Influence of the [2.1.1]-(2,6)-pyridinophane macrocycle ring size constraint on the structure and reactivity of copper complexes

Article abstract of DOI:10.1021/ic049886o

The macrocycle [2.1.1]-(2,6)-pyridinophane (L) binds to CuCl to give a monomeric molecule with tridentate binding of the ligand but in a distorted tetrahedral 3 + 1 geometry, where one nitrogen forms a longer (by 0.12 ?) bond to Cu. In dichloromethane solvent this pyridine donor undergoes facile site exchange with a second pyridine in the macrocycle, to give time-averaged mirror symmetry. Both experimental and density functional theory studies of the product of chloride abstraction, using NaBAr^F ₄ in CH₂Cl₂, show that the Cu⁺ binds in a trigonal pyramidal, not planar, arrangement in LCu⁺. This illustrates the ability of macrocyclic ligand constraint to impose an electronically unfavorable geometry on 3-coordinate Cu(I). LCuBAr ^F₄ and a triflate analogue LCu^I(OTf) readily react with oxygen in dichloromethane to produce, in the latter case, a hydroxo-bridged dimer [LCu^II(μ-OH)]₂(OTf)₂, of the intact (unoxidized) ligand L. Since the analogous LCuCl does not react as fast with O₂ in CH₂CI₂, outer-sphere electron transfer is concluded to be ineffective for oxidation of cuprous ion here.

Full text of DOI:10.1021/ic049886o

Inorg. Chem. 2004, 43, 4300−4305
Influence of the [2.1.1]-(2,6)-Pyridinophane Macrocycle Ring Size
Constraint on the Structure and Reactivity of Copper Complexes
A. N. Vedernikov, M. Pink, and K. G. Caulton*
Department of Chemistry and Molecular Structure Center, Indiana UniVersity,
Bloomington, Indiana 47405
Received January 28, 2004
The macrocycle [2.1.1]-(2,6)-pyridinophane (L) binds to CuCl to give a monomeric molecule with tridentate binding
of the ligand but in a distorted tetrahedral “3 + 1” geometry, where one nitrogen forms a longer (by 0.12 Å) bond
to Cu. In dichloromethane solvent this pyridine donor undergoes facile site exchange with a second pyridine in the
macrocycle, to give time-averaged mirror symmetry. Both experimental and density functional theory studies of the
product of chloride abstraction, using NaBArF4 in CH Cl2, show that the Cu+ binds in a trigonal pyramidal, not
2
planar, arrangement in LCu+. This illustrates the ability of macrocyclic ligand constraint to impose an electronically
I
unfavorable geometry on 3-coordinate Cu(I). LCuBArF4 and a triflate analogue LCu(OTf) readily react with oxygen
II
in dichloromethane to produce, in the latter case, a hydroxo-bridged dimer [LCu (µ-OH)]2(OTf)2, of the intact
(unoxidized) ligand L. Since the analogous LCuCl does not react as fast with O in CH Cl2, outer-sphere electron
2
2
transfer is concluded to be ineffective for oxidation of cuprous ion here.
Introduction
necessarily facial (i.e. all cis) on the metal. Even the free
211L ligand itself does not have the nitrogen lone pairs
coplanar or converging.2,3 In some cases, the facial geometry
will lead to cis-N-M-N angles highly distorted from those
preferred by the metal, which in turn may make η2-binding
to the metal more favorable than η3-binding. This is the case
The pyridinophane ligand L is a newly synthesized ligand
designed to complement centrally connected tripodal ligands
carrying three neutral nitrogen donors such as HB(pz)3- and
HC(pz)3, the tris(pyrazolyl)borates and -methanes.1 The
design element of the [n.m.p]-pyridinophanes of special
+ 4
for divalent Pt in LPtCH3 , LPtH+,5 and also in LMX2 (M
) Pd, Pt, X ) Cl, CH3),2,3 and the pendant nitrogen can
then play an active and stabilizing role in chemical reactiv-
ity.6 It influences both reactant and product. In brief, it has
been shown to form a bond to PtIV, after oxidation, and thus
provide product stability (octahedral PtIV) for even the
challenging cleavage of an alkane C-H bond. Because it is
an intramolecular effect, it occurs without the entropy penalty
involved with coordinating a truly free (i.e. solution phase)
ligand to product. It can also accept a proton during the
heterolytic cleavage of an arene C-H bond,6 again providing
some enthalpic benefit at minimal entropy cost. The resulting
pendant pyridinium functionality can then hydrogen bond
importance is the length of the linker elements (CH2)q, which
governs the macrocycle size and thus controls the coordina-
tion geometry of the three pyridyl nitrogens around a single
metal.2 Thus, for smaller macrocyles, the metal cannot fit
into the 211L macrocycle (i.e. cannot become coplanar with
all three N), and the geometry of the pyridinophane will be
II
2
to the filled dz orbital of a planar 4-coordinate Pt product,
* Author to whom correspondence should be addressed. E-mail:
caulton@indiana.edu.
(3) Vedernikov, A. N.; Huffman, J. C.; Caulton, K. G. Inorg. Chem. 2002,
41, 6867.
(1) Trofimenko, S. Scorpionates: The Coordination Chemistry of Poly-
(4) Vedernikov, A. N.; Huffman, J. C.; Caulton, K. G. New J. Chem.
2003, 27, 665-667.
pyrazolylborate ligands; Imperial College Press: River Edge, NJ,
1999.
(2) Vedernikov, A. N.; Pink, M.; Caulton, K. G. J. Org. Chem. 2003, 68,
(5) Vedernikov, A. N.; Caulton, K. G. Angew. Chem., Int. Ed. 2002, 41,
4102-4104.
4806.
(6) Vedernikov, A. N.; Caulton, K. G. Chem. Commun. 2003, 358.
4300 Inorganic Chemistry, Vol. 43, No. 14, 2004
10.1021/ic049886o CCC: $27.50 © 2004 American Chemical Society
Published on Web 05/25/2004

Influence of Macrocycle Ring Size Constraints
or it can participate in an intramolecular redox equilibrium
via the hydrogen transfer illustrated in eq 1.2 We have also
It is with an eye to exploring reactivity different from the
reported Pt chemistry that we selected copper for testing of
the macrocyclic constraint idea. This has led to the recent
demonstration that copper complexes of [2.1.1]-2,6 pyridi-
nophane catalyze rapid olefin aziridination.11 They also
catalyze oxidation of saturated hydrocarbons to olefins by
PhINSO2(tolyl) a reaction in which nitrogen receives the
hydrogens removed in oxidizing alkane to olefin.12 This
report examines the structural outcome of conflicting steric
and electronic preferences when the [2.1.1]-(2,6)-pyridi-
nophane, L, is bound to CuCl where CuI would normally
prefer tetrahedral geometry. The origin of distortions is
analyzed both by comparison to the structure of an acyclic
analogue of the pyridinophane bound to CuCl13 and also by
computational experiments where structures of systematically
modified LCu+ species are determined by geometry opti-
mization using density functional theory (DFT14 (PBE15)).
Reactivity of the LCuI fragment toward dioxygen is also
reported and shows the oxidation to occur at the metal but
apparently only by electron transfer in an inner-sphere
process which depends on O2 first coordinating to Cu(I).
This study of the geometric distortions imposed by this
pyridinophane macrocycle on Cu(I) and even Cu(II) therefore
has enhanced importance in the exploitation of the macro-
cycle constraint idea.
established2 that such ligand constraints can impose metal
complex reactant destabilization, which has the obvious effect
of improving reaction thermodynamics (via enthalpy); this
can be especially important in trying to cleave strong and
poor donor bonds (e.g. alkane C-H) where we have shown
computationally that CH4 bond cleavage by a typical planar
4 coordinate PtII species is endergonic but reactant destabi-
lization by progressive ring constraints by nmpL can lead to
an exergonic methane oxidative addition reaction. Experi-
mental determination of the structure of some four-coordinate
211LMX2 species (M ) Pd, Pt; X ) Cl, CH3) has actually
revealed more: the nonbonded pyridine has an N/M distance
(2.658 Å) short enough to reveal3 a filled/filled repulsion
2
between the pendant N lone pair and the filled dz orbital.
This has the result of increasing the reducing power of the
species because, as in hydrazine and other diamines which
also have filled/filled repulsion, the filled antibonding
combination of the two lone pairs has a lower ionization
potential than does either lone pair alone. This then traces
the origin of generalized reactant destabilization to its orbital
origin.7 While the Bronsted basicity functionality has been
demonstrated for other multidentate amine or imine ligands
when they bind to a metal with less than their full comple-
ment of nitrogen lone pairs,8-10 comparison of those ex-
amples to pyridinophane Bronsted basicity can begin to
establish how ligand internal connectivity and consequent
angular constraints influences the effectiveness of such
behavior. This paper begins such comparisons.
Results and Discussion
LCuCl. Synthesis, Structure, and Solution NMR Be-
havior. The complex was synthesized by reaction of L with
CuCl in benzene under anaerobic conditions. The molecule
crystallizes (Figure 1) with no rigorous symmetry in a polar
space group as a molecular species (Tables 1 and 2). In the
LCuCl species, four coordinate Cu(I) is distorted from
tetrahedral symmetry in a way which makes the pyridine
rings involving N1 and N2 equivalent (by angles Cl-Cu-N
and by their Cu-N distances) and N3 distinct (longer Cu-N
distance by 0.12 Å and a smaller Cl-Cu-N angle, by over
14°). All of these distortions can be ascribed to the N2
pyridine ring not being able to direct both ortho CH2 groups
simultaneously in a way that will accommodate the N1 and
N3 pyridine rings, and apparently it is the ring with the
longer, more flexible CH2CH2 linker that is left to “suffer.”
The largest N-Cu-N angle involves the CH2CH2 linker (and
thus a seven-membered NCuNC4 ring).
We have more recently tried to evaluate the suitability of
the 211L macrocycle to dictate reactivity enhancement on
other metals. This objective represents significant new “ter-
rain” since it involves other preferred coordination geometries
than the planar/octahedral pair relevant to our exploration
of d8/d6 platinum chemistry. We are, moreover, interested
in moving to the less expensive 3d metals, and these im-
mediately encounter (1) the relative inapplicability of the
16/18 electron rule and (2) the frequent occurrence of one-
and not two-electron redox changes. These are of course
related characteristics but ones which were not true of our
use of pyridinophane on platinum. Moreover, it would appear
that some alkane C-H cleavage reactions by metalloproteins
can occur without formation of any metal-carbon bond: the
classic oxidative addition (eq 2) or even electrophilic
cleavage reactions (eq 3) are not relevant.
An excellent comparison compound is the CuCl com-
plex of the nonmacrocyclic tpdm ligand, which shows that
the greater ligand flexibility/adaptability permits a larger
(11) Vedernikov, A. N.; Caulton, K. G. Org. Lett. 2003, 5, 2591.
(12) Vedernikov, A. N.; Caulton, K. G. Chem. Commun. 2004, 162.
(13) Vedernikov, A. N.; Huffman, J. C.; Caulton, K. G. Inorg. Chem. 2002,
41, 6244-6248.
(14) Parr, R. G.; Yang, W. Density-functional theory of atoms and
molecules; Oxford University Press: Oxford, U.K., 1989.
(15) Perdew, J. P.; Burke, K.; Ernzerhof, M. Phys. ReV. Lett. 1996, 77,
3865.
(7) Canty, A. J.; van Koten, G. Acc. Chem. Res. 1995, 28, 406.
(8) Jenkins, H. A.; Klempner, M. J.; Prokopchuk, E. M.; Puddephatt, R.
J. Inorg. Chim. Acta 2003, 72.
(9) Prokopchuk, E. M. P., R. J. Organometallics 2003, 22, 787.
(10) Puddephatt, R. J. Coord. Chem. ReV. 2001, 157 and references therein.
(16) Que, L.; Tolman, W. B. Angew. Chem., Int. Ed. 2002, 41, 1114.
Inorganic Chemistry, Vol. 43, No. 14, 2004 4301

Vedernikov et al.
Figure 1. ORTEP drawing of LCuCl, omitting hydrogen atoms and
showing selected atom labeling (50% probability level).
Table 1. Crystal Data and Structure Refinement for LCuCl and
[LCu(µ-OH)]2(OTf)2
LCuCl
[LCu(µ-OH)]2(OTf)2
empirical formula
fw
C
19.33H17.67Cl1.67CuN3
C41H38Cl2Cu2F6N6O8S2
1118.87
414.65
cryst color, shape,
yellow needle,
0.40 × 0.10 × 0.10
136(2)
0.710 73
hexagonal, P61
blue plate,
Figure 2. Two views of the experimentally determined structure, (a)
viewing the ClCuN2N1 plane edge-on and (b) rotated 90° about the Cu-
Cl vector from (a).
size (mm3)
0.12 × 0.10 × 0.03
temp (K)
113(2)
wavelength (Å)
cryst system,
space group
unit cell dimens
(Å, deg)
0.710 73
triclinic, P1h
a ) 21.0593(16)
b ) 21.0593
c ) 7.5642(6)
R ) 90
â ) 90
γ ) 120
2905.2(3)
6
a ) 9.9327(9)
b ) 10.6687(10)
c ) 11.3605(10)
R ) 83.677(2)
â ) 69.113(2)
γ ) 74.624(2)
1084.33(17)
1
V (Å3)
Z
Figure 3. Calculated (DFT) geometry-optimized structure of the three-
coordinate cation LCu+: Cu-N3, 1.975 Å; Cu-N2, 1.940 Å; Cu-N1,
1.927 Å; N3-Cu-N2, 104.7°; N1-Cu-N3, 102.9°; N2-Cu-N1, 122.7°.
d(calcd) (Mg/mm3) 1.422
1.713
1.288
abs coeff (mm-1
final R indices
[I > 2σ(I)]
)
1.364
R1 ) 0.0350,
R1 ) 0.0536,
wR2 ) 0.0811a,b
wR2 ) 0.1264a,b
(d) The angular distortion from tetrahedral (or 3-fold) sym-
metry involves Cl bending toward N3, which is the weakly
bound nitrogen.
a R1 ) ∑(|Fo| - |Fc|)/∑|Fo|. b wR2 ) [∑[w(Fo2 - Fc )2]/∑[w(Fo )2]]1/2
.
2
2
Table 2. Selected Bond Lengths (Å) and Angles (deg) for LCuCl
All parameters of the DFT-calculated structure of LCuCl
match closely (e2% for bond lengths, e2° for bond angles)
those found by X-ray diffraction in the solid state.
Despite the lack of mirror symmetry in the solid state
caused by the staggered gauche conformation of the mac-
rocycle ethylene bridge, LCuCl behaves as a mirror sym-
metric species on the 1H NMR time scale in CD2Cl2 solution
at 22 °C. The macrocycle exhibits two doublets for the
methylene bridges at 4.11 and 4.71 ppm, three doublets of
equal intensity for the meta-protons of the macrocycle
pyridine residues, and two triplets of corresponding para-
protons at 7.50 and 7.60 in 2:1 ratio. The spectral pattern
does not change at -45 °C. The 13C NMR spectrum is also
consistent with this conclusion about the symmetry of LCuCl
in solution.
Cu1-N1
Cu1-N2
2.052(3)
2.069(2)
Cu1-N3
Cu1-Cl1
2.190(3)
2.2305(8)
N1-Cu1-N2
N1-Cu1-N3
N2-Cu1-N3
90.30(10)
99.65(11)
90.32(10)
N1-Cu1-Cl1
N2-Cu1-Cl1
N3-Cu1-Cl1
129.28(8)
126.53(7)
111.96(8)
Nt-Cu-Nt, more equal Cu-N distances, and thus ap-
proximate mirror symmetry.13
Three-Coordinate Nonplanar LCu+? a. Computed
Geometry. To establish the CuI binding preference of L in
the absence of a fourth ligand, the structure of LCu+ was
also optimized by a DFT (PBE) method.14,15 The result
(Figure 3) shows clearly that CuI cannot fit into the N3 plane,
and so the trigonal planar geometry found for a three-
coordinate d10 metal must be distorted. The N-Cu-N angles
are all larger than in LCuCl, and all Cu-N distances are
shorter, all consistent with Cu moving closer to the N3 plane.
As in LCuCl, the N-Cu-N distance involving the longer
Figure 2a,b illustrates the following:
(a) A “3 + 1” designation of the coordination geometry
is appropriate with the long Cu-N3 distance symbolized by
“1”.
(b) Cu is close to coplanar with the “3” ligands designated
above, Cl, N2, and N1.
(c) The more weakly bound py (N3) is displaced angularly
(toward N2 and thus off the idealized mirror plane relating
N2 and N1) by the longer link, (CH2)2.
4302 Inorganic Chemistry, Vol. 43, No. 14, 2004

Influence of Macrocycle Ring Size Constraints
-CH2CH2- linker is larger, and there is enough flexibility
that the hydrogens in this linker can adopt a staggered
conformation. The Cu/N distances in LCu+, although they
are distinctly different (by 0.05 Å), do not show as great a
variation as those in LCuCl (0.137 Å).
b. Experiment. The formally three coordinated cation
LCu+ could be generated by abstraction of chloride ligand
by NaBArF (BArF ) tetrakis[3,5-bis(trifluoromethyl)-
4
4
phenyl]borate) in the low coordinating solvent dichlo-
romethane. Stirring a slight excess of this reagent with a
dichloromethane solution of LCuCl at room temperature
leads to its fast dissolution and then the appearance of another
precipitate. The resulting light-yellow solution exhibited 1H
and 13C NMR spectra consistent with a formula for the
Figure 4. ORTEP drawing of the centrosymmetric dication in L2Cu2(µ-
OH)2(OTf)2, omitting hydrogen atoms and showing selected atom labeling
(50% probability level).
Table 3. Selected Bond Lengths (Å) and Angles (deg) in the Ion
2+
soluble product as LCuBArF , containing a cation of mirror
[LCu(µ-OH)]2
4
symmetry at both 22 and -45 °C. The signals of its
methylene bridges appeared as a broadened AB quartet while
protons of the ethylene bridge showed two broad singlets at
22 °C, which resolved at -45 °C into multiplets. At both
temperatures the aromatic signals showed three sharp
doublets and two triplets. Thus, in accordance with DFT
calculations, the resulting three-coordinated copper complex,
which might weakly coordinate dichloromethane, remains
pyramidal and does not display any evidence of the copper
atom to migrate through the macrocycle hole, which would
lead to coalescence of signals of the (CH2)n bridges.
Cu1-O1
Cu1-O1#1
Cu1-N1
1.930(5)
1.962(5)
2.108(3)
Cu1-N3
2.111(4)
2.135(4)
3.0013(11)
Cu1-N2
Cu1-Cu1#1
O1-Cu1-O1#1
O1-Cu1-N1
O1#1-Cu1-N1
O1-Cu1-N3
O1#1-Cu1-N3
N1-Cu1-N3
79.1(2)
90.3(2)
167.5(2)
150.09(19)
91.75(18)
94.37(16)
O1-Cu1-N2
O1#1-Cu1-N2
N1-Cu1-N2
N3-Cu1-N2
Cu1-O1-Cu1#1
123.86(18)
103.68(18)
87.66(16)
85.90(14)
100.9(2)
(forming CuII) or two (forming CuIII), as well as about the
ability of the pyridinophane macrocycle attached to the metal
to resist oxidative degradation. The latter might occur first
at the thermodynamically most vulnerable benzylic carbons
linking three pyridine rings together.17 However, there are
also many examples of pendant aryl “hydroxylation.”16 The
effect, if any, of the ability of X to remain coordinated to
copper on this reactivity can shed some light on the nature
of plausible reaction intermediates. Indeed, three copper(I)
Reactivity under Oxidizing Conditions. a. General
Comments. The observed11 copper-catalyzed olefin aziri-
dination must begin by coordination of PhINTs, to Cu, and
perhaps even loss of PhI to leave behind a copper/NTs
species which contains oxidized copper (CuIII or CuII); this
conclusion follows because there is no detectable olefin
coordination to the L/Cu species. Similarly, spectroscopic
and ESI-mass spectrometric observation11 of the L/Cu +
alkane + PhINTs reacting system has shown the presence
of NTs complexed to one, and also two copper centers, and
with oxidation states higher than +1. DFT calculations
show11 that several Cu(II) species seen by ESI-MS and
pyridinophane complexes studied here (X ) BArF , OTf, Cl)
4
showed significantly different reactivities.
i. BArF Salt. When a solution of LCuBArF in dichlo-
4
4
romethane was exposed at -90 °C to oxygen (2:1 mole
ratio), an immediate dark brown color appeared which turned
to green after the sample was allowed to warm to room
temperature. Removal of solvent produced a dark green oil,
which resisted attempts at crystallization.
1
detected by H NMR spectroscopy are at energies low
enough to make them candidates for participants in the
oxidation of alkane to olefin.12 It thus becomes important to
begin to define the reaction chemistry of LCuX with aprotic
oxidants in non-Bronsted acidic medium. This takes on
enhanced importance because these two examples of catalysis
by 211LCu catalyst precursors both show evidence of limited
catalyst lifetime, suggesting (oxidative) catalyst degradation.
We therefore studied the reactivity of LCuX species toward
O2, an oxidant for which there is a vast resource of literature
precedent.16 Only studies in the most recent 10 years have
used low-temperature conditions to detect primary products
(ligated CuO2+ and Cu2O22+, the latter as redox isomers) and
show the complexity of these reactions that ultimately
produce secondary products which frequently involve (in-
tramolecular) attack on ligand C-H bonds.
-
ii. CF3SO3 Salt. A similar experiment performed with
LCu(OTf) showed similar but much slower reactivity. The
reaction mixture, warmed to room temperature, changed its
color from dark brown to yellow-green. All LCu(OTf) was
consumed in 30 min as evidenced by 1H NMR. Standing 12
h at 22 °C lead to formation of small bluish-green crystals
of [(η3-L)Cu(µ-OH)]2(OTf)2 suitable for X-ray structure
determination in 40% isolated yield.
The product is a dimer, centrosymmetric in the unit cell,
with two Cu(II) bridged by two hydroxides (Tables 1 and 3
and Figure 4). Each hydroxide proton18 is hydrogen bonded
to a triflate oxygen with an O/O distance of 2.58 Å. The
Cu-OH distances, 1.930(5) and 1.962(5) Å, are too long to
belong to an oxo ligand on Cu(III) (typically 1.80 Å16,19
)
b. Reactions of LCuX (X ) BArF , OTf, Cl) with
4
Oxygen. Reactivity of complexes of general formula LCuX
toward oxygen may provide important information about the
ability of this new species to be oxidized by one electron
(17) Spodine, E.; Manzur, J.; Garland, M. T.; Fackler, J. P., Jr.; Staples,
R. J.; Trzcinska-Bancroft, B. Inorg. Chim. Acta 1993, 203, 73.
(18) The origin of the hydroxyl proton was not determined.
Inorganic Chemistry, Vol. 43, No. 14, 2004 4303

Vedernikov et al.
2+
and fall in the range typical for CuII (µ-OH)2 units. The
Solvents were dried and distilled by following standard protocols
and stored in gastight bulbs under argon. All reagents for which a
synthesis is not given are commercially available from Aldrich or
Pressure Chemicals and were used as received without further
purification. All NMR solvents were dried, vacuum-transferred, and
stored in an argon-filled glovebox. [2.1.1]-(2,6)-pyridinophane was
synthesized according to the published procedure.3 1H and 13C NMR
spectra were recorded on Inova 700 spectrometer (1H 400 MHz;
2
near-3-fold symmetry of the pyridinophane leaves all three
Cu-N distances nearly identical (2.108(3)-2.135(4) Å); the
N-Cu-Cu′ angles fall in a narrow range (120.86(10)-
129.86(15)°), which indicates that “square pyramidal” is an
imperfect description of the five coordinate polyhedron. As
seen in Figure 4, the closest approach to square pyramidal
geometry would have N2 as the apical group. The “bite”
angles of the pyridinophane (N-Cu-N) range from 85.90-
(14) to 94.37(16)°, which is generally 5° smaller than in
LCuCl.
iii. Cl Complex. Finally, a dichloromethane solution of
the chloro complex exposed to oxygen reacted very slowly
even at 22 °C so that its transformation was not complete
even in 1 month to produce an amorphous green precipitate.
Thus, stronger coordination of X to the metal in (formally)
cationic LCu+ inhibits oxidation of copper(I) in these
pyridinophane complexes. These facts imply that precoor-
dination of oxygen to copper is necessary for redox reaction
and outer-sphere electron transfer is not involved. The
observed color changes, in particular, the initial brown color
1
13C 100.62 MHz). H and 13C NMR chemical shifts are reported
in ppm and referenced to residual solvent resonance peaks.
Computational Details. Theoretical calculations in this work
have been performed using the density functional theory (DFT)
method,14 specifically functional PBE,15 implemented in an original
program package “Priroda”.20 The applicability of the PBE func-
tional for calculation of geometry of Cu(I) and Cu(II) complexes
as well as its comparison with B3LYP was described before.21 In
PBE calculations, relativistic Stevens-Basch-Krauss (SBK) ef-
fective core potentials (ECP)22-24 optimized for DFT calculations
have been used. The basis set was 311-split for main group elements
with one additional polarization p-function for hydrogen and an
additional two polarization d-functions for elements of higher
periods. Full geometry optimization has been performed without
constraints on symmetry. For all species under investigation
frequency analysis has been carried out. All minima have been
checked for the absence of imaginary frequencies.
LCuCl, C19H17ClN3Cu. To a dry flask containing a magnetic
stirring bar was added, in a glovebox, CuCl (99 mg, 1.0 mmol)
and 20.0 mL of dry benzene. To the stirred mixture was added
[2.1.1]-(2,6)-pyridinophane (287 mg, 1.0 mmol) dissolved in 5.0
mL of benzene. Stirring was continued at room temperature for 5
h. The orange precipitate formed was filtered off, washed twice
with 1.0 mL portions of benzene, and dried. Yield: 320 g (80%).
Long yellow needles suitable for X-ray structural determination
have been obtained by slow evaporation of a saturated solution
LCuCl in a dichloromethane-benzene mixture.
2+
and ultimate formation of [LCu(µ-OH)]2 species, are
similar to those previously observed for tacn complexes19
and indicate possible formation of an intermediate like
[LCuIII(µ-O)]22+. Finally, isolation and characterization of
[LCu(µ-OH)]2(OTf)2, whose benzylic protons are intact,
shows some resistance of the macrocycle toward oxidation.
Conclusions
Systematic modification of the nucleophilicity of the anion
X in LCuX has revealed that chloride, the least likely ligand
to dissociate to free X- in CH2Cl2 solvent, is in fact quite
protected against reaction with O2. This suggests that O2
oxidizes these Cu(I) species only by an inner-sphere mech-
anism, via coordinated O2, and not effectively by an outer-
sphere mechanism. Clearly this demonstration hinges on a
coordination number of 4 being retained, so it indicates not
only retention of the Cu-Cl bond but also the persistence
of η3 binding of the pyridinophane, despite the observed
variation of Cu-N distances in LCuCl. This suggests that
the impact of the macrocycle here is not so much in reactant
destabilization but that destabilization occurs instead when
copper reaches higher oxidation states. The structure of
1H NMR (CD2Cl2, 22 °C): 3.25 (m, 2H, C2H4), 4.11 (d, J )
13.0 Hz, 2H, CH2), 4.38 (m, 2H, C2H4), 4.71 (d, J ) 13.0 Hz, 2H,
CH2), 7.10 (d, J ) 8.0 Hz, 2H, meta-CH), 7.12 (d, J ) 7.8 Hz,
2H, meta-CH), 7.25 (d, J ) 7.8 Hz, 2H, meta-CH), 7.50 (d, J )
7.8 Hz, 2H, para-CH), 7.60 (d, J ) 7.8 Hz, 1H, para-CH).
13C NMR (CD2Cl2, -45 °C): 35.91 (CH2), 46.54 (C2H4), 122.03,
122.26, 123.05 (meta-C, py), 137.25, 137.87 (para-C, py), 155.02,
155.51, 159.70 (ortho-C, py).
Reactions of LCuCl with NaBArF . In an argon-filled glovebox
4
a Teflon-capped NMR Young tube was charged with 7.7 mg (20
µmol) of LCuCl, 18 mg (>20 µmol) of NaBArF , and 0.6 mL of
4
CD2Cl2. The tube was closed, removed from the glovebox, and
put into a rotating clamp to allow constant tumbling of the reaction
mixture. In few minutes large chunks of NaBArF4 dissolved, a new
fine precipitate appeared, and the liquid color turned light-yellow.
According to NMR data, all starting complex was consumed
2+
[LCuII(OH)]2 shows that the macrocycle nicely accom-
modates the 90° ( 4° angles of a facial ligand geometry, so
it is perhaps at CuIII (d8 planar) that the biggest impact of a
macrocycle constraint should be sought in the future. When
the ligand is dissociation prone or dissociated (CF3SO3- and
suggesting that the yield of LCuBArF is almost quantitative.
4
F
1H NMR (CD2Cl2, -45 °C): 3.32 (m, 2H, C2H4), 3.91 (m, 2H,
C2H4), 4.16 (d, J ) 13.8 Hz, 2H, CH2), 4.24 (d, J ) 13.8 Hz, 2H,
CH2), 7.17 (d, J ) 7.8 Hz, 2H, meta-CH), 7.22 (d, J ) 7.8 Hz,
2H, meta-CH), 7.32 (d, J ) 7.8 Hz, 2H, meta-CH), 7.52 (br s, 4H,
BAr4 , respectively), then the fact that copper is in an unusual
pyramidal geometry for coordination number 3 (Figure 3)
will, however, enhance the enthalpy of its reactions. Note
also the flexibility of the 2.1.1-pyridinophane, in that N-
Cu-N varies from 90° (in (LCuOH)22+) to 123° (in LCu+).
(20) Ustynyuk, Y. A.; Ustynyuk, L. Y.; Laikov, D. N.; Lunin, V. V. J.
Organomet. Chem. 2000, 597, 182.
(21) Vedernikov, A. N.; Wu, P.; Huffman, J. C.; Caulton, K. G. Inorg.
Chim. Acta 2002, 330, 103.
Experimental Section
General Methods. All manipulations were carried out under
(22) Cundari, T. R.; Stevens, W. J. J. Chem. Phys. 1993, 98, 5555.
(23) Stevens, W. J.; Basch, H.; Krauss, M. J. Chem. Phys. 1984, 81, 6026.
(24) Stevens, W. J.; Basch, H.; Krauss, M.; Jasien, P. Can. J. Chem. 1992,
70, 612.
purified argon using standard Schlenk and glovebox techniques.
(19) Aboelella, N. W.; Lewis, E. A.; Reynolds, A. M.; Brennessel, W. W.;
Cramer, C. J.; Tolman, W. B. J. Am. Chem. Soc. 2002, 124, 10660.
4304 Inorganic Chemistry, Vol. 43, No. 14, 2004

Influence of Macrocycle Ring Size Constraints
para-CH, BArF ), 7.57 (d, J ) 7.8 Hz, 2H, para-CH), 7.66 (d, J )
0.10 × 0.10 mm3) was placed onto the tip of a 0.1 mm diameter
glass capillary and mounted on a SMART6000 (Bruker) at 136 K.
The data collection was carried out using Mo KR radiation (graphite
monochromator) with a frame time of 30 s and a detector distance
of 5.1 cm. Three major sections of frames were collected with 0.30°
steps in ω at three different φ settings and a detector position of
-43° in 2θ. An additional set of 50 frames was collected to model
decay. Data to a resolution of 0.77 Å were considered in the
reduction. Final cell constants were calculated from the xyz centroids
of 6999 strong reflections from the actual data collection after
integration (SAINT)25 The intensity data were corrected for
absorption (SADABS).26 The space group P61 was determined on
the basis of intensity statistics and systematic absences. A direct-
methods solution (SIR-9227) was calculated which provided most
non-hydrogen atoms from the E-map. Full-matrix least-squares/
difference Fourier cycles were performed which located the
remaining non-hydrogen atoms (SHELXL-9728). All non-hydrogen
atoms were refined with anisotropic displacement parameters except
for the atoms belonging to the disordered solvent. The hydrogen
atoms were placed in ideal positions and refined as riding atoms
with relative isotropic displacement parameters. The remaining
electron density is located in the solvent channel and in the vicinity
of the copper and chlorine atoms of the complex.
4
7.8 Hz, 1H, para-CH), 7.71 (br s, 8H, ortho-CH, BArF ).
4
13C NMR (CD2Cl2, -45 °C): 35.97 (CH2), 46.55 (C2H4), 117.73
(m, para-C, BArF ), 123.02, 123.28, 123.97 (meta-C, py), 124.66
4
(quart, 1JC-F ) 272.7 Hz, CF3, BArF ), 128.88 (quart, 2JC-F ) 31.7
4
Hz, meta-C, BArF ), 134.89 (br s, ortho-C, BArF ), 138.89, 139.63
4
4
(para-C, py), 155.12, 155.73, 159.64 (ortho-C, py), 161.92 (quart,
1JB-C ) 50 Hz, B-C, BArF ).
4
LCu(OTf), C20H17F3N3SO3Cu. In a glovebox, to a dry flask
containing a magnetic stirring bar were added CuOTf‚0.5(PhMe)
(258 mg, 1.0 mmol) and 20.0 mL of dry benzene. To the stirred
mixture, [2.1.1]-(2,6)-pyridinophane (287 mg, 1.0 mmol) dissolved
in 5.0 mL of benzene was added. Stirring was continued at room
temperature for 5 h. The light yellowish-green precipitate formed
was filtered off, washed twice with 1.0 mL portions of benzene,
and dried. Yield: 450 g (90%).
Signals of LCu(OTf) in its 1H NMR spectrum are broad at room
temperature and become more narrow at -30 °C.
1H NMR (CD2Cl2, -30 °C): 3.27 (m, 2H, C2H4), 4.11 (m, 2H,
C2H4), 4.12 (d, J ) 13.0 Hz, 2H, CH2), 4.47 (d, J ) 13.0 Hz, 2H,
CH2), 7.14 (d, J ) 7.8 Hz, 2H, meta-CH), 7.17 (d, J ) 7.8 Hz,
2H, meta-CH), 7.28 (d, J ) 7.8 Hz, 2H, meta-CH), 7.54 (d, J )
7.7 Hz, 2H, para-CH), 7.62 (d, J ) 7.8 Hz, 1H, para-CH).
13C NMR (CD2Cl2, -30 °C): 35.90 (CH2), 46.64 (C2H4), 122.43,
122.61, 123.42 (meta-C, py), 128.44 (CF3), 137.68, 138.30 (para-
C, py), 155.21, 155.64, 159.33 (ortho-C, py).
X-ray Diffraction Structure Determination of [L2Cu2(OH)2-
(O3SCF3)2‚CH2Cl2]. A blue crystal (approximate dimensions 0.12
× 0.10 × 0.03 mm3) was placed onto the tip of a 0.1 mm diameter
glass capillary and mounted on a SMART6000 (Bruker) at 113(2)
K. Data collection, solution, and refinement were generally as
discussed above. A frame time of 80 s was used. Data to a resolution
of 0.84 Å were considered in the reduction, and the intensity data
were corrected for absorption (SADABS).26
Reactions of LCuX (X ) BArF , OTf, Cl) with O2. In an argon-
4
filled glovebox, a Teflon-capped NMR Young tube was charged
with 40 µmol of corresponding LCuX dissolved in CD2Cl2 (1 mL,
X ) BArF ; 1.5 mL, X ) Cl; 2.0 mL, X ) OTf). The tube was
4
closed, removed from the glovebox, and cooled to -90 °C in an
acetone bath. Argon was removed under vacuum, and the tube was
filled with air (2.5 mL at 1 atm and 25 °C; 20 µmol of O2).
Case a. In the case of X ) BArF4 a dark-brown color developed
immediately; the change of color was slower in the case of X )
OTf; no color changes was observed in the case of X ) Cl. The
tube was allowed to warm to 22 °C; the color turned green (X )
There is a disorder of the pyridine ring that is attached to two
different bridges, CH2 and C2H4; it involves interchange of the CH2
and C2H4 groups and consequent displacement of the pyridine ring
(occupancy ratio 71:29). There is likewise a disorder of the
hydroxide to a second position (68:32 occupancy). Disordered
alternatives are labeled with a suffix “D”. Given the close approach
of these Cu2(µ-O) oxygens to triflate oxygens, there is no doubt
that there are protons on the bridging atoms causing the short
contact.
BArF ) or yellow-green (X ) OTf). According to NMR data, the
4
conversion of the starting complex was complete (X ) BArF ) in
4
less than 10 or 30 (X ) OTf) min. Removal of solvent afforded
green noncrystallizable oil (X ) BArF ).
4
Case b. In the case of X ) OTf, after 12 h, small bluish-green
crystals of [(η3-L)Cu(µ-OH)]2(OTf)2 suitable for X-ray diffraction
were collected and washed with a few drops of dichloromethane.
Yield: 7.5 mg (40%). The 1H NMR spectrum of the product
exhibited broad but distinct signals of [(η3-L)Cu(µ-OH)]2(OTf)2.
1H NMR (CD2Cl2, 22 °C): -0.52 (br s, 2H, CH2), 1.39 (br s,
2H, CH2), 3.03-3.49 (m, 2H, C2H4), 3.70-4.17 (m, 2H, C2H4),
9.55 (br s, 2H, para-CH), 10.15 (br s, 1H, para-CH), 13.99 (br s,
2H, meta-CH), 16.61 (br s, 2H, meta-CH), 17.34 (br s, 2H, meta-
CH).
Case c. Oxidation of LCuCl was not complete (50% conversion)
even in 1 month. A small amount of amorphous bluish-green
precipitate of unknown composition had then formed on the bottom
of the vessel.
X-ray Diffraction Structure Determination of [LCuCl‚
0.333CH2Cl2]. A yellow crystal (approximate dimensions 0.40 ×
Acknowledgment. This work was supported by the DOE.
A.N.V. is on leave from the Chemical Faculty, Kazan State
University, Kazan, Russia. This work has been made possible
in part due to support from the Russian Foundation for Basic
Research (Grant No. 01.03.32692).
Supporting Information Available: Full crystallographic
details (CIF and pdf files). This material is available free of charge
via the Internet at http://pubs.acs.org.
IC049886O
(25) Bruker Analytical X-ray Systems, Madison, WI.
(26) Blessing, R. Acta Crystallogr. 1995, A51, 33.
(27) SIR-92: Altomare, A.; Cascarno, G.; Giacovazzo, C.; Gualardi, A. J.
Appl. Crystallogr. 1993, 26, 343.
(28) Bruker Analytical Systems, Madison, WI.
Inorganic Chemistry, Vol. 43, No. 14, 2004 4305