Effects of cation-anion interactions on the structures and photophysical properties of anionic d0 tungsten-benzylidyne complexes

Article abstract of DOI:10.1016/S0020-1693(02)01292-6

The structures and photophysical properties of anionic tungsten-benzylidyne complexes of the type [Na(L)][W(≡CPh)(OBu^t)₄] ([Na(L)]1; L=blank, 15-crown-5, crypt-2,2,2) are strongly dependent on the nature of the [Na(L)]⁺ ion. The structures of the 1^- ions are qualitatively similar, consisting of square-pyramidal tungsten centers with short W≡C bonds, but differ as a function of cation in the extent of their Na-OBu^t interactions, the geometries of their OBu^t ligands, and their W-O bond distances. The 1^- ions exhibit long-lived luminescence (τ>1 μs) in fluid solution and the solid state at room temperature. The emission lifetimes and energies are cation dependent even in polar solvents, indicating the presence of [Na(L)]⁺/1^- ion pairs in solution for some L. The emissive state is a spin triplet of either [π(W≡CPh)]¹[d_xy]¹ or [π(W≡CPh)]¹[π*(W≡CPh)]¹ configuration.

Full text of DOI:10.1016/S0020-1693(02)01292-6

Inorganica Chimica Acta 345 (2003) 309ꢀ319
/
www.elsevier.com/locate/ica
Effects of cationꢀ
photophysical properties of anionic d0 tungstenꢀ
complexes
/
anion interactions on the structures and
/
benzylidyne
Cheslan K. Simpson a, Ryan E. Da Re a, Timothy P. Pollagi b, Ian M. Steele a,
Richard F. Dallinger c, Michael D. Hopkins a,*
a Department of Chemistry, The University of Chicago, Chicago, IL 60637, USA
b Department of Chemistry, The University of Pittsburgh, Pittsburgh, PA 15260, USA
c Department of Chemistry, Wabash College, Crawfordsville, IN 47933, USA
Received 13 June 2002; accepted 2 September 2002
Dedicated to Professor Richard R. Schrock
Abstract
The structures and photophysical properties of anionic tungstenꢀ
([Na(L)]1; Lꢁ
blank, 15-crown-5, crypt-2,2,2) are strongly dependent on the nature of the [Na(L)]ꢂ ion. The structures of the 1ꢃ
ions are qualitatively similar, consisting of square-pyramidal tungsten centers with short WꢀC bonds, but differ as a function of
cation in the extent of their Naꢀ
OBut interactions, the geometries of their OBut ligands, and their WꢁO bond distances. The 1ꢃ
ions exhibit long-lived luminescence (t ꢀ1 ms) in fluid solution and the solid state at room temperature. The emission lifetimes and
energies are cation dependent even in polar solvents, indicating the presence of [Na(L)]ꢂ/1ꢃ ion pairs in solution for some L. The
emissive state is a spin triplet of either [p(Wꢀ
CPh)]1[dxy]1 or [p(WꢀCPh)]1[p*(WꢀCPh)]1 configuration.
/
benzylidyne complexes of the type [Na(L)][W(ꢀ
/
CPh)(OBut)4]
/
/
/
/
/
/
/
/
# 2002 Elsevier Science B.V. All rights reserved.
Keywords: Alkylidyne complexes; Luminescence; Ion pairing
1. Introduction
molecules [4] and conjugated organic [5] and organo-
metallic [6] polymers.
In contrast to the extensively developed thermal
The discovery by Schrock two decades ago of high-
chemistry of high-oxidation-state metalꢀalkylidyne
/
oxidation-state metalꢀalkylidyne complexes was a mile-
/
complexes, little is known about their electronic excited
states and photophysics [7]. Our interest in this aspect of
their chemistry was motivated by a proposal that
stone in organometallic chemistry that opened a number
of new and still evolving areas of research [1]. Much of
this research has been motivated by Schrock’s discovery
that certain derivatives of this class of complexes
catalyze alkyne-metathesis reactions [1]. The ease with
(pseudo)tetragonal d0 complexes with multiple metalꢁ
/
ligand (ME) bonds are good candidates for luminescent
LMCT chromophores [8], of which derivatives that
luminesce in fluid solution are rare [8,9] compared to
the ubiquitous MLCT chromophores. The basis for this
which alkyne metathesis and related triple-bond (Mꢀ
CꢀE) metathesis reactions [2] allow the formation of Cꢀ
C and MꢀC (MꢁMo [3], W) bonds has encouraged the
application of this chemistry to the synthesis of organic
/
M,
/
/
/
/
proposal is that the p(ME)0
/
M(dxy) (denoted p0
/
n)
electronic transition of d0 complexes (which, in many
such compounds, should produce the lowest-energy
excited state) is qualitatively similar, in terms of
symmetry and expected excited-state distortion, to the
* Corresponding author. Tel.: ꢂ
0805
E-mail address: mhopkins@uchicago.edu (M.D. Hopkins).
/
1-773-702 6490; fax: ꢂ1-773-702
/
M(dxy)0
/
p*(ME) (n0p*) transition [10] of the well
/
established class of emissive d2 multiply bonded com-
0020-1693/02/$ - see front matter # 2002 Elsevier Science B.V. All rights reserved.
PII: S 0 0 2 0 - 1 6 9 3 ( 0 2 ) 0 1 2 9 2 - 6

310
C.K. Simpson et al. / Inorganica Chimica Acta 345 (2003) 309ꢀ319
/
plexes. In this context, the reports by Floriani and
coworkers on the chemistry of square-pyramidal d0
reagents were obtained from commercial sources and
used as received.
tungstenꢀ
type [{p-But-calix[4]ꢁ
attracted our attention [11ꢀ
by these workers on [W(CMe)(OR)4]ꢃ model ions
[14,16] indicate that the HOMO and LUMO are p(Wꢀ
C) and W(dxy), respectively. Thus, a study of the
photophysical properties of complexes of the
[W(CR)(OR?)4]ꢃ type would afford an opportunity to
/
alkylidyne calix[4]arene complexes of the
(O)4}W(CR)]ꢃ (Rꢁ
Ph, SiMe3)
17]. Theoretical calculations
NMR spectra were recorded at ambient temperature.
Chemical shifts are referenced to Me4Si using residual
1H or natural abundance 13C resonances of the solvent.
Electronic-absorption spectra were recorded of samples
contained in sealed cuvettes [21]. Emission and emis-
sion-excitation spectra were recorded using a Spex
Fluorolog II spectrometer equipped with a Xenon
lamp and an R406 photomultiplier tube. Emission
spectra were corrected for instrument response using
standard methods [22]. Emission-excitation spectra were
corrected for instrument response using the internal
correction hardware and software of the instrument.
Emission quantum yields were measured relative to
[Ru(bpy)3][PF6]2 in water [23] and corrected for solvent
refractive index [24]. Emission lifetimes were measured
using the 355-nm (third harmonic) line of a Spectra
Physics GCR-11 Nd:YAG laser operating at 10 Hz as
the excitation source. Samples for lifetime measurements
were placed in sealed quartz cuvettes (solution samples)
or epr tubes (solid samples). A Hamamatsu R928
photomultiplier tube attached to an ISA 0.22 m mono-
chromator was used to detect emission. Emission decay
traces were recorded on a Tektronix 2431L digital
oscilloscope, downloaded to a computer, and fit to
exponential functions. Plots of ln(Iem) versus t were
linear over at least three lifetimes for all samples except
[Na(crypt-2,2,2)]1 in the solid state; this sample con-
tained two structural forms of the compound that
possess different emission lifetimes.
/
/
/
/
further test whether lowest-lying p0
in general, luminescent.
We report here the syntheses, structures, and photo-
/n excited states are,
physical properties of salts of the anionic tungstenꢀ
/
benzylidyne complex [W(CPh)(OBut)4]ꢃ (1ꢃ) with
several sodium counterions (Naꢂ, [Na(15-crown-5)]ꢂ,
[Na(crypt-2,2,2)]ꢂ).
Complexes
of
the
type
[W(CR)(OR?)4]ꢃ, with monodentate alkoxide ligands,
are potentially advantageous for the purpose of studying
excited-state energies and properties compared to calix-
arene derivatives because the electronic properties of
these ligands are more broadly tunable. We have
discovered that the 1ꢃ ion is strongly luminescent in
both fluid solution and the solid state at room tempera-
ture, and that its structure and photophysical properties
are significantly dependent on the nature of the counter-
ion.
2. Experimental
Elemental analyses were performed either by Micro-
lytics (South Deerfield, MA) or by Midwest Microlabs
(Indianapolis, IN). Despite multiple attempts the results
of the carbon analyses are, in general, in borderline
agreement with expectation. The complex [Na(15-
crown-5)]1 was noted to darken over time in the solid
state at r.t. in a glovebox, undoubtedly accounting for
its particularly poor analysis.
2.1. General procedures
All syntheses and manipulations were performed
under dinitrogen atmosphere using standard glovebox
or standard Schlenk techniques. HPLC-grade solvents,
stored under dinitrogen in stainless-steel cylinders, were
purified by passing them under dinitrogen pressure
through a stainless-steel system consisting of either
two 4.5 in.ꢄ24 in. (1 gal) columns of activated A2
/
2.2. Synthesis of the complexes
alumina (MeCN, Et2O, CH2Cl2, and THF) or one
column of activated A2 alumina and one column of
activated BASF R3-11 catalyst (C6H5CH3, C5H12) [18].
Deuteriated NMR solvents were sparged with dinitro-
gen and stirred over CaH2 (CD2Cl2) or Na/K (1:2) alloy
(C6D6, THF-d8), from which they were transferred
under vacuum. Solvents used for photophysical mea-
surements were purified by stirring for two days over
Na/K (1:2) alloy (C5H12, dimethoxyethane and THF) or
CaH2 (CH2Cl2), from which they were transferred in
vacuo. The compound W(CPh)(OBut)3 [19] was pre-
pared from the reaction between W2(OBut)6 [20] and
diphenylacetylene at room temperature (r.t.) in C5H12
solution in the presence of a small amount of 3-hexyne,
and isolated using standard procedures. All other
2.2.1. [Na][W(CPh)(OBut)4] ([Na]1)
To a stirred suspension of NaOBut (0.10 g, 1.0 mmol)
in C5H12 (10 ml) was added W(CPh)(OBut)3 (0.52 g, 1.1
mmol) in C5H12 (5 ml). This resulted in the immediate
precipitation of a bright yellow solid. The mixture was
allowed to stand at ꢃ40 8C for 4 h, after which the
/
solid was collected by filtration, washed with C5H12
(approximately 25 ml), and dried under vacuum to give
[Na]1 (0.58 g, 95% yield). Yellow crystals of [Na]1-d5×
2C5H12 (in which the benzylidyne ligand is deuterium
labeled) suitable for X-ray diffraction measurements
were obtained by layering a suspension of NaOBut in
C6H5CH3 with a solution of W(CPh-d5)(OBut)3 in
/1/

C.K. Simpson et al. / Inorganica Chimica Acta 345 (2003) 309ꢀ
/
319
311
C6H5CH3 and C5H12 (1:1) and allowing the resulting
mixture to stand at ꢃ40 8C for 1 day.
2.3. Crystal structure determinations
/
Anal. Calc. (Found) for C23H41O4NaW: C, 46.95
(47.42); H, 7.02% (6.97). 1H NMR (300.1 MHz,
CD2Cl2): d 7.24 (t, 2H, m-Ph), 6.99 (d, 2H, o-Ph),
6.78 (t, 1H, p-Ph), 1.46 (s, 36H, OBut). 1H NMR (400.1
MHz, THF-d8): d 7.09 (t, 2H, m-Ph), 6.88 (d, 2H, o-
Ph), 6.55 (t, 1H, p-Ph), 1.38 (s, 36H, OBut). 13C{1H}
X-ray crystal structures were determined using a
Siemens R3m/V (for [Na]1-d5×1/2C5H12) or a Bruker
AXS SMART APEX diffractometer (for [Na(15-crown-
5)]1, a-[Na(crypt-2,2,2)]1×THF, and b-[Na(crypt-
/
/
2,2,2)]1) with graphite-monochromated Mo Ka (lꢁ
/
˚
0.71073 A) radiation. Crystals were attached to fine
NMR (100.6 MHz, THF-d8): d 258.7 (Wꢀ
/
C), 149.2
glass fibers with Fluorolube and placed in a cold N2
stream for data collection. All computer programs used
in the structure solution and refinement are contained in
the SHELXTL (v 5.1) program suite [25]. Crystal data and
data collection and refinement parameters are set out in
(Ph), 133.8 (Ph), 127.0 (Ph), 123.7 (Ph), 75.79
(OC(CH3)3), 33.61 (OC(CH3)3).
2.2.2. [Na(15-crown-5)][W(CPh)(OBut)4] ([Na(15-
crown-5)]1)
To [Na]1 (0.58 g, 0.99 mmol) in THF (10 ml) was
added 15-crown-5 (0.22 g, 1.0 mmol) in THF (5 ml). The
Table 1. For [Na(15-crown-5)]1, a-[Na(crypt-2,2,2)]1×
/
THF, and b-[Na(crypt-2,2,2)]1, initial cell constants
were calculated using approximately 1000 reflections
taken from data frames collected at three values of f (0,
90, 1808). The data were processed using SAINT (v 6.02)
and were corrected for X-ray absorption using SADABS
resulting solution was shaken, cooled to ꢃ
layered with Et2O and C5H12, and allowed to stand at
35 8C for 2 days. Yellow crystals of [Na(15-crown-
/
35 8C,
ꢃ
/
(v 2.01). For [Na]1-d5×
determined from a least-squares fit of the angular
settings of 24 reflections in the range 20852u 5258.
/1/2C5H12 lattice parameters were
5)]1 (0.65 g, 82% yield) suitable for X-ray diffraction
measurements were collected.
/
/
Anal. Calc. (Found) for C33H61O9NaW: C, 49.01
(47.90); H, 7.60% (7.42). 1H NMR (300.1 MHz,
CD2Cl2): d 7.17 (t, 2H, m-Ph), 6.93 (d, 2H, o-Ph),
6.61 (t, 1H, p-Ph), 3.71 (s, 20H, 15-crown-5), 1.36 (s,
36H, OBut). 13C{1H} NMR (75.5 MHz, CD2Cl2): d
The intensities of three representative reflections were
measured every 197 reflections during the data collec-
tion to ascertain that the sample did not decay. An
empirical absorption correction, based on the azimuthal
(c) scans of six reflections, was applied.
260.2 (Wꢀ
/
C), 148.9 (Ph), 134.2 (Ph), 126.8 (Ph), 122.6
Direct methods were employed to determine the
positions of heavy atoms. Remaining non-hydrogen
atoms were located from subsequent difference Fourier
syntheses and refined using full-matrix least-squares
methods. Idealized hydrogen-atom positions and ther-
mal parameters were derived from the atoms to which
they are bonded (except for those of the interstitial
(Ph), 75.48 (OC(CH3)3), 69.52 (15-crown-5), 33.70
(OC(CH3)3).
2.2.3. [Na(crypt-2,2,2)][W(CPh)(OBut)4]
([Na(crypt-2,2,2)]1)
To [Na]1 (0.16 g, 0.27 mmol) in THF (5 ml) was
added a solution of crypt-2,2,2 (0.10 g, 0.27 mmol) in
THF (5 ml). The solution was shaken, layered with
C5H12 molecule of [Na]1-d5×
/
1/2C5H12, vide infra). The
final cycles of refinement for each structure used
anisotropic thermal parameters for all non-hydrogen
atoms, with the following two exceptions. For b-
[Na(crypt-2,2,2)]1, attempts to refine the para carbon
atom (C(5)) of the benzylidyne ligand with anisotropic
thermal displacement parameters led to poor refine-
ment, so this atom was refined isotropically. (This
problem was observed for data from two different
Et2O, and cooled to ꢃ80 8C for 2 days, giving yellow
/
crystals of a-[Na(crypt-2,2,2)]1×
/
THF (0.23 g, 82% yield).
(The a designation refers to the structural type of the 1ꢃ
ion.) If the reaction mixture is instead cooled to
ꢃ35 8C, orange crystals of the THF-free polymorph
/
b-[Na(crypt-2,2,2)]1 are obtained. Both types of crystals
were suitable for X-ray diffraction studies.
crystals). For a-[Na(crypt-2,2,2)]1×
/THF, all non-hydro-
Anal. Calc. (Found) for C41H77N2NaO10W: C, 51.04
(50.52); H, 8.04 (8.11); N, 2.90% (2.86). 1H NMR (300.1
MHz, CD2Cl2): d 7.14 (t, 2H, m-Ph), 6.94 (d, 2H, o-Ph),
6.58 (t, 1H, p-Ph), 3.62 (s, 12H, crypt), 3.58 (t, 12H,
crypt), 2.65 (t, 12H, crypt), 1.38 (s, 36H, OBut). 1H
NMR (300.1 MHz, C6D6): d 7.59 (d, 2H, o-Ph), 7.38 (t,
2H, m-Ph), 6.72 (t, 1H, p-Ph), 3.22 (s, 12H, crypt), 3.16
(t, 12H, crypt), 2.19 (t, 12H, crypt), 1.96 (s, 36H, OBut).
gen atoms except those of THF were refined anisotro-
pically. The THF molecule was disordered; assignment
of the oxygen atom was based on geometric considera-
tions.
Two details concerning the refinement of the crystal
structure of [Na]1×
refinements revealed that one t-butyl group is rotation-
ally disordered about the CꢁO bond; two orientations
/1/2C5H12 should be noted. Successive
/
13C{1H} NMR (75.5 MHz, CD2Cl2): d 257.1 (Wꢀ
/
C),
were observed that are 608 out of phase. The methyl
carbon atom positions were refined using a shared free
variable that relates site occupancies of the two con-
formations to the observed electron density (with total
149.7 (Ph), 133.1 (Ph), 126.7 (Ph), 121.8 (Ph), 74.55
(OC(CH3)3), 68.96 (crypt), 68.10 (crypt), 53.25 (crypt),
33.48 (OC(CH3)3).

312
C.K. Simpson et al. / Inorganica Chimica Acta 345 (2003) 309ꢀ319
/
Table 1
Crystallographic data for [Na(L)][W(CPh)(OBut)4]×([Na(L)]1) complexes
Parameter
[Na]1-d5×1/2C5H12
25.5H42D5NaO4W
[Na(15-crown-5)]1
a-[Na(crypt-2,2,2)]1×THF
b-[Na(crypt-2,2,2)]1
Empirical formula
Formula weight
T (K)
C
C33H61NaO9W
808.66
173
C45H85N2NaO11
1036.99
100
W
C41H77N2NaO10
964.89
100
W
629.47
213
triclinic
Crystal system
Space group
Unit cell dimensions
monoclinic
P21 (No. 4)
orthorhombic
Pbca (No. 61)
orthorhombic
Pcca (No. 54)
¯
P1 (No. 2)
/
˚
a (A)
9.931(2)
11.784(2)
13.848(3)
66.70(2)
78.59(2)
78.87(2)
1447.1(5)
2
10.818(3)
15.553(4)
11.930(3)
90
21.395(9)
19.143(8)
24.571(11)
90
19.680(8)
19.396(7)
23.974(9)
90
˚
b (A)
˚
c (A)
a (8)
b (8)
g (8)
112.870(4)
90
90
90
90
90
3
˚
V (A )
Z
Crystal size (mm)
rcalc (g cmꢃ3
m (cmꢃ1
F(000)
1849.4(9)
2
10063(8)
8
9151(6)
8
0.24ꢄ0.26ꢄ0.34
1.433
40.31
0.30ꢄ0.30ꢄ0.25
1.452
31.82
0.20ꢄ0.12ꢄ0.10
0.15ꢄ0.15ꢄ0.15
)
1.369
23.60
4336
1.401
25.88
4015
)
634
832
u Range (8)
2.11ꢀ
/
26.02
2.04ꢀ/25.04
1.65ꢀ/28.28
1.70ꢀ25.07
/
Reflections collected
Independent reflections (Rint
Index ranges
5973
9399
59 878
44 109
)
5683 (0.0507)
ꢃ125h 512
05k 514
ꢃ155l 517
0.0281
5308 (0.0234)
ꢃ115h 512
ꢃ155k 518
ꢃ135l 514
0.0198
12 108 (0.0317)
ꢃ275h 527
ꢃ155k 525
ꢃ295l 532
0.0293
8106 (0.0580)
ꢃ215h 523
ꢃ235k 522
ꢃ285l 523
0.0543
R1 [I ꢀ2s(I)]
wR2 [I ꢀ2s(I)]
R1 (all data)
wR2 (all data)
0.0671
0.0366
0.0705
0.0506
0.0200
0.0507
0.0761
0.0393
0.0790
0.1417
0.0652
0.1463
Data/restraints/parameters
Max/min transmission
Goodness-of-fit
5683/0/324
1.00/0.712
1.011
5308/1/397
1.00/0.776
1.038
12 108/0/516
1.00/0.819
0.984
8106/0/492
1.00/0.599
1.165
ꢃ3
˚
Largest difference peak/hole (e A
)
1.419/ꢃ0.814
1.088/ꢃ0.598
2.167/ꢃ0.778
2.937/ꢃ2.017
occupancy constrained to unity); site-occupancy factors
of 0.638 and 0.362 were found for the two orientations.
In addition, the interstitial pentane molecule is disor-
Derivatives of [Na]1 in which the sodium ion is
coordinated by 15-crown-5 or crypt-2,2,2 were prepared
in
equivalent of these ligands in THF. The reaction
between [Na]1 and 15-crown-5 at ꢃ35 8C provides
ꢀ/80% yield by treatment of [Na]1 with one
¯
dered, residing on a 1 site.
/
crystals of the salt [Na(15-crown-5)]1. The reaction
between [Na]1 and crypt-2,2,2 provides the salt
3. Results and discussion
[Na(crypt-2,2,2)]1. At ꢃ
is predominantly a morphology that lacks interstitial
solvent, while at ꢃ80 8C the main crystalline form is
[Na(crypt-2,2,2)]1×
THF; the structures of the 1ꢃ ions in
these salts differ significantly (vide infra). In addition to
X-ray crystallography, these salts were characterized by
1H and 13C NMR spectroscopy and elemental analysis.
The [Na(L)]1 salts are generally more soluble in polar
organic solvents than is [Na]1; [Na(crypt-2,2,2)]1 is even
slightly soluble in benzene and toluene.
The 1H and 13C NMR spectra of [Na]1, [Na(15-
crown-5)]1, and [Na(crypt-2,2,2)]1 are consistent with
the formulation of the 1ꢃ ions, exhibiting resonances
attributable to the benzylidyne ligand and 4 equiv. tert-
butoxide ligands. Principally diagnostic for the presence
of the benzylidyne ligand is the characteristic 13C NMR
/
35 8C the crystalline product
3.1. Synthesis and characterization of
[Na(L)][W(CPh)(OBut)4] complexes
/
/
The complex [Na][W(CPh)(OBut)4] ([Na]1) can be
prepared in ꢀ90% yield from the reaction between
/
equimolar amounts of W(CPh)(OBut)3 and NaOBut at
25 8C in pentane, from which it precipitates as a yellow
solid. The compound is air sensitive in the solid state
and in solution. This salt is slightly soluble in dichlor-
omethane and more so in acetonitrile, tetrahydrofuran
and 1,2-dimethoxyethane, but shows evidence of slow
decomposition in solution (vide infra). Compound [Na]1
was characterized by elemental analysis, 1H and 13C
NMR spectroscopy, and X-ray crystallography (vide
infra).

C.K. Simpson et al. / Inorganica Chimica Acta 345 (2003) 309ꢀ
/
319
313
resonance of the triply bonded carbon at approximately
d 260. This and the other resonances are shifted
downfield from those of the starting material
W(CPh)(OBut)3. In CD2Cl2 solution, the resonances of
[Na]1 are shifted from those of [Na(15-crown-5)]1 and
[Na(crypt-2,2,2)]1, suggesting that the former is (more
strongly) ion paired in this solvent.
The NMR spectra of freshly prepared solutions of
[Na(L)]1 in CD2Cl2 and THF-d8 also show weak
resonances attributable to W(CPh)(OBut)3, which
slowly increase in intensity over time. Since this starting
material is easily removed from [Na(L)]1 during pur-
ification of the product (by washing with pentane), the
presence of these signals is attributed to decomposition
of 1ꢃ in these solvents. In contrast, the 1H-NMR
spectrum of [Na(crypt-2,2,2)]1 in C6D6 does not exhibit
resonances due to W(CPh)(OBut)3. It is not certain
whether [Na(L)]1 decomposes to form W(CPh)(OBut)3
via dissociation of t-butoxide or by the reaction between
1ꢃ and the solvent.
Fig. 1. ORTEP drawing (30% probability ellipsoids) of the structure of
the [W(CPh)(OBut)4]ꢃ ion of a-[Na(crypt-2,2,2)][W(CPh)(OBut)4]×
THF. Hydrogen atoms are omitted for clarity.
/
The [Na(L)]1 compounds are members of the general
class of [W(CR)X4]ꢃ ions, of which the prior examples
are [W(CBut)Cl4]ꢃ [26] and the calixarene derivatives
[{p-But-calix[4]ꢁ
/
(O)4}W(CR)]ꢃ noted above [11ꢀ
/
17].
The 1ꢃ ion is also the conjugate base of the alkylidene
complex of which related
W(CHPh)(OBut)4,
W(CHR)(OR?)4 derivatives have been prepared from
the reaction between W(CR)(OR?)3 and HOR? [27]. It is
probable that 1ꢃ could also be prepared by deprotona-
tion of the conjugate-acid benzylidene complex, but this
has not been investigated given the simplicity and high
yield of the present synthetic procedure.
3.2. Crystal structures of [Na(L)][W(CPh)(OBut)4]
complexes
Fig. 2. ORTEP drawing (30% probability ellipsoids) of the structure of
the [W(CPh)(OBut)4]ꢃ ion of b-[Na(crypt-2,2,2)][W(CPh)(OBut)4].
Hydrogen atoms are omitted for clarity.
The crystal structures of [Na]1×
/
1/2C5H12, [Na(15-
crown-5)]1, a-[Na(crypt-2,2,2)]1×
/
THF, and b-[Na-
t
˚
2.04 A) are similar to those found for [{p-Bu -calix[4]ꢁ
/
(crypt-2,2,2)]1 were determined through single-crystal
X-ray diffraction studies. The structures of these com-
(O)4}W(CPh)]ꢃ (1.96ꢀ
some significant differences among Wꢁ
within individual compounds that result from specific
features of the structures (vide infra). In general, the Wꢁ
O bond distances are longer than those for W(CPh)(O-
/
2.00 A) [14], although there are
˚
/
O bond lengths
plexes are shown in Figs. 1ꢀ4 and selected bond
/
distances and bond angles are set out in Table 2.
The structures of the 1ꢃ ions of the [Na(L)]1
complexes are qualitatively similar in several respects.
The geometry about the tungsten center is roughly
square-pyramidal, with an apical benzylidyne ligand
/
But)3 (1.86ꢀ
1.87 A) [28], which presumably reflects the
˚
/
stronger trans influence of these ligands in 1ꢃ than in
the trigonal WO3 geometry of W(CPh)(OBut)3.
and equatorial alkoxide ligands. As expected for metalꢀ
/
Several structural features of the 1ꢃ ions are sensitive
functions of the nature of the counter ion and crystal
morphology. The structures of these ions in a-[Na(crypt-
alkylidyne complexes, the Wꢀ
/
C bonds are short
˚
1.780(4) A), similar to those for W(CPh)(O-
(1.755(4)ꢀ
/
But)3 (d(Wꢀ
/
C)ꢁ
(O)4}W(CPh)]ꢃ (d(Wꢀ
WꢀCꢁC linkages are approximately linear (ꢂ
1758). As is generally observed for the equatorial ligands
of square-pyramidal complexes with multiple metalꢁ
ligand bonds [29], the WO4 fragment is pyramidal (ꢂ
CꢀWꢁOꢁ99ꢀ1108). The WꢁO bond distances (1.90ꢀ
/
1.758(5) A) [28] and [{p-Bu -calix[4]ꢁ
/
t
˚
˚
2,2,2)]1×
first because they lack the close anionꢀ
found for [Na]1 and [Na(15-crown-5)]1 (vide infra). In
/
THF and b-[Na(crypt-2,2,2)]1 are considered
/
C)ꢁ
/
1.728(7) A) [14], and the
/
cation contacts
/
/
/
WCC]
/
a-[Na(crypt-2,2,2)]1×
/
THF (Fig. 1) the t-butyl groups of
1ꢃ lie on the same side of the approximate O4 plane as
/
/
/
/
/
/
/
/
the alkylidyne ligand. This orientation of alkoxide

314
C.K. Simpson et al. / Inorganica Chimica Acta 345 (2003) 309ꢀ319
/
Fig. 4. ORTEP drawing (20% probability ellipsoids) of the structure of
a dimer of [Na][W(CPh-d5)(OBut)4]×
1/2C5H12. Hydrogen atoms and
the interstitial pentane are omitted for clarity.
/
Fig. 3. ORTEP drawing (30% probability ellipsoids) of the structure of
[Na(15-crown-5)][W(CPh)(OBut)4]. Hydrogen atoms are omitted for
clarity.
Table 2
Selected bond distances and bond and torsion angles for [Na(L)][W(CPh)(OBut)4] ([Na(L)]1) complexes
[Na]1
[Na(15-crown-5)]1
a-[Na(crypt-2,2,2)]1
b-[Na(crypt-2,2,2)]1
˚
Bond distances (A)
WꢁC(1)
WꢁO(1)
WꢁO(2)
WꢁO(3)
WꢁO(4)
NaꢁO(1)
NaꢁO(2)
NaꢁO(3)
NaꢁO(4)
1.755(4)
1.985(3)
1.899(3)
1.985(3)
2.037(3)
2.313(3)
1.780(4)
1.950(3)
1.944(2)
1.994(3)
2.019(2)
1.772(3)
1.971(2)
1.756(8)
1.980(5)
1.961(5)
1.979(5)
1.945(5)
1.9533(19)
1.9652(19)
1.952(2)
a
2.272(3)
2.387(3) a, 2.505(3)
2.746(3)
2.426(3)
Bond angles (8)
WꢁC(1)ꢁC(2)
C(1)ꢁWꢁO(1)
C(1)ꢁWꢁO(2)
C(1)ꢁWꢁO(3)
C(1)ꢁWꢁO(4)
WꢁO(1)ꢁC(8)
WꢁO(2)ꢁC(12)
WꢁO(3)ꢁC(16)
WꢁO(4)ꢁC(20)
177.5(4)
175.3(3)
177.0(2)
178.5(6)
99.9(2)
104.68(17)
106.67(18)
105.50(17)
110.05(17)
139.8(3)
102.67(15)
104.10(15)
105.49(16)
105.16(14)
141.0(2)
104.27(10)
108.85(11)
102.75(10)
108.67(10)
135.74(17)
142.32(18)
138.88(17)
142.25(18)
108.0(2)
99.1(2)
107.3(2)
131.7(4)
137.2(4)
126.6(4)
137.6(4)
150.0(3)
138.3(3)
137.9(3)
143.3(2)
136.2(3)
136.6(2)
Torsion angles (8)
C(1)ꢁWꢁO(1)ꢁC(8)
C(1)ꢁWꢁO(2)ꢁC(12)
C(1)ꢁWꢁO(3)ꢁC(16)
C(1)ꢁWꢁO(4)ꢁC(20)
ꢃ6.3(5)
21.3(7)
25.8(5)
11.6(4)
16.9(4)
40.6(6)
26.3(4)
22.4(4)
ꢃ24.8(3)
ꢃ34.0(3)
ꢃ26.3(3)
ꢃ40.0(3)
ꢃ179.3(6)
ꢃ2.0(7)
ꢃ179.2(5)
1.0(7)
a
Distance to Na(#1).

C.K. Simpson et al. / Inorganica Chimica Acta 345 (2003) 309ꢀ
/
319
315
ligands, denoted structure type a (see in-text figure) is
similar to that seen for W(CPh)(OBut)3 [28] but inverted
group is oriented over the E t-butoxy ligands; this
orientation is similar to that found for [{p-But-calix[4]ꢁ
/
from that observed for [{p-But-calix[4]ꢁ
/
(O)4}W(CR)]ꢃ
(O)4}W(CPh)]ꢃ [14] but rotated approximately 458
from a-[Na(crypt-2,2,2)]1.
[14], for which the R groups are constrained to lie
opposite the benzylidyne ligand relative to the O4 plane.
This latter arrangement of alkoxide R groups is denoted
structure type g (see in-text figure). The benzylidyne
phenyl group approximately bisects the cis W(OBut)2
linkages. The W(OBut)4 fragment is slightly puckered,
with the two trans OBut ligands that are distorted
The structures of [Na(15-crown-5)]1, shown in Fig. 3,
and [Na]1, shown in Fig. 4, exhibit strong interactions
between the Naꢂ ion and oxygen atoms of 1ꢃ. The 1ꢃ
ions of [Na]1 and [Na(15-crown-5)]1 are of structural
type a, although, as for a-[Na(crypt-2,2,2)]1, the t-butyl
groups are not rigorously eclipsed with the Wꢀ
/
C bond
downward (O(2) and O(4): ꢂ
108.67(10)8) exhibiting larger internal bond angles (ꢂ
WꢁOꢁCꢁ142.32(18), 142.35(18)8) than those that are
higher (O(1) and O(3): WꢁOꢁ104.27(10),
Cꢀ
102.75(10)8; WꢁOꢁCꢁ135.74(17), 138.88(17)8).
None of the t-butyl groups are eclipsed with the Wꢀ
bond, exhibiting CꢀWꢁOꢁC torsion angles of 25ꢀ408.
These differences in bond and torsion angles are
reflected in the WꢁO bond distances, which fall into
two statistically distinguishable (s ]2.58) sets: the Wꢁ
O(1) and WꢁO(3) bonds (1.971(2), 1.9652(19) A) are
longer than the WꢁO(2) and WꢁO(4) bonds
/
Cꢀ
/
Wꢁ
/
Oꢁ
/
108.85(11),
(CꢀWꢁOꢁC torsion anglesꢁ6ꢀ418). In both [Na(15-
/
/
/
/
/
/
crown-5)]1 and [Na]1 the Naꢂ ions coordinate to the
oxygen atoms on the O4 face opposite the t-butyl
groups: for [Na(15-crown-5)]1, a single Naꢂ ion co-
ordinates to two oxygen atoms (O(3) and O(4)), while
for [Na]1, which crystallizes as a dimer with bridging
sodium ions, each Naꢂ ion is four coordinate, bonding
to O(3) and O(4) of one 1ꢃ ion and to O(1) and O(4) of
the other 1ꢃ ion. Neither 1ꢃ ion exhibits puckering of
the WO4 unit to the same extent as do the a- and b-
[Na(crypt-2,2,2)]1 structures (although O(4) of [Na]1
/
/
/
ꢂ
/
/
/
/
ꢂ
/
/
/
/
/C
/
/
/
/
/
/
/
˚
/
/
/
has an unusually large Cꢀ
/
Wꢁ
/
O
bond angle,
˚
(1.9533(19), 1.952(2) A).
110.05(17)8, due to the fact that it is the only t-butoxide
The structure of b-[Na(crypt-2,2,2)]1 (Fig. 2) differs
from that of a-[Na(crypt-2,2,2)]1 principally in that the
t-butyl groups are oriented alternately above (Z: O(2)
and O(4)) and below (E: O(1) and O(3)) the O4 plane,
relative to the benzylidyne ligand, instead of all lying
above the plane. This orientation of alkoxide ligands is
denoted as structure type b (see in-text figure). As
observed for a-[Na(crypt-2,2,2)]1 there is puckering of
the WO4 unit, with similar trends in bond distances and
angles. Specifically, the Z t-butoxy ligands have larger
ligand that is coordinated to two Naꢂ ions).
In general, the length of a given Wꢁ
crown-5)]1 and [Na]1 depends on the strengths of its
NaꢁO interactions and those of the t-butoxide ligand
trans to it. Taking the average WꢁO bond distance of
the a- and b-[Na(crypt-2,2,2)]1 complexes (1.963 A) as
normal for the 1ꢃ ion, the Wꢁ
O bonds of t-butoxide
ligands with coordinated Naꢂ ions are 0.022ꢀ
/O bond in [Na(15-
/
/
˚
/
˚
0.074 A
/
longer than normal. For both complexes, the Wꢁ
/O
bonds that lack coordinated Naꢂ ions are trans to the
˚
O bonds; these bonds are 0.013ꢀ0.064 A
elongated Wꢁ
/
/
Cꢀ
ligands (99.9(2), 99.1(2)8) as well as larger Wꢁ
angles (Z: 137.2(4), 137.6(4)8; E: 131.7(4), 126.6(4)8)
/
Wꢁ
/
O angles (108.0(2), 107.3(2)8) than do the E
shorter than normal. The structure of [Na]1 provides the
extreme example of this phenomenon: the WꢁO(4)
bond, which is coordinated by two Naꢂ ions, exhibits
/
Oꢁ/C
/
˚
and shorter Wꢁ
/
O bonds (Z: 1.961(5), 1.945(5) A; E:
˚
˚
the longest distance (2.037(3) A) among the four
1.980(5), 1.979(5) A), although the comparatively low
quality of the crystal structure reduces the confidence
level of this latter observation. The benzylidyne phenyl
[Na(L)]1 structures, while the WꢁO(2) bond that is
/
˚
trans to it is the shortest (1.899(3) A).

316
C.K. Simpson et al. / Inorganica Chimica Acta 345 (2003) 309ꢀ319
/
Fig. 5. Electronic-absorption and -emission spectra of [Na(15-crown-
5)][W(CPh)(OBut)4] at room temperature: (*
) THF solution; (---)
microcrystalline sample.
/
3.3. Photophysical properties of
[Na(L)][W(CPh)(OBut)4] complexes
Fig. 6. Molecular-orbital diagrams for [W(CR)(OR?)4]ꢃ (Rꢁ
ions.
/H, Ph)
The complexes [Na]1, [Na(15-crown-5)]1, and
[Na(crypt-2,2,2)]1 are luminescent in fluid solution and
the solid state at room temperature. These complexes
constitute one of the few classes of d0 transition-metal
complexes that luminesce in fluid solution [9]. The
electronic-absorption and -emission spectra of [Na(15-
crown-5)]1, which are representative of those of the
other [Na(L)]1 complexes, are shown in Fig. 5 and
photophysical data for the [Na(L)]1 complexes are set
out in Table 3 [30].
The electronic-absorption spectra of [Na(L)]1 com-
plexes are dominated by an intense band in the near-UV
that maximizes at 323 nm (30 960 cmꢃ1) for [Na(15-
crown-5)]1 (Fig. 5) and [Na]1 and at 332 nm (30 120
cmꢃ1) for [Na(crypt-2,2,2)]1 in THF solution. To the
red of this band the absorption slowly tails off without
revealing additional distinct bands. Assignments for
these absorption features can be contemplated on the
basis of the molecular-orbital diagrams shown in Fig. 6.
Theoretical calculations on g-[W(CMe)(OR)4]ꢃ [14,16]
and simple symmetry considerations indicate that the
HOMO of a complex with strict C4v symmetry is the
degenerate p(Wꢀ
C) orbital, of dxz, dyz parentage,
should lie next highest in energy. The dxy orbital is
/
C) level and that the LUMO is dxy; the
degenerate p*(Wꢀ
/
nonbonding (d symmetry) with respect to the alkylidyne
ligand and p*(Wꢁ
b, and g, with the Wꢁ
(jcos uj) of the CꢀWꢁOꢁ
of a cylindrically symmetric alkylidyne R group with
phenyl, as for 1ꢃ, should significantly split the p(Wꢀ
C)
and p*(WꢀC) orbitals due to p conjugation between the
WꢀC and C6H5 moieties. In view of these considera-
/
OR) in character for structure types a,
O p-overlap being a function
R torsion angles. Replacement
/
/
/
/
/
/
/
tions, the strong absorption band in the spectra of the
1ꢃ ions is undoubtedly attributable to the dipole- and
spin-allowed 1[p(Wꢀ
/
CPh)0
The absorption tail to lower energy of this band may
arise from 3[p(Wꢀ
CPh)0p*(WꢀCPh)] and/or p(Wꢀ
/
p*(Wꢀ
/
CPh)] transition.
/
/
/
/
Table 3
Photophysical data for [Na(L)][W(CPh)(OBut)4] ([Na(L)]1) complexes at room temperature
[Na]1
n¯max (cmꢃ1
[Na(15-crown-5)]1
n¯max (cmꢃ1
[Na(crypt-2,2,2)]1
n¯max (cmꢃ1
Solvent
/
)
t (ms) f ꢄ104
/
)
t (ms) f ꢄ104
/
)
t (ms)
f ꢄ104
b
c
Solid
Acetonitrile
16 350
6.8
1.6
3.3
15 850
14 900
15 000
15 200
10.2
2.1
14 000 a; 15 625
Â0.30
d
d
Â14 700
14 900
14 850
Dichloromethane
Tetrahydrofuran
1,2-Dimethoxyethane
Toluene
4.6
3.7
3.0
7.7
2.1
1.4
14 800
14 800
14 400
2.8
4.6
4.0
0.7
15 700
e
15 200
e
4.3
e
e
a
Microcrystals grown from THF/ether at ꢃ80 8C, which provided mostly a-[Na(crypt-2,2,2)]1; some b form is present.
Microcrystals grown from THF/ether at ꢃ35 8C, which provided mostly b-[Na(crypt-2,2,2)]1; some a form is present.
Excited-state decay is biexponential.
b
c
d
e
Emission is too weak to measure.
Compound is insoluble in this solvent.

C.K. Simpson et al. / Inorganica Chimica Acta 345 (2003) 309ꢀ
/
319
317
1
CPh)0
/
dxy transitions. The relative energies of these
ment of the emission to the triplet [p(Wꢀ
/
CPh)]1[dxy
]
transitions will depend on the geometry of the alkoxide
ligands, and should be tunable by changing the alkyli-
dyne and alkoxide R groups. Definitive assignments will
require polarized single-crystal spectroscopic measure-
ments; the initial experiments of this type have not yet
resolved this question [31].
state, on the basis that the dxy orbital should be higher
in energy for the b form than the a form as a result of
the signficantly larger Cꢀ
/
Wꢁ
/
Oꢁ
/
C torsion angles of the
latter.
Photoexcitation of the [Na(L)]1 complexes in fluid
solution and the solid state produces easily visible
4. Conclusion
orangeꢀred luminescence. The blue tail of the emission
/
band overlaps the end of the red tail of the absorption
spectrum (Fig. 5), suggesting that the absorption band
that produces the emissive state is very weak. Consistent
with this, the photophysical data indicate that the
emission is a spin-forbidden process. Specifically, the
emission lifetimes of the [Na(L)]1 complexes in both the
solid state and solution exceed 1 ms (with the exception
of [Na(crypt-2,2,2)]1 in the solid state), and the emission
quantum yields in tetrahydrofuran solution are on the
order of 10ꢃ4. From these data, the radiative-decay rate
Anionic square-pyramidal tungstenꢀbenzylidyne
/
complexes of the type [Na(L)][W(CPh)(OBut)4] consti-
tute rare examples of d0 complexes that luminesce in
fluid solution. The structures and photophysical proper-
ties of the [W(ꢀ
are strongly dependent on the nature of the [Na(L)]ꢂ
counterion. The long emission lifetimes (]1 ms) of the
1ꢃ ions open the way for studying the bimolecular
photochemistry of high-oxidation-state metalꢀalkyli-
/
CPh)(OBut)4]ꢃ ions in these complexes
/
/
dyne complexes, which has not previously been ex-
plored. Although the present data do not allow the two
constants are in the range krꢁ
/
25ꢀ
/
70 sꢃ1 and the
2.7ꢀ3.6ꢄ
nonradiative decay rate constants are knrꢁ
/
/
/
plausible candidates for the emissive state, [p(Wꢀ
/
105 sꢃ1. Because of the uncertainty regarding the
transition responsible for the absorption tail, it is not
yet established whether the emission originates from a
1
CPh)]1[dxy
]
and [p(Wꢀ
/
CPh)]1[p*(Wꢀ
/
CPh)]1, to be
distinguished, both should be amenable to tuning
through counterion and alkylidyne and alkoxide R
group variation. Indeed, preliminary investigations of
the related ions [W(CBut)(OBut)4]ꢃ and [{p-But-
1
triplet [p(Wꢀ
/
CPh)]1[dxy
]
or [p(Wꢀ
/
CPh)]1[p*(Wꢀ
/
CPh)]1 state. Preliminary measurements of emission
spectra at low temperature reveal complicated vibronic
structure, with particularly intense progressions with
approximately 1200 cmꢃ1 spacings being evident.
Although definitive assignment of these modes is
calix[4]ꢁ
/
(O)4}W(ꢀ
/
CR)]ꢃ indicate that they are strongly
luminescent as well.
difficult due to the complexity of the Wꢀ/CPh oscillator
[32], the high frequencies are consistent with modes that
have significant contributions from phenyl coordinates.
Vibronic activity of such modes would be expected for
electronic transitions involving either excited state.
The emission properties of the 1ꢃ complexes are a
function of the nature of the [Na(L)]ꢂ ion. The emission
of [Na(L)]1 in the solid state shows modest variations in
5. Supplementary material
CIF format crystallographic files for [Na][W(CPh-
d5)(OBut)4]×
/
1/2C5H12,
But)4], a-[Na(crypt-2,2,2)][W(CPh)(OBut)4]×
[Na(15-crown-5)][W(CPh)(O-
THF, and
/
b-[Na(crypt-2,2,2)][W(CPh)(OBut)4] have been depos-
ited with the Cambridge Crystallographic Data Center,
band maximum with cation, with a range of 14 000ꢀ
/
16 350 cmꢃ1 (612ꢀ
/
712 nm), and, qualitatively, in
CCDC nos. 187044ꢀ187047. Copies of this information
/
may be obtained free of charge from The Director,
CCDC, 12 Union Road, Cambridge, CB2 1EZ, UK
intensity, with the emission of [Na]1 and [Na(15-
crown-5)]1 appearing considerably stronger to the eye
than that of [Na(crypt-2,2,2)]. This latter observation is
consistent with the trend in emission lifetimes in the
(fax: ꢂ44-1223-336-033; e-mail: deposit@ccdc.cam.ac.
/
uk or www: http://www.ccdc.cam.ac.uk).
solid state (t [Na]1ꢁ
ms, t [Na(crypt-2,2,2)]1$
/
6.8 ms, t [Na(15-crown-5)]1ꢁ
/
10.2
/
0.3 ms). The fact that the
emission properties of [Na]1 and [Na(15-crown-5)]1
also differ from those of [Na(crypt-2,2,2)]1 in solution
is strong evidence that the former two complexes
Acknowledgements
maintain some cationꢀ
It is also noteworthy that the emission band maxima of
a- and b-[Na(crypt-2,2,2)]1 differ by ꢀ
1500 cmꢃ1 in
the solid state [33]. A plausible case can be made that the
blue shift of the emission band of b-[Na(crypt-2,2,2)]1
relative to that of a-[Na(crypt-2,2,2)]1 supports assign-
/
anion interactions in solution.
This research was supported by National Science
Foundation Grant CHE 9700451 (MDH) and the
McLain-McTurnan-Arnold Research Fund at Wabash
College (RFD). We thank Steve Geib and Kevin John
for assistance with collecting some of the X-ray diffrac-
tion data.
/

318
C.K. Simpson et al. / Inorganica Chimica Acta 345 (2003) 309ꢀ319
/
(j) B.W. Pfennig, M.E. Thompson, A.B. Bocarsly, J. Am. Chem.
Soc. 111 (1989) 8947.
References
[10] V.M. Miskowski, H.B. Gray, M.D. Hopkins, Adv. Trans. Met.
Coord. Chem. 1 (1996) 159.
[1] (a) R.R. Schrock, Chem. Rev. 102 (2002) 145;
(b) J.S. Murdzek, R.R. Schrock, in: H. Fischer, P. Hofmann, F.R.
Kreissl, R.R. Schrock, U. Schubert, K. Weiss (Eds.), Carbyne
Complexes, VCH, Weinheim, 1988, p. 147;
[11] L. Gianni, E. Solari, A. Zanotti-Gerosa, C. Floriani, A. Chiesi-
Villa, C. Rizzoli, Angew. Chem., Int. Ed. Engl. 35 (1996) 2825.
[12] L. Giannini, E. Solari, C. Floriani, A. Chiesi-Villa, C. Rizzoli, J.
Am. Chem. Soc. 120 (1998) 823.
(c) R.R. Schrock, Acc. Chem. Res. 19 (1986) 342.
[2] M.H. Chisholm, Chem. Rec. 1 (2001) 12.
[13] S. Dovesi, E. Solari, R. Scopelliti, C. Floriani, Angew. Chem., Int.
Ed. Engl. 38 (1999) 2388.
[3] Y.-C. Tsai, P.L. Diaconescu, C.C. Cummins, Organometallics 19
(2000) 5260.
[4] A. Furstner, Angew. Chem., Int. Ed. Engl. 39 (2000) 3012.
¨
[5] (a) U.H.F. Bunz, Acc. Chem. Res. 34 (2001) 998;
(b) U.H.F. Bunz, Chem. Rev. 100 (2000) 1605.
[14] L. Giannini, E. Solari, S. Dovesi, C. Floriani, N. Re, A. Chiesi-
Villa, C. Rizzoli, J. Am. Chem. Soc. 121 (1999) 2784.
[15] L. Giannini, G. Guillemot, E. Solari, C. Floriani, N. Re, A.
Chiesi-Villa, C. Rizzoli, J. Am. Chem. Soc. 121 (1999) 2797.
[16] S. Fantacci, A. Sgamellotti, N. Re, C. Floriani, J. Chem. Soc.,
Dalton Trans. (2001) 1718.
[6] (a) K.D. John, M.D. Hopkins, Chem. Commun. (1999) 589;
(b) H.A. Brison, T.P. Pollagi, T.C. Stoner, S.J. Geib, M.D.
Hopkins, Chem. Commun. (1997) 1263;
[17] (a) C. Floriani, R. Floriani-Moro, Adv. Organomet. Chem. 47
(2001) 167;
(b) C. Floriani, R. Floriani-Moro, in: A. Asfari, V. Bohmer, J.
¨
(c) T.P. Pollagi, S.J. Geib, M.D. Hopkins, J. Am. Chem. Soc. 116
(1994) 6051.
[7] The luminescence properties of several lower-oxidation-state
Harrowfield, J. Vicens (Eds.), Calixarenes 2001, Kluwer, Dor-
drecht, 2001, p. 536.
metalꢀalkylidyne complexes have been reported; these reports
/
include Ref. [6a] and the following:
[18] A.B. Pangborn, M.A. Giardello, R.H. Grubbs, R.K. Rosen, F.J.
Timmers, Organometallics 15 (1996) 1518.
(a) C.C.S. Cavalheiro, K.E. Torraca, K.S. Schanze, L. McElwee-
White, Inorg. Chem. 38 (1999) 3254;
[19] M.L. Listemann, R.R. Schrock, Organometallics 4 (1985) 74.
[20] (a) M. Akiyama, M.H. Chisholm, F.A. Cotton, M.W. Extine,
D.A. Haitko, D. Little, P.E. Fanwick, Inorg. Chem. 18 (1979)
2266;
(b) A. Mayr, M.P.Y. Yu, V.W.W. Yam, J. Am. Chem. Soc. 121
(1999) 1760;
(c) F.W. Lee, M.C.W. Chan, K.K. Cheung, C.M. Che, J.
Organomet. Chem. 552 (1998) 255;
(b) M.H. Chisholm, J.D. Martin, J.E. Hill, I.P. Rothwell, Inorg.
Synth. 29 (1992) 137.
(d) F.W. Lee, M.C.W. Chan, K.K. Cheung, C.M. Che, J.
Organomet. Chem. 563 (1998) 191;
[21] J.L. Marshall, M.D. Hopkins, H.B. Gray, in: A.L. Wayda, M.Y.
Darensbourg (Eds.), Experimental Organometalic Chemistry,
ACS Symp. Ser. 357, Oxford, 1987, p. 254.
(e) W.M. Xue, Y. Wang, M.C.W. Chan, Z.M. Su, K.K. Cheung,
C.M. Che, Organometallics 17 (1998) 1946;
(f) W.M. Xue, M.C.W. Chan, T.C.W. Mak, C.M. Che, Inorg.
Chem. 36 (1997) 6437;
[22] C.A. Parker, W.T. Rees, Analyst 85 (1960) 587.
[23] J.V. Caspar, T.J. Meyer, J. Am. Chem. Soc. 105 (1983) 5583.
[24] G.A. Crosby, J.N. Demas, J. Phys. Chem. 75 (1971) 991.
[25] G. Sheldrick, SHELXTL Version 5.1, Bruker Analytical X-ray
Systems, Madison, WI.
(g) T.K. Schoch, A.D. Main, R.D. Burton, L.A. Lucia, E.A.
Robinson, K.S. Schanze, L. McElwee-White, Inorg. Chem. 35
(1996) 7769;
[26] J.H. Wengrovius, J. Sancho, R.R. Schrock, J. Am. Chem. Soc.
103 (1981) 3932.
(h) W.M. Xue, Y. Wang, T.C.W. Mak, C.M. Che, J. Chem. Soc.,
Dalton Trans. (1996) 2827;
[27] (a) J.H. Freudenberger, R.R. Schrock, Organometallics 4 (1985)
1937;
(i) S. Trammell, B.P. Sullivan, L.M. Hodges, W.D. Harman, S.R.
Smith, H.H. Thorp, Inorg. Chem. 34 (1995) 2791;
(j) J. Manna, T.M. Gilbert, R.F. Dallinger, S.J. Geib, M.D.
Hopkins, J. Am. Chem. Soc. 114 (1992) 5870;
(k) J.D. Carter, K.B. Kingsbury, A. Wilde, T.K. Schoch, C.J.
Leep, E.K. Pham, L. McElwee-White, J. Am. Chem. Soc. 113
(1991) 2947;
(b) J.A. Heppert, S.D. Dietz, N.W. Eilerts, R.W. Henning, M.D.
Morton, F. Takusagawa, F.A. Kaul, Organometallics 12 (1993)
2565.
[28] (a) F.A. Cotton, W. Schwotzer, E.S. Shamshoum, J. Organomet.
Chem. 296 (1985) 55;
(b) F.A. Cotton, W. Schwotzer, E.S. Shamshoum, Organometal-
lics 3 (1984) 1770.
[29] W.A. Nugent, J.M. Mayer, Metal-Ligand Multiple Bonds, Wiley,
(l) A.B. Bocarsly, R.E. Cameron, H.D. Rubin, G.A. McDermott,
C.R. Wolff, A. Mayr, Inorg. Chem. 24 (1985) 3976.
[8] K.S. Heinselman, M.D. Hopkins, J. Am. Chem. Soc. 117 (1995)
12340.
New York, 1988, pp. 157ꢀ158.
/
[30] It must be noted that the interpretation of the photophysical data
for solution samples is complicated by the fact that the 1ꢃ ion
decomposes in polar solvents to give W(CPh)(OBut)3 as the main
identifiable product, as described in Section 3.1. Consistent with
the NMR spectroscopic measurements, monitoring of solution
samples of [Na(L)]1 by electronic-absorption spectroscopy before
and after photophysical measurements indicated that features
[9] (a) G.V. Loukova, V.A. Smirnov, Chem. Phys. Lett. 329 (2000)
437;
(b) V.W.W. Yam, G.Z. Qi, K.K. Cheung, Organometallics 17
(1998) 5448;
(c) D.S. Williams, A.V. Korolev, Inorg. Chem. 37 (1998) 3809;
(d) V.W.W. Yam, G.Z. Qi, K.K. Cheung, J. Chem. Soc., Dalton
Trans. (1998) 1819;
attributable to 1ꢃ decreased in intensity by 1ꢀ
10%, depending on
/
(e) K.S. Heinselman, V.M. Miskowski, S.J. Geib, L.C. Wang,
M.D. Hopkins, Inorg. Chem. 36 (1997) 5530;
(f) D.S. Williams, D.W. Thompson, A.V. Korolev, J. Am. Chem.
Soc. 118 (1996) 6526;
the sample concentration and the duration of the experiment.
However, two lines of evidence argue against the possibility that
the photophysical data in solution are strongly perturbed by the
decomposition products. First, the decomposition product
W(CPh)(OBut)3 does not luminesce in solution or the solid state
(g) D.L. Thorn, R.L. Harlow, Inorg. Chem. 31 (1992) 3917;
(h) S. Paulson, B.P. Sullivan, J.V. Caspar, J. Am. Chem. Soc. 114
(1992) 6905;
at temperatures above 100
K [T.P. Pollagi, Ph.D. Thesis,
University of Pittsburgh (1985)]; thus, it can be excluded as the
source of the emission. Moreover, emission-excitation spectra of
solution samples of [Na(L)]1 exhibit the same features as the
(i) B.W. Pfennig, M.E. Thompson, A.B. Bocarsly, Organometal-
lics 12 (1993) 649;

C.K. Simpson et al. / Inorganica Chimica Acta 345 (2003) 309ꢀ
/
319
319
electronic-absorption spectra, indicating that the emission arises
from these chromophores. Nonetheless, the possibility that
unidentified decomposition products quench the emissive state
of the [Na(L)]1 complexes cannot be excluded; emission lifetimes
and quantum yields in solution should be considered to be lower
limits.
[33] It is likely that the solid a- and b-[Na(crypt-2,2,2)]1 samples are
each contaminated with the other isomer, based on the shapes of
their emission bands and the fact that the emission does not decay
according to a single-exponential function. Unfortunately, the
emission from single crystals of a- and b-[Na(crypt-2,2,2)]1 was
not sufficiently strong to allow the emission spectra and lifetimes
of the pure isomers to be measured.
[31] R.E. Da Re, unpublished results.
[32] J. Manna, R.F. Dallinger, V.M. Miskowski, M.D. Hopkins, J.
Phys. Chem., B 104 (2000) 10928.