Coordination of nitroxide and nitronyl-nitroxide organic radicals to trimeric perfluoro-o-phenylene mercury

Article abstract of DOI:10.1021/ic050528a

The interaction of trimeric perfluoro-o-phenylene mercury (1) with TEMPO (1,1,5,5-tetramethylpentamethylene nitroxide) in CH₂Cl₂ leads to the formation of the 1:1 adduct [1·TEMPO] (2). The same reaction carried out with NIT-Ph (2-(phenyl)-4,4,5,5-tetramethylimidazoline-1-oxyl-3- oxide) leads to the formation of either [1·NIT-Ph·1] (3) or [1·NIT-Ph]_n (4), depending on the amount of NIT-Ph present in solution. Adducts 2, 3, and 4 have been fully characterized and their crystal structures determined. The solid-state structure of 2 contains molecules of [1·TEMPO] in which the nitroxide oxygen atom is triply coordinated to the mercury centers of 1. A similar situation is encountered in the structure of 3 where each oxygen atom of the NIT-Ph molecule interacts with the mercury centers of an adjacent molecule of 1. The structure of 4 consists of extended helical polymeric chains that contain alternating molecules of 1 and NIT-Ph. As in 2 and 3, the interactions responsible for the formation of these chains involve the triple coordination of the oxygen atoms of the NIT-Ph molecule to the mercury centers of 1. DFT calculations suggest that the bonding in adducts such as 2, 3, and 4 is most likely dominated by electrostatic rather than covalent interactions. In agreement with this view, magnetic susceptibility measurements carried out on these adducts indicate that 1 does not mediate significant coupling between organic radicals coordinated on either side of the trinuclear core.

Full text of DOI:10.1021/ic050528a

Inorg. Chem. 2005, 44, 6248−6255
Coordination of Nitroxide and Nitronyl-nitroxide Organic Radicals to
Trimeric Perfluoro-o-phenylene Mercury
Mason R. Haneline and Franc¸ois P. Gabba1ı *
Texas A&M UniVersity, Chemistry TAMU 3255, College Station, Texas 77843-3255
Received April 6, 2005
The interaction of trimeric perfluoro-o-phenylene mercury (1) with TEMPO (1,1,5,5-tetramethylpentamethylene nitroxide)
in CH2Cl2 leads to the formation of the 1:1 adduct [1 TEMPO] (2). The same reaction carried out with NIT-Ph
(2-(phenyl)-4,4,5,5-tetramethylimidazoline-1-oxyl-3-oxide) leads to the formation of either [1 NIT-Ph 1] (3) or [1 NIT-
Ph]n (4), depending on the amount of NIT-Ph present in solution. Adducts 2, 3, and 4 have been fully characterized
and their crystal structures determined. The solid-state structure of 2 contains molecules of [1 TEMPO] in which
•
•
•
•
•
the nitroxide oxygen atom is triply coordinated to the mercury centers of 1. A similar situation is encountered in the
structure of 3 where each oxygen atom of the NIT-Ph molecule interacts with the mercury centers of an adjacent
molecule of 1. The structure of 4 consists of extended helical polymeric chains that contain alternating molecules
of 1 and NIT-Ph. As in 2 and 3, the interactions responsible for the formation of these chains involve the triple
coordination of the oxygen atoms of the NIT-Ph molecule to the mercury centers of 1. DFT calculations suggest
that the bonding in adducts such as 2, 3, and 4 is most likely dominated by electrostatic rather than covalent
interactions. In agreement with this view, magnetic susceptibility measurements carried out on these adducts indicate
that 1 does not mediate significant coupling between organic radicals coordinated on either side of the trinuclear
core.
Introduction
Such derivatives have been investigated for the multiple
electrophilic complexation of both anionic and neutral
electron-rich substrates. Perfluoro-o-phenylene mercury (1)
is a prototypical example of such polyfunctional Lewis
acids.3,4 It exhibits a fascinating coordination chemistry with
anions and has been employed in the formation of hexa-
coordinated halide complexes.6 This trinuclear complex also
interacts with neutral electron-rich substrates7 including
organic carbonyls,8-10 nitriles,11 and sulfoxides12 to form
Polyfunctional organomercurials featuring proximal metal
centers have been extensively studied as polydentate Lewis
acids. Typical examples of such complexes include 1,8-bis-
(mercurio)naphathalenes,1 1,2-bis(mercurio)benzenes,2 and
various macrocyclic3,4 species such as mercuracarborands.5
* E-mail: gabbai@mail.chem.tamu.edu.
(1) Schmidbaur, H.; O¨ ller, H.-J.; Wilkinson, D. L.; Huber, B. Chem. Ber.
1989, 122, 31-36.
(2) (a) Beauchamp, A. L.; Olivier, M. J.; Wuest, J. D.; Zacharie, B.
Organometallics 1987, 6, 153-156. (b) Tschinkl, M.; Schier, A.;
Riede, J.; Gabba¨ı, F. P. Organometallics 1999, 18, 1747-1753. (c)
Vaugeois, J.; Wuest, J. D. J. Am. Chem. Soc. 1998, 120, 13016-
13022. (d) Vaugeois, J.; Simard, M.; Wuest, J. D. Organometallics
1998, 17, 1215-1219. (e) Simard, M.; Vaugeois, J.; Wuest, J. D. J.
Am. Chem. Soc. 1993, 115, 370-372. (f) Wuest, J. D.; Zacharie, B.
J. Am. Chem. Soc. 1987, 109, 4714-4715. (g) Beauchamp, A. L.;
Olivier, M. J.; Wuest, J. D.; Zacharie, B. J. Am. Chem. Soc. 1986,
108, 73-77. (h) Wuest, J. D.; Zacharie, B. J. Am. Chem. Soc. 1985,
107, 6121-6123. (i) Wuest, J. D.; Zacharie, B. Organometallics 1985,
4, 410-411.
(6) (a) Chistyakov, A. K.; Stankevich, I. V.; Gambaryan, N. P.; Struchkov,
Y. T.; Yanovsky, A. I.; Tikhonova, I. A.; Shur, V. B. J. Organomet.
Chem. 1997, 536, 413-424. (b) Shur, V. B.; Tikhonova, I. A.;
Yanovskii, A. I.; Struchkov, Y. T.; Petrovskii, P. V.; Panov, S. Y.;
Furin, G. G.; Vol’pin, M. E. J. Organomet. Chem. 1991, 418, C29-
C32. (c) Shur, V. B.; Tikhonova, I. A.; Yanovskii, A. I.; Struchkov,
Y. T.; Petrovskii, P. V.; Panov, S. Y.; Furin, G. G.; Vol’pin, M. E.
Dokl. Akad. Nauk. SSR 1991, 321, 1002-1004. (d) Shur, V. B.;
Tikhonova, I. A.; Dolgushin, F. M.; Yanovsky, A. I.; Struchkov, Y.
T.; Volkonsky, A. Y.; Solodova, E. A.; Panov, S. Y.; Petrovskii, P.
V.; Vol’pin, M. E. J. Organomet. Chem. 1993, 443, C19-C21. (e)
Koomen, J. M.; Lucas, J. E.; Haneline, M. R.; Beckwith King, J. D.;
Gabba¨ı, F.; Russell, D. H. Int. J. Mass Spectrom. 2003, 225, 225-
231.
(3) Shur, V. B.; Tikhonova, I. A. Russ. Chem. Bull. 2003, 52, 2539-
2554.
(4) Haneline, M. R.; Taylor, R.; Gabba¨ı, F. P. Chem. Eur. J. 2003, 9,
5188-5193.
(5) (a) Hawthorne, M. F.; Zheng, Z. Acc. Chem. Res. 1997, 30, 267-
276. (b) Wedge, T. J.; Hawthorne, M. F. Coord. Chem. ReV. 2003,
240 (1-2), 111-128.
(7) Tsunoda, M.; Gabba¨ı, F. P. J. Am. Chem. Soc. 2003, 125, 10492-
10493.
(8) Ball, M. C.; Brown, D. S.; Massey, A. G. J. Organomet. Chem. 1981,
206, 265-277.
6248 Inorganic Chemistry, Vol. 44, No. 18, 2005
10.1021/ic050528a CCC: $30.25
© 2005 American Chemical Society
Published on Web 08/04/2005

Organic Radicals Coordinated to Perfluoro-o-phenylene Mercury
Chart 1. Schematic Structures of 1, the 1:1 and 1:2 Complexes Formed by 1, and Electron-Rich Substrates (L)
Chart 2
discrete [1•L] and [1•L2] complexes in which the electron
rich terminus of the substrates interacts simultaneously with
the three mercury atoms of 1 (Chart 1). In the [1•L2]
complexes, two molecules of the donor are coordinated to
the mercury centers of 1 on either side of the molecular plane.
Work carried out in our laboratory also demonstrates that 1
is a remarkable building block for the construction of
supramolecular materials.3,4 In particular, 1 readily interacts
with various arenes to form extended binary stacks in which
the arene is weakly π-coordinated to the mercury centers of
1.13 These supramolecules feature distinct physical properties
and display intense room-temperature phosphorescence of
the arene as a result of a mercury heavy atom effect.13 As
part of our continuing interest in the chemistry of this
molecule, we are actively exploring its potential for the
elaboration of novel materials.
magnetic interactions, the organic radicals could couple in
either a ferromagnetic (A) or an antiferromagnetic fashion
(B). Also, it can be envisaged that 1 does not mediate
magnetic interactions, leading to a situation in which the
organic radicals remain uncoupled (C). To determine which
of these situations occurs, we have now prepared and studied
a series of adducts involving 1 and 1,1,5,5-tetramethylpen-
tamethylene nitroxide (TEMPO) or 2-(phenyl)-4,4,5,5-tet-
ramethylimidazoline-1-oxyl-3-oxide (NIT-Ph) nitronyl radi-
cals.
Experimental Section
General: Due to the toxicity of the mercury compounds discussed
in these studies, extra care was taken at all times to aVoid contact
with solid, solution, and airborne particulate mercury compounds.
All experiments were carried out in a well-aerated fume hood.
Atlantic Microlab, Inc., Norcross, GA, performed the elemental
analyses. Infrared spectra were recorded as KBr pellets on a Mattson
Genesis Series FTIR. NIT-Ph was synthesized according to the
published procedure.15 Commercially available starting materials
and solvents were purchased from Aldrich Chemical and were used
as provided. Compound 1 was prepared according to the published
procedure outlined by Sartori and Golloch.16 Magnetic susceptibility
and magnetization measurements were carried out with a Quantum
Design SQUID magnetometer MPMS-XL. dc magnetic measure-
ments were performed with an applied field of 1000 G in the 2-300
K temperature range. Data were corrected for the diamagnetic
contributions calculated from the Pascal constants.17 The small
discontinuity observed in the thermal variation of the XmT product
for compound 2 and 3 (Figure 6) is caused by a subtraction of the
magnetization obtained for the sample holder (diamagnetic) from
the magnetization obtained for the actual sample. This discontinuity
is accentuated by the small size of the sample and the fact that the
compounds are magnetically dilute (see MPMS Application Note
1014-213 - Quantum Design - available on the Internet at
http://www.qdusa.com/).
Nitroxides and nitronyl nitroxides are some of the most
stable organic radicals.14 Like organic carbonyls, these
molecules feature an electron-rich terminal oxygen atom that
should readily coordinate to the mercury centers of 1. If 1:2
complexes are accessible, three distinct magnetic situations
(A-C) can be envisioned (Chart 2). If 1 is able to mediate
(9) King, J. B.; Tsunoda, M.; Gabba¨ı, F. P. Organometallics 2002, 21,
4201-4205.
(10) King, J. B.; Haneline, M. R.; Tsunoda, M.; Gabba¨ı, F. P. J. Am. Chem.
Soc. 2002, 124, 9350-9351.
(11) (a) Tikhonova, I. A.; Dolgushin, F. M.; Yanovsky, A. I.; Starikova,
Z. A.; Petrovskii, P. V.; Furin, G. G.; Shur, V. B. J. Organomet. Chem.
2000, 613, 60-67. (b) Tikhonova, I. A.; Dolgushin, F. M.; Tugashov,
K. I.; Furin, G. G.; Petrovskii, P. V.; Shur, V. B. Russ. Chem. Bull.
2001, 50, 1673-1678. (c) Tikhonova, I. A.; Dolgushin, F. M.;
Tugashov, K. I.; Ellert, O. G.; Novotortsev, V. M.; Furin, G. G.;
Antipin, M. Yu.; Shur, V. B. J. Organomet. Chem. 2004, 689, 82-
87. (d) Baldamus, J.; Deacon, G. B.; Hey-Hawkins, E.; Junk, P. C.;
Martin, C. Aust. J. Chem. 2002, 55, 195-198.
(12) Tikhonova, I. A.; Dolgushin, F. M.; Tugashov, K. I.; Petrovskii, P.
V.; Furin, G. G.; Shur, V. B. J. Organomet. Chem. 2002, 654, 123-
131.
(13) (a) Haneline, M. R.; Tsunoda, M.; Gabba¨ı, F. P J. Am. Chem. Soc.
2002, 124, 3737-3742. (b) Haneline, M. R.; King, J. B.; Gabba¨ı, F.
P. J. Chem. Soc., Dalton Trans. 2003, 13, 2686-2690. (c) Omary,
M. A.; Kassab, R. M.; Haneline M. R.; Elbjeirami, O.; Gabba¨ı, F. P.
Inorg. Chem. 2003, 42, 2176-2178.
(15) Ullman, E. F.; Osiecki, J. H.; Boocock, D. G. B.; Darcy, R. J. Am.
Chem. Soc. 1972, 94, 7049-7059.
(16) Sartori, P.; Golloch, A. Chem. Ber. 1968, 101, 2004-2009.
(17) Theory and Applications of Molecular Paramagnetism; Boudreaux,
E. A., Mulay, L. N., Eds.; John Wiley & Sons: New York, 1976.
(14) Volodarsky, L. B.; Reznikov, V. A.; Ovcharenko, V. I. Synthetic
Chemistry of Stable Nitroxides; CRC Press: Boca Raton, FL, 1994.
Inorganic Chemistry, Vol. 44, No. 18, 2005 6249

Haneline and Gabba1ı
Scheme 1
Synthesis of [1•TEMPO] (2). A solution of compound 1 (100
mg, 96 µmol) in CH2Cl2 was mixed with a solution of TEMPO
(15 mg, 98 µmol) in CH2Cl2. Upon slow evaporation of the solvent,
pale yellow crystals of compound 2 were observed (109 mg,
yield: 95%). mp 239 °C decomposition. Anal. Calcd for C27H18F12-
Hg3NO: C, 27.01; H, 1.51. Found: C, 27.03; H, 1.42.
Synthesis of [1•NIT-Ph•1] (3). A solution of compound 1 (100
mg, 96 µmol) in CH2Cl2 was mixed with a solution of NIT-Ph
(22.4 mg, 96 µmol) in CH2Cl2. Upon evaporation of the solvent,
the crystals were washed with hexanes to remove excess NIT-Ph.
The remaining pink crystals were washed quickly with 0.5 mL of
CH2Cl2, affording 40 mg of compound 3 (yield: 36%). mp 220 °C
decomposition. Anal. Calcd for C49H17F24Hg6N2O2: C, 25.35; H,
0.74. Found: C, 25.57; H, 0.72.
Synthesis of [1•NIT-Ph] (4). A solution of compound 1 (100
mg, 96 µmol) in CH2Cl2 was mixed with a solution of NIT-Ph
(100 mg, 0.529 mmol) in CH2Cl2. Upon evaporation of the solvent,
the crystals were washed with hexanes to remove excess NIT-Ph.
The remaining purple crystals were washed quickly with 0.5 mL
of CH2Cl2, affording 90 mg of compound 4 (yield: 73%). mp 190
°C decomposition. Anal. Calcd for C31H17F12Hg3N2O2: C, 29.10;
H, 1.34. Found: C, 28.99; H, 1.32.
ratio in CH2Cl2, slow evaporation of the solvent leads to the
formation of pink crystals of the 2:1 adduct [1•NIT-Ph•1]
(3) and purple needles of the 1:1 adduct [1•NIT-Ph] (4)
(Scheme 1). High yields of the latter were obtained when a
large excess of NIT-Ph was employed. The composition of
2-4 was confirmed by elemental analysis. Each adduct was
found to be air-stable and decomposed at temperatures above
190 °C. Keeping in mind that organomercurials are some-
times used to generate organic radicals, the observed stability
of 1 in the presence of a radical trap such as TEMPO is
noteworthy. The EPR spectra of these compounds in CH2-
Cl2 correspond to those of the free radical, suggesting
dissociation of the adducts in solution. A similar conclusion
was derived from NMR spectroscopic measurements for
aldehyde or ketone adducts of 1 which do not subsist in
solution.9 The lability of this type of complexes points to
the weakness of the bonding interactions involving 1 and
organic substrates containing terminal oxo-ligands. The N-O
stretching frequency in 3 (1353 cm-1, KBr) and 4 (1354
cm-1, KBr) is lower than that in pure NIT-Ph (1367 cm-1,
KBr), suggesting a moderate weakening of the N-O bond.23
In the case of 2, the N-O stretching vibration could not be
detected because of overlap with bands from the trinuclear
mercury complex.
Single-Crystal X-ray Analysis. X-ray data for 2-4 were
collected on a Bruker Smart-CCD diffractometer using graphite-
monochromated Mo KR radiation (λ ) 0.71073 Å). Specimens of
suitable size and quality were selected and mounted onto a glass
fiber with Apezion grease and run at 110 K. The structures were
solved by direct methods, which successfully located most of the
non-hydrogen atoms. Subsequent refinement on F2 using the
SHELXTL/PC package (version 6.1) allowed location of the
remaining non-hydrogen atoms.
Calculations. Geometry optimization and single-point energy
calculations were performed using density functional theory (DFT)
in the Amsterdam density functional package (ADF).18-20 The
Becke exchange functional and the Lee-Yang-Parr correlation
functional (BLYP) were utilized in the calculation.21,22 The triple-
ú, double-polarization (TZ2P) basis function was used. The scalar
zero-order-regular-approximation (ZORA) was applied to account
for relativistic effects. The cores of atoms were frozen, C and F up
to the 1s level and Hg up to the 4d level. The Dirac utility was
used to generate relativistic frozen core potentials for the scalar
ZORA calculations. All quoted electronic structure data from
optimized structures and single-point energy data use an integration
of 6.0. The energy convergence criterion was set at 10-6 au, and
the geometry was constrained to be D3.
Compound 2 crystallizes in the triclinic space group P1h
with one molecule of [1•TEMPO] in the asymmetric unit
(Figure 1, Table 1). The resulting Hg-O distances range
from 2.889(11) to 3.141(12) Å and are well within the sum
of the van der Waals radii for mercury (rvdw ) 1.75 Å)24,25
and oxygen (rvdw ) 1.54 Å).26 As a result of these
interactions, the oxygen atom is essentially equidistant from
the three Lewis acidic sites and sits at a distance, d, of 2.17
Å from the plane defined by the three mercury atoms. The
N-O bond of the nitroxide is nearly perpendicular to this
plane and forms an angle, R, of 86.9° (Figure 1). The metrical
Results and Discussion
Combining equimolar CH2Cl2 solutions of 1 and TEMPO
affords pale yellow crystals of [1•TEMPO] (2) upon slow
evaporation (Scheme 1). The same experiment carried out
in the presence of excess TEMPO does not yield a 2:1
complex. When 1 and NIT-Ph are combined in an equimolar
(18) ADF2003.01. SCM, Theoretical Chemistry, Vrije Universiteit,
Amsterdam, The Netherlands.
(19) Guerra, C. F.; Snijders, J. G.; Te Velde, G.; Baerends, E. J. Theor.
Chem. Acc. 1998, 99, 391-403.
(20) Te Velde, G.; Bickelhaupt, F. M.; Baerends, E. J.; Fonseca Guerra,
C.; Van Gisbergen, S. J. A.; Snijders, J. G.; Ziegler, T. J. Comput.
Chem. 2001, 22, 931-967.
(23) Ullman, E. F.; Osiecki, J. H.; Boocock, D. G. B.; Darcy, R. J. Am.
Chem. Soc. 1972, 94, 7049-7059.
(24) Canty, A. J.; Deacon, G. B. Inorg. Chim. Acta 1980, 45, L225-L227.
(25) Pyykko¨, P.; Straka, M. Phys. Chem. Chem. Phys. 2000, 2, 2489-
2493.
(21) Becke, A. D. Phys. ReV. A: At., Mol., Opt. Phys. 1988, 38, 3098-
3100.
(22) Lee, C.; Yang, W.; Parr, R. G. Phys. ReV. B: Condens. Matter 1988,
37, 785-789.
(26) Nyburg, S. C.; Faerman, C. H. Acta Crystallogr., Sect. B 1985, 41,
274-279.
6250 Inorganic Chemistry, Vol. 44, No. 18, 2005

Organic Radicals Coordinated to Perfluoro-o-phenylene Mercury
Figure 1. Compound 2 with 50% thermal ellipsoids (fluorine and hydrogen
atoms omitted for clarity). Selected bond distances (Å) and angles (°):
Hg(1)-O(1) 3.141(12), Hg(2)-O(1) 2.989(12), Hg(3)-O(1) 2.889(11),
Hg(1)-C(1) 2.109(16), Hg(1)-C(8) 2.096(18), Hg(2)-C(7) 2.096(18),
Hg(2)-C(14) 2.115(16), Hg(3)-C(2) 2.034(19), Hg(3)-C(13) 2.06(2),
O(1)-N(1) 1.306(18), C(8)-Hg(1)-C(1) 175.3(7), C(7)-Hg(2)-C(14)
176.5(7), C(2)-Hg(3)-C(13) 177.1(7). The inset (top, right) shows the
angle R and the distance d used to describe the orientation of an electron-
rich functionality (X-Y) with respect to the trinuclear core of 1.
Figure 2. Extended structure of 2. Dashed lines show short distances
between the TEMPO molecules.
Figure 3. Compound 3 with 50% thermal ellipsoids (fluorine and hydrogen
atoms omitted for clarity). Two molecules are shown. Selected bond
distances (Å) and angles (°): Hg(1)-O(1) 2.877(6), Hg(2)-O(1)
2.975(6), Hg(3)-O(1) 2.846(6), Hg(2)-C(23) 3.328(10), Hg(1)-C(1)
2.071(9), Hg(1)-C(8) 2.066(9), Hg(2)-C(7) 2.071(9), Hg(2)-C(14)
2.088(9), Hg(3)-C(2) 2.065(9), Hg(3)-C(13) 2.066(8), O(1)-N(1)
1.277(9), C(8)-Hg(1)-C(1) 176.1(4), C(7)-Hg(2)-C(14) 176.4(4), C(2)-
Hg(3)-C(13) 173.8(4).
and angular parameters observed in 2 resemble those
encountered in [1•acetone] (R ) 85.6°, d ) 1.945 Å, Hg-O
) 2.810(12)-2.983(12) Å),10 [1•acetaldehyde] (R ) 66.3°,
d ) 2.086 Å, Hg-O ) 2.912(13)-2.965(8) Å),9 [1•(DMF)2]
(R ) 67.7 and 88.3°, d ) 1.990 and 2.062 Å, Hg-O )
2.799(5)-3.042(5) Å).11,12 An examination of the extended
structure of 2 reveals that the shortest intermolecular
distances between neighboring TEMPO molecules do not
involve the N-O functionality but instead occur between
C-H groups (Figure 2). Based on the calculated position of
the hydrogen atoms, these CH‚‚‚HC distances can be
estimated to be in the 2.5-2.7 Å range. Neighboring
molecules of [1•TEMPO] form offset π-π stacking inter-
actions. These interactions (not pictured) occur between the
phenylene ring containing C(1), its symmetry equivalent
generated by [-x + 1, -y + 1, -z + 3] (centroid distance
) 3.70 Å), and the phenylene ring containing C(13) in a
neighboring molecule generated by [-x + 1, -y + 1, -z +
2] (centroid distance ) 4.03 Å). Similar π-π stacking
interactions have been observed in the crystal structure of
other perfluoro-o-phenylene mercury derivatives.27
Compound 3 crystallizes in the monoclinic space group
C2/c (Table 1). The crystal structure of this derivative
consists of isolated C2 symmetric molecules of [1•NIT-Ph•1]
in which a NIT-Ph molecule is sandwiched by two molecules
of 1 (Figure 3). The two oxygen atoms of the NIT-Ph are
triply coordinated to the mercury centers provided by the
juxtaposed molecules of 1. The Hg-O distances (Hg-O )
2.846(6)-2.975(6) Å) as well as the orientation of the linear
N-O functionality with respect to the plane of the trinuclear
mercury complex (R ) 85.1°, d ) 2.025 Å) are similar to
those found in 2. Additional interactions between 1 and NIT-
Ph include two symmetrically equivalent Hg-Caromatic con-
tacts of 3.33(1) Å which occur between Hg(2)/Hg(2a) and
C(23)/C(23A), respectively. As in the case of 2, neighboring
molecules of [1•NIT-Ph•1] form offset π-π stacking
interactions (not pictured), which involve the phenylene ring
containing C(7) and the phenylene ring containing C(13) in
(27) Tschinkl, M.; Schier, A.; Riede, J.; Gabba¨ı F. P. Angew. Chem., Int.
Ed.. 1999, 38, 3547-3549.
Inorganic Chemistry, Vol. 44, No. 18, 2005 6251

Haneline and Gabba1ı
ecules involve C-H groups and not the N-O functionalities.
These CH‚‚‚HC distances (not pictured), in the range of 2.6-
2.8 Å, occur between the phenyl and methyl groups of two
NIT-Ph molecules belonging to parallel chains.
The N-O bond distance (2: O(1)-N(1) 1.31(2); 3:
N(1)-O(1) 1.277(9); 4: N(1)-O(1) 1.28(3) N(2)-O(2)
1.27(3) Å) is similar to that observed in pristine TEMPO
for 2 (1.29 Å)28 and pristine NIT-Ph for 3 and 4 (1.26 and
1.27 Å).29 The absence of significant lengthening of the N-O
bonds is indicative of the weakness of the Hg-O interaction.
In agreement with this view, we note that the C-Hg-C
angle at each mercury atom does not substantially deviate
from linearity.
To provide a better description of the interaction involving
1 and the oxygen atom of the nitroxides or nitronyl nitroxides
radicals, we have studied the electronic structure of 1 using
density functional theory (DFT). Examination of the literature
indicates that few theoretical studies have been carried out
on compound 1. Most studies available to date were obtained
using AM1,30,31 MNDO,32 or molecular mechanics33 methods,
and some of the results do not always agree with experi-
mental data. For these reasons, geometry optimization and
single-point energy calculations were performed using
density functional theory (DFT) in the Amsterdam density
functional package (ADF).34-36 The optimized structure of
1, which has D3h symmetry, features bond distances and
angles that are comparable to those determined for pure 1
by X-ray diffraction (Table 2).37 The HOMO is largely based
on the perfluorophenylene rings of 1 (Figure 5). In contrast,
the LUMO bears large contributions of mercury 6p orbitals
(44%) and exhibits a large lobe in the center of 1 (Figure
5). This large lobe of the LUMO located in the middle of
the three mercury centers suggests that this particular region
of the molecule is where Lewis acidity is at a maximum. In
agreement with this view, this large lobe appears directly
aligned with the direction along which Lewis basic substrates,
including TEMPO and NIT-Ph, bind to the molecule. These
results lend support to the claim that the Lewis acidic
character of 1 is dominated by the empty 6p orbitals on the
mercury. However, the computed magnitude of the HOMO-
LUMO gap, which is equal to 3.357 eV, indicates that the
Figure 4. Portion of a polymeric chain observed in the structure of
compound 4 (30% thermal ellipsoids, fluorine and hydrogen atoms omitted
for clarity). Selected bond distances (Å) and angles (°): Hg(1)-O(1)
3.020(17), Hg(2)-O(1) 2.974(19), Hg(3)-O(1) 2.90(2), Hg(1A)-O(2)
2.91(2), Hg(2A)-O(2) 2.97(2), Hg(3A)-O(2) 2.99(2), Hg(1)-C(1)
2.11(3), Hg(1)-C(8) 2.08(2), Hg(2)-C(7) 2.09(3), Hg(2)-C(14) 2.12(3),
Hg(3)-C(2) 2.10(2), Hg(3)-C(13) 2.09(2), O(1)-N(1) 1.28(3), O(2)-N(2)
1.27(3), C(1)-Hg(1)-C(8) 176.4(10), C(7)-Hg(2)-C(14) 177.1(9), C(2)-
Hg(3)-C(13) 176.3(9).
(28) Bordeaux, D.; Lajzerowicz-Bonneteau, J.; Briere, R.; Lemaire, H.;
Rassat, A. Org. Magn. Reson. 1973, 5, 47-52.
a neighboring molecule generated by [-x + 1, -y, -z]
(centroid distance ) 4.14 Å).
(29) Zheludev, A.; Barone, V.; Bonnet, M.; Delley, B.; Grand, A.;
Ressouche, E.; Rey, P.; Subra, R.; Schweizer, J. J. Am. Chem. Soc.
1994, 116, 2019-2027.
Compound 4 crystallizes in the hexagonal space group P61
(Table 1). Examination of the crystal structure reveals
extended helical binary polymeric chains with alternating
molecules of 1 and NIT-Ph that propagate parallel to one
another (Figure 4). The interactions responsible for the
formation of these infinite chains involve the triple coordina-
tion of the oxygen atoms of the NIT-Ph molecule to the
mercury centers of 1. Thus, the coordination about the
molecules of 1 is similar to that encountered in [1•(L)2]
complexes. As a result, the environment of the NIT-Ph
molecule as well as the coordination geometry about the NO
functionalities is virtually identical to that found in 3 (R )
87.3°, d ) 2.123 Å, Hg-O ) 2.90(2)-3.020(17) Å). The
shortest intermolecular distances between the NIT-Ph mol-
(30) Chistyakov, A. L.; Stankevich, I. V.; Gambaryan, N. P.; Struchkov,
Yu. T.; Yanovsky, A. I.; Tikhonova, I. A.; Shur, V. B. J. Organomet.
Chem. 1997, 536, 413-424.
(31) Castro, E. A. Molecules 2003, 8, 418-429.
(32) Saitkulova, L. N.; Bakhmutova, E. V.; Shubina, E. S.; Tikhonova, I.
A.; Furin, G. G.; Bakhmutov, V. I.; Gambaryan, N. P.; Chistyakov,
A. L.; Stankevich, I. V.; Shur, V. B.; Epstein, L. M. J. Organomet.
Chem. 1999, 585, 201-210.
(33) Castro, E. A. J. Mol. Struct. (THEOCHEM) 2002, 619, 45-50.
(34) ADF2003.01. SCM, Theoretical Chemistry, Vrije Universiteit, Am-
sterdam, The Netherlands.
(35) Guerra, C. F.; Snijders, J. G.; Te Velde, G.; Baerends, E. J. Theor.
Chem. Acc. 1998, 99, 391-403.
(36) Te Velde, G.; Bickelhaupt, F. M.; Baerends, E. J.; Fonseca Guerra,
C.; Van Gisbergen, S. J. A.; Snijders, J. G.; Ziegler, T. J. Comput.
Chem. 2001, 22, 931-967.
(37) Haneline, M. R.; Gabba¨ı, F. P. Z. Naturforsch. 2004, 59B, 1483-
1487.
6252 Inorganic Chemistry, Vol. 44, No. 18, 2005

Organic Radicals Coordinated to Perfluoro-o-phenylene Mercury
Table 1. Crystal Data, Data Collection, and Structure Refinement for 2, 3, and 4
2
3
4
Crystal Data
formula
Mr
C27H18F12Hg3NO
1202.19
0.36 × 0.18 × 0.06
triclinic
C49H17F24Hg6N2O2
2325.19
0.44 × 0.41 × 0.32
monoclinic
C2/c
17.963(4)
15.835(3)
20.652(4)
C31H17F12Hg3N2O2
1279.24
crystal size (mm3)
crystal system
space group
a (Å)
0.34 × 0.28 × 0.22
hexagonal
P?1
P6(1)
11.559(3)
11.559(3)
44.627(14)
10.263(2)
12.009(2)
12.669(3)
87.16(3)
70.83(3)
74.77(3)
1421.9(5)
2
b (Å)
c (Å)
R (°)
â (°)
115.67(3)
γ (°)
V (Å3)
5294.8(18)
4
2.917
17.460
4148
5164(2)
6
2.468
13.441
3486
Z
F
calc (g cm-3
)
2.808
16.259
1086
µ(Mo KR) (mm-1
F(000) (e)
)
Data Collection
T/K
110(2)
110(2)
110(2)
scan mode
hkl range
measured reflns
w
w
w
-11f11, -13f13, -14f14
8204
-21f21, -18f16, -24f24
19319
-12f12, -12f12, -49f47
32596
unique reflns [Rint
reflns used for refinement
abs correction
]
3989 [0.0404]
3989
SADABS
0.266545
4659 [0.0386]
4659
SADABS
0.313625
4893 [0.0737]
4893
Empirical
Tmin/Tmax
0.352/0.827
Refinement
refined params
361
376
397
R1,a wR2b [I > 2σ(I)]
0.0596, 0.1329
3.906, -3.468
0.0272, 0.0643
1.001, -0.988
0.0478, 0.1215
2.348, -1.302
F
fin (max/min) (e Å-3
)
a R1 ) Σ(Fo - Fc)/ΣFo. b wR2 ) {[Σw(Fo - Fc )2]/[Σw(Fo )2]}1/2; w ) 1/[σ2(Fo ) + (ap)2 + bp]; p ) (Fo + 2Fc )/ 3; a ) 0.02 (2), 0.0485 (3), 0.07
(4); b ) 100 (2), 0 (3), 80 (4).
2
2
2
2
2
2
Table 2. Experimental and Calculated Bond Distances (Å) and Angles (°) for the Structure of 1
a Average metrical parameters derived from the crystal structures of four polymorphs of 1.37
Figure 5. Views of the HOMO and LUMO of compound 1.
LUMO might be too high in energy to efficiently mix with
the donor orbitals of the ligand. This conclusion is also
supported by the cyclic voltammogram of 1 in THF using
nBu4PF6 as a supporting electrolyte which does not show
any reduction in the solvent window. As a result, the bonding
in adducts such as 2, 3, and 4 is most likely dominated by
electrostatic rather than covalent interactions. This conclusion
is in agreement with the work of Fackler38 who showed, by
Inorganic Chemistry, Vol. 44, No. 18, 2005 6253

Haneline and Gabba1ı
magnetic interactions. This observation appears to be in
agreement with the large separation of the spin-bearing NIT-
Ph molecules which do not form intermolecular contacts
below 4.8 Å. It also indicates that the molecules of 1 act as
efficient magnetic insulators. It is important to note that the
behavior of [1•NIT-Ph•1] parallels that of solid [1•Cp-
NiCp•1] in which the sandwiched nickelocene molecules
appear insulated from one another.39 In the case of 2 and 4,
ÌmT remains independent of temperature until approximately
10 K, at which point a sharp decrease is observed, indicating
probable long-range anti-ferromagnetic coupling. The Weiss
constant for 2 (θ ) -0.17 K) is in agreement with this
extremely weak coupling. In the case of 4, the Weiss constant
(θ ) -1.41 K) is, in fact, similar to that observed for pristine
NIT-Ph.40 Because of the weakness of these antiferro-
magnetic interactions, a definitive assignment of the coupling
pathways cannot be unequivocally established. In the case
of 2, these weak interactions possibly result from intermo-
lecular interactions between proximal TEMPO molecules
(Figure 2). In the case of 4, the weak anti-ferromagnetic
interactions could be mediated by the trinuclear mercury
complex. However, the observed orientation places the
SOMO of the NIT-Ph molecule, which has π-symmetry,41
almost orthogonal to the LUMO of 1. As a result, the
mediation of magnetic interactions by the trinuclear mercury
complex can only be minimal. Hence, the presence of short
intermolecular distances between NIT-Ph molecules belong-
ing to parallel chains as discussed above could also be
associated with this weak magnetic interaction.
Concluding Remarks
We have demonstrated that 1 forms adducts with organic
nitronyl radicals where the electron-rich terminus of the N-O
moiety interacts with the three mercury centers of 1. The
structure of adducts 2-4 is reminiscent of that observed for
other adducts involving 1 and donors such as organic
carbonyls,8-10 nitriles,11 and sulfoxides.12 The DFT calcula-
tions that we have performed suggest that the LUMO of 1
might be too high in energy to significantly interact with
filled orbitals of the donors. In turn, these investigations
suggest that the interaction between 1 and the radical is
mostly electrostatic rather than covalent. A similar conclusion
can be extended to adducts of 1 with other donors including
organic carbonyls, nitriles, and sulfoxides.3,4,6,8-12 Finally, 1
does not appear to mediate effective magnetic coupling of
nitronyl nitroxides coordinated on either side of its molecular
plane. The high energy of the unoccupied orbitals of 1 as
well as the perpendicular orientation of the nitronyl nitroxide
ring with respect to the plane of the trinuclear complex are
most likely responsible for the absence of significant
coupling. Altogether, these results suggest that 1 may serve
to insulate molecules with unpaired electrons. The properties
of [1•NIT-Ph•1], which do not show any magnetic coupling,
certainly corroborate this view. To further support this
Figure 6. Thermal variation of the XmT product for compound 2 (A),
compound 3 (B), and compound 4 (C) in the 2-300 K range.
a single-point energy DFT calculation, that the electrostatic
potential surface at the center of the trinuclear macrocycle
is positive while the periphery is negative.
Variable-temperature (2-300 K) magnetic susceptibility
data were collected on crushed single crystals of compounds
2-4 (Figure 6). In all three cases the room temperature ÌmT
(0.4 emu K mol-1) is in good agreement with the expected
value for an isolated molecule with one unpaired electron
(S ) 1/2). In the case of 3, which contains isolated sandwiches
of [1•NIT-Ph•1], the ÌmT remains constant over the entire
temperature range, indicating the absence of detectable
(39) Haneline, M. R.; Gabba¨ı, F. P. Angew. Chem., Int. Ed. 2004, 43, 5471-
5474.
(40) Awaga, K.; Maruyama, Y. J. Chem. Phys. 1989, 91, 2743-2747.
(41) D’Anna, J. A.; Wharton, J. H. J. Chem. Phys. 1970, 53, 4047-52.
(38) Burini, A.; Fackler, J. P., Jr.; Galassi, R.; Grant, T. A.; Omary, M.
A.; Rawashdeh-Omary, M. A.; Pietroni, B. R.; Staples, R. J. J. Am.
Chem. Soc. 2000, 122, 11264-11265.
6254 Inorganic Chemistry, Vol. 44, No. 18, 2005

Organic Radicals Coordinated to Perfluoro-o-phenylene Mercury
argument, we are currently focusing on the synthesis of
complexes in which the mercury centers of 1 are directly
facing the SOMO of π-delocalized radicals.
calculations. Acknowledgment is made to the donors of the
American Chemical Society Petroleum Research Fund for
support of this research (Grant ACS PRF# 38143 -AC 3).
M.R.H. thanks Proctor and Gamble for a graduate fellowship.
Acknowledgment. We would like to thank Curtis
Berlinguette, Eric Schelter, and Professor Kim Dunbar for
their assistance with the magnetic measurements and dis-
cussions; James Gardinier for the synthesis of NIT-Ph; Tim
Hugbanks and Lindsay Roy for their help with the ADF
Supporting Information Available: X-ray crystallographic data
for 2, 3, and 4 in CIF format. This material is available free of
charge via the Internet at http://pubs.acs.org.
IC050528A
Inorganic Chemistry, Vol. 44, No. 18, 2005 6255