Potassium and well-defined neutral and cationic calcium fluoroalkoxide complexes: Structural features and reactivity

Article abstract of DOI:10.1021/om500343w

The fluorinated aminoether alcohols (1-aza-12-crown-4)CH₂C(CF₃)₂OH ({RO₁^F}H), (MeOCH₂CH₂)₂NCH₂C(CF₃)₂OH ({RO₂^F}H

Full text of DOI:10.1021/om500343w

Article
pubs.acs.org/Organometallics
Potassium and Well-Defined Neutral and Cationic Calcium
Fluoroalkoxide Complexes: Structural Features and Reactivity
Sorin-Claudiu Rosç a, Thierry Roisnel, Vincent Dorcet, Jean-Franco̧ is Carpentier,* and Yann Sarazin*
Institut des Sciences Chimiques de Rennes, UMR 6226 CNRS, Universite
France
́
de Rennes 1, Campus de Beaulieu, 35042 Rennes Cedex,
S
* Supporting Information
F
ABSTRACT: The fluorinated aminoether alcohols (1-aza-12-crown-4)CH2C(CF3)2OH ({RO1 }H), (MeOCH2CH2)2NCH2C-
F
F
(CF3)2OH ({RO2 }H), and (MeOCH2CH2)(Me)NCH2C(CF3)2OH ({RO3 }H) have been synthesized and used to prepare the
F
heteroleptic calcium amido complexes {ROx }Ca(N(SiMe2R)2) (4−7; x = 2, 3, R = H, Me). The ability to form stable complexes
varies with the chelating and electron-donating ability of the aminoether alkoxide ligand, as exemplified by our failure to isolate
cleanly the elusive {RO1 }Ca(N(SiMe2R)2). X-ray diffraction studies show that, in the solid state, intramolecular Ca···F
interactions help reach coordinative saturation in the dimeric [{RO2 }Ca(N(SiMe3)2)]2 ([4]2), [{RO2 }Ca(N(SiMe2H)2)]2
F
F
F
F
F
([5]2), [{RO3 }Ca(N(SiMe3)2)]2 ([6]2), and [{RO3 }Ca(N(SiMe2H)2)]2 ([7]2), which crystallized free of solvent coligands.
F
F
Similar stabilizing K···F patterns were found in the polymetallic potassium fluoroalkoxides [{RO1 }K]2 ([1]2), [{RO2 }K]4
F
([2]4), and [{RO3 }K]4 ([3]4); [2]4 and [3]4 form heterocubanes in the solid state. Examination of the XRD data for [1]2−[7]2
shows that metal···F interactions can be favored over binding of Oether atoms for calcium and potassium. Pulse-gradient spin−
echo NMR spectroscopy shows that the complexes remain aggregated in aromatic solvents. The solvent-free salts [{ROx }Ca+]·
F
−
+
−
[H2N{B(C6F5)3}2 ] (x = 1 (8), 2 (9), 3 (10)) are obtained by treating 4−7 with [H(OEt2)2 ]·[H2N{B(C6F5)3}2 ] or by
reacting Ca(N(SiMe3)2)2 with [{ROxF}HH+]·[H2N{B(C6F5)3}2 ]; the solid-state structures of [8·H2O]2 and [9·H2O]2 again
−
showed the presence of Ca···F contacts. Complexes 5−7 are promising catalysts for the regiospecific anti-Markovnikov
hydrophosphination of styrene with diphenylphosphine, affording TOF values as high as 52 molsubst molCa−1 h−1 with up to 400
equiv of substrates within 1−2 h at 60 °C.
has enabled Roesky5 and Harder6 to prepare a collection of
INTRODUCTION
■
heteroleptic {DippNacNac}Ca(Nu)(S)x complexes, while Hill
used the ubiquitous {DippNacNac}Ca(N(SiMe3)2)(THF) to
catalyze a variety of reactions.2,7 Nevertheless, in spite of these
major achievements along with some others,2 the need for new
ligand platforms and routes to robust heteroleptic calcium
complexes remains.
Calcium is a large (rionic = 1.00 Å), electropositive, and
oxophilic metal that forms d0 complexes. Heteroleptic {LX}-
Ca(Nu)(S)x complexes ({LX} = ancillary monoanionic ligand;
Nu = monoanionic nucleophile; S = solvent) are of interest for
catalysis, but they are labile and readily engage in deleterious
Schlenk equilibria. As a result, the coordination chemistry of
calcium has been regarded as challenging. Yet, after Hanusa’s
seminal efforts,1 it has experienced a surge of interest and
synthetic strategies that enable the utilization of calcium
complexes for applications in polymerization, fine chemical
catalysis, and material science have been devised.2 West-
erhausen has shown that the organometallic chemistry of
calcium is diverse and extends further than that of a heavy
Grignard analogue.3 Ruhlandt-Senge has reported low-coor-
dinate [Ca]-amide complexes bearing bulky amides.4 Not least,
the bulky {DippNacNac}− ligand (Dipp = diisopropylphenyl)
In a landmark article, Ruhlandt-Senge discussed that
secondary (noncovalent) interactions such as M···Cπ, M···Nπ,
M···F, and agostic bonds are, in some cases, key for the
stabilization of alkaline-earth (Ae = Ca, Sr, Ba) complexes.8 For
instance, Ae···H−Si agostic interactions allow the synthesis of
Special Issue: Catalytic and Organometallic Chemistry of Earth-
Abundant Metals
Received: March 31, 2014
Published: May 22, 2014
© 2014 American Chemical Society
5630
dx.doi.org/10.1021/om500343w | Organometallics 2014, 33, 5630−5642

Organometallics
Article
Chart 1
Scheme 1. Synthesis of Potassium (1−3) and Heteroleptic Calcium Amide (4−7) Alkoxides
stable {LX}Ae(N(SiMe2H)2)(S)x complexes. This strategy has
proved valuable to isolate heteroleptic complexes,9 but the
benefits are limited in catalysis, for instance when the
N(SiMe2H)2 amide is replaced by another reactive group
prior to entering the catalytic cycle, or because of the intrinsic
(detrimental) reactivity of the Si−H moiety.10 Intramolecular
Ae···F interactions between the metal atom and ancillary ligand
are attractive, because they are both strong and permanent
features of the complex. Fluorinated alkoxide ligands have been
used to prepare volatile compounds suitable for metal organic
chemical vapor deposition.11 We have in the past employed
fluorinated alkoxo ligands bearing CF3 groups in positions α to
the Oalkoxide atom to synthesize group 3, 4, and 13 precatalysts
for the polymerization of α-olefins and lactide.12 These
alkoxides are comparable to phenolates in terms of electronic
properties, and the steric bulk and electron-withdrawing effect
of the α-CF3 prevent the formation of polymetallic species.
F
Well-defined cationic Ae complexes of the type [{RO0 }Ae+]·
F
[X−], where {RO0 }− is a fluorinated alkoxide bearing a 1-aza-
15-crown-5 tether and X− is a weakly coordinating anion, are
stable thanks to strong (ca. 30 kcal mol−1) intramolecular Ae···
F interactions,13 but attempts to prepare the parent charge-
F
neutral {RO0 }Ae(N(SiMe3)2) complexes failed because of
kinetic lability, leading to ligand redistribution in solution.
We report here the use of fluorinated alkoxide {ROx }−
F
ligands with aminoether tethers for the synthesis of heteroleptic
calcium amide complexes {ROxF}Ca(N(SiMe2R)2) (x = 1−3;
R = H, Me; Chart 1). It is shown that Ae···F interactions help
toward the stabilization of these species. Their performance as
catalysts in the intermolecular hydrophosphination of styrene is
illustrated, and their ability to act as precursors for highly
electron-deficient [{ROxF}Ca+]·[X−] is presented.
5631
dx.doi.org/10.1021/om500343w | Organometallics 2014, 33, 5630−5642

Organometallics
Article
both cases) do not provide compelling evidence for the
presence of intramolecular Ca···H−Si interactions; again, low-
temperature NMR brought no improvement, and decoales-
cence of the resonance that could potentially be expected did
not occur at −80 °C. Consistent with this, the FTIR spectra of
5 and 7 show a unique, symmetric absorption band for Si−H
RESULTS AND DISCUSSION
■
Synthesis of Potassium and Calcium Charge-Neutral
Complexes. The protio ligands {ROx }H (prepared in high
yields by an equimolar reaction of the appropriate aminoether
with 2,2-bis(trifluoromethyl)oxirane)14,15 react with KN-
(SiMe3)2 or KN(SiMe2H)2 in diethyl ether to afford the
F
stretching at ν
̃
2016 and 2015 cm−1, i.e. in the region expected
F
corresponding solvent-free potassium salts {ROx }K (x = 1 (1),
for SiH moieties not interacting with a metal center; Ae···H−Si
contacts (Ae = Ca−Ba) are characterized by a band at lower
energy, typically 1900−1980 cm−1, depending on the strength
of the interaction and identity of the metal. The 29Si{1H} NMR
2 (2), 3 (3)) in satisfactory yields (Scheme 1). Complexes 1−3
have been fully characterized; they form a dimer or tetramers in
the molecular solid state, and their polymetallic structure
persists in solution in aromatic solvents (vide infra). These
complexes failed to react cleanly with 1:1 mixtures of CaI2 and
KN(SiMe2R)2 to give heteroleptic amido calcium complexes
during salt metathesis reactions (R = H, Me), and one-pot
29
data for 5 and 7 (δ −25.1 and −25.4 ppm, respectively) are
Si
comparable to those for related {LO}Ca(N(SiMe2H)2)(THF)x
aminoether phenolate complexes.9
Solid-State Structural Investigations. Single crystals of
1−7 were grown typically from concentrated hydrocarbon
(toluene, benzene, and/or pentane) or diethyl ether solutions,
and their structures were determined by X-ray diffraction
studies.
F
reactions of {ROx }H, CaI2, and 2 equiv of KN(SiMe3)2 were
also unsuitable. Analytically pure samples of the targeted
F
heteroleptic complexes {ROx }Ca(N(SiMe2R)2) (x = 2, R =
Me, 4; x = 2, R = H, 5; x = 3, R = Me, 6; x = 3, R = H, 7) were
eventually obtained in moderate yields by protonolysis of
Ca(N(SiMe3)2)2, Ca(N(SiMe3)2)2(THF)2, or Ca(N-
(SiMe2H)2)2(THF) with the suitable protio ligand (Scheme
1). Whereas we anticipated that the protio ligand of highest
denticity should also yield heteroleptic complexes, we were
F
The complex {RO1 }K (1) recrystallized as the centrosym-
metric dimer [1]2, where each metal atom is 8-coordinated and
the two K atoms are directly bridged through the Oalkoxide atoms
(Figure 1). The coordination sphere on each metal is
F
unable to obtain {RO1 }Ca(N(SiMe2R)2) for R = H, Me; this
was reminiscent of the difficulties encountered in our attempts
13
F
at making {RO0 }Ca(N(SiMe2R)2). The identities of 4−7,
which all form solvent-free bimetallic dimers in the solid state
regardless of the nature of the starting material, have been
authenticated by structural analysis and NMR spectroscopy,
and their purity was corroborated by combustion analyses.
Complexes 5−7 are soluble in common nonprotic organic
solvents (ethers, hydrocarbons) and decompose within minutes
in chlorinated solvents (CD2Cl2, CDCl3). Their NMR data
were recorded in benzene-d6; 4 is only sparingly soluble in
hydrocarbons, and THF-d8 was used as the NMR solvent.
The hexamethyldisilazide compounds 4 and to a lesser extent
6 are less stable than the congeneric tetramethyldisilazides 5
and 7; in particular, 4 rapidly shows signs of decomposition in
solution. This was tentatively attributed to the presence of
stabilizing Ca···H−Si interactions in 5 and 7, although their
existence could not be firmly demonstrated by spectroscopic
methods.
The 19F{1H} NMR spectra for the potassium complexes 1−3
and the calcium complexes 4 and 5 exhibit a single, sharp
singlet for the CF3 groups typically in the range δ −77.2 to
F
Figure 1. ORTEP representation of the molecular structure of {RO1 }
K (1), recrystallized as the centrosymmetric dimer [1]2. Hydrogen
atoms are omitted for clarity. Ellipsoids are drawn at the 50%
probability level. Selected bond lengths (Å) and angles (deg): K(1)−
O(15) = 2.630(2), K(1)−O(15)#1 = 2.682(2), K(1)−O(3) =
2.777(2), K(1)−O(6) = 2.821(2), K(1)−O(9) = 2.753(2), K(1)−
N(12) = 2.990(2), K(1)−F(21)#1 = 2.928(2), K(1)−F(23) =
3.035(2), K(1)−K(1)#1 = 3.842(1); K(1)−O(15)−K(1)#1 = 92.65(5).
19F
−81.8 ppm, indicating the equivalence of all CF3 moieties in
solution for these complexes. On the other hand, two quartets
of equal intensity are detected in the spectra of the Ca
complexes 6 (centered at −73.4 and −75.1 ppm; 4JFF = 9.4 Hz)
4
and 7 (centered at −76.5 and −76.8 ppm; JFF = 9.4 Hz),
indicating that the CF3 groups are in these cases nonequivalent.
Hence, the 19F{1H} NMR spectra for 1−7 recorded at 25 °C
provided no evidence for the metal···F intramolecular
interactions detected in all these complexes in the molecular
solid state (vide infra). Attempts to observe splitting of the
resonances by recording low-temperature data (down to −80
°C) did not improve on this situation; evidently the CF3 groups
are highly fluxional, rotating and exchanging positions too easily
for NMR detection methods. In their 1H NMR spectra,
completed by the four heteroelements from the macrocyclic
tether and by two fluorine atoms situated at a short distance
from the metals (average K···F = 2.98 Å). These F···K contacts
fall in the same range as those reported for K[Cu(OC4F9)2]
complexes (average distance K···F = 2.96 Å),16 for the K(18-
crown-6)+ salt of germanium fluorinated pentoxide (K···F =
3.02 and 3.28 Å),17 and for heteropolymetallic [KM(OC-
(CF3)3)3(THF)]4 compounds used for MOCVD (M = Sr,
average K···F = 2.94 Å; M = Ba, average K···F = 3.04 Å ; M =
Eu; average K···F = 2.95 Å).11b In addition to the bridging
1
complexes 5 and 7 exhibit a resonance at δ H 4.88 and 4.86
ppm, respectively, assigned to the SiH moieties. These chemical
shifts and the corresponding 1JSiH coupling constant (162 Hz in
O
alkoxide, each fluorinated alkoxide dispenses one K···F contact
5632
dx.doi.org/10.1021/om500343w | Organometallics 2014, 33, 5630−5642

Organometallics
Article
F
Figure 2. (left) ORTEP representation of the molecular structure of {RO2 }K (2), recrystallized as the tetramer [2]4. Hydrogen atoms are omitted
for clarity. (right) Simplified representation of the K4O4 heterocubane in [2]4, showing the coordination sphere around K atoms and the atom-
numbering scheme. Color scheme: F atoms, green; K atoms, purple; C atoms, black; O atoms, red; N atoms, blue. Ellipsoids are drawn at the 50%
probability level. Selected bond lengths (Å): K(1)−O(8) = 2.815(2), K(1)−O(20) = 2.642(1), K(1)−O(22) = 2.707(1), K(1)−O(41) = 2.707(1),
K(1)−F(14) = 2.879(1), K(1)−F(25) = 3.076(1), K(1)−F(28) = 2.862(1), K(1)−F(50) = 3.155(1), K(1)−N(1) = 3.060(2), K(2)−O(22) =
2.639(1), K(2)−O(39) = 2.777(1), K(2)−O(41) = 2.800(1), K(2)−O(70) = 2.699(1), K(2)−F(30) = 2.830(1), K(2)−F(44) = 3.138(1), K(2)−
F(45) = 2.857(1), K(2)−N(35) = 2.979(2), K(3)−O(20) = 2.627(1), K(3)−O(41) = 2.620(1), K(3)−O(67) = 2.738(2), K(3)−O(70) = 2.775(1),
K(3)−F(74) = 2.865(1), K(3)−F(75) = 3.238(2), K(3)−N(60) = 3.060(2), K(4)−O(20) = 2.692(1), K(4)−O(22) = 2.634(1), K(4)−O(70) =
2.599(1), K(4)−O(88) = 2.708(2), K(4)−F(15) = 2.761(1), K(4)−F(17) = 2.962(1), K(4)−N(85) = 2.974(12).
F
Figure 3. (left) ORTEP representation of the molecular structure of {RO3 }K (3), recrystallized as the tetramer [3]4. Only one of the two
independent and comparable molecules found in the asymmetric unit is depicted. Hydrogen atoms are omitted for clarity. (right) Simplified
representation of the K4O4 heterocubane in [3]4, showing the coordination sphere around K atoms and the atom-numbering scheme. Color scheme:
F atoms, green; K atoms, purple; C atoms, black; O atoms, red; N atoms, blue. Ellipsoids are drawn at the 50% probability level. Selected bond
lengths (Å): K(1)−O(1) = 2.774(6), K(1)−O(13) = 2.666(5), K(1)−O(53) = 2.709(5), K(1)−O(73) = 2.708(5), K(1)−N(5) = 2.911(7), K(1)−
F(11) = 2.908(7), K(1)−F(52) = 2.839(5), O(13)−K(2) = 2.744(5), K(2)−O(21) = 2.810(6), K(2)−O(33) = 2.658(5), K(2)−O(53) = 2.643(5),
K(2)−N(25) = 2.910(7), F(10)−K(2) = 2.777(6), F(17)−K(2) = 3.180(7), O(33)−K(3) = 2.699(5), K(3)−O(41) = 2.734(5), K(3)−O(53) =
2.677(5), K(3)−O(73) = 2.739(5), K(3)−N(45) = 2.875(6), K(3)−F(50) = 2.923(5), K(3)−F(77) = 2.931(5), K(3)−F(76) = 3.011(6), O(13)−
K(4) = 2.635(5), O(33)−K(4) = 2.681(5), K(4)−O(61) = 2.765(6), K(4)−O(73) = 2.623(5), K(4)−N(26) = 2.921(6), F(30)−K(4) = 3.020(7),
F(32)−K(4) = 3.081(7), K(1)−K(4) = 3.786(2), K(2)−K(3) = 3.710(2), K(2)−K(4) = 3.767(2).
with each of the metal atoms, and therefore they both act
overall as a μ2:κ6,κ2 chelate. The K(1)···K(1)#1 distance of 3.84
Å is much shorter than that in elemental potassium (4.54 Å)
and smaller than double the van der Waals radius of potassium
(rvdW = 2.75 Å); this coincides with a narrow K(1)−O(15)−
K(1)#1 angle of 92.6°. Similar or even stronger geometric
constraints were for instance seen in the complex
K2Ru2(Co)4(PPh3)2(4-tBu-C6H5O)6 (K···K = 3.60 Å, ∠(K−
O−K) = 78.3°),18 in the potassium calix[4]arene complex
{[calix[4](O)4K]2(μ-K)6(THF)10} (K···K = 3.27 Å, ∠(K−O−
5633
dx.doi.org/10.1021/om500343w | Organometallics 2014, 33, 5630−5642

Organometallics
Article
K) = 75.6°),19 and in [{RO0 }K]2 (K···K = 3.68 Å, ∠(K−O−
K) = 89.3°).
F
F
The complex {RO2 }K (2) recrystallized as the tetramer
[2]4, with a central K4O4 core adopting a slightly distorted
cubic arrangement where each metal atom is adjacent to three
Oalkoxide atoms (Figure 2). The formation of heterocubanes is
well-known for alkali-metal alkoxides, as for instance seen in
[MOC4F9]4 (M = Na, K).20 Each metal atom engages in two to
four nonequivalent K···F interactions with neighboring CF3
groups (average distance K···F (Å): K(1), 2.99; K(2), 2.94;
K(3), 3.05; K(4), 2.86; the limit for significant K···F
interactions was set at 3.23 Å),21 and the coordination sphere
is completed by the Nside arm atom and only one Oside arm atom,
with total coordination numbers varying from to 7 to 9. The
other metric parameters around the metals in [2]4 are
unexceptional. Note that, for each metal, the binding of only
one of the two OMe moieties in the aminoether tether is
required to ensure coordinative saturation, as the addition of
multiple K···F interactions turns out to be more stabilizing than
the coordination of the second OMe, a possibility already
suggested by Ruhlandt-Senge.8a As a matter of fact, the second
methoxy does not interact with any metal atom in the crystal
lattice, and the coordination spheres about the metal atoms in
[2]4 actually resemble closely those in [3]4, also a heterocubane
Figure 4. ORTEP representation of the molecular structure of
[{RO2 }Ca(N(SiMe2H)2)]2 ([5]2). Hydrogen atoms (except SiH) are
F
obtained upon recrystallization of {RO3 }K (2) where the
F
dangling CH2CH2OMe fragment is replaced by a simple methyl
group on the N atom (Figure 3). The heterocubane structure
imposes short K···K distances (ca. 3.60−4.00 Å) and narrow
K−O−K angles in these two complexes. Formation of
omitted for clarity. Ellipsoids are drawn at the 50% probability level.
Selected bond lengths (Å) and angles (deg): Ca(1)−O(11) =
2.295(1), Ca(1)−O(11)#1 = 2.302(1), Ca(1)−N(1) = 2.310(1),
Ca(1)−O(25)#1 = 2.419(1), Ca(1)−N(22)#1 = 2.597(1), Ca(1)−
F(14) = 2.833(1); Si(1)−N(1)−Ca(1) = 103.27(7), Si(2)−N(1)−
Ca(1) = 126.68(8), Ca(1)−O(11)−Ca(1)#1 = 104.63(4), Si(1)−
N(1)−Si(2) = 130.03(9).
F
F
tetramers [2]4 and [3]4 with the ligands {RO2 } and {RO3 },
F
in contrast to the dimer seen with {RO1 } in [1]2 (and with
{RO0 }), can be linked to the lower intrinsic coordinating
21
F
ability of these ligands.
The molecular structure of the dimeric calcium complex [5]2
depicted in Figure 4 shows a peculiar arrangement about the
metal atoms. The positions of the SiH hydrogen atoms were
not idealized but were determined from the electron density
map. In this dimer, which features an inversion center, the
metal atoms are 6-coordinated or, if Ca···H−Si is taken into
account, 7-coordinated. The large discrepancies between the
two Ca−N−Si angles (103.3 and 126.7°) and corresponding
Ca···Si distances (3.15 and 3.58 Å to Si(1) and Si(2),
respectively) indeed suggest that the Si(1)−H(1S) moieties
are involved in agostic bonding with the metal atoms. The
coplanarity of Ca(1)−N(1)−Si(1)−H(1S) is in agreement
with this interpretation.9 Yet, one may also note that Si(2) and
H(2S) also fit in this very plane, even if other metric parameters
do not support the claim of Ca···H(2S)−Si(2) agostic
interactions. Pertinent metric parameters for [5]2 and other
related calcium complexes are collected in Table 1.
whereas the bridging Oalkoxide atoms are also substantially closer
to the metal atoms (ca. 2.30 Å) than the sole Omethoxy atom
involved in the coordination sphere of the metal (Ca(1)−
O(25) = 2.42 Å). Remarkably, in line with the above
observations on [2]4, the other methoxy side arm (viz.
O(29)) in [5]2 does not interact with any metal atom, as
coordinative saturation is filled by the N(SiMe2H)2 moiety and
a Ca···F secondary interaction in addition to the ancillary
ligand.
The geometry in the centrosymmetric dimer [4]2 (Figure 5)
is similar to that in [5]2 (Table 1), except that the potential
Ca···H−Si agostic interaction in the latter in now replaced by a
second, albeit weaker, Ca···F contact. The metal atoms in [4]2
are 7-coordinated, with relatively intense (Ca···F(20) = 2.92 Å)
and loose (Ca···F(15) = 3.11 Å) secondary interactions, while
the rest of the coordination sphere is occupied by the ancillary
ligand. The metric parameters about the metal in [4]2 are
comparable to those measured in [5]2, and again one of the
Omethoxy atoms (viz. O(25)) is not interacting with the metal
atoms. On the other hand, the geometric features around the
Namide atoms are different; the Si−N−Si angle of 121.5° is
narrower than that in [5]2 (130.0°), and the difference between
the two types of Ca−N−Si angles in [4]2 (112.5 and 125.6°) is
not as great as in [5]2. These observations, as well as the
coordination of a second F atom to achieve a coordination
number of 7, seem to militate in favor of a Ca···H−Si agostic
interaction in [5]2 in the solid state.
The structural features presented here may indeed argue in
favor of (at least) Ca···H(1S)−Si(1) agostic bonding, but
spectroscopic methods did not provide supporting evidence
(vide supra), and we are unable to conclude as to the existence
of such agostic bonding on the basis of these experimental data.
There is no ambiguity regarding the Ca(1)···F(14) inter-
action22 characterized by a short distance of 2.83 Å between
these two atoms, even if the intensity of this secondary
interaction is weaker than that detected in {DippNacNac(CF3)2
}
-
Ca(N(SiMe3)2)(THF)2 (Ca···F = 2.47 and 2.49 Å)23 and in the
F
−
F
cationic complex [{RO0 }Ae+]·[H2N{B(C6F5) }2 ] (Ca···F =
The bonding pattern in the centrosymmetric dimer [{RO3 }-
3
2.64−2.68 Å).13 Expectedly, the Ca−Namide (2.31 Å) bond
length is much shorter than the Ca−Namine length (2.60 Å),
CaN(SiMe2H)2]2 ([7]2) matches that in [5]2, except for the
fact that the dangling, noncoordinating CH2CH2OMe side arm
5634
dx.doi.org/10.1021/om500343w | Organometallics 2014, 33, 5630−5642

Organometallics
Article
Table 1. Summary of Metric Data for the Heteroleptic Calcium Complexes [4]2−[7]2
complex
Ca atom
Ca···F (Å)
Ca···Si (Å)
Ca−N−Si (deg)
N−Si−N (deg)
F
[{RO2 }Ca(N(SiMe3)2)]2 ([4]2)
Ca(1) = Ca(1)#1
2.92
3.11
2.83
3.37
3.60
3.15
3.58
3.42
3.62
3.35
3.65
3.29
3.47
112.5
125.6
103.3
126.7
114.4
125.2
109.9
128.0
110.8
120.5
121.5
F
[{RO2 }Ca(N(SiMe2H)2)]2 ([5]2)
Ca(1) = Ca(1)#1
Ca(1)
130.0
119.7
121.6
128.2
F
[{RO3 }Ca(N(SiMe3)2)]2 ([6]2)
3.08
Ca(2)
2.71
3.06
3.00
F
[{RO3 }Ca(N(SiMe2H)2)]2 ([7]2)
Ca(1) = Ca(1)#1
Figure 6. ORTEP representation of the molecular structure of
F
[{RO3 }Ca(N(SiMe2H)2)]2 ([7]2). Hydrogen atoms (except SiH, the
position of which was accurately localized from the electron density
map) are omitted for clarity. Ellipsoids are drawn at the 50%
probability level. Selected bond lengths (Å) and angles (deg): Ca(1)−
N(20) = 2.300(1), Ca(1)−O(17) = 2.311(1), Ca(1)−O(17)#1
=
2.318(1), Ca(1)−O(2) = 2.408(1), Ca(1)−N(5) = 2.571(1), Ca(1)−
F(12)#1 = 3.005(1); Si(2)−N(20)−Si(1) = 128.16(8), Si(2)−N(20)−
Ca(1) = 110.83(6), Si(1)−N(20)−Ca(1) = 120.49(7), Ca(1)−
O(17)−Ca(1)#1 = 105.03(4).
Figure 5. ORTEP representation of the molecular structure of
[{RO2 }Ca(N(SiMe3)2)]2 ([4]2). Hydrogen atoms are omitted for
F
clarity. Ellipsoids are drawn at the 50% probability level. Selected bond
lengths (Å) and angles (deg): Ca(1)−O(11) = 2.332(1), Ca(1)−N(1)
= 2.332(1), Ca(1)−F(15) = 3.113(2), Ca(1)−O(11)#1 = 2.343(1),
Ca(1)−O(29)#1 = 2.375(1), Ca(1)−N(22)#1 = 2.719(2), Ca(1)−
F(20)#1 = 2.920(1); Si(1)−N(1)−Si(2) = 121.48(9), Si(1)−N(1)−
Ca(1) = 112.46(7), Si(2)−N(1)−Ca(1) = 125.64(8), Ca(1)−O(11)−
Ca(1)#1 = 105.64(5).
NMR Diffusion Measurement Experiments. The
nuclearity of the potassium complexes 2 and 3 and calcium
complexes 4−7 in solution was assessed by pulse gradient
spin−echo (PGSE) NMR spectroscopy, following protocols
developed for related alkaline and alkaline-earth complexes.24
All measurements were performed at 298 K, using 13.0−30.0
mM solutions in benzene-d6.25 The validity of the method was
assessed using Si(SiMe3)4 (TMSS) as a reference. From the
PGSE experiments, the translational coefficient Dt was acquired
for all compounds from the plot of ln(I/I0) vs −γ2δ2G2(Δ − δ/
3)Dt (see the Experimental Section for details). The values of
the hydrodynamic radius of the metal complexes (rH,PGSE) thus
determined are collected in Table 2.26 For all complexes, the
value of the hydrodynamic radius determined from the solid-
state structures (ellipsoidal model)24 matches well that
estimated by PGSE experiments, suggesting that they maintain
their polynuclear structure in aromatic hydrocarbons.
that was in [5]2 has now been replaced by a simple methyl
group. As can be seen in Figure 6 and in Table 1, this bears no
incidence on the coordination about the metal atoms. There is
one Ca···F secondary interaction for each metal in [7]2, with a
corresponding distance of 3.00 Å.
F
The arrangement is different in [{RO3 }Ca(N(SiMe3)2)]2
([6]2), displayed in Figure 7. Here, the two calcium atoms are
not equivalent (Table 1). The atom Ca(1) is 6-coordinated,
having only one Ca···F interaction (with F(26), 3.08 Å); the
atom Ca(2) is 7-coordinated as found in [4]2, with two Ca···F
contacts of 2.71 and 3.06 Å. The Ca2O2 central core is not
symmetrical, and the interaction of a single F atom on Ca(1)
generally induces slight deviations from symmetry throughout
the dimer. Note, however, that the values of the two N−Si−N
angles are very close.
These measurements also confirmed that the heteroleptic
F
complexes [{ROx }Ca(N(SiMe2R)2)]2 (R = H, Me) do not
decompose entirely or in part to a mixture of homoleptic
5635
dx.doi.org/10.1021/om500343w | Organometallics 2014, 33, 5630−5642

Organometallics
Article
where X = O, F, N; Met = K, Ca; B = 0.37; RK−O = 2.132, RK−F
= 1.992, RK−N = 2.260; RCa−O = 1.967, RCa−F = 1.842, RCa−N
=
2.140.28,29
For the calcium complex [4]2 (Table 3), coordinative
saturation is not reached if only coordination of the N and O
side atoms is considered. Further extension by inclusion of the
Ca···F interactions affords near coordinative saturation (∑ν =
1.959), and in this case the mean contribution of Ca···F
interactions to the overall coordination sphere approaches 4%.
For [2]4, the mean contribution of K···F interactions to the
coordination sphere (averaged over the four metal atoms)
amounted to a considerable 18% (see the Supporting
Information). This value compares well with the average values
F
determined for [{RO3 }K]4 ([3]4) (17%; see the Supporting
Information) and Ruhlandt-Senge’s [K(PFTB)(THF)4] (20%;
PFTB = perfluoro-tert-butoxide).11b Higher K···F contributions
were found in the tetrameric alkoxides KOC(Ph)(CF3)2 (34%)
and KOCMe(CF3)2 (32%), respectively,30 which is unsurpris-
ing, as the K atoms are not supported by side N/O atoms in
these compounds.
Synthesis of Calcium Cations. Cationic calcium com-
plexes devoid of coordinated solvent are rare because of their
intrinsic sensitivity, but they present an interest in catalysis
owing to their high electrophilicity.13 The compounds
Figure 7. ORTEP representation of the molecular structure of
[{RO3 }Ca(N(SiMe3)2)]2 ([6]2). Hydrogen atoms and noninteracting
F
benzene molecules are omitted for clarity. Ellipsoids are drawn at the
50% probability level. Selected bond lengths (Å) and angles (deg):
Ca(1)−O(2) = 2.305(5), Ca(1)−N(1) = 2.353(7), Ca(1)−O(1) =
2.371(5), Ca(1)−O(12) = 2.397(6), Ca(1)−N(15) = 2.615(7),
Ca(1)−F(26) = 3.084(6), Ca(1)−Ca(2) = 3.760(2), O(1)−Ca(2) =
2.317(5), O(2)−Ca(2) = 2.403(5), F(22)−Ca(2) = 3.056(6), Ca(2)−
N(31) = 2.370(7), Ca(2)−O(42) = 2.475(6), Ca(2)−N(45) =
2.627(7), Ca(2)−F(51) = 2.712(5); Si(4)−N(31)−Si(3) = 121.6(4),
Si(1)−N(1)−Si(2) = 119.7(4).
F
−
F
[{RO1 }Ca+]·[H2N{B(C6F5)3}2 ] (8), [{RO2 }Ca+]·[H2N{B-
(C6F5)3}2 ] (9), and [{RO3 }Ca+]·H2N{B(C6F5)3}2 ] (10)
were obtained as colorless solids by equimolar reactions of
Ca(N(SiMe3)2)2 with the appropriate doubly acidic com-
−
F
−
pounds [{ROxF}HH+]·[H2N{B(C6F5)3}2 ] (x = 1−3; these are
−
F
prepared by treatment of {ROx }H with Bochmann’s acid,
31
[H(OEt2)2}+]·[H2N{B(C6F5)3}2 ]; see the Supporting In-
formation) in chlorobenzene (Scheme 2). Solvents such as
THF must be avoided because of the risk of coordination onto
the metal, and hence the standard synthetic precursor
Ca(N(SiMe3)2)2(THF)2 should not be used. Alternatively, 9
and 10 were also prepared by addition of Bochmann’s acid to 5
and 6, respectively, but this is less convenient, as it entails the
preliminary synthesis of the charge-neutral heteroleptic parent
−
F
species {ROx }2Ca and Ca(N(SiMe3)2)2, as a single diffusion
coefficient was measured in all cases.27
Bond Valence Analysis. The contribution of secondary
interactions (K···F and Ca···F) to the coordination sphere of
the metal atoms can be estimated using bond valence sum
analysis (BVSA). This theoretical model, first introduced by
Brown and Altermatt28 and later developed by O’Keefe and
Brese,29 makes use of bond distances obtained from single-
crystal X-ray diffraction to assign the contribution of each
neighboring atom toward the metallic center through tradi-
tional interactions and secondary interactions. Bond valences
(ν) were calculated for each environment in the complexes
−
complexes. The choice of the counterion H2N{B(C6F5)3}2
was prompted by synthetic considerations. This large (538
Å3),32 weakly coordinating anion possesses a dipolar moment
which provides good crystallization properties, and indeed salts
of this anion are known to crystallize more easily than with the
F
F
[{RO2 }K]4 ([2]4) and [{RO2 }Ca(N(SiMe3)2)]2 ([4]2) using
eq 1, with experimental bond lengths dMet−X and tabulated
spherical, more conventional B(C6F5)4 borate derivative.33
−
The compositions of 8−10 were authenticated by NMR
spectroscopy, and their purity was confirmed by microanalytical
measurements. They are poorly soluble in nonpolar solvents,
bond lengths RMet−X
:
ν = exp[(RMet−X − dMet−X)/B]
(1)
a
Table 2. PGSE NMR Measurements and X-ray Crystallographic Data for Complexes 2−7
X-ray
b
c
complex
Dt (10−9 m2 s−1
)
rH,PGSE (Å)
a (Å)
b (Å)
rH,X‑ray (Å)
F
[{RO2 }K]4 ([2]4)
0.88
0.77
1.33
1.13
0.65
1.49
5.74
6.45
5.25
6.01
6.14
4.75
7.65
5.59
6.93
6.95
7.60
5.56
4.65
5.23
4.48
4.87
4.78
4.68
6.48
5.47
5.99
6.17
6.51
5.25
F
[{RO3 }K]4 ([3]4)
F
[{RO2 }Ca(N(SiMe3)2)]2 ([4]2)
F
[{RO3 }Ca(N(SiMe3)2)]2 ([6]2)
F
[{RO2 }Ca(N(SiMe2H)2)]2 ([5]2)
F
[{RO3 }Ca(N(SiMe2H)2)]2 ([7]2)
a
All NMR spectra were recorded in C6D6 at 298 K using 13.0−30.0 mM solutions of complex and of the external reference Si(SiMe3)4 (TMSS).
b
c
1
Average of the values of Dt found for three or more separate peaks in the H PGSE NMR spectrum of the complex. Calculated according to
rH,X‑ray= (a2b)1/3 where a and b, respectively, the major and minor semiaxes of the prolate ellipsoid formed by the complex, are determined from the
solid-state structures.
5636
dx.doi.org/10.1021/om500343w | Organometallics 2014, 33, 5630−5642

Organometallics
Table 3. Bond Valence Analysis Calculations for [{RO2 }Ca(N(SiMe3)2)]2 ([4]2; Met = Ca(1))
Article
F
a
a
a
b
b
b
d(Ca−O)
d(Ca−F)
d(Ca−N)
ν(Ca−O)
ν(Ca−F)
ν(Ca−N)
∑ν
2.331
2.343
2.374
2.920
3.110
2.332
2.719
0.374
0.362
0.333
1.069
0.054
0.032
0.595
0.209
0.086
0.804
1.959
a
b
Data from the X-ray structure of [4]2. Calculated using eq 1.
Scheme 2. Synthesis of Cationic Calcium Fluoroalkoxide
Complexes
Figure 8. ORTEP representation of one of the two nonequivalent
F
dimeric cations in the asymmetric unit of [{RO2 }Ca+(H2O)]2·
−
2[H2N{B(C6F5)3}2 ] ([9·H2O]2), showing the coordinating H2O
molecules and Ca···F contacts. Counterions, hydrogen atoms, and
noncoordinating solvent molecules are omitted for clarity. O(91) and
O(92) correspond to the O atoms of coordinated water molecules.
The Ca(2)···F(63) and Ca(3)···F(90) distances are in the range 2.70−
2.75 Å.
and we found that they could only be dissolved in THF. 19F
and 11B NMR in THF-d8 testified to the integrity of the anion,
but the existence of Ca···F interactions could not be probed by
NMR spectroscopy owing to the mandatory use of this
coordinating solvent. Attempts to grow single crystals of 8−10
suitable for X-ray diffraction studies were thwarted by their
extreme sensitivity (electrophilicity) and affinity for water.
Attempts using various crystallization techniques only afforded
recrystallization of 8 in dichloromethane to afford the adduct
[8·H2O]2. The structure of the dimeric cationic fragment
depicted in Figure 9 shows the presence of a coordinated water
molecule and a strong intramolecular Ca···F−C interaction
(Ca(1)−F(21) = 2.650(3) Å; Ca(2)−F(67) = 2.651(3) Å)
between each of the metal atoms and CF3 groups of the ligand;
there is no contact with the counterion. All attempts to grow
water-free crystals of these species proved unsuccessful; note
that under the same experimental conditions, the cationic
F
crystals of the water adducts [{RO1F}Ca+(H2O)]2·2[H2N{B-
−
(C6F5)3}2−] ([8·H2O]2) and [{RO2 }Ca+(H2O)]2·2[H2N{B-
[{RO0 }Ca+]2·2[H2N{B(C6F5)3}2 ], where the ligand {RO0 }−
F
−
F
(C6F5)3}2 ] ([9·H2O]2), where the water molecules presum-
ably came from the moisture background level.34 Unsatisfactory
refinement (final R1 = 9.52%) precludes precise discussion of
the metric parameters in [9·H2O]2 (Figure 8), but several
features can nonetheless be mentioned with confidence: (i)
each asymmetric unit contains four anions and two structurally
contains one more Oheterocycle atom than does {RO0 }− (see
F
Chart 1), crystallized without binding water.13
Hydrophosphination Catalysis. Well-defined heteroleptic
calcium complexes have emerged as potent precatalysts in
atom-efficient hydroelementation reactions such as hydro-
amination and hydrophosphination of alkenes.2 Yet, reports of
calcium complexes in the hydrophosphination reactions of
activated alkenes are confined to the use of precatalysts bearing
bulky {DippNacNac}−,7a tridentate amidinate,35 and imino-
anilide ligands.36 The fluorinated alkoxide calcium complexes
5−7 catalyze competently the addition of diphenylphosphine
across the CC vinylic bond in styrene under mild conditions
(Table 4). Complex 4 was not included in this screening, as we
found it was not sufficiently stable under catalysis conditions.
Characteristically for these calcium systems (and their
heavier alkaline-earth congeners), this hydrophosphination
reaction proceeds with strict 100% anti-Markovnikov regiose-
F
nonidentical dications of the general formula [{RO2 }-
Ca+(H2O)]2, and Ca···F interactions (ca. 2.70−2.75 Å) are
seen in only one of these, (ii) each of the four metal atoms is
coordinated by a molecule of water, (iii) there is no contact
between the cations and the anions, and (iv) unlike what is
F
found in the charge-neutral parents, for each {RO2 } ligand the
coordination of both OCH3 moieties onto the metal centers is
seen. Evidently these observations point at the greater need for
electron density in these cationic complexes, as highlighted by
their ability to bind water from surrounding traces of moisture.
Even with the more chelating and electron-donating ligand
F
{RO1 }, the Ca cations trapped residual water during the
5637
dx.doi.org/10.1021/om500343w | Organometallics 2014, 33, 5630−5642

Organometallics
Article
Ca(N(SiMe3)2)2(THF)2 and Ca(N(SiMe3)2)2 (entries 1 and
2), showing the beneficial role of the fluorinated alkoxides.
Precatalysts with the N(SiMe2H)2 moiety perform somewhat
better than those having a N(SiMe3)2 group. This observation,
which is counterintuitive on the basis of pKa considerations
alone, is explained by the reduced stability of compounds 4 and
6 in solution (signs of decomposition of the metal complexes
can be observed in a matter of hours under catalysis
conditions), while on the other hand, complexes 5 and 7 are
visually more stable (the first signs of decomposition occurring
only after 24 h). This is tentatively attributed to the presence of
Si−H moieties in 5 and 7, which can impart stabilization to the
metal through weak Ca···H−Si interactions. With zinc, the
higher catalytic activity of cationic compounds with respect to
their neutral analogues has been observed for the cyclo-
hydroamination of aminoalkenes.38 We found instead that 9
and 10 do not catalyze well the addition of HPPh2 to styrene
(entry 12), possibly because they are hampered by their limited
solubility.
F
Figure 9. ORTEP representation of the dication in [{RO1 }-
CONCLUSION
■
−
Ca+(H2O)]2·2[H2N{B(C6F5)3}2 ] ([8·H2O]2), showing the coordi-
nating H2O molecules and Ca···F contacts. Counterions, hydrogen
atoms, and noncoordinating solvent molecules are omitted for clarity.
O(31) and O(81) are O atoms of coordinated water molecules.
Potassium and heteroleptic calcium complexes supported by
monoanionic fluorinated alkoxide ligands of varying denticity
afford unusual coordination patterns in the solid state, forming
polymetallic species which remain aggregated in solution. The
calcium complexes 4−7 represent new additions to the small
family of kinetically inert heteroleptic Ca mono(alkoxide)
complexes, the most prominent example being Hanusa’s
Ca(clox)(X)(THF)x (clox = OC(C6H5)2CH2C6H4-Cl-4; X =
I, x = 4; X = N(SiMe3)2, x = 3).1c Both potassium and calcium
complexes are stabilized by a pattern of intramolecular M···F
secondary interactions which contribute significantly to reach
coordinative saturation around the metal atoms. In addition,
although definitive evidence for agostic Ca···H−Si interactions
could not be found, it is undeniable that the [Ca]−
N(SiMe2H)2 complexes 5 and 7 are less labile than their
[Ca]−N(SiMe3)2 derivatives 4 and 6. The heteroleptic
complexes 4−7 can serve as synthetic precursors to solvent-
free discrete cationic Ca complexes which exhibit extreme
lectivity at 60 °C. With as little as 1 mol % catalyst loading, the
turnover numbers and frequencies observed with 5−7 for
catalyzed reactions performed in the presence of solvent
(entries 3−5) or with neat substrates (entries 6−11, apparent
TOF ≈ 50 molsubst molCa h−1) compare well with those
−1
reported with other calcium-based precatalysts7a,35,36 and place
them among some of the most efficient systems reported to
date for this challenging reaction.35−37 Valuably, these
complexes featured the same catalytic performances when
higher substrate loadings were used (see e.g. entry 11). Under
comparable conditions (entries 1−5), complexes 5−7 are
generally more active than the simple bis(amido) complexes
Table 4. Hydrophosphination of Styrene with Ph2PH Catalyzed by the Calcium Amide Compounds 5−7 and Cationic Complex
10
a
TOF (molsubst molCa h−1
)
−1
entry
catalyst
[styrene]0:[HPPh2]0:[Ca]0
t (h)
conversion (%)
b
1
Ca(N(SiMe3)2)2(THF)2
Ca(N(SiMe3)2)2
50:50:1
50:50:1
50:50:1
50:50:1
50:50:1
50:50:1
50:50:1
100:100:1
50:50:1
100:100:1
400:400:1
50:50:1
24
24
24
24
24
1
53
55
85
46
69
99
87
99
82
99
26
8
1.1
1.3
1.8
1.0
1.4
50
b
2
b
F
3
{RO2 }Ca(N(SiMe2H)2) (5)
b
F
4
{RO3 }Ca(N(SiMe3)2) (6)
b
F
5
{RO3 }Ca(N(SiMe2H)2) (7)
c
F
6
{RO2 }Ca(N(SiMe2H)2) (5)
c
F
7
{RO3 }Ca(N(SiMe3)2) (6)
1
44
c
F
8
{RO3 }Ca(N(SiMe3)2) (6)
2
45
c
F
9
{RO3 }Ca(N(SiMe2H)2) (7)
1
41
c
F
10
{RO3 }Ca(N(SiMe2H)2) (7)
2
45
c
F
11
{RO3 }Ca(N(SiMe2H)2) (7)
2
52
c
{RO3 }Ca+·A− (10)
2
2.0
F
12
a
b
c
Conversion determined by 1H NMR spectroscopy. Conditions: 10 μmol of precatalyst, [Ca] = 16.67 mM, 0.6 mL of C6D6, 60 °C. Conditions: 10
μmol of precatalyst, no solvent, 60 °C.
5638
dx.doi.org/10.1021/om500343w | Organometallics 2014, 33, 5630−5642

Organometallics
Article
F
{RO1 }K (1). KN(SiMe3)2 (56 mg, 0.56 mmol) was added in
electrophilicity and capture even traces of residual water
molecules. They also act as potent precatalysts for the 100%
anti-Markovnikov hydroelementation of styrene with diphenyl-
phosphine, displaying activities in the upper range of those
reported for this reaction. Following these encouraging results,
we are now aiming at exploiting these new Ca mono(alkoxide)
complexes for the catalysis of a variety of atom-efficient organic
transformations.
F
portions with a bent glass finger to a solution of {RO1 }H (100 mg,
0.56 mmol) in Et2O (10 mL). A white precipitate formed immediately.
The reaction mixture was stirred overnight at room temperature. The
volatiles were removed in vacuo, and the resulting solid was washed
with pentane (4 × 5 mL) to yield 1 as a colorless solid (88 mg, 80%).
The product displayed poor solubility in hydrocarbon and ether
solvents and was soluble only in CH2Cl2. X-ray-quality crystals were
obtained by recrystallization from a concentrated CD2Cl2 solution at
1
room temperature. H NMR (CD2Cl2, 500.13 MHz, 25 °C): δ 3.73−
3.51 (m, 12H, all crown ether), 2.71 (overlapping m, 6H, NCH2CH2
and CH2C(CF3)2) ppm. 13C{1H} NMR (CD2Cl2, 125.75 MHz, 25
°C): δ 127.81 (q, 1JCF = 298.2 Hz, CF3), 67.14, 66.63, 66.25 (all crown
ether), 59.88 (CH2C(CF3)2), 54.80 (NCH2CH2) ppm; the resonance
for C(CF3)2) could not be detected. 19F{1H} NMR (CD2Cl2, 376.52
MHz, 25 °C): δ −78.41 (s, 6F, CF3) ppm. Anal. Calcd for
C12H18F6KNO4 (393.36): C, 36.6; H, 4.6; N, 3.6. Found: C, 36.6;
H, 3.8; N, 3.8.
EXPERIMENTAL SECTION
■
General Procedures. All manipulations were performed under an
inert atmosphere using standard Schlenk techniques or in a dry,
solvent-free glovebox (Jacomex; O2 <1 ppm, H2O <5 ppm) for catalyst
loading. CaI2 (Aldrich, 99.999 anhydrous beads %) and HPPh2 were
used as received. HN(SiMe3)2 (Acros) and HN(SiMe2H)2 were dried
over CaH2 and distilled prior to use. Styrene was dried and distilled
over CaH2 and stored over 3 Å molecular sieves. The compounds
F
{RO2 }K (2). KN(SiMe2H)2 (0.10 g, 0.59 mmol) was added in
31
[H(OEt2)2 ]·[H2N{(B(C6F5)3}2−], Ca(N(SiMe3)2)2,39 Ca(N-
(SiMe3)2)2(THF)2,40 and Ca(N(SiMe2H)2)2(THF)9 were prepared
following literature protocols. Solvents (THF, Et2O, CH2Cl2, pentane,
and toluene) were purified and dried (water contents all below 10
ppm) over alumina columns (MBraun SPS). THF was further distilled
under argon from sodium mirror/benzophenone ketyl prior to use. All
deuterated solvents (Eurisotop, Saclay, France) were stored in sealed
ampules over activated 3 Å molecular sieves and were thoroughly
degassed by several freeze−thaw−vacuum cycles.
+
F
portions with a bent glass finger to a solution of {RO2 }H (0.18 g, 0.59
mmol) in Et2O (5 mL). The reaction mixture was stirred overnight at
room temperature. The volatiles were removed in vacuo to give a
colorless solid. Extraction with pentane (20 mL) and evaporation of
the volatiles afforded 2 as a colorless solid (0.16 g, 78%). X-ray-quality
crystals were obtained by recrystallization from a concentrated pentane
solution. 1H NMR (C6D6, 300.13 MHz, 25 °C): δ 3.23 (t, 4H, 3JHH
=
5.1 Hz, CH2OCH3), 3.07 (s, 6H, OCH3), 3.00 (s, 2H, CH2C(CF3)2),
3
2.78 (t, 4H, JHH = 5.1 Hz, NCH2CH2) ppm. 13C{1H} NMR (75.46
NMR spectra were recorded on Bruker AM-400 and AM-500
MHz, C6D6, 25 °C): δ 127.61 (q, 1JCF = 297.2 Hz, CF3), 82.43 (hept,
2JCF = 22.8 Hz, C(CF3)2), 69.20 (CH2OCH3), 58.19 (OCH3), 57.92
(NCH2CH2), 53.77 (CH2C(CF3)2) ppm. 19F{1H} NMR (C6D6,
376.52 MHz, 25 °C): δ −77.17 (s, 6F, CF3) ppm. Anal. Calcd for
C10H16F6KNO3 (351.33): C, 34.2; H, 4.6; N, 4.0. Found: C, 34.3; H,
4.7; N, 4.1.
1
spectrometers. All H and 13C{1H} chemical shifts were determined
using residual signals of the deuterated solvents and were calibrated vs
SiMe4. 29Si{1H} chemical shifts were determined against Si(Si-
(CH3)3)4. Assignment of the signals was carried out using 1D (1H,
13C{1H}) and 2D (COSY, HMBC, HMQC) NMR experiments.
Coupling constants are given in hertz. 19F{1H} chemical shifts were
determined by external reference to an aqueous solution of NaBF4.
PGSE NMR experiments were carried out on a Bruker Avance III 400
MHz spectrometer equipped with a BBOF pulsed field-gradient probe
using a bipolar gradient pulse stimulated echo sequence. Each
experiment was performed on a 0.1 M solution at 298 K using a
spectral width of 4807 Hz, a 90° pulse width of 11.5 μs, a diffusion
delay time of 0.05 s, and a total diffusion-encoding pulse width of
0.0016 s. The diffusion encoding pulse strength was arrayed from 0 to
35 G cm−2 over 12 or 16 increments with 4 dummy scans and 8 scans
per increment. The method used to determine hydrodynamic radii was
identical with that reported elsewhere.24
F
{RO3 }K (3). KN(SiMe2H)2 (0.55 g, 3.20 mmol) was added in
F
portions with a bent glass finger to a solution of {RO3 }H (0.90 g, 3.34
mmol) in Et2O (30 mL). The reaction mixture was stirred overnight at
room temperature. The volatiles were removed in vacuo to yield 3 as a
colorless solid (0.95 g, 93%). X-ray-quality crystals were obtained by
recrystallization from a concentrated pentane solution at room
1
temperature. H NMR (C6D6, 400.13 MHz, 25 °C): δ 3.03 (s, 3H,
3
OCH3), 2.92 (t, 2H, JHH = 4.6 Hz, CH2OCH3), 2.75 (s, 2H,
CH2C(CF3)2), 2.32 (br, 2H, NCH2), 2.17 (s, 3H, NCH3) ppm.
13C{1H} NMR (100.62 MHz, C6D6, 25 °C): δ 126.46 (q, 1JCF = 296.1
Hz, CF3), 82.86 (hept, 2JCF = 23.1 Hz, C(CF3)2), 69.76 (OCH3), 62.28
(CH2OCH3), 59.74 (CH2C(CF3)2), 58.25 (NCH2CH2), 43.15
(NCH3) ppm. 19F{1H} NMR (C6D6, 376.52 MHz, 25 °C): δ
−77.30 (s, 6F, CF3) ppm. Anal. Calcd for C8H12F6KNO2 (307.28): C,
31.3; H, 3.9; N, 4.6. Found: C, 31.2; H, 3.1; N, 3.9.
Elemental analyses were performed on a Carlo Erba 1108 Elemental
Analyzer instrument at the London Metropolitan University by
Stephen Boyer and were the average of a minimum of two
independent measurements.
F
F
F
{RO2 }Ca(N(SiMe3)2) (4). A solution of {RO2 }H (0.23 g, 0.73
mmol) in pentane (10 mL) was added at −78 °C over a period of 1 h
to a solution of Ca(N(SiMe3)2)2(THF)2 (0.44 g, 0.88 mmol) in
pentane (10 mL). The mixture was warmed to room temperature and
stirred overnight, and the volatiles were removed under vacuum. The
resulting powder was stripped with pentane (3 × 4 mL) and dried in
vacuo to give a brown powder, which was dissolved in toluene (5 mL).
Layering of the resulting solution with pentane (15 mL) resulted in
the formation of a brown oil. The oil was removed by filtration, and
the remaining solution was stored in a freezer at −28 °C to afford 5 as
{RO1 }H. A solution of 1-aza-12-crown-4 (0.49 g, 2.81 mmol) in
Et2O (5 mL) was added dropwise at 0 °C to a solution of 2,2-
bis(trifluoromethyl)oxirane (0.55 g, 3.05 mmol) in Et2O (5 mL). The
reaction mixture was warmed to room temperature and stirred for 2
days. The volatiles were then evaporated under reduced pressure to
F
afford {RO1 }H as a colorless solid (0.88 g, 88%). X-ray-quality
crystals were obtained by slow concentration of a solution of the title
compound in Et2O. 1H NMR (C6D6, 500.13 MHz, 25 °C): δ 6.80 (s,
1H, OH), 3.34−3.30 (m, 4H, OCH2 moieties), 3.29−3.25 (m, 4H,
OCH2 moieties), 3.24−3.20 (m, 4H, OCH2 moieties), 2.79 (s, 2H,
CH2C(CF3)2), 2.41−2.31 (m, 4H, NCH2CH2) ppm. 13C{1H} NMR
(C6D6, 125.75 MHz, 25 °C): δ 124.61 (q, 1JCF = 288.2 Hz, CF3), 73.97
1
colorless crystals (50 mg, 13%; not optimized). H NMR (THF-d8,
3
500.13 MHz, 25 °C): δ 3.70 (t, JHH = 5.3 Hz, 4H, NCH2CH2), 3.63
2
(s, 6H, OCH3), 2.85−2.80 (overlapping m, 6H, NCH2CH2 and
CH2C(CF3)2), −0.05 (s, 18H, NSi(CH3)3) ppm. 13C{1H} NMR
(THF-d8, 125.73 MHz, 25 °C): δ 127.24 (q, CF3,1JCF = 292.0 Hz),
(hept, JCF = 28.3 Hz, C(CF3)2), 70.60, 70.39, 69.05 (all OCH2
moieties), 57.85 (CH2C(CF3)2), 55.68 (NCH2CH2) ppm. 19F{1H}
NMR (C6D6, 376.52 MHz, 25 °C): δ −77.21 (s, 6F, CF3) ppm. Anal.
Calcd for C12H19F6NO4 (355.27): C, 40.5; H, 5.4; N, 3.9. Found: C,
40.7; H, 5.3; N, 4.0. Mass spectrometry ESI: [M + Na+]
(C12H19F6NO4Na) calcd m/z 378.1116, found 378.1112.
2
82.33 (hept, C(CF3)2, JCF = 25.9 Hz), 70.31 (NCH2CH2), 61.16
(OCH3), 56.23 (CH2C(CF3)2), 53.68 (NCH2CH2), 5.92 (NSi(CH3)3)
ppm. 19F{1H} NMR (THF-d8, 376.45 MHz, 25 °C): δ −81.78 (s, 6F,
CF3) ppm. Anal. Calcd for C16H34CaF6N2O3Si2 (512.69): C, 37.5; H,
6.7; N, 5.5. Found: C, 37.6; H, 6.5; N, 5.6.
F
F
The protio ligands {RO2 }H and {RO3 }H were prepared in the
same way (Supporting Information).
5639
dx.doi.org/10.1021/om500343w | Organometallics 2014, 33, 5630−5642

Organometallics
Article
F
F
{RO2 }Ca(N(SiMe2H)2) (5). A solution of {RO2 }H (0.18 g, 0.57
mmol) in Et2O (10 mL) was added at −78 °C over a period of 1 h to a
solution of Ca(N(SiMe2H)2)2(THF) (0.28 g, 0.76 mmol) in Et2O (10
mL). The mixture was warmed to room temperature and stirred
overnight, and the volatiles were removed under vacuum. The
resulting powder was stripped with pentane (3 × 4 mL) and dried in
vacuo to give analytically pure 3 as an off-white powder (0.22 g, 79%).
Single crystals of 5 suitable for X-ray diffraction crystallography were
obtained by recrystallization from Et2O at room temperature. 1H
solid was purified by reprecipitation from CH2Cl2 with pentane (three
times). The final colorless powder was dried in vacuo to afford
analytically pure 8 (96 mg, 70%). 1H NMR (THF-d8, 400.13 MHz, 25
°C): δ 5.74 (br, 2H, NH2), 4.06−3.93 (m, 8H, OCH2 moieties), 3.93−
3.82 (m, 4H, OCH2 moieties), 3.12−2.75 (overlapping m, 6H,
NCH2CH2 and CH2C(CF3)2) ppm. 13C{1H} NMR (THF-d8, 100.62
MHz, 25 °C): δ 150.08, 147.69, 141.39, 138.85, 136.36 (all C6F5),
1
126.99 (q, JCF = 292.8 Hz, CF3), 72.36, 71.16, 70.75 (all OCH2
moieties), 59.51 (CH2C(CF3)2), 54.01 (NCH2CH2) ppm; the
resonance for C(CF3)2) was not observed. 19F NMR (THF-d8,
1
NMR (C6D6, 400.13 MHz, 25 °C): δ 4.88 (m, 2H, JSiH = 162 Hz,
3
SiH), 3.22−3.14 (m, 2H, CH2OCH3), 3.03−2.89 (overlapping m,
376.52 MHz, 25 °C): δ −79.49 (s, 6F, CF3), −133.13 (d, 12F, JFF
=
2
3
12H, CH2OCH3, CH2C(CF3)2 and NCH2CH2), 2.47 (d, 2H, JHH
=
18.4 Hz, o-F), −161.14 (t, 6F, JFF = 20.2 Hz, p-F), −166.44 (t, 12F,
3JFF = 19.3 Hz, m-F) ppm. 11B NMR (THF-d8, 128.38 MHz, 25 °C): δ
−8.31 ppm. Anal. Calcd for C48H20B2CaF36N2O4 (1434.33): C, 40.2;
H, 1.4; N, 2.0. Found: C, 40.1; H, 1.3; N, 2.1.
3
12.4 Hz, NCH(H)), 0.47 (d, 12H, JHH = 2.5 Hz, Si(CH3)2H) ppm.
13C{1H}NMR (100.62 MHz, C6D6, 25 °C): δ 125.84 (q, 1JCF = 290.1
2
Hz, CF3), 79.30 (hept, JCF = 25.7 Hz, C(CF3)2), 69.35 (CH2OCH3),
F
F
59.46 (OCH3), 55.45 (CH2C(CF3)2), 53.45 (NCH2CH2), 5.02
(Si(CH3)2H) ppm. 19F{1H} NMR (376.52 MHz, C6D6, 25 °C): δ
−77.64 (s, 6F, CF3) ppm. 29Si{1H} NMR (C6D6, 79.49 MHz, 25 °C):
[{RO2 }Ca+]·[H2−N{B(C6F5)3}2−] (9). Method A. [{RO2 }HH+]·
[H2N{(B(C6F5)3}2 ] (0.15 g, 0.11 mmol) was added in portions
with a bent glass finger to a solution of Ca(N(SiMe3)2)2 (37 mg, 0.11
mmol) in C6H5Cl (10 mL). The stirring was continued at room
temperature for 2 days. The solution was evaporated under dynamic
vacuum, and the resulting colorless solid was purified by repeated
reprecipitation from CH2Cl2 with pentane (three times). Drying under
vacuum to constant weight afforded 9 as a colorless powder (97 mg,
63%).
δ −25.1 ppm. FTIR (Nujol in KBr plates): νSi−H
2016 (s) cm−1. Anal.
̃
Calcd for C14H30CaF6N2O3Si2 (484.64): C, 34.7; H, 6.2; N, 5.8.
Found: C, 34.4; H, 6.4; N, 5.7.
F
F
{RO3 }Ca(N(SiMe3)2) (6). A solution of {RO3 }H (0.34 g, 1.26
mmol) in Et2O (15 mL) was added at −78 °C over a period of 1 h to a
solution of Ca(N(SiMe3)2)2 (0.45 g, 1.26 mmol) in Et2O (20 mL).
The reaction mixture was warmed to room temperature and stirred
overnight, and the volatiles were removed under vacuum. The
resulting powder was stripped with pentane (3 × 4 mL) and dried in
vacuo to give 6 as a yellow powder (0.42 g, 71%). Single crystals of 2
suitable for X-ray diffraction crystallography were obtained by
+
−
Method B. [H(OEt2)2 ]·[H2N{(B(C6F5)3}2 ] (97 mg, 0.08 mmol)
was added in portions to a solution of 5 (40 mg, 0.08 mmol) in Et2O
(10 mL). A colorless precipitate formed after a few minutes. The
stirring was continued overnight at room temperature. The solution
was then filtered to isolate a solid, which was purified by repeated
reprecipitation from CH2Cl2 with pentane (three times). The title
compound was isolated as a colorless powder after drying in vacuo (85
1
recrystallization from C6D6. H NMR (C6D6, 500.13 MHz, 25 °C):
δ 3.03 (overlapping m, 5H, OCH3 and CH2OCH3), 2.39 (ABq, 2H,
ΔδAB = 0.08, JAB = 14.8 Hz, CH2C(CF3)2), 2.12 (s, 3H, NCH3), 2.02
(m, 2H, NCH2CH2), 0.35 (s, 18H, NSi(CH3)3) ppm. 13C{1H} NMR
(C6D6, 125.75 MHz, 25 °C): δ 125.70 (q, 1JCF = 290.8 Hz, C(CF3)2),
125.46 (q, JCF = 289.6 Hz, CF3), 79.66 (hept, JCF = 25.3 Hz,
C(CF3)2), 69.53 (CH2OCH3), 60.90 (OCH3), 59.75 (NCH2CH2),
58.43 (CH2C(CF3)2), 45.70 (NCH3), 6.43 (Si(CH3)3) ppm. 19F{1H}
1
mg, 75%). H NMR (THF-d8, 500.13 MHz, 25 °C): δ 5.74 (br, 2H,
NH2), 3.78 (t, 4H, JHH = 4.8 Hz, CH2OCH3), 3.49 (s, 6H, OCH3),
3
3
2.94 (s, 2H, CH2C(CF3)2), 2.90 (t, 4H, JHH = 5.1 Hz, NCH2CH2)
ppm. 13C{1H} NMR (THF-d8, 125.75 MHz, 25 °C): δ 149.90, 148.01,
1
2
1
141.18, 139.23, 138.64, 136.70 (all C6F5), 127.06 (q, JCF = 292.0 Hz,
CF3), 70.87 (CH2OCH3), 60.26 (OCH3), 56.89 (CH2C(CF3)2), 54.41
(NCH2CH2) ppm; the resonance for C(CF3)2) was not observed. 19F
NMR (THF-d8, 376.52 MHz, 25 °C): δ −79.68 (s, 6F, CF3), −133.13
4
NMR (C6D6, 376.52 MHz, 25 °C): δ −73.39 (q, 3F, JFF = 9.4 Hz,
CF3), −75.14 (q, 3F, 4JFF = 9.4 Hz, CF3) ppm. 29Si{1H} NMR (C6D6,
79.49 MHz, 25 °C): δ −14.3 ppm. Satisfactory elemental analysis for
C14H30CaF6N2O2Si2 (468.64; C, 35.9; H, 6.5; N, 6.0) could not be
obtained in spite of repeated attempts.
3
3
(d, 12F, JFF = 18.4 Hz, o-F), −161.14 (t, 6F, JFF = 20.2 Hz, p-F),
−166.44 (t, 12F, JFF = 19.3 Hz, m-F) ppm. 11B NMR (THF-d8,
128.38 MHz, 25 °C):
3
δ −8.35 ppm. Anal. Calcd for
F
F
{RO3 }Ca(N(SiMe2H)2) (7). A solution of {RO3 }H (0.15 g, 0.55
mmol) in Et2O (10 mL) was added at −78 °C over a period of 1 h to a
solution of Ca(N(SiMe2H)2)(THF) (0.28 g, 0.74 mmol) in Et2O (10
mL). The reaction mixture was warmed to room temperature and
stirred overnight, and the volatiles were removed under vacuum. The
resulting powder was stripped with pentane (3 × 4 mL) and dried in
vacuo to give pure 4 as an off-white powder (0.18 g, 73%). Single
crystals of 7 suitable for X-ray diffraction crystallography were
obtained by recrystallization from Et2O at room temperature. 1H
C46H18B2CaF36N2O3 (1392.29): C, 39.7; H, 1.3; N, 2.0. Found: C,
39.6; H, 1.3; N, 2.1.
F
F
[{RO3 }Ca+]·[H2N{B(C6F5)3}2−] (10). Method A. [{RO3 }HH+]·
−
[H2N{(B(C6F5)3}2 ] (200 mg, 0.15 mmol) was added in portions to a
solution of Ca(N(SiMe3)2)2 (54 mg, 0.15 mmol) in C6H5Cl (5 mL).
The stirring was continued at room temperature for 2 days. The
volatile fraction was evaporated under vacuum to give a solid which
was purified by repeated reprecipitation from CH2Cl2 with pentane
(three times). The colorless powder was dried in vacuo to afford
analytically pure 10 (164 mg, 80%).
1
NMR (C6D6, 400.13 MHz, 25 °C): δ 4.86 (m, 2H, JSiH = 162 Hz,
+
−
SiH), 3.12 (s, 3H, OCH3), 2.84−2.61 (overlapping m, 4H, CH2OCH3
Method B. [H(OEt2)2 ]·[H2N{(B(C6F5)3}2 ] (76 mg, 0.06 mmol)
was added in portions to a solution of 6 (30 mg, 0.06 mmol) in Et2O
(10 mL). A colorless precipitate formed within a few minutes. Stirring
was continued overnight at room temperature. The solution was
filtered out, and the colorless solid was purified by repeated
reprecipitation from CH2Cl2 with pentane (three times). The crude
sample contained residual Et2O, which could not be removed. The
powder was dried in vacuo to afford 10 (66 mg, 76%) as a colorless
2
and NCH2CH2), 2.51 (ABq, 2H, ΔδAB = 0.12, JAB = 15.1 Hz,
CH2C(CF3)2), 2.03 (s, 3H, NCH3), 0.45 (d, 12H, J = 2.5 Hz,
Si(CH3)2H) ppm. 13C{1H} NMR (C6D6, 100.62 MHz, 25 °C): δ
1
1
125.92 (q, JC2F = 289.9 Hz, CF3), 125.36 (q, JCF = 290.1 Hz, CF3),
79.45 (hept, JCF = 25.8 Hz, C(CF3)2), 69.39 (CH2OCH3), 60.15
(OCH3), 59.26 (CH2C(CF3)2), 58.60 (NCH2CH2), 43.28 (NCH3),
5.09 (Si(CH3)2H), 4.82 (Si(CH3)2H) ppm. 19F{1H} NMR (C6D6,
376.52 MHz, 25 °C): δ −76.48 (q, 3F, 4JFF = 9.2 Hz, CF3), −76.81 (q,
1
powder. H NMR (THF-d8, 400.13 MHz, 25 °C): δ 5.74 (br, 2H,
3
3F, JFF = 9.3 Hz, CF3) ppm. 29Si{1H} NMR (C6D6, 79.49 MHz, 25
4
NH2), 3.78 (t, 2H, JHH = 5.2 Hz, CH2OCH3), 3.55 (s, 3H, OCH3),
°C): δ −25.4 ppm. FTIR (Nujol in KBr plates): νSi−H
2015 (s) cm−1.
̃
2.90−2.70 (overlapping m, 4H, NCH2CH2 and CH2C(CF3)2), 2.38 (s,
3H, NCH3) ppm. 13C{1H} NMR (THF-d8, 100.62 MHz, 25 °C): δ
Anal. Calcd for C12H26CaF6N2O2Si2 (440.59): C, 32.7; H, 6.0; N, 6.4.
Found: C, 32.3; H, 6.0; N, 6.2.
1
150.08, 147.71, 141.40, 138.87, 136.39 (all C6F5), 126.88 (q, JCF
=
[{RO1 }Ca+]·[H2N{B(C6F5)3}2−] (8). [{RO1 }HH+]·[H2N{(B-
(C6F5)3}2]− (132 mg, 0.09 mmol) was added in portions with a
bent glass finger to a solution of Ca(N(SiMe3)2)2 (34 mg, 0.09 mmol)
in C6H5Cl (5 mL). Stirring was continued at room temperature for 2
days. The solution was evaporated in vacuo, and the resulting colorless
F
F
293.8 Hz, CF3), 71.53 (CH2OCH3), 61.21 (OCH3), 60.01 (CH2C-
(CF3)2), 58.92 (NCH2CH2), 43.66 (NCH3) ppm; the resonance for
C(CF3)2) was not observed. 19F NMR (THF-d8, 376.52 MHz, 25 °C):
δ −79.16 (s, CF3), −133.13 (d, 3JFF = 17.7 Hz, 12F, o-F), −161.15 (t,
3
3JFF = 20.2 Hz, 6F, p-F), −166.45 (t, JFF = 19.0 Hz, 12 F, m-F) ppm.
5640
dx.doi.org/10.1021/om500343w | Organometallics 2014, 33, 5630−5642

Organometallics
Article
11B NMR (THF-d8, 128.38 MHz, 25 °C): δ −10.21 ppm. Anal. Calcd
for C44H14B2CaF36N2O2 (1348.24): C, 39.2; H, 1.0; N, 2.1. Found: C,
39.2; H, 1.1; N, 2.2.
Typical Procedure for Hydrophosphination Reactions. In the
glovebox, the precatalyst was loaded in an NMR tube. All subsequent
operations were carried out on a vacuum manifold using Schlenk
techniques. The required amount of solvent was added with a syringe
to the precatalyst, followed by addition of styrene and HPPh2. The
NMR tube was immerged in an oil bath set at the desired temperature,
and the reaction time was measured from this point. The reaction was
terminated by addition of “wet” C6D6 to the reaction mixture.
(c) Pillai Sarish, S.; Jana, A.; Roesky, H. W.; Schulz, T.; John, M.;
Stalke, D. Inorg. Chem. 2010, 49, 3816.
(6) (a) Harder, S.; Brettar, J. Angew. Chem., Int. Ed. 2006, 45, 3474.
(b) Ruspic, C.; Harder, S. Inorg. Chem. 2007, 46, 10426. (c) Spielmann,
J.; Buch, F.; Harder, S. Angew. Chem., Int. Ed. 2008, 47, 9434.
(d) Spielmann, J.; Harder, S. J. Am. Chem. Soc. 2009, 131, 5064.
(7) (a) Crimmin, M. R.; Casely, I. J.; Hill, M. S. J. Am. Chem. Soc.
2005, 127, 2042. (b) Crimmin, M. R.; Barrett, A. G. M.; Hill, M. S.;
Hitchcock, P. B.; Procopiou, P. A. Organometallics 2007, 26, 2953.
(c) Crimmin, M. R.; Arrowsmith, M.; Barrett, A. G. M.; Casely, I. J.;
Hill, M. S.; Procopiou, P. A. J. Am. Chem. Soc. 2009, 131, 9670.
(d) Liptrot, D. J.; Hill, M. S.; Mahon, M. F.; MacDougall, D. J. Chem.
Eur. J. 2010, 16, 8508. (e) Brinkmann, C.; Barrett, A. G. M.; Hill, M.
S.; Procopiou, P. A. J. Am. Chem. Soc. 2012, 134, 2193.
ASSOCIATED CONTENT
■
S
* Supporting Information
(8) (a) Buchanan, W. D.; Allis, D. G.; Ruhlandt-Senge, K. Chem.
Commun. 2010, 46, 4449. For a recent illustration, see: (b) Loh, C.;
Text, figures, tables, and CIF files giving details of the synthesis
and characterization of {RO0 }K, {ROx }H, and [{ROx }-
F
F
F
Seupel, S.; Gorls, H.; Krieck, S.; Westerhausen, M. Organometallics
̈
HH+]·[H2N{B(C6F5)3}2 ], X-ray structures of {RO0 }K,
−
F
2014, 33, 1480.
F
F
−
{RO1 }H, and [{ROx }HH+]·[H2N{B(C6F5)3}2 ], crystallo-
graphic data for 1−3, [4]2−[7]2, [8·H2O]2, [9·H2O]2, and
bond valence analysis. This material is available free of charge
via the Internet at http://pubs.acs.org.
(9) (a) Sarazin, Y.; Rosc
Maron, L.; Carpentier, J.-F. Organometallics 2010, 29, 6569. (b) Michel,
O.; Tornroos, K. W.; Maichle-Mossmer, C.; Anwander, R. Chem. Eur.
̧ a, D.-A.; Poirier, V.; Roisnel, T.; Silvestru, A.;
̈
̈
J. 2011, 17, 4964. (c) Michel, O.; Tornroos, K. W.; Maichle-Mossmer,
̈
̈
C.; Anwander, R. Eur. J. Inorg. Chem. 2012, 44.
(10) (a) Liu, B.; Roisnel, T.; Carpentier, J.-F.; Sarazin, Y. Chem. Eur.
J. 2013, 19, 2784. (b) Hill, M. S.; Liptrot, D. J.; MacDougall, D. J.;
Mahon, M. F.; Robinson, T. P. Chem. Sci. 2013, 4, 4212.
(11) (a) Buchanan, W. D.; Guino-o, M. A.; Ruhlandt-Senge, K. Inorg.
Chem. 2010, 49, 7144. (b) Buchanan, W. D.; Ruhlandt-Senge, K.
Chem. Eur. J. 2013, 19, 10708.
AUTHOR INFORMATION
Corresponding Authors
*E-mail for J.-F.C.: jean-francois.carpentier@univ-rennes1.fr.
*E-mail for Y.S.: yann.sarazin@univ-rennes1.
Notes
■
(12) (a) Lavanant, L.; Chou, T.-Y.; Chi, Y.; Lehmann, C. W.; Toupet,
L.; Carpentier, J.-F. Organometallics 2004, 23, 5450. (b) Carpentier, J.-
F. Dalton Trans. 2010, 39, 37 and references cited therein.
(c) Bouyahyi, M.; Roisnel, T.; Carpentier, J.-F. Organometallics
2010, 29, 491. (d) Dagorne, S.; Bouyahyi, M.; Vergnaud, J.;
Carpentier, J.-F. Organometallics 2010, 29, 1865. (e) Bouyahyi, M.;
Roisnel, T.; Carpentier, J.-F. Organometallics 2012, 31, 1458.
(f) Normand, M.; Kirillov, E.; Roisnel, T.; Carpentier, J.-F.
Organometallics 2012, 31, 1448.
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
The authors thank the Agence Nationale de la Recherche for
financial support (Ph.D. grant to S.-C.R., ANR-11-BS07-009-01
■
́
GreenLAkE). Assistance from J.-P. Guegan (PGSE NMR, ISC
Rennes) and S. Boyer (elemental analysis, London Metropol-
itan University) is most appreciated.
(13) Sarazin, Y.; Liu, B.; Roisnel, T.; Maron, L.; Carpentier, J.-F. J.
REFERENCES
Am. Chem. Soc. 2011, 133, 9069.
(14) Like {RO0 }H, the protio ligands {RO2 }H and {RO3 }H are
isolated as colorless oils. On the other hand, {RO1 }H is a crystalline
■
13
F
F
F
(1) See for instance: (a) Williams, R. A.; Hanusa, T. P.; Huffman, J.
C. Organometallics 1990, 9, 1128. (b) Hanusa, T. P. Polyhedron 1990,
9, 1345 (review article). (c) Williams, R. A.; Tesh, K. F.; Hanusa, T. P.
J. Am. Chem. Soc. 1991, 113, 4843. (d) Harvey, M. J.; Hanusa, T. P.;
Young, V. G., Jr. Angew. Chem., Int. Ed. 1999, 38, 217. (e) Hanusa, T.
P. Coord. Chem. Rev. 2000, 210, 329.
F
material, and its molecular solid-state structure was elucidated. See the
Supporting Information for details.
F
(15) The protio ligand {RO2 }H was first employed by Chi and co-
F
workers for the synthesis of the homoleptic compounds {RO2 }2Ae
(Ae = Sr, Ca), which act as potent source reagents for the chemical
vapor deposition of SrF2 and BaF2 thin films: Chi, Y.; Ranjan, S.;
Chou, T.-Y.; Liu, C.-S.; Peng, S.-M.; Lee, G.-H. Dalton Trans. 2001,
2462.
(16) Lum, J. S.; Tahsini, L.; Golen, J. A.; Moore, C.; Rheingold, A. L.;
Doerrer, L. H. Chem. Eur. J. 2013, 19, 6374.
(17) Kawashima, T.; Iwama, N.; Tokitoh, N.; Okazaki, R. J. Org.
Chem. 1994, 59, 491.
(18) Snelgrove, J.; Fogg, D. Private communication, CCDC No.
183980.
(19) Guillemot, G.; Solari, E.; Rizzoli, C.; Floriani, C. Chem. Eur. J.
2002, 8, 2072.
(2) (a) Harder, S. Chem. Rev. 2010, 110, 3852. (b) Barrett, A. G. M.;
Crimmin, M. R.; Hill, M. S.; Procopiou, P. A. Proc. R. Soc. A 2010, 466,
927. (c) Topics in Organometallic Chemistry; Harder, S., Vol. Ed.;
Springer: Berlin, 2013; Vol. 45.
(3) See for instance: (a) Westerhausen, M. Inorg. Chem. 1991, 30, 96.
(b) Westerhausen, M. Angew. Chem., Int. Ed. 2001, 40, 2975.
(c) Fischer, R.; Gaertner, M.; Goerls, H.; Westerhausen, M. Angew.
Chem., Int. Ed. 2006, 45, 609. (d) Westerhausen, M.; Gaertner, M.;
Fischer, R.; Langer, J. Angew. Chem., Int. Ed. 2007, 46, 1950.
(e) Krieck, S.; Gorls, H.; Yu, L.; Reiher, M.; Westerhausen, M. J. Am.
Chem. Soc. 2009, 131, 2977.
(4) (a) Vargas, W.; Englich, U.; Ruhlandt-Senge, K. Inorg. Chem.
2002, 41, 5602. (b) Alexander, J. S.; Ruhlandt-Senge, K. Eur. J. Inorg.
Chem. 2002, 2761. (c) Hitzbleck, J.; O’Brien, A. Y.; Forsyth, C. M.;
Deacon, G. B.; Ruhlandt-Senge, K. Chem. Eur. J. 2004, 10, 3315.
(d) Gillett-Kunnath, M.; Teng, W.; Vargas, W.; Ruhlandt-Senge, K.
Inorg. Chem. 2005, 44, 4862. (e) Torvisco, A.; Decker, K.; Uhlig, F.;
Ruhlandt-Senge, K. Inorg. Chem. 2009, 48, 11459. (f) Torvisco, A.;
O’Brien, A. Y.; Ruhlandt-Senge, K. Coord. Chem. Rev. 2011, 255, 1268.
(5) (a) Ruspic, C.; Nembenna, S.; Hofmeister, A.; Magull, J.; Harder,
S.; Roesky, H. W. J. Am. Chem. Soc. 2006, 128, 15000. (b) Nembenna,
S.; Roesky, H. W.; Nagendran, S.; Hofmeister, A.; Magull, J.;
Wilbrandt, P.-J.; Hahn, M. Angew. Chem., Int. Ed. 2007, 46, 2512.
(20) (a) Chisholm, M. H.; Drake, S. R.; Naiini, A. A.; Streib, W. E.
Polyhedron 1991, 10, 337. (b) Samuels, J. A.; Folting, K.; Huffman, J.
C.; Caulton, K. G. Chem. Mater. 1995, 7, 929.
(21) The minimum and maximum distances for K···F interactions
were defined by Plenio as 2.98 and 3.47 Å: Plenio, H. Chem. Rev. 1997,
97, 3363. They correspond respectively to the sum of the ionic radius
of potassium (1.51 Å) and the van der Waals radius of fluorine (1.47
Å), and the sum of the van der Waals radii of fluorine and the ionic van
der Waals radii of potassium by Kollman. As the maximum tolerated
distance seemed excessive to us in view of recent examples available
from the CCDC database, we have set it as the average of the
5641
dx.doi.org/10.1021/om500343w | Organometallics 2014, 33, 5630−5642

Organometallics
Article
minimum and maximum distances defined by Plenio, even if we are
aware that this may be somewhat restrictive.
(22) The limit for Ca···F interactions was set at 3.13 Å, as
recommended by Plenio.21
(23) Barrett, A. G. M.; Crimmin, M. R.; Hill, M. S.; Hitchcock, P. B.;
Procopiou, P. A. Angew. Chem., Int. Ed. 2007, 46, 6339.
(24) (a) Rosç a, S.-C.; Rosç a, D.-A.; Dorcet, V.; Kozak, C. M.; Kerton,
F. M.; Carpentier, J.-F.; Sarazin, Y. Dalton Trans. 2013, 42, 9361. For
other examples with zinc or yttrium, see: (b) Williams, C. K.;
Breyfogle, L. E.; Kyung Choi, S.; Nam, W.; Young, V. G., Jr.; Hillmyer,
M. A.; Tolman, W. B. J. Am. Chem. Soc. 2003, 125, 11350. (c) Platel, R.
H.; Hodgson, L. M.; White, A. J. P.; Williams, C. K. Organometallics
2007, 26, 4955.
F
(25) Measurements for [{RO1 }K]2 ([1]2) could not be performed
in benzene-d6 owing to the poor solubility of the complex in this
solvent. The diffusion coefficient (Dt = 3.60 × 10−9 m2 s−1) measured
in dichloromethane-d2 shows that the complex remains dimeric in this
solvent (rH,PGSE = 5.66 Å, rH,X‑ray = 5.04 Å).
(26) The following values were determined for TMSS (considered
spherical): DtTMSS,benzene‑d6 = (1.0750−1.8430) × 10−9 m2 s−1
(depending on the concentration ot TMSS that was used) and
c
TMSS, benzene‑d6 = 4.8223 Å, using f TMSS = 1.00 and rHTMSS = 4.28 Å. The
value of cTMSS,benzene‑d was calculated according to cTMSS = (6/(1 +
6
0.695)(rHsolv/rH
)
2.234) using rH
= 2.68 Å as found in:
TMSS
benzene‑d6
Zuccaccia, D.; Macchioni, A. Organometallics 2005, 24, 3476.
(27) Partial or total decomposition to {ROxF}2Ca and Ca(N(-
SiMe3)2)2 would have resulted in the detection of three and two clearly
distinct diffusion coefficients, respectively.
(28) Brown, I. D.; Altermatt, D. Acta Crystallogr. Sect. B 1985, B41,
244.
(29) Brese, N. E.; O’Keefe, M. Acta Crystallogr., Sect. B 1991, B47,
192.
(30) Sum, J. S.; Tahsini, L.; Golen, J. A.; Moore, C.; Rheingold, A. L.;
Doerrer, L. H. Chem. Eur. J. 2013, 19, 6374.
(31) Lancaster, S. J.; Rodriguez, A.; Lara-Sanchez, A.; Hannant, M.
D.; Walker, D. A.; Hughes, D. L.; Bochmann, M. Organometallics 2002,
21, 451.
(32) Hannant, M. H.; Wright, J. A.; Lancaster, S. J.; Hughes, D. L.;
Horton, P. N.; Bochmann, M. Dalton Trans. 2006, 2415.
(33) Bochmann, M. Coord. Chem. Rev. 2009, 253, 2000.
(34) These crystals were grown by recrystallization from saturated
dichloromethane (moisture level <5 ppm) solutions prepared and
stored in a glovebox with a moisture level lower than 4 ppm.
Crystallization outside the glovebox and/or using silylated glassware or
Teflon vials gave the same outcome.
(35) Hu, H.; Cui, C. Organometallics 2012, 31, 1208.
(36) (a) Liu, B.; Roisnel, T.; Carpentier, J.-F.; Sarazin, Y. Angew.
Chem., Int. Ed. 2012, 51, 4943. (b) Liu, B.; Roisnel, T.; Carpentier, J.-
F.; Sarazin, Y. Chem. Eur. J. 2013, 19, 13445.
(37) Koshti, V.; Gaikwad, S.; Chikkali, S. H. Coord. Chem. Rev. 2014,
265, 52.
(38) (a) Zulys, A.; Dochnahl, M.; Hollmann, D.; Lohnwitz, K.;
̈
Herrmann, J.-S.; Roesky, P. W.; Blechert, S. Angew. Chem., Int. Ed.
2005, 44, 7794. (b) Dochnahl, M.; Lohnwitz, K.; Pissarek, J.-W.;
̈
Biyikal, M.; Schulz, S. R.; Schon, S.; Meyer, N.; Roesky, P. W.;
̈
Blechert, S. Chem. Eur. J. 2007, 13, 6654.
(39) He, X.; Noll, B. C.; Beatty, A.; Mulvey, R. E.; Henderson, K. W.
J. Am. Chem. Soc. 2004, 126, 7444.
(40) Boncella, J. M.; Coston, C. J.; Cammack, J. K. Polyhedron 1991,
10, 769.
5642
dx.doi.org/10.1021/om500343w | Organometallics 2014, 33, 5630−5642