C, o-CH), 134.7 (1C, o-CH), 124.4 (1C, Hg‚‚‚CH),
118.0 (1C, Hg‚‚‚CH), 117.0 (1C, Hg‚‚‚CH), 113.8 (1C, Hg‚‚‚CH), 23.4
(2C, CH3), 21.8 (1C, CH3), 20.6 (1C, CH3), 16.9 (1C, CH3), 15.6 (1C,
CH3).
Acknowledgment. Financial support for this work is pro-
vided by the Robert A. Welch Foundation, including the Bruker
CCD Smart System Diffractometer of the Texas Center for
Crystallography at Rice University. The Bruker Avance 200
and 500 NMR spectrometers were purchased with funds from
ONR Grant N00014-96-1-1146 and NSF Grant CHE-9708978,
respectively.
Liquid Clathrates. To a solid mixture of HgCl2 (1.00 g, 3.68 mmol)
and AlCl3 (0.984 g, 7.37 mmol) was added the appropriate arene (10
mL). The resulting yellow-orange solution was stirred vigorously at
60 °C until all of the HgCl2 dissolved. In approximately 30 min stirring
was halted, and within a further 5 min the reaction mixture separated
into two layers. The bottom chlathrate layer has the general formula,
Hg(arene)8.5(AlCl4)2. Similar chlathrates are observed with GaCl3. UV-
visible spectral data are given in Table 3.
C6H6. Bottom layer, 27Al NMR (52.15 MHz, C6D6 as an external
lock solvent): δ 105 (W1/2 ) 1170 Hz).
Supporting Information Available: Full listings of bond
length and angles, anisotropic thermal parameters, and hydrogen
atom parameters; tables of calculated and observed structure
factors; optimized structural parameters for DFT calculations;
elemental analysis; energies for optimized structures; calculated
UV-visible absorption spectra and oscillator strengths (PDF).
This material is available free of charge via the Internet at
http://pubs.acs.org.
m-C6H4Me2. Bottom layer, 27Al NMR (52.15 MHz, C6D6 as an
external lock solvent): δ 104 (W1/2 ) 1800 Hz). Top layer, 27Al NMR
(52.15 MHz, C6D6 as an external lock solvent): δ 105 (W1/2 ) 1210
Hz).
p-C6H4Me2. Bottom layer, 27Al NMR (52.15 MHz, C6D6 as an
external lock solvent): δ 105 (W1/2 ) 1720 Hz).
Computational Methods. All density functional calculations were
carried out using a Gaussian-98 suite.38 Complete geometry optimiza-
tions were performed at B3LYP39 level using the 6-31G** basis set
for C and H and Stuttgart RLC ECP basis set for Hg, Cl, and Al. C2
JA011003U
Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P.
M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Andres, J. L.; Gonzalez, C.;
Head-Gordon, M.; Replogle, E. S.; Pople, J. A. Gaussian 98, revision A.9;
Gaussian, Inc.: Pittsburgh, PA, 1998.
(39) (a) Becke, A. D. J. Chem. Phys. 1993, 98, 5648. (b) Stephens, P.
J.; Devlin, C. F.; Chabalowski, C. F.; Frisch M. J. J. Phys. Chem. 1994,
98, 11623. (c) Lee, C.; Yang, W.; Parr, R. G. Phys. ReV. 1988, B37, 785.
(40) Foresman, J. B.; Head-Gordon, M.; Pople, J. A.; Frisch, M. J. J.
Phys. Chem. 1992, 96, 135.
(38) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb,
M. A.; Cheeseman, J. R.; Zakrzewski, V. G.; Montgomery, J. A., Jr.;
Stratmann, R. E.; Burant, J. C.; Dapprich, S.; Millam, J. M.; Daniels, A.
D.; Kudin, K. N.; Strain, M. C.; Farkas, O.; Tomasi, J.; Barone, V.; Cossi,
M.; Cammi, R.; Mennucci, B.; Pomelli, C.; Adamo, C.; Clifford, S.;
Ochterski, J.; Petersson, G. A.; Ayala, P. Y.; Cui, Q.; Morokuma, K.; Malick,
D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Cioslowski, J.;
Ortiz, J. V.; Baboul, A. G.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz,
P.; Komaromi, I.; Gomperts, R.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-
(41) Sheldrick, G. M. SHELXTL; Bruker AXS, Inc.: Madison, Wisconsin,
1997.