Rearrangement in a Tripodal Nitroxide Ligand to Modulate the Reactivity of a Ti-F Bond

Article abstract of DOI:10.1021/acs.inorgchem.5b01687

The tripodal nitroxide ligand [(2-^tBuNO)C₆H₄CH₂)₃N]^3- (TriNOx^3-) binds the Ti(IV) cation and prevents inner-sphere coordination of chloride in the complex [Ti(TriNOx)]Cl (1). The ligand undergoes an η²-NO to κ¹-O rearrangement to enable a fluoride ion to bind in the related complex Ti(TriNOx)F (2). Computational and reactivity studies demonstrated that the ligand rearrangement contributed to the enthalpy change in the transfer of a fluoride anion.

Full text of DOI:10.1021/acs.inorgchem.5b01687

Article
pubs.acs.org/IC
Rearrangement in a Tripodal Nitroxide Ligand To Modulate the
Reactivity of a Ti−F Bond
Michael A. Boreen, Justin A. Bogart, Patrick J. Carroll, and Eric J. Schelter*
P. Roy and Diana T. Vagelos Laboratories, Department of Chemistry, University of Pennsylvania, 231 South 34th Street, Philadelphia,
Pennsylvania 19104, United States
*
S Supporting Information
t
ABSTRACT: The tripodal nitroxide ligand [(2- BuNO)-
3
−
3−
C H CH ) N] (TriNOx ) binds the Ti(IV) cation and
6
4
2 3
prevents inner-sphere coordination of chloride in the complex
2
1
[
Ti(TriNOx)]Cl (1). The ligand undergoes an η -NO to κ -O
rearrangement to enable a fluoride ion to bind in the related
complex Ti(TriNOx)F (2). Computational and reactivity studies
demonstrated that the ligand rearrangement contributed to the
enthalpy change in the transfer of a fluoride anion.
INTRODUCTION
■
New methods of fluorinating small molecules are critical to
modern chemical industries. Approximately 20% and 30% of
pharmaceutical and agricultural compounds, respectively,
contain fluorine atoms. Late-stage fluorination chemistries
are of particular interest, as these methods expand the synthetic
routes available for the synthesis of pharmaceuticals and
1
,2
1
8
Figure 1. H TriNOx.
3
agrochemicals and afford better ways to synthesize F positron
3
,4
emission tomography (PET) tracers. It is of interest to
develop simple and selective nucleophilic fluorination proce-
dures.
RESULTS AND DISCUSSION
Synthesis and Characterization of Complexes. The
■
Ideal fluoride sources are soluble in organic solvents.
Considering metal-based sources, the fluoride salts KF, CsF,
and AgF are common nucleophilic fluoride sources but have
limitations due to poor solubility. Recently, our group reported
complex [Ti(TriNOx)]Cl (1) was synthesized in 53% yield by
adding a THF solution of H TriNOx and Et N to a−78 °C
3
3
THF solution of TiCl (THF) (Scheme 1). Complex 1 was air-
4
2
5
the first example of a organo-cerium(IV) fluoride complex as
Scheme 1. Synthesis of [Ti(TriNOx)]Cl (1)
part of a broader interest in the coordination chemistry of metal
6
−8
fluorides and fluoride-containing ligands.
In the current
2
1
contribution, an η -NO to κ -O rearrangement process in a
multidentate nitroxide ligand was shown to increase the
IV
reactivity of a Ti −F moiety due to steric pressure at the
metal cation and a favorable enthalpy change upon rearrange-
1
2
ment of the ligand arm from a κ -O to η -NO coordination
mode.
Recently, our group showed that the coordination chemistry
t
3−
of the tripodal nitroxide ligand [(2- BuNO)C H CH ) N]
6
4
2 3
and water-stable and water-soluble. High-resolution mass+
3−
(
TriNOx ) (Figure 1) with a series of rare-earth ions was
highly dependent upon cation size. Since Ti(IV) is smaller
than the rare-earth ions, we expected TriNOx would form a
sterically hindered Ti(IV) complex. Herein, we demonstrate
spectrometry showed the presence of the [Ti(TriNOx)]
9
cation. Cooling a saturated DCM solution of 1 to −78 °C
yielded X-ray-quality yellow crystals, enabling determination of
the solid-state structure (Figure 2).
3
−
3
−
that the TriNOx
framework undergoes a change in
coordination mode depending on anion size at the Ti(IV)
cation.
Received: July 26, 2015
Published: September 23, 2015
©
2015 American Chemical Society
9588
DOI: 10.1021/acs.inorgchem.5b01687
Inorg. Chem. 2015, 54, 9588−9593

Inorganic Chemistry
Article
Figure 2. Thermal ellipsoid plot of 1 at the 50% probability level with
hydrogen atoms and Bu groups omitted for clarity. Selected bond
Figure 3. Thermal ellipsoid plot of 2 at the 50% probability level with
hydrogen atoms and Bu groups omitted for clarity. Selected bond
distances (Å): Ti(1)−F(1) 1.8536(11), Ti(1)−O(1) 1.8017(14),
Ti(1)−O(2) 1.8857(14), Ti(1)−O(3) 1.8929(14), Ti(1)−N(2)
2.1937(16), Ti(1)−N(3) 2.2458(16), N(1)−O(1) 1.400(2), N(2)−
O(2) 1.4278(19), N(3)−O(3) 1.444(2).
t
t
distances (Å): Ti(1)−O(1) 1.8912(14), Ti(1)−O(2) 1.8720(13),
Ti(1)−O(3) 1.8900(14), Ti(1)−N(1) 2.1682(17), Ti(1)−N(2)
2
.1852(16), Ti(1)−N(3) 2.1959(16), Ti(1)−N(4) 2.2775(16),
N(1)−O(1) 1.440(2), N(2)−O(2) 1.436(2), N(3)−O(3) 1.441(2).
The crystal structure of 1 showed that the chloride ion was
1
indicating that the κ -O-bound nitroxide group was not
outer sphere, leaving an ion-paired, effectively C symmetric
3
oxidized. The NMR spectra of 2 (Figures S3−S5 in the
Supporting Information) were also consistent with this
assignment. Additionally, the distance from the titanium cation
to the apical nitrogen, N(4) in Figures 2 and 3, increased from
2.2775(16) Å in 1 to 3.073(1) Å in 2. Majumdar and co-
workers reported the lengthening and concurrent weakening of
bond when V(III) complexes of tris(2-
aminoethyl)amine (tren) derivatives were converted to their
respective V −oxo complexes. Lengthening or dissociation of
the apical nitrogen in tripodal ligand frameworks can be a
significant thermodynamic factor when an additional ligand
bonds to the metal cation. Both of the observed structural
changes in the TriNOx ligand opened the coordination
sphere, allowing the fluoride to bind the titanium cation.
To compare the sterics of the TriNOx ligand in complexes
and 2 quantitatively, the program SambVca was used to
calculate the percent buried volume (%VBur) for each
complex. Percent buried volume values have primarily been
used to examine the steric demand of phosphines and NHCs,
but they have also been applied to a variety of ligands, including
guanidinates, aryloxides, and 1,4,7-triazacyclononanes. We
calculated %V values of 92.8% for TriNOx in 1 and 82.0%
for TriNOx in 2 (see the Experimental Section for full
details). Thus, the change in coordination mode of TriNOx
changed its steric profile significantly, consistent with the
coordination of a fluoride ligand in complex 2.
Solution-phase cyclic voltammetry experiments were per-
formed to assess the differences in electronic properties
between complexes 1 and 2 (Figures S8−S10 in the Supporting
Information). Complex 1 exhibited a ligand-based, quasi-
+
[
Ti(TriNOx)] cation in the solid state. Evidently due to
coordination through three η -NO groups, the titanium center
2
was too sterically encumbered to accommodate a chloride
1
0
ligand. The related homoleptic compounds Ti(ONMe ) and
2
4
1
1
Ti(ONEt ) were reported to be eight-coordinate, with four
2
4
2
η -NO groups in each case. The average Ti−O bond length of
the V−N
axial
1
(
.844(14) Å in 1 was shorter than those in Ti(ONMe )
2
4
10
11
average 1.947(1) Å) and Ti(ONEt ) (1.980(3) Å). The
V
20
2
4
average Ti−N(nitroxide) bond length in 1 was 2.1831(16) Å,
slightly longer than those of Ti(ONMe ) (average 2.1625(1)
2
4
1
0
11
Å) and Ti(ONEt ) (2.108(5) Å). The average N−O bond
20
2
4
length of 1.439(2) Å in 1 was slightly longer than those in
3−
1
0
Ti(ONMe2) (average 1.428(1) Å)
and Ti(ONEt2)
4
4
1
1
(
1.402(7) Å). Bond distances and NMR spectra (Figures
3−
S1 and S2 in the Supporting Information) for 1 are consistent
1
with a Ti(IV) cation bonded to three fully reduced nitroxide
1
0−19
groups.
21
We hypothesized that the TriNOx3− ligand could change its
coordination mode to accommodate an additional ligand at the
titanium center. With this concept in mind, we targeted the
small, negatively charged fluoride ion. Indeed, the reaction of
complex 1 with AgF in −25 °C acetonitrile overnight caused X-
ray-quality orange crystals of Ti(TriNOx)F (2) to form
22
23
24
25
3−
Bur
3−
3−
(
Scheme 2, and Figure 3). The crystal structure of complex 2
Scheme 2. Synthesis of Ti(TriNOx)F (2) from Complex 1
reversible redox couple with E = 0.58 V versus ferrocene and
pc
Epa = 0.36 V and a chemically irreversible titanium reduction
feature with Epc = −1.67 V. Complex 2 exhibited two ligand-
based irreversible oxidation features at E = −0.28 and 0.23 V
pa
and an irreversible metal reduction feature at E = −1.89 V. As
pc
expected, the electron-donating fluoride ligand stabilized the
redox features on complex 2 to more negative potentials. The
shift in the potential of the ligand oxidation had important
confirmed the hypothesis that TriNOx3 could undergo
−
structural rearrangement in order to accommodate another
ligand at the titanium center. A comparison of the structures of
+
synthetic repercussions, since the reduction potential of Ag in
2
26
1
and 2 (Figures 2 and 3) showed that one of the η -NO
acetonitrile is 0.04 V versus ferrocene. In solution at room
1
nitroxide groups in complex 1 switched to κ -O coordination in
complex 2. The Ti(1)−O(1) and N(1)−O(1) bond lengths in
the structure of 2, 1.8017(14) and 1.400(2) Å, respectively,
temperature, complex 2 was observed to be oxidized rapidly by
silver(I) salts. Solvent and temperature were thus important
aspects of the synthesis of 2; isolation of 2 was enabled by its
10−19
+
were consistent with a reduced, anionic nitroxide group,
precipitation from solution before reaction with Ag .
9
589
DOI: 10.1021/acs.inorgchem.5b01687
Inorg. Chem. 2015, 54, 9588−9593

Inorganic Chemistry
Article
Scheme 3. Computational Scheme Used To Calculate the ΔG Value of Ligand Rearrangement (ΔGLR)
−
Reactions of 1 with CsF in acetonitrile at room temperature
also led to formation of 2, but the product decomposed under
these conditions, preventing its isolation in good yield. We
avoided using common acidic fluoride sources such as
Et N(HF) and pyridine(HF) to synthesize 2, as the presence
TriNOx3 ligand to return to a lower energy conformation. In
an initial reactivity experiment, complex 2 was reacted with
(CH ) SiCl in CDCl , resulting in instant, nearly quantitative
3
3
3
conversion to 1 and (CH ) SiF. Since we had shown 2 could
3 3
perform fluorination chemistry and cleanly return to 1, we
decided to perform computational studies of fluorination with 2
(see the Supporting Information for full details).
DFT calculations were performed at the B3LYP level of
theory with the 6-31G* basis set; calculations with solvent
continuum were performed using the CPCM SCRF method.
Complexes 2 and 1 (1 = [Ti(TriNOx)] ) were optimized in
each solvent continuum on the basis of crystal structures or
calculated gas-phase coordinates. Complex 1 -open repre-
sented complex 2 with the fluoride ion removed, and it was not
3
3
of free H TriNOx in some reactions with 2 suggested that 2
3
was sensitive to acidic protons. BF ·Et O was also not used due
3
2
to concerns it would react further by abstracting the fluoride
−
ligand from 2 to form the BF4 salt. Complex 1 reacted only
slowly with CsF in acetontrile at −25 °C, giving very low yields
of 2. This result suggested that reactions with other poorly
+
+
+
acetonitrile-soluble fluorides, such as [Me N]F, would not be
4
+
productive for synthesizing 2. At room temperature and at −25
°C, 1 reacted with AgF much more quickly than with CsF. The
+
optimized (Scheme 3). Complex 1 -open had the same
high reactivity of AgF was evidently important for the reaction
to occur at −25 °C; under the same conditions, the reaction
with CsF was too slow to achieve a reasonable yield. AgF, AgCl,
and 2 were all insoluble in −25 °C acetonitrile; thus, once 2
formed, it precipitated as crystals that were not oxidized by the
silver salts also present. On the other hand, the solubility of 1
under these conditions allowed the reaction to occur despite
not having AgF in solution. The combination of the solubility
of 1 and the lack of solubility of AgF, AgCl, and 2 under the
reaction conditions were thus all key for the reaction to occur
but without the product decomposing.
+
number of atoms and charge as complex 1 but had the
TriNOx3 ligand in the same conformation as in complex 2,
−
3−
enabling comparison of the free energy of TriNOx in each
conformation at the Ti(IV) cation (Scheme 3).
On the basis of Scheme 3, ΔG , the free energy of ligand
rearrangement, was calculated in the gas phase and in benzene,
chloroform, and acetonitrile by subtracting the total free energy
LR
+
+
of 1 -open from the free energy of 1 . ΔG approximated the
LR
3
−
energy difference between the two TriNOx coordination
modes. The heterolytic Ti−F bond strength was estimated by
subtracting the free energy of 2 from the sum of the free
1
9
F NMR spectroscopy supported the claim that the fluoride
+
1
9
energies of 1 and a fluoride anion.
The calculated ΔG values in each solvent continuum are
shown in Table 1. The negative values of ΔG indicated that
ion was inner sphere in solution. The F chemical shift of 2
was 114.6 ppm in CD Cl , while free fluoride ions (from
LR
2
2
LR
tetramethylammonium fluoride) in CH Cl appeared at −97
2
2
2
7
19
28
3
ppm. The reported F chemical shifts of (C H Me)TiF
5
4
29
Table 1. Calculated ΔGLR Values and Heterolytic Ti−F bond
and {[Me Si(C H ) ]TiF } in CDCl were 131.0 and 68.8
a
2
5
4 2
2
3
1
13
Strengths
ppm, respectively. The H and C NMR spectra of 1 and 2
were similar; both sets of data indicated a C -symmetric
solution
3
complex (Figures S1−S4 in the Supporting Information). The
gas
phase
+
[
Ti(TriNOx)] cation of 1 was expected to show C symmetry,
C H
CHCl3
CH CN
3
6
6
3
but the NMR data for 2 were unexpected, on the basis of the
rearrangement of one nitroxide arm in the solid-state structure.
A gas-phase geometry optimization of 2 predicted a minimum
energy structure close to that shown in Figure 3 (vide supra),
showing that the results of the calculations were consistent with
the asymmetric coordination mode observed in the solid state.
However, the NMR data indicated that 2 was symmetric in
solution at 300 K with the three nitroxide groups rapidly
ΔGLR (kJ/mol)
−179.7
754.3
−151.7
474.6
−135.8
365.5
−121.6
277.1
heterolytic Ti−F bond strength
(
kJ/mol)
a
+
+
Calculated free energies of 1 , 1 -open, and 2 in each solvent appear
in Table S1 in the Supporting Information.
1+
was more stable than 1 -open, as expected. ΔG increased
+
LR
from −179.7 kJ/mol in the gas phase to −121.6 kJ/mol in
acetonitrile as the solvent polarity increased. As expected, the
heterolytic Ti−F bond strength was weaker in more polar
solvents. On the basis of our calculations, the ligand
rearrangement energy change was comparable to that of
heterolytic cleavage of the Ti−F bond in acetonitrile and
amounted to a significant contribution to fluorine transfer
reactions with 2.
2
1
interconverting between the η -NO and κ -O coordination
1
19
modes. Variable-temperature H and F NMR experiments on
2
in CD Cl showed that the spectra retained C symmetry to
2
2
3
200 K (Figures S6 and S7 in the Supporting Information).
Thus, the energetic barrier to this fluxional process was small.
Computational Studies of Complexes. An important
consequence of these observations was that the higher energy
3
−
conformation of TriNOx would be expected to destabilize the
For comparison we calculated the heterolytic Ti−F bond
strength of the azatitanatrane [(HNCH CH ) N]TiF. Calcu-
2 2 3
30
fluoride ligand, as loss of the fluoride ligand would allow the
9
590
DOI: 10.1021/acs.inorgchem.5b01687
Inorg. Chem. 2015, 54, 9588−9593

Inorganic Chemistry
Article
lations showed that this system exhibited only very minor
additional factor that drove the Ti−F bond cleavage was
−
−
−
ligand rearrangement after removal of the fluoride ion and
precipitation of [Ti(TriNOx)]X (X = Br , Cl , PF ) in THF.
6
+
optimization of {[(HNCH CH ) N]Ti} , mainly due to
Fluorination reactions with titanium−fluoride complexes and
organic substrates are uncommon in the literature, although
examples of C−F bond activation with titanium complexes and
2
2 3
shortening of the Ti−N bonds. By the same method used to
calculate the values in Table 1 (see the Supporting Information
for full details), the heterolytic Ti−F bond strength of
2
9,31−35
organofluorine compounds have been reported.
There
[
(HNCH CH ) N]TiF was calculated to be 992.6 kJ/mol in
are also many examples of reactions with organosilicon
2
2 3
31,36−39
the gas phase, almost 250 kJ/mol larger than that in complex 2.
This result supported the hypothesis that ligand rearrangement
between 2 and 1 contributed to the heterolytic Ti−F bond
scission.
substrates.
Taw et al. reported reaction of Cp Ti(CF )-
2 3
(F) with 5 equiv of (CH ) SiCl to form Cp Ti(CF )(Cl). In
3
3
2
3
contrast, 2 instantly reacted with (CH ) SiCl to completion.
3
3
Thus, while the high Ti−F bond strength disfavored fluoride
3−
Fluorine Transfer Reactions with Ti(TriNOx)F. On the
transfer, the high-energy conformation of TriNOx in 2
increased the reactivity of the fluoride.
basis of the instantaneous reaction of 2 with (CH ) SiCl to
3
3
form (CH ) SiF in 98% yield (entry 1, Table 2), we decided to
3
3
CONCLUSIONS
■
a
Table 2. Fluorine Transfer Reactions with 2
Two complexes of titanium with a tripodal nitroxide ligand
were synthesized. Percent buried volume calculations con-
firmed that ligand rearrangement upon replacing the chloride
counterion with fluoride permitted coordination of the fluoride
ligand to the titanium center. DFT calculations supported the
hypothesis that the ligand rearrangement stored energy,
providing driving force for fluorine transfer reactions. Reactivity
studies confirmed that the titanium fluoride complex could
rapidly fluorinate highly electrophilic substrates. On the basis of
these results, high-energy ligand rearrangements may contrib-
ute to new strategies for the development of fluoride transfer
reagents.
b
entry
substrate
temp (°C)
time
conversn (%)
c
d
1
2
3
4
5
6
7
(CH ) SiCl
23
23
60
60
60
60
60
5 min
5 min
1 h
98
42
98
3
3
e
d
[Ph C][PF ]
3
6
Ph CBr
3
,
fh
Ph CCl
24 h
24 h
1 h
dec
dec
18
4
3
tBuBr
,
gh
EXPERIMENTAL SECTION
General Methods. Unless otherwise noted, all reactions and
manipulations were performed under an inert atmosphere (N ) using
■
PhCOCl
4-(CH )-C H SO Cl
2 h
3
6
4
2
2
a
standard Schlenk techniques or in a Vacuum Atmospheres, Inc. Nexus
II drybox equipped with a molecular sieve 13X/Q5 Cu-0226S catalyst
purifier system. Glassware was oven-dried for at least 3 h at 150 °C
prior to use. H and C NMR spectra were obtained on a Bruker
DMX-300 Fourier transform NMR spectrometer at 300 and 75.4
MHz, respectively. 19F NMR spectra were obtained on a Bruker DMX-
Reactions were performed in a sealed J-Young NMR tube with 10 mg
of complex 2 and 2 equiv of the substrate, unless specified otherwise.
b
19
Percent conversion from d F NMR with PhF internal standard.
1
13
c
e
Substrate was added in air. The reaction was complete instantly. 1.4
f
g
equiv of substrate. 81% of 2 remained after 24 h. 41% of 2 remained
after 24 h. 90% of 2 remained after 24 h at 60 °C without substrate.
h
3
3
00 or DMX-360 Fourier transfer NMR spectrometer at 282.2 or
38.7 MHz, respectively. Chemical shifts were recorded in units of
1
13
parts per million (for H and C spectra) downfield from residual
solvent signals or (for 19F spectra) referenced to an external CFCl3
reference at 0 ppm, an internal fluorobenzene standard at −113.15
ppm (for fluorine transfer reactions), or an internal hexafluorobenzene
standard at −164.9 ppm (for variable-temperature NMR). All coupling
constants were reported in Hz. H and F variable-temperature NMR
measurements were carried out at 300 and 282 MHz, respectively.
Infrared spectra were obtained from 400 to 4000 cm− using a
PerkinElmer Spectrum 100 FT-IR spectrometer. High-resolution mass
spectra were measured using a Waters 2695 Separations Module
(1S0RR23444).
investigate the scope of fluorine transfer reactions with 2 in
THF (see the Experimental Section for details of fluorine
transfer reaction procedures). Complex 2 was initially reacted
with 1-iodobutane, benzyl chloride, and 4-nitrobenzyl bromide,
substrates ideal for associative reaction pathways. However, no
fluorine-transfer products were observed in these cases. We
then elected to react 2 with [Ph C][PF ] (entry 2), an ideal
1
19
1
3
6
substrate for dissociative chemistry, as it acts as a source of the
trityl carbocation. A precipitate, presumably [Ti(TriNOx)]-
19
[
PF ], formed instantly from this reaction, and F NMR
6
Materials. Tetrahydrofuran, acetonitrile, hexanes, diethyl ether,
fluorobenzene, and dichloromethane were purchased from Fisher
showed 42% conversion to Ph CF. No other products were
3
observed by 19F NMR; it remains unclear what other products
Scientific. The solvents were sparged for 20 min with dry N and dried
2
were formed. Complex 2 reacted cleanly with Ph CBr at 60 °C,
giving 98% conversion to Ph CF (entry 3). However, complex
3
using a commercial two-column solvent purification system compris-
ing columns packed with Q5 reactant and neutral alumina, respectively
3
2
slowly decomposed with no conversion to the fluorinated
(for hexanes), or two columns of neutral alumina (for THF, Et O,
2
t
product on reaction with Ph CCl or BuBr at 60 °C (entries 4
CH
Cl , fluorobenzene, and CH CN). Deuterated solvents were
2 2 3
3
purchased from Cambridge Isotope Laboratories, Inc., and stored over
and 5, respectively). These results indicated that the generation
of a carbocation was necessary for 2 to fluorinate an alkyl
halide.
9
4
Å molecular sieves for several days prior to use. H TriNOx,
3
40
n
TiCl (THF) , and the supporting electrolyte [ Pr N][B(3,5-(CF ) -
4
2
4
3 2
41
C H ) ] were prepared according to literature procedures. All other
6
3 4
Poor yields of the corresponding fluorinated products were
materials were purchased from commercial sources. Triethylamine
obtained when 2 was reacted with PhCOCl and 4-(CH )-
C H SO Cl (Table 2, entries 6 and 7, respectively). These
fluorination reactions indicated that the Ti−F bond of 2 could
only be broken using highly electrophilic molecules. An
3
(Et N) was used after three freeze−pump−thaw cycles. Benzyl
3
6
4
2
chloride and hexafluorobenzene were distilled under vacuum and
used after three freeze−pump−thaw cycles. 2-Bromo-2-methylpropane
t
( BuBr) was washed with CaCl and K CO and dried over molecule
2
2
3
9
591
DOI: 10.1021/acs.inorgchem.5b01687
Inorg. Chem. 2015, 54, 9588−9593

Inorganic Chemistry
Article
sieves after three freeze−pump−thaw cycles prior to use. Benzoyl
separatory funnel with distilled water to remove Et N·HCl. The
3
chloride (PhCOCl) was washed with CaCl and dried over molecular
organic layer was extracted three times with DCM, and the combined
organic portions were dried under reduced pressure to yield a crude
yellow solid. Complex 1 was isolated as yellow crystals by layering
2
sieves after three freeze−pump−thaw cycles prior to use. 1-Iodobutane
was washed with MgSO and filtered through activated alumina after
4
three freeze−pump−thaw cycles prior to use. 4-Methylbenzenesul-
Et O onto a saturated DCM solution at 0 °C. Yield: 0.302 g, 52.9%. X-
2
fonyl chloride (4-(CH )-C H SO Cl) and ferrocene (Fc) were
ray-quality crystals of 1 were obtained by cooling a saturated DCM
solution of 1 to −25 °C. Anal. Calcd for C H ClN O Ti: C, 63.01;
3
6
4
2
sublimed under vacuum at 80 and 23 °C, respectively, prior to use.
Cesium fluoride (CsF) was dried under vacuum at 90 °C prior to use.
Silver fluoride (AgF), trimethylsilyl chloride ((CH ) SiCl), 4-nitro-
33
45
4
3
H, 7.21; N, 8.91. Found: C, 62.76; H, 7.56; N, 8.91. FT-IR (KBr,
−
1
cm ): 3326, 2976, 2940, 1959, 1601, 1488, 1456, 1397, 1367, 1314,
1278, 1261, 1235, 1220, 1174, 1117, 1033, 1003, 965, 946, 876, 856,
3
3
benzyl bromide, triphenylmethyl bromide (Ph CBr), triphenylmethyl
3
1
chloride (Ph CCl), and triphenylmethyl hexafluorophosphate
838, 799, 782, 763, 730, 697, 642, 629, 617, 561, 544, 524. H NMR
3
(
[Ph C][PF ]) were used as received.
(300 MHz, CDCl
3
): δ 7.89 (1 H, d, J = 6.6 Hz, Ar-H), 7.45−7.34 (3
3
6
General Procedure for Fluorine Transfer Reactions. In a J.
Young NMR tube was placed a protio THF solution of ∼10 mg of
complex 2 and fluorobenzene as an internal standard. An initial 19
NMR spectrum was recorded, and then the corresponding substrate
H, m, Ar-H), 4.60 (1 H, d, J = 12.6 Hz, −CHH−), 4.12 (1 H, d, J =
13
12.6 Hz, −CHH−), 0.84 (9 H, s, −C(CH
CDCl ): δ 143.3 (Ar-C), 133.9 (Ar-C), 132.6 (Ar-C), 129.8 (Ar-C),
129.6 (Ar-C), 129.4 (Ar-C), 67.5 (−CH
) ). C NMR (75.4 MHz,
3 3
F
3
−), 59.9 (−C(CH
Ti, [1 − Cl]
)
), 27.0−
2
3
3
19
was added under N . The reaction was monitored by F NMR
(−C(CH
3
)
3
). HRMS (ESI): m/z calcd for C33
H
45
N
O
4
3
2
spectroscopy and was considered complete when the signal at 163
ppm (the position of the 19F resonance of 2 in THF) disappeared
completely. Percent conversion values were calculated on the basis of
integration of the product against the internal standard.
Electrochemistry. Cyclic voltammetry (CV) experiments were
performed using a CH Instruments 620D Electrochemical Analyzer/
Workstation, and the data were processed using CHI software v 9.24.
593.2971, found 593.2975.
t
Synthesis of [((2- BuNO)C
H
CH
)
N]TiF (2). A vial with
6
4
2
3
complex 1 (0.0408 g, 0.0645 mmol, 1 equiv) and AgF (0.0403 g,
0.317 mmol, 4.9 equiv) was cooled to −25 °C. In the vial was quickly
placed 1.5 mL of −25 °C MeCN. The reaction mixture was briefly
stirred and then placed in the dark at −25 °C for 20 h, after which
time X-ray-quality crystals of 2 had formed. The products were filtered
and washed with −25 °C MeCN and then hexanes. Complex 2 was
quickly (to mitigate oxidation of 2 by AgCl and unreacted AgF) rinsed
into a clean vial with THF and dried under reduced pressure. Complex
2 was isolated as a pure yellow-orange powder by redissolving the
All experiments were performed in an N atmosphere drybox using
2
electrochemical cells that consisted of a 4 mL vial, glassy-carbon (3
mm diameter) working electrode, a platinum-wire counter electrode,
and a silver wire plated with AgCl as a quasi-reference electrode. The
working electrode surfaces were polished prior to each set of
experiments and were periodically replaced on scanning >0 V versus
ferrocene (Fc) to prevent the buildup of oxidized product on the
electrode surfaces. Potentials were recorded in CH Cl2 and were
referenced versus Fc, which was added as an internal standard for
calibration at the end of each run. Solutions employed during CV
studies were ∼3 mM in analyte and 100 mM in [ Pr N][B(3,5-(CF ) -
C H ) ] ([ Pr N][BAr ]). All data were collected in a positive-
feedback IR compensation mode. The CH Cl solution cell resistances
were measured prior to each run to ensure resistances ≤∼500 Ω.
Scan rate dependences of 50−1000 mV/s were performed to
determine electrochemical reversibility.
X-ray Crystallography. X-ray intensity data were collected on a
Bruker APEXII CCD area detector employing graphite-monochro-
mated Mo Kα radiation (λ = 0.71073 Å) at a temperature of 100(1) K.
In all cases, rotation frames were integrated using SAINT, producing
a listing of unaveraged F and σ(F ) values that were then passed to
the SHELXTL program package for further processing and structure
solution on a Dell Pentium 4 computer. The intensity data were
corrected for Lorentz and polarization effects and for absorption using
TWINABS or SADABS. The structures were solved by direct
methods (SHELXS-97). Refinement was by full-matrix least squares
based on F using SHELXL-97. All reflections were used during
refinements. Non-hydrogen atoms were refined anisotropically, and
hydrogen atoms were refined using a riding model.
dried solid in Et
0.0231 g, 58.1%. Anal. Calcd for C33
O, filtering, and drying under reduced pressure. Yield:
Ti: C, 64.70; H, 7.40;
2
H
45FN O
4 3
−1
N, 9.15. Found: C, 64.65; H, 7.53; N, 9.04. FT-IR (KBr, cm ): 3370,
3061, 2969, 2932, 2799, 2719, 1959, 1597, 1582, 1483, 1460, 1448,
1392, 1367, 1358, 1307, 1259, 1231, 1185, 1125, 1114, 1092, 1051,
1039, 1032, 1016, 986, 956, 932, 892, 861, 839, 795, 781, 759, 743,
707, 659 (vs, Ti−F), 636, 619, 602, 581, 560, 539, 521, 503, 488. H
NMR (300 MHz, CD Cl ): δ 7.42 (1 H, d, J = 8.1 Hz, Ar-H), 7.30−
2 2
2
n
1
4
3 2
n
F
4
6
3
4
4
7
.21 (2 H, m, Ar-H), 7.17−7.12 (1 H, m, Ar-H), 4.25 (1 H, d, J = 12.0
2
2
4
1
Hz, −CHH−), 2.69 (1 H, d, J = 12.0 Hz, −CHH−), 0.79 (9 H, s,
13
−
C(CH ) ). C NMR (75.4 MHz, CD Cl ): δ 148.9 (Ar-C), 133.6
3 3 2 2
(
(
Ar-C), 133.3 (Ar-C), 128.3 (Ar-C), 128.0 (Ar-C), 126.6 (Ar-C), 66.2
19
−CH −), 60.7 (−C(CH ) ), 26.8 (−C(CH ) ). F NMR (282.2
MHz, CD Cl ): δ 114.6.
2
3
3
3 3
2
2
Fluorine Transfer Reactions. Triphenylmethyl Fluoride. The
General Procedure for Fluorine Transfer Reactions” was followed
4
2
“
2
2
using 1.4 equiv of triphenylmethyl hexafluorophosphate (42%
conversion) or 2 equiv of triphenylmethyl bromide (98% conversion)
as substrates. 19F{ H} NMR (282 MHz, THF): δ − 126.2.
Trimethylsilyl Fluoride. The “General Procedure for Fluorine
Transfer Reactions” was followed, except that 2 equiv of a stock
solution of the substrate trimethylsilyl chloride (0.1 M in THF) was
added in air (98% conversion). 19F{ H} NMR (282 MHz, THF): δ
43
1
44
45
46
1
2
46
−
157.1.
Benzoyl Fluoride. The “General Procedure for Fluorine Transfer
Reactions” was followed, except that 2 equiv of a stock solution of the
Percent Buried Volume Calculations. SambVca was used to
2
1
substrate benzoyl chloride (0.1 M in THF) was added under N (18%
calculate the percent buried volume (%VBur) for each complex. The
amine atom N(4) in each structure (see Figures 2 and 3) was
designated as the atom coordinated to the metal, and the Ti(1)−N(4)
distance in each complex was designated as the distance from the
center of the sphere (2.28 Å for 1 and 3.07 Å for 2). The calculations
used the following parameters: sphere radius, 3.5 Å; mesh spacing,
2
conversion). 19F{ H} NMR (282 MHz, THF): δ 17.2.
1
4
-Methylbenzenesulfonyl Fluoride. The “General Procedure for
Fluorine Transfer Reactions” was followed using 2 equiv of 4-
methylbenzenesulfonyl chloride as a substrate (4% conversion).
19
1
F{ H} NMR (282 MHz, THF): δ 66.4.
0
.05 Å; H atoms included; Bondi radii scaled by 1.17.
t
Synthesis of {Ti[((2- BuNO)C H CH ) N]}Cl (1). To a stirred
6
4
2 3
ASSOCIATED CONTENT
■
THF solution (15 mL) of TiCl (THF) (0.303 g, 0.907 mmol, 1
4
2
* Supporting Information
S
equiv) was added a THF solution (15 mL) of H TriNOx (0.500 g,
3
0
.911 mmol, 1.0 equiv) and Et N (0.51 mL, 3.7 mmol, 4.0 equiv)
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.inorg-
chem.5b01687.
3
dropwise at −78 °C. Stirring was continued at −78 °C for 1 h. The
reaction mixture was then warmed to room temperature and was
stirred at that temperature for 22 h. Workup was performed in air. The
resulting solid precipitate was isolated by vacuum filtration and rinsed
with THF. The filtered solid was dissolved in DCM and added to a
Computational details, H, 13C, and 19F NMR data, H
1
1
19
and F VT-NMR data, electrochemical data, DFT-
9
592
DOI: 10.1021/acs.inorgchem.5b01687
Inorg. Chem. 2015, 54, 9588−9593

Inorganic Chemistry
optimized coordinates for{[(HNCH CH ) N]Ti} and
Article
+
(
23) Maity, A. K.; Fortier, S.; Griego, L.; Metta-Magan
̃
a, A. J. Inorg.
2
2 3
Chem. 2014, 53, 8155−8164.
[
(HNCH CH ) N]TiF in the gas phase, and DFT-
2 2 3
+
́
(24) Conley, M. P.; Forrest, W. P.; Mougel, V.; Coperet, C.; Schrock,
optimized coordinates for 1 and 2 in the gas phase,
benzene, chloroform, and acetonitrile (PDF)
Crystallographic data for compound 1 (CIF)
Crystallographic data for compound 2 (CIF)
R. R. Angew. Chem., Int. Ed. 2014, 53, 14221−14224.
(
25) Thangavel, A.; Wieliczko, M.; Bacsa, J.; Scarborough, C. C.
Inorg. Chem. 2013, 52, 13282−13287.
(26) Connelly, N. G.; Geiger, W. E. Chem. Rev. 1996, 96, 877−910.
27) Christe, K. O.; Wilson, W. W.; Wilson, R. D.; Bau, R.; Feng, J. A.
(
J. Am. Chem. Soc. 1990, 112, 7619−7625.
AUTHOR INFORMATION
Corresponding Author
E-mail for E.J.S.: schelter@sas.upenn.edu.
■
(28) Herzog, A.; Liu, F.-Q.; Roesky, H. W.; Demsar, A.; Keller, K.;
Noltemeyer, M.; Pauer, F. Organometallics 1994, 13, 1251−1256.
̈
(29) Kuehnel, M. F.; Holstein, P.; Kliche, M.; Kruger, J.; Matthies, S.;
*
Nitsch, D.; Schutt, J.; Sparenberg, M.; Lentz, D. Chem. - Eur. J. 2012,
8, 10701−10714.
30) Rioux, F.; Schmidt, M. W.; Gordon, M. S. Organometallics 1997,
6, 158−162.
(31) Nikiforov, G. B.; Roesky, H. W.; Koley, D. Coord. Chem. Rev.
014, 258−259, 16−57.
32) Santamaría, C.; Beckhaus, R.; Haase, D.; Saak, W.; Koch, R.
Chem. - Eur. J. 2001, 7, 622−626.
33) Burk, M. J.; Staley, D. L.; Tumas, W. J. Chem. Soc., Chem.
Commun. 1990, 809−810.
34) Fout, A. R.; Scott, J.; Miller, D. L.; Bailey, B. C.; Pink, M.;
Notes
1
(
1
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
2
(
We thank the U.S. DOE, Office of Science, Early Career
Research Program (DE-SC0006518), Research Corporation for
Science Advancement (Cottrell Scholar Award to E.J.S.) and
the University of Pennsylvania for support of this work. This
work also used the Extreme Science and Engineering Discovery
Environment (XSEDE), which is supported by U.S. National
Science Foundation Grant No. ACI-1053575.
(
(
Mindiola, D. J. Organometallics 2009, 28, 331−347.
(35) Piglosiewicz, I. M.; Kraft, S.; Beckhaus, R.; Haase, D.; Saak, W.
Eur. J. Inorg. Chem. 2005, 2005, 938−945.
(
36) Gru
Goesmann, H. Z. Anorg. Allg. Chem. 1996, 622, 1091−1096.
37) Haiges, R.; Boatz, J. A.; Schneider, S.; Schroer, T.; Yousufuddin,
M.; Christe, K. O. Angew. Chem., Int. Ed. 2004, 43, 3148−3152.
38) Shah, S. A. A.; Dorn, H.; Gindl, J.; Noltemeyer, M.; Schmidt, H.-
G.; Roesky, H. W. J. Organomet. Chem. 1998, 550, 1−6.
39) Taw, F. L.; Clark, A. E.; Mueller, A. H.; Janicke, M. T.; Cantat,
̈
̈
n, M.; Harms, K.; Kocker, R. M. Z.; Dehnicke, K.;
REFERENCES
■
(
(
(
1) Isanbor, C.; O’Hagan, D. J. Fluorine Chem. 2006, 127, 303−319.
2) Mu
3) Campbell, M. G.; Ritter, T. Org. Process Res. Dev. 2014, 18, 474−
(
̈
ller, K.; Faeh, C.; Diederich, F. Science 2007, 317, 1881−1886.
(
4
80.
4) Neumann, C. N.; Ritter, T. Angew. Chem., Int. Ed. 2015, 54,
216−3221.
5) Williams, U. J.; Robinson, J. R.; Lewis, A. J.; Carroll, P. J.; Walsh,
P. J.; Schelter, E. J. Inorg. Chem. 2014, 53, 27−29.
6) Yin, H.; Lewis, A. J.; Williams, U. J.; Carroll, P. J.; Schelter, E. J.
Chem. Sci. 2013, 4, 798−805.
7) Yin, H.; Lewis, A. J.; Carroll, P.; Schelter, E. J. Inorg. Chem. 2013,
2, 8234−8243.
8) Yin, H.; Robinson, J. R.; Carroll, P. J.; Walsh, P. J.; Schelter, E. J.
Chem. Commun. 2014, 50, 3470−3472.
9) Bogart, J. A.; Lippincott, C. A.; Carroll, P. J.; Schelter, E. J. Angew.
Chem., Int. Ed. 2015, 54, 8222−8225.
10) Mitzel, N. W.; Parsons, S.; Blake, A. J.; Rankin, D. W. H. J.
Chem. Soc., Dalton Trans. 1996, 2089−2093.
11) Wieghardt, K.; Tolksdorf, I.; Weiss, J.; Swiridoff, W. Z. Anorg.
Allg. Chem. 1982, 490, 182−190.
12) Mahanthappa, M. K.; Cole, A. P.; Waymouth, R. M.
Organometallics 2004, 23, 836−845.
13) Mahanthappa, M. K.; Huang, K.-W.; Cole, A. P.; Waymouth, R.
M. Chem. Commun. 2002, 502−503.
14) Hughes, D. L.; Jimenez-Tenorio, M.; Leigh, G. J.; Walker, D. G.
J. Chem. Soc., Dalton Trans. 1989, 2389−2395.
15) Dove, A. P.; Xie, X.; Waymouth, R. M. Chem. Commun. 2005,
152−2154.
16) Willner, A.; Niemeyer, J.; Mitzel, N. W. Dalton Trans. 2009,
473−4480.
17) Schroder, K.; Haase, D.; Saak, W.; Beckhaus, R.; Kretschmer, W.
P.; Lutzen, A. Organometallics 2008, 27, 1859−1868.
18) Huang, K.-W.; Han, J. H.; Cole, A. P.; Musgrave, C. B.;
Waymouth, R. M. J. Am. Chem. Soc. 2005, 127, 3807−3816.
19) Barroso, S.; Madeira, F.; Calhorda, M. J.; Ferreira, M. J.; Duarte,
M. T.; Martins, A. M. Inorg. Chem. 2013, 52, 9427−9439.
20) Majumdar, S.; Stauber, J. M.; Palluccio, T. D.; Cai, X.; Velian,
(
3
(
(
T.; Scott, B. L.; Hay, P. J.; Hughes, R. P.; Kiplinger, J. L.
Organometallics 2012, 31, 1484−1499.
(40) Manxzer, L. E.; Deaton, J.; Sharp, P.; Schrock, R. R. Inorg. Synth.
(
1
(
982, 21, 135−140.
41) Thomson, R. K.; Scott, B. L.; Morris, D. E.; Kiplinger, J. L. C. R.
(
5
(
Chim. 2010, 13, 790−802.
(
(
(
Go
(
42) SAINT; Bruker AXS Inc., Madison, WI, 2009.
43) SHELXTL; Bruker AXS Inc., Madison, WI, 2009.
44) Sheldrick, G. M. TWINABS; University of Go
ttingen, Germany, 2008.
45) Sheldrick, G. M. SADABS; University of Go
Germany, 2007.
46) Sheldrick, G. M. Acta Crystallogr., Sect. A: Found. Crystallogr.
008, 64, 112−122.
̈
ttingen,
(
̈
̈ ̈
ttingen, Gottingen,
(
(
2
(
(
(
(
(
2
(
4
(
̈
̈
(
(
(
A.; Rybak-Akimova, E. V.; Temprado, M.; Captain, B.; Cummins, C.
C.; Hoff, C. D. Inorg. Chem. 2014, 53, 11185−11196.
(
21) Poater, A.; Cosenza, B.; Correa, A.; Giudice, S.; Ragone, F.;
Scarano, V.; Cavallo, L. Eur. J. Inorg. Chem. 2009, 2009, 1759−1766.
(
22) Clavier, H.; Nolan, S. P. Chem. Commun. 2010, 46, 841−861.
9
593
DOI: 10.1021/acs.inorgchem.5b01687
Inorg. Chem. 2015, 54, 9588−9593