Protonolysis of Fe-C bonds of a diiminopyridineiron(II) dialkyl complex by acids of different strengths: Influence of monoanionic ligands on the spectroscopic properties of diiminopyiridine-FeY2 complexes

Article abstract of DOI:10.1016/j.ica.2013.11.028

The reaction of the dialkyliron complex [Fe(CH₂SiMe ₃)₂(^MesBIP)] (^MesBIP = 2,6-bis((N-mesityl)acetimidoyl)pyridine) with protic acids (HY) of different strengths (Y = C₆F₅O, CF₃CO₂, Cl, CF₃SO₃) invariably leads to the cleavage of both Fe-C bonds, independent of the Fe/HY ratio used (either 1:2 or 1:1), affording the corresponding complexes [FeY₂(^MesBIP)]. Relevant spectroscopic features of these compounds, such as paramagnetic ¹H NMR shifts and UV-Vis absorption bands, exhibit a marked dependence on the nature of Y.

Full text of DOI:10.1016/j.ica.2013.11.028

Inorganica Chimica Acta 412 (2014) 73–78
Contents lists available at ScienceDirect
Inorganica Chimica Acta
journal homepage: www.elsevier.com/locate/ica
Protonolysis of Fe–C bonds of a diiminopyridineiron(II) dialkyl complex
by acids of different strengths: Influence of monoanionic ligands on the
spectroscopic properties of diiminopyiridine-FeY2 complexes
⇑
M. Ángeles Cartes, Pilar Palma, John J. Sandoval, Juan Cámpora , Eleuterio Álvarez
Instituto de Investigaciones Químicas, CSIC – Universidad de Sevilla, c/Américo Vespucio, 49, 41092 Sevilla, Spain
a r t i c l e i n f o
a b s t r a c t
The reaction of the dialkyliron complex [Fe(CH2SiMe3)2(MesBIP)] (MesBIP = 2,6-bis((N-mesityl)acetimi-
doyl)pyridine) with protic acids (HY) of different strengths (Y = C6F5O, CF3CO2, Cl, CF3SO3) invariably
leads to the cleavage of both Fe–C bonds, independent of the Fe/HY ratio used (either 1:2 or 1:1), afford-
ing the corresponding complexes [FeY2(MesBIP)]. Relevant spectroscopic features of these compounds,
such as paramagnetic 1H NMR shifts and UV–Vis absorption bands, exhibit a marked dependence on
the nature of Y.
Article history:
Received 10 October 2013
Received in revised form 26 November 2013
Accepted 27 November 2013
Available online 5 December 2013
Keywords:
2,6-Bisiminopyridine ligands
Iron
Ó 2013 Elsevier B.V. All rights reserved.
Alkyl complexes
Olefin polymerization
1. Introduction
(BIP)] complexes with organoaluminum reagents gives rise to both
neutral and cationic bimetallic Fe/Al species that very likely have
Olefin polymerization or oligomerization catalysts based on
iron complexes of 2,6-bisiminopyridine (BIP) ligands have at-
tracted much interest due to their high activity and the abundance
and low toxicity and of iron [1]. In addition, the modular design of
BIP ligands facilitates the variation of the stereoelectronic environ-
ment of the active centre, enabling a precise control of the molec-
ular weight of the polymers. It is usually assumed that, similarly to
other polymerization systems, activation of [FeX2(BIP)] complexes
with alumoxanes or other organoaluminum-based co-catalysts
give rise to catalytically active alkyliron species. Such a classic
Ziegler–Natta mechanism gained strong support in 2005, when
Chirik prepared cationic complexes of the type [Fe(R)(S)(iPrBIP)]+
an active participation in the polymerization process [4]. In addi-
tion, it has been recognised that the counteranion that balances
the electric charge of active cationic species plays a role of crucial
importance in the performance of most homogeneous Ziegler–Nat-
ta polymerization catalysts [5]. Thus, it seems very likely that iron
complexes of the type [Fe(R)(Y)(BIP)] (where Y symbolizes an anio-
nic ligand) should exhibit significant differences in their ability to
act as polymerization catalysts, as the coordinating strength of Y or
its ability to interact with the co-catalysts can be varied widely.
Several years ago, we reported a general methodology that pro-
vides access to iron dialkyl complexes of the type [Fe(CH2SiMe3)2
(BIP)] [6]. and we wondered whether these complexes could react
selectively with protic acids HY of different strengths to afford the
desired [Fe(CH2SiMe3)(Y)(BIP)] complexes. As we show in this con-
tribution, it turned out that such mixed ligand compounds are not
stable or cannot be produced through this route. Instead, the
protonation reaction affords symmetrical [FeY2(BIP)] derivatives.
This allowed us to compare some of the key spectroscopic features
of these compounds and analyse how these properties are influ-
enced by the nature of the anionic Y ligand.
(
iPrBIP = BIP ligand with 2,6-diisopropylphenyl as aryl substituent
in the imine; S = OEt2, THF, or none) by protonation of dialkyl pre-
cursors with [HNPhMe2]+ [BPh4]ꢀ [2], and demonstrated that such
cationic iron alkyls behave as highly active single-component cat-
alysts for ethylene polymerization. However, the precise nature of
the active species on the real catalysts generated with the aid of
alkylaluminum co-catalysts is still the subject of some controversy.
The latter are known to play a very important role in the catalytic
process, influencing both the activity and the molecular weight
distribution of the polyolefinic products [3]. Indeed, spectroscopic
investigations of the aluminium-activated iron catalysts by
Bryliakov and Talsi have revealed that the interaction of [FeX2
2. Results and discussion
We investigated the reactions of the readily available dialkyl
complex [Fe(CH2SiMe3)2(MesBIP)] (1) with four protic acids of dif-
ferent strengths: Pentafluorophenol (pKa = 5.4), trifluoroacetic acid
⇑
Corresponding author. Tel.: +34 954489555; fax: +34 954460565.
E-mail address: campora@iiq.csic.es (J. Cámpora).
0020-1693/$ - see front matter Ó 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.ica.2013.11.028

74
M. Ángeles Cartes et al. / Inorganica Chimica Acta 412 (2014) 73–78
Scheme 1.
(pKa = 0.2), hydrogen chloride (pKa = ꢀ7) and triflic acid (pKa = ꢀ12)
[7].1 Stoichiometric (1:1) amounts of the acids diluted in THF were
slowly added to the solutions of the iron dialkyl in the same solvent
at ꢀ80 °C, and then allowed to slowly warm to room temperature. In
spite of the care taken to control the reaction conditions, these reac-
tions invariably led to products 2–5 resulting from the cleavage of
both Fe–C bonds of 1 (Scheme 1). The products were precipitated
by addition of hexane, leaving purple mother liquors containing
unreacted 1. As expected, higher yields of all four products were
obtained when the acids and the iron alkyl were reacted in 2:1 ratio.
These results suggest that the non-symmetrical alkyl complexes
[Fe(CH2SiMe3)(Y)(MesBIP)] are unstable and rapidly disproportionate
in solution affording a mixture of the corresponding symmetrical
complexes. This conclusion is also supported by Chirik’s observation
that the reaction of the mixed complex [Fe(CH2SiMe3)(Cl)(Py)2] with
iPrBIP does not afford the intended (chloro)trimethylsilylmethyl
derivative [Fe(CH2SiMe3)2(Cl)(iPrBIP)], but a mixture of the dialkyl
and dichloro complexes [8].
Apart from the well-known chloro derivative 4 [3], none of the
rest of the products, 2. 3 and 5, has been described previously. They
are all paramagnetic with
leff = 5.0–5.6 lB at room temperature,
consistent with a high-spin configuration with four unpaired elec-
trons. Crystals of 2 and 5 suitable for X-ray analysis were obtained
by recrystallization. Figs. 1 and 2 show ORTEP views of these two
complexes, and Table 1 collects selected bond lengths and angles.
Remarkably, very few iron bisiminopyridine complexes containing
alkoxo or aryloxo ligands have been reported before [9], and to the
best of our knowledge, no dialkoxo or diaryloxo derivatives are
known so far. Similarly to the analogous halide complexes
[FeX2(MesBIP)] (X = Cl, Br), the iron centre of the aryloxide 2 exhib-
its a distorted trigonal bipyramidal geometry, with the imine nitro-
Fig. 1. ORTEP view of the X ray structure of compound 2.
chemical bonding. As it is usually found in this type of compounds,
the Fe–N bonds involving the imino groups are somewhat differ-
ent, decreasing the overall molecular symmetry from C2v (if both
Fe–imine bonds were identical) to Cs. The structure of complex 5
contains, in addition to the triflate ligands, a molecule of water at-
tached to the iron centre. The presence of an aqua ligand is also re-
vealed by a strong absorption at 3345 cmꢀ1 in the IR spectrum of
this compound. Very likely, this water comes from adventitious
traces of moisture in the solvents during the synthesis or the
recrystallization of this complex, and its presence reveals the
strong Lewis acidity of the corresponding bis-triflate precursor.
Britovsek has reported related Fe(II) and Mn(II) bis-triflato com-
plexes with BIP ligands. Interestingly, the Mn(II) derivative con-
taining the Me2BIP ligand (N-aryl groups = 2,6-dimethylphenyl)
was also isolated as the monohydrate [Mn(OTf)2(OH2)(Me2BIP)]
[11]. Goldberg has recently reported a mixed iron(II) thiolate-tri-
flate complex. [Fe(OTf)(SPh)(iPrBIP)] [12]. The geometry of the iron
centre in 5 is approximately octahedral, with the aqua ligand and
one of the triflate ligands occupying ‘‘axial’’ positions, i.e., perpen-
dicular to the main coordination plane defined by the three N
atoms of the BIP ligand and the second triflate group. This
gen atoms occupying the axial positions. The
describes the distortion degree between perfect bipyramidal trigo-
nal ( = 1) and square pyramidal ( = 0) geometries, takes the value
s parameter [10], that
s
s
0.84 for this compound. A crystallographically imposed mirror
plane bisects the molecule through the iron and the three nitrogen
atoms and relates the pentafluroroaryloxide moieties. The latter
are oriented in such a way that one of the ortho fluorine substitu-
ents (F5) approaches to the iron atom. This conformation could be
favoured by an attractive electrostatic interaction, but the Fe–F5
distance (3.0149(12) Å) is too long to mean any significant
1
pKa values in water. Note that the acid strength of HY is inversely related to the
basicity of Yꢀ ligands, which correlates with their ability to donate electron density to
the metal centre. Aqueous pKa were selected because they are available for the four
acids used in this work, but note that the acid strengths of HOTf, HCl, acetic acid and
phenol holds in water, acetonitrile and dichloroethane, see Ref. [7a].

M. Ángeles Cartes et al. / Inorganica Chimica Acta 412 (2014) 73–78
75
Fig. 3. UV–Vis spectra of complexes 1 – 5.
coordinated to the iron centre (Fig. 3). The absorption maximum
(kmax) appears at 550 nm for 1 and shifts to longer wavelengths
for the products arising from the protonation with acid, in the or-
der 1 < 2 (580 nm) < 3 (625 nm) < 4 (680 nm). This band has also
been observed in the visible spectra of other iron-BIP complexes
and assigned to a charge transfer transition involving the BIP li-
gand [17]. The observed trend indicates that the energy gap be-
tween the orbitals responsible for the transition decreases as the
acid HY from which the complex arises becomes stronger, or what
is the same, as the anion Yꢀ becomes more electronegative. This
points to a ligand to metal (LMCT) rather than a metal to ligand
charge transfer band (MLCT), since electron withdrawing from
the metal is expected to lower the energy of metal-centred orbitals,
while leaving those of the BIP ligand relatively unaffected. Triflate
complex 5 is the exception to the mentioned trend, as its visible
absorption occurs at 550 nm, when it would be expected to be
the lowest in energy for the series. This apparent exception is read-
ily explained by the presence of the additional aqua ligand, which
contributes to compensate the electronic deficiency caused by the
strongly electron-withdrawing triflate ligands on the metal centre.
High spin Fe(II) complexes with BIP ligands give rise to useful
NMR spectra. Typically, the 1H signals have linewidths between a
few and several hundred hertz and can be readily observed, except
those belonging to groups directly attached to the metal centre.
These signals can be assigned on the basis of their intensity, width
and their characteristically large chemical shifts. Recognising
trends in the spectra of series of related complexes is highly helpful
in the task of assigning resonances, and constitutes a useful tool for
the identification of new complexes. The 1H NMR spectra of com-
plexes 2–5 provide some of such useful trends (see Fig. 4). Since
Fig. 2. ORTEP view of the X ray structure of compound 5.
Table 1
Selected bond distances (Å) and bond angles (°) for compounds 2 and 5.
Distances (Å)
Fe–N1
2
5
Angles (deg)
2
5
2.0943(17) 2.089(2)
O1–Fe–O10/
O4
97.06(7) 84.32(7)
Fe–N2
Fe–N3
Fe–O1
Fe–O4
2.2923(17) 2.236(2)
2.2368(17) 2.210(2)
1.9840(11) 2.1084(18) O1–Fe–N1
N1–Fe–N3
N2–Fe–N3
73.97(6) 74.50(8)
147.58(6) 147.71(8)
131.36(3) 62.26(7)
–
–
–
2.2277(19) O4–Fe–N1
2.1288(18) O7–Fe–N1
–
–
93.82(7)
Fe–O7
110.88(8)
O7ꢁ ꢁ ꢁ(H)ꢁ ꢁ ꢁO3
2.804(3)
O1–Fe–N2
O1–Fe–N3
103.14(5) 118.47(7)
98.14(5) 93.82(7)
configuration contrasts with that preferred by the analogous man-
ganese complex [Mn(OTf)2(OH2)(Me2BIP)], which shows both tri-
flate groups in the axis and the aqua ligand sharing the
equatorial plane with the BIP ligand [8]. A hydrogen bond links
the aqua ligand of 5 to the cis triflate group. The distance between
the oxygen atoms involved in this interaction, O3 and O7,
(2.804(3) Å) suggests that the H bridge is relatively weak, as it is
longer than those observed in typical Oꢁ ꢁ ꢁHꢁ ꢁ ꢁO hydrogen bonds,
for instance those formed in water or in carboxylic acids (2.6–
2.7 Å) [13]. The Fe–O bonds (either those involving the triflate or
aqua ligands) are 2.1–2.2 Å long, significantly longer than the Fe–
O bonds in the aryloxide 2 (1.9840(11) Å). This difference is due
in part to the higher coordination number in 5, but they also reflect
the weaker nature of the Fe–O bonds involving the aqua and tri-
flate anions, the latter predominantly ionic in character. Although
we were unable to grow X-ray quality crystals of carboxylate 3,
this compound probably has a pentacoordinated structure with
terminally bound carboxylate ligands, as observed for related car-
boxylate compounds containing terpy [14] or related tridentate
N,N,N ligands [15]. Terminal coordination of the trifluoroacetate li-
gands is supported by the large separation between the IR absorp-
tions for the
m
s and
m
as modes of the carboxylate group, observed at
(Dm
1697 and 1260 cmꢀ1
= 437 cmꢀ1) [16].
The colours of complexes 1–5 are strikingly varied. While the
dialkyl 1 is purple, perfluorophenolate 2 is green, the trifluoroace-
tate and triflate derivatives 3 and 5 are burgundy-red and magenta,
respectively, and the chloro complex 4 is dark blue. The origin of
these colours is a broad absorption band in the visible spectrum,
whose position varies depending on the anionic ligands
Fig. 4. 1H NMR spectra of complexes 2–5. Residual peaks of deuterated solvent
(C6D6 for 2 and CD2Cl2 for 3, 4 and 5) are marked with asterisk (*).

76
M. Ángeles Cartes et al. / Inorganica Chimica Acta 412 (2014) 73–78
none of the ligands Y contain H atoms, these spectra show only
signals corresponding to the MesBIP ligand. They all exhibit the
same number of signals, corresponding to apparent C2v molecular
symmetry. In the case of the triflate complex, the simplicity of
the spectrum is not consistent with the lower symmetry observed
in the solid state, indicating that in solution the aquo land triflate
ligands exchange their relative positions, probably via a mecha-
nism involving OTf dissociation. This is supported by the 19F spec-
spectra of 5. Magnetic susceptibility measurements were made
with a Sherwood Scientific balance model MSB-auto. Magnetic mo-
ments are reported at room temperature (298 K) and have been
corrected for diamagnetic contributions estimated from Pascal
constants [18]. Complex 1 was prepared as described in Ref. [6].
Pentafluorophenol, hydrogen chloride (1.1 M solution in ether), tri-
fluoroacetic acid and triflic acid were purchased from Aldrich and
used as received.
trum, which shows
a
single resonance at
d
ꢀ14.6 ppm in
dichloromethane, which shifts to ꢀ71.0 ppm in acetonitrile, close
to the position expected for an uncoordinated triflate anion [11].
One of the most evident features of these spectra is the strong
sensitivity of the chemical shift of some of the MesBIP signals to
the nature of ligands Y. The signal H4 is particularly noteworthy,
as it shifts to higher field in the order Y = Cl (complex 4,
34.8 ppm) ꢂ C6F5O (2, 35.5 ppm) > triflate (5, 22.7 ppm) > trifluo-
4.2. General procedure for the reaction of 1 with protic acids in
stoichiometric ratio 1:1
A
solution containing approx. 440 mg of compound 1
(0.7 mmol) in 40 ml of THF was stirred at ꢀ80 °C, and a second
solution containing exactly the equimolar amount of the corre-
sponding acid (HY = perfluorophenol, 0.46 M in hexane; trifluoro-
acetic acid, 0.88 M in toluene; hydrogen chloride, 1.15 M in
diethyl ether; triflic acid, 0.47 M in diethyl ether) was added
drop-wise. No colour change was immediately observed, except
in the case of triflic acid, which causes the mixture to turn to a red-
dish hue. The cooling bath was removed and the mixture was al-
lowed to warm slowly. With the rest of the acid reagents
(perfluoprophenol, hydrogen chloride and trifluoroacetic acid), col-
our changes were observed only when the temperature rises to ca.
0 °C. After stirring the mixture for 1 h at room temperature, 40 ml
of hexane were added. This caused the precipitation of the product,
which was isolated by filtration and dried in vacuum. The remain-
ing solution was taken to dryness, extracted with hexane (ca.
40 ml) and filtered. The resulting extract has the characteristic pur-
ple colour of 1, and after concentration and cooling to ꢀ20 °C, a
small amount of crystals of this compound can be recovered. The
solids were dried under vacuum, affording crude yields of products
2–5 that were below 50% of the initial amount of 1.
roacetate (3, ꢀ11.2 ppm). The signal for the
a-methyl of the imino
group shows the same tendency, but with the opposite sign: 2
(-22.6 ppm) ꢂ 4 (-22.2) < 5 (ꢀ3.6 ppm) < 3 (+4.1 ppm). The inverse
correlation between the chemical shifts of the H4 and a-Me signals
holds for the dialkyl 1, for which they take extreme values, +284.0
and ꢀ184.3, respectively. The analogous signals in other Fe(II) dial-
kyl complexes containing BIP ligands exhibit similar extreme para-
magnetic shifts [6,8,9]. The observed H4/a-Me sequences bear no
simple relationship with the donor capacity of Y, since paramag-
netic chemical shifts depend on the spin density on the observed
nucleus rather than on partial electric charge effects. However,
the much larger paramagnetic shifts of the position of H4 and
a-Me signals in 1 enables us to discriminate between simple
coordination complexes containing mild
r-donor Y groups from
organometallic species.
3. Conclusions
4.3. General procedure for the reaction of 1 with protic acids in
stoichiometric ratio 1:2
Our attempts to selectively cleave one Fe–C bonds of the dialkyl
1 with a variety of protic acids HY of different strengths have been
unsuccessful, as these reactions invariably lead to [FeY2(MesBIP)]
complexes arising from the cleavage of both Fe–C bonds. We eval-
uated the influence of ligands Y on the spectroscopic properties of
this series of complexes. The charge transfer absorption band
responsible for the colour of these complexes shifts to longer
wavelengths as the electron density on the Fe atom, controlled
by the Y ligands, decreases. Although the chemical shift of some
1H NMR signals of the MesBIP ligand (in particular, those due to pyr-
This is the same procedure described above, but double
amounts of the HY reagents were used. After precipitation with
hexane the products were filtrated leaving a nearly colourless solu-
tion, which evinces total consumption of 1. The solids were dried
under vacuum and recrystallized as indicated below.
4.4. Complex 2
idine H4 and the imine a-Me group) are very sensitive to the nat-
Green crystals. Recrystallized from cold THF (ꢀ20 °C), Isolated
yield, 61% (0.46 mmol from 0.75 mmol of 1). 1H NMR (C6D6,
ure of Y, no clear-cut relationship is observed. These exhibit much
larger paramagnetic shifts in the organometallic parent compound
1 than in the coordination complexes 2–5.
298 K, 300 MHz), d (ppm): ꢀ22.6 (
9.8 ( 1/2 = 107 Hz, 12 H, o-CH3 Aryl); 12.7 (
Aryl); 17.7 ( 1/2 = 8, 6H, p-CH3 Aryl); 34.8 (
H4py); 72.2 (
1/2 = 40, 2H, H3py). UV–Vis (CH2Cl2, C = 10ꢀ4M):
k(max) = 295 nm ( = 8200); 350 nm( = 2600), 476 nm ( = 950),
515 nm ( = 1000), 580 nm (shoulder, = 830). IR (Nujol mull):
1641, 1569 (C@N, C@C py) 1259,1214 (Car–O); 1014, 995 (C–
F). eff (magnetic balance): 5.5 B. Anal. Calc. for C39H31F10FeN3O2:
D
m
1/2 = 53, 6H, CH3–C@NAr);
1/2 = 15, 4H, m-CH
1/2 = 27, 1H,
D
m
Dm
D
m
Dm
4. Experimental
Dm
e
e
e
4.1. General considerations
e
e
m
m
m
All manipulations were carried out under inert atmosphere by
using conventional Schlenk techniques or a nitrogen-filled glove-
box. Solvents were rigorously dried and degassed prior to use.
Microanalysis were performed by the Analytical Service of the
Instituto de Investigaciones Químicas. IR spectra were recorded
with Bruker Vector 22 or Tensor 27 spectrometers, and UV–Vis
spectra with a Perkin–Elmer Lambda 12 spectrophotometer, using
special cuvettes provided with a gastight YoungÒ Teflon valve.
NMR spectra were recorded with Bruker 300 and 400 MHz spec-
trometers. Chemical shifts are expressed relative to TMS. The 1H
NMR residual resonance of the solvent was used as internal stan-
dard and an external reference of CF3CO2H was used for the 19F
l
l
C, 57.16; H, 3.81; N, 5.13. Found: C, 56.70; H, 3.95; N, 5.03%.
4.5. Complex 3
Burgundy-red solid. Recrystallized from CH2Cl2. Isolated yield,
40% (0.26 mmol from 0.65 mmol of 1). 1H NMR (CD2Cl2, 298 K,
300 MHz), d (ppm): ꢀ11.2 (
62 Hz, 6 H, CH3–C@NAr); 13.8 (
15.1 ( 1/2 = 23 Hz, 4 H, m-CH Aryl); 19.2 (
p-CH3 Aryl); 84.6 ( 1/2 = 49 Hz, 2 H, H3py). leff (magnetic bal-
ance): 5.13 = 5500);
B. UV–Vis (CH2Cl2, C = 10ꢀ4M): 295 nm (
D
m
1/2 = 49 Hz, 1 H, H4py); 4.1 (
1/2 = 107 Hz, 12 H, o-CH3 Aryl);
1/2 = 19 Hz, 6 H,
Dm1/2 =
D
m
D
m
Dm
D
m
l
e

M. Ángeles Cartes et al. / Inorganica Chimica Acta 412 (2014) 73–78
77
350 nm (shoulder,
1697 s(CO2), 1635
e
m
= 2500), 625 nm (
e
= 540). IR (Nujol mull):
as(CO2), 1202
perfluoropolyether were mounted on a glass fiber and fixed under
a cold nitrogen stream. The Intensity data were collected on a Bru-
ker-Nonius X8ApexII CCD area detector diffractometer using Mo
m
(C@N), 1588 (C@C py), 1260 m
m
m
(C–F). Anal. Calc. for C32H33Cl2F6FeN3O4: C, 54.80; H, 4.60; N,
6.18. Found: C, 54.56; H, 4.67; N, 5.54%.
Ka
radiation source (k = 0.71073 Å) and graphite monochromator.
The data collection strategy used was / and rotations with nar-
x
row frames. Instrument and crystal stability were evaluated from
the measurement of equivalent reflections at different measuring
times and no decay was observed. The data were reduced using
SAINT [19] and corrected for Lorentz and polarisation effects, and a
semiempirical absorption correction was applied using SADABS
4.6. Complex 4 [3]
Dark blue solid. Washed with 40 ml of hexane and dried under
vacuum. Isolated yield, 53 % (0.34 mmol from 0.65 mmol of 1). 1H
NMR (CD2Cl2, 298 K, 300 MHz), d (ppm): ꢀ22.2 (
CH3–C@NAr); 12.5 1/2 = 179 Hz, 12 H, o-CH3 Aryl); 15.5
1/2 = 73 Hz, H, m-CH Aryl); 21.7 1/2 = 66 Hz, H,
p-CH3 Aryl); 35.5 ( 1/2 = 84 Hz, 1 H, H4py); 82.2 ( 1/2 = 94 Hz,
H, H3py). UV–Vis (CH2Cl2 C = 10ꢀ4 M): 285 nm
= 7700),
= 331). IR (Nujol mull):
Dm1/2 = 86 Hz, 6 H,
[20]. The structures were solved by direct methods using SIR
-
(Dm
2002 [21] and refined against all F2 data by full-matrix least-
squares techniques using SHELXTL-6.12 [22] minimizing w[Fo2ꢀFc2]2.
All the non-hydrogen atoms were refined with anisotropic dis-
placement parameters. The hydrogen atoms of compounds 2–5
were included from calculated positions and allowed to ride on
the attached atoms with isotropic temperature factors (Uiso values)
fixed at 1.2 times (1.5 times for methyl groups) those Ueq values of
the corresponding attached atoms. Complete structural data have
been deposited with CCDC Reference Nos. 964359 (2) and
964360 (5).
(D
m
4
(
D
m
6
D
m
Dm
2
(e
325 nm (shoulder,
1620 (C@N), 1587
e
= 4700), 675 nm (e
m
m
(C@Cpy).
4.7. Complex 5
Magenta crystals. Recrystallized from cold CH2Cl2. Isolated
yield, 57% (0.30 mmol from 0.52 mmol of 1). 1H NMR (CD2Cl2,
298 K, 300 MHz), d (ppm): ꢀ3.6 (
0.8 ( 1/2 = 133 Hz, 12 H, o-CH3 Aryl); 11.6 (
m-CH Aryl); 20.3 ( 1/2 = 11 Hz, 6 H, p-CH3 Aryl); 22.3 (
63 Hz, 1 H, H4py); 78.7 (
1/2 = 69 Hz, H3py). 19F NMR (CD2Cl2,
298 K, 376 MHz): ꢀ14.6 (
1/2 = 965 Hz). 19F NMR (MeCN, 298 K,
376 MHz): ꢀ71.0 (
1/2 = 877 Hz). UV–Vis (CH2Cl2, C = 10ꢀ4M):
k(max) = 299 nm ( = 5500); 355 nm (shoulder, = 1800); 550 nm
= 820). eff: 4.9 lB (magnetic balance, 25 °C). IR (Nujol mull):
(OH), 1638, 1619 (C@N), 1588 (C@Cpy), 1304, 1214,
D
m
1/2 = 52 Hz, 6 H, CH3–C@NAr);
Acknowledgements
D
m
Dm
1/2 = 24 Hz, 4 H,
D
m
Dm1/2 =
This work was supported by the Government of Spain (project
number CTQ2012-30962), the Junta de Andalucía (project number
FQM5074) and the European Union (EU) (FEDER funds). MAC
thanks CSIC for a predoctoral research fellowship (I3P program).
D
m
D
m
D
m
e
e
(e
l
3345
m
m
m
Appendix A. Supplementary material
1165
m
(C–F). Anal. Calc. for C29H33F6FeN3O7S2ꢁ0.5CH2Cl2: C, 43.64;
H, 4.22; N, 5.17. Found: C, 43.55; H, 4.084; N, 5.36%.
CCDC 964359 and 964360 contains the supplementary crystal-
lographic data for 2 and 5. These data can be obtained free of
charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif. Supplementary data associ-
ated with this article can be found, in the online version, at http://
dx.doi.org/10.1016/j.ica.2013.11.028.
4.8. X-ray crystal structure analyses of 2 and 5
A summary of the crystallographic data and the structure
refinement is reported in Table 2. Crystals coated with dry
Table 2
Crystallographic data collection, intensity measurements and structure refinement
parameters for 2 and 5.
References
[1] (a) L. Li, P.T. Gomes, in: G Giambastiani, J. Cámpora (Eds.), Olefin Upgrading
Catalysis with Nitrogen-based Metal Complexes, State of the Art and
Perspectives, vol. II, Springer, Dordrecht, 2011, p. 77.;
(b) C. Bianchini, G. Giambastiani, I. Guerrero Rios, G. Montovani, A. Meli, A.M.
Segarra, Coord. Chem. Rev. 250 (2007) 1391.
[2] M.W. Bouwkamp, E. Lobkowsky, P.J. Chirik, J. Am. Chem. Soc. 127 (2005) 9660.
[3] G.J.P. Britovsek, M. Bruce, V.C. Gibson, B.S. Kimberley, P.J. Maddox, S.
Mastroianni, S.J. McTavish, C. Redshaw, S. Strömberg, A.J.P. White, D.J.
Williams, J. Am. Chem. Soc. 121 (1999) 8728.
[4] (a) K.P. Bryliakov, N.V. Semikolenova, V.A. Zakharov, K.P. Talsi, Organometallics
23 (2004) 5375;
(b) K.P. Bryliakov, E.P. Talsi, N.V. Semikolenova, V.A. Zakharov, Organometallics
28 (2009) 3225.
[5] (a) M. Bochmann, J. Organomet. Chem. 689 (2004) 3982;
(b) M. Bochmann, Organometallics 29 (2010) 4711.
2
5
Formula
C
39H31F10FeN3O2ꢁ2(C4H8O) C29H33F6FeN3O7S2
Formula weight
Crystal system
Space group
a (Å)
b (Å)
c (Å)
963.73
orthorhombic
Pnma
15.7849(5)
19.3433(6)
14.5566(5)
90.00
90.00
90.00
4444.6(2)
100(2)
4
769.55
monoclinic
P21/n
15.0793(6)
8.8111(4)
25.3883(11)
90.00
102.926(2)
90.00
3287.7(2)
100(2)
4
1584
1.555
0.670
30.6
a
(°)
b (°)
c
(°)
V (Å3)
T (K)
Z
[6] J. Cámpora, A.M. Naz, P. Palma, E. Álvarez, Organometallics 24 (2006) 4878.
[7] (a) E. Raamat, K. Kaupmees, G. Ovsjannikov, A. Tummal, A. Kütt, J. Saame, I.
Koppel, I. Kaljurand, L. Lipping, T. Rodima, V. Pihl, I.A. Koppel, I. Leito, J. Phys.
Org. Chem. 26 (2013) 162;
F(0015840)
1992
1.440
0.429
30.5
Dcalc (Mgmꢀ3
)
l
(mmꢀ1
)
(b) J.F.J. Dippy, S.R.C. Hughes, A. Rozanski, J. Chem. Soc. (1959) 2492;
(c) D. Stefanidis, S. Cho, S. Dhe-Paganon, W.P. Jenks, J. Am. Chem. Soc. 115
(1993) 1650.
hmax (°)
Number of reflections
collected
106012
86096
[8] I. Fernández, R.J. Trovitch, E. Lobkovsky, P.J. Chirik, Organometallics 27 (2008)
109.
[9] R.J. Trovitch, E. Lobkovsky, M.W. Bouwkamp, P.J. Chirik, J. Am. Chem. Soc. 27
(2009) 6264.
[10] W.A. Addison, T.N. Rao, J. Reedijk, J. van Rijn, G.C. Verschoor, J. Chem. Soc.,
Dalton Trans. (1984) 1349.
Number of reflections used 6976
10003
441
0.0540
0.1592
1.062
Number of parameters
343
R1(F) [F2 > 2
r
(F2)]a
0.0394
0.1151
1.071
wR2(F2)b (all data)
Sc (all data)
P
P
[11] G.J.P. Britovsek, J. England, S.K. Spitzmesser, A.J.P. White, D.J. Williams, Dalton
Trans. (2005) 945.
[12] Y.M. Baidei, M.A. Siegler, D.P. Goldberg, J. Am. Chem. Soc. 133 (2011) 1274.
[13] F.A. Cotton, G. Wilkinson, C.A. Murillo, M. Bochmann, Advanced Inorganic
Chemistry, sixth ed., John Wiley & Sons, New York, 1999.
a
b
c
R1(F) = (|Fo| ꢀ |Fc|)/ |Fo| for the observed reflections [F2 > 2
r
(F2)].
P
P
wR2(F2) = { [w(Fo ꢀ Fc2)2]/ w(Fo2)2}1/2
.
2
P
2
S = { [w(Fo ꢀ Fc2)2]/(n–p)}1/2
; (n = number of reflections, p = number of
parameters).

78
M. Ángeles Cartes et al. / Inorganica Chimica Acta 412 (2014) 73–78
[14] K.L. Tokarev, M.A. Kiskin, A.A. Sidorov, G.G. Aleksandrov, V.N. Ikorskii, I.P.
Suzdalev, V.N. Novotortsev, I.L. Eremenko, Russ. Chem. Bull. 57 (2008)
1209.
[17] G.J.P. Britovsek, V.C. Gibson, S.K. Spitzmesser, K.P. Tellmann, A.J.P. White, D.J.
Williams, J. Chem. Soc., Dalton Trans. (2002) 1159.
[18] G.A. Bain, J.F. Berry, J. Chem. Ed. 85 (2008) 532.
[15] D. Lee, S.J. Lippard, Inorg. Chim. Acta 341 (2002) 1.
[19] SAINT, APEX2 Suite, v. 2.1. Bruker AXS Inc., Madison, Wisconsin, USA, 2006.
[20] G. Sheldrick, SADABS, APEX2 Suite, v. 2.1. Bruker AXS Inc, Madison, Wisconsin,
USA, 2006.
[21] SIR-2002, APEX2 Suite, v., 2.1, Bruker AXS Inc., Madison, Wisconsin, USA, 2006.
[22] SHELXTL, v. 6.14. Bruker AXS Inc., Madison, Wisconsin, USA, 2000–2003.
[16] (a) R.C. Mehrotra, R. Bohra, Metal Carboxylates, Academic Press, London, 1983;
(b) A. Behr, R. He, K.D. Juszak, C. Krüger, Y.H. Tsay, Chem. Ber. 119 (1986) 991;
(c) E. Nakamoto, Infrared Spectra of Inorganic and Coordination Compounds,
fourth ed., John Wiley & Sons, New York, 1986.