Highly Modular Piano-Stool N-Heterocyclic Carbene Iron Complexes: Impact of Ligand Variation on Hydrosilylation Activity

Article abstract of DOI:10.1021/acs.organomet.1c00200

The piano-stool configuration combined with N-heterocyclic carbene (NHC) ligation constitutes an attractive scaffold for employing iron in catalysis. Here, we have expanded this scaffold by installing a pentamethyl cyclopentadienyl (Cp*) ligand as a strong electron donor compared to the traditionally used unsubstituted cyclopentadiene (Cp). Moreover, decarboxylation is introduced as a method to prepare these iron(II) NHC complexes, which avoids the isolation of air-sensitive free carbenes. In addition to the Cp/Cp? variation, the complexes have been systematically modulated at the NHC scaffold, the NHC wingtip groups, and the ancillary ligands in order to identify critical factors that govern the catalytic activity of the iron center in the hydrosilylation of aldehydes. These modulations reveal the importance of steric tailoring and optimization of electron density for high catalytic performance. The data demonstrate a critical role of the NHC scaffold with triazolylidenes imparting consistently higher activity than imidazolylidenes and a correlation between catalytic activity and steric rather than electronic factors. Moreover, the implementation of steric bulk is strongly dependent on the nature of the NHC and severely limited by the Cp? iron precursor. The best performing catalytic systems reach turnover frequencies, TOFmax's, of up to 360 h-1 at 60 °C. Mechanistic investigations by 1H NMR and in situ IR spectroscopies indicate a catalyst activation that involves CO release and aldehyde coordination to the [Fe(Cp)(NHC)I] fragment.

Full text of DOI:10.1021/acs.organomet.1c00200

pubs.acs.org/Organometallics
Article
Highly Modular Piano-Stool N‑Heterocyclic Carbene Iron
Complexes: Impact of Ligand Variation on Hydrosilylation Activity
Pamela V. S. Nylund, Nathalie C. Ségaud, and Martin Albrecht*
Cite This: Organometallics 2021, 40, 1538−1550
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: The piano-stool configuration combined with N-heterocyclic
carbene (NHC) ligation constitutes an attractive scaffold for employing iron in
catalysis. Here, we have expanded this scaffold by installing a pentamethyl
cyclopentadienyl (Cp*) ligand as a strong electron donor compared to the
traditionally used unsubstituted cyclopentadiene (Cp). Moreover, decarbox-
ylation is introduced as a method to prepare these iron(II) NHC complexes,
which avoids the isolation of air-sensitive free carbenes. In addition to the Cp/
Cp* variation, the complexes have been systematically modulated at the NHC
scaffold, the NHC wingtip groups, and the ancillary ligands in order to identify
critical factors that govern the catalytic activity of the iron center in the hydrosilylation of aldehydes. These modulations reveal the
importance of steric tailoring and optimization of electron density for high catalytic performance. The data demonstrate a critical
role of the NHC scaffold with triazolylidenes imparting consistently higher activity than imidazolylidenes and a correlation between
catalytic activity and steric rather than electronic factors. Moreover, the implementation of steric bulk is strongly dependent on the
nature of the NHC and severely limited by the Cp* iron precursor. The best performing catalytic systems reach turnover
1
frequencies, TOFmax’s, of up to 360 h−1 at 60 °C. Mechanistic investigations by H NMR and in situ IR spectroscopies indicate a
catalyst activation that involves CO release and aldehyde coordination to the [Fe(Cp)(NHC)I] fragment.
synthesis of NHC iron complexes is traditionally accomplished
via the versatile free carbene route.22−25 The need of base and
inert conditions however poses limitations, and especially,
NHCs with small wingtip groups risk rearrangement or
decomposition under basic conditions.24−26 Here, we present
a new synthetic method to prepare NHC iron(II) piano-stool
complexes. Moreover, we have used this and traditional
methods to modularize NHC iron piano-stool complexes to
evaluate the effects of variation in the cyclopentadiene (Cp)
unit, the NHC scaffold, and the wingtip groups on catalytic
hydrosilylation activity, demonstrating the importance of
considering even slightly different catalysts depending on the
substrate.
INTRODUCTION
■
Iron is the second most abundant metal in the Earth’s crust,
easy to extract, and consequently cheap. Additionally, it is
biologically relevant and known to be nontoxic to both humans
and the environment. Despite these facts, iron was for a long
time scarce in the catalytic literature compared to some of the
more expensive, rare, and toxic metals in the periodic system.
The application of iron as a molecular catalyst has been
catching up with other metals over the last 20 years, however,
and is now covering a plethora of organic reactions.1−3
Hydrosilylation, one of the fields that has received much
attention in iron catalysis during the past decades,4−8 is a
versatile methodology for the reduction of a variety of
substrates under base-free conditions and without the need
for highly reducing conditions as imparted by H2 pressure.8−10
In particular, hyydrosilylation of alkenes is industrially
important due to the increasing demand for silanes and
siloxanes.11
In the past two decades, iron complexes have emerged as
attractive hydrosilylation catalysts for alkenes in an approach to
make industrial hydrosilylation more benign. While these
systems normally deploy NN or NNN type ligands,12−16
piano-stool iron complexes with N-heterocyclic carbene
(NHC) ligands have demonstrated efficiency in the hydro-
silylation of carbonyls rather than alkenes under fairly mild
conditions.17−21 These iron(II) complexes offer an excellent,
cheap, and nontoxic alternative to systems based on noble
metals such as rhodium, palladium, and platinum.8 The
RESULTS AND DISCUSSION
■
Synthesis of Imidazole-Derived NHC Complexes. The
known imidazolylidene iron Cp complexes 3a, 3b, and
3d22,23,27 and the new complex 3c were prepared by the
established free base route involving the deprotonation of the
corresponding imidazolium salts 1a−d with KOtBu followed
Received: April 1, 2021
Published: May 13, 2021
https://doi.org/10.1021/acs.organomet.1c00200
© 2021 American Chemical Society
Organometallics 2021, 40, 1538−1550
1538

Organometallics
pubs.acs.org/Organometallics
Article
a
Scheme 1. Synthesis of Iron Imidazolylidene Complexes
a
Reagents and conditions: (i) KOtBu, THF, rt, 60 min, then [Cp(*)Fe(CO)2I], toluene, rt, 16 h; (ii) KOtBu, THF, rt, 60 min, then CO2, toluene,
rt, 1 h; (iii) [Cp(*)Fe(CO)2I], toluene, 80 °C, 16 h; (iv) hν, CH2Cl2, 16 h.
a
Scheme 2. Synthesis of Iron Triazolylidene Complexes
a
Reagents and conditions: (i) KOtBu, THF, rt, 60 min, then [CpFe(CO)2I], toluene, rt, 16 h; (ii) KOtBu, THF, rt, 60 min, then CO2, toluene, rt,
1 h; (iii) [CpFe(CO)2I], toluene, 80 °C, 16 h; (iv) hν, CH2Cl2, 16 h.
by addition of the iron precursor [FeCp(CO)2I] (Scheme
1).22,23 Using the pentamethyl cyclopentadienyl (Cp*)
precursor [FeCp*(CO)2I] instead afforded complexes 4a and
4b in good yields. However, the analogous complex from
imidazolium salt 1c with isobutyl wingtip groups did not form,
suggesting higher steric congestion imparted by the Cp* ligand
as compared to the Cp system. Notably, two- and three-legged
piano-stool complexes with mesityl-substituted NHC ligands
were successfully prepared by the groups of Tatsumi and Song
starting from the CO-free iron precursors [Cp*FeN(SiMe3)2],
[Cp*FeCl(TMEDA)], and [FeCp*(HMDS)] (TMEDA =
Me2NCH2CH2NMe2, HMDS = N(SiMe3)2).28−31
Iron NHC complexes are generally prepared from the air
and moisture sensitive free carbene, which requires inert
conditions.32−34 Here, we demonstrate that the decarbox-
ylation method offers an attractive alternative for the synthesis
of these iron complexes. While this strategy has previously
been applied predominantly for platinum group metals,35 its
application to first-row transition metals is much rarer, mostly
involves copper, and has never been applied to iron(II)
precursors so far.36−40 The reaction of the imidazolium
carboxylate 2a with [FeCpI(CO)2] induced the release of
CO2 and afforded the known complex 3a in similar yields as
via the deprotonation route (43% vs 58%, Scheme 1).23
Attractively, the metalation via decarboxylation proceeds in a
distinct step that is not air or moisture sensitive and avoids the
tedious isolation of the air-sensitive free carbene prior to
transfer to the iron precursor. Likewise, the new Cp*
complexes 4a and 4b were also generated via decarboxylation
of the imidazolium carboxylates 2a and 2b, respectively.35,41
The formation of the new cationic iron complexes 3c and
4a,b was supported in 1H NMR spectroscopy by the expected
relative integral ratio of the ligand signals and those of the Cp
or Cp* fragments at 5.31 and 1.83 0.01 ppm, respectively,
and by the characteristic Fe−Ccarbene resonance at δC 162.5
10.5 (Figures S1−S26). UV irradiation in CH2Cl2 induced CO
release and iodide coordination to form the neutral complexes
5 and 6. Ligand exchange was indicated by a characteristic
color change from yellow to green as well as a diagnostic
upfield shift of the Cp resonance in the 13C NMR spectra from
δC ∼88 to ∼80 in complexes 5 and from ∼99 to ∼90 in the
Cp* complexes 6 as a consequence of the substitution of the π-
accepting CO ligand with the stronger donating iodide
1
(Figures S27−S38 and S39−S44, respectively). All H and
13C NMR shifts are in the same range as those of the related
complexes.17,19,22
Analogous triazolylidene complexes were obtained from the
triazolium salts 1e−g. The formation of previously reported
1539
https://doi.org/10.1021/acs.organomet.1c00200
Organometallics 2021, 40, 1538−1550

Organometallics
pubs.acs.org/Organometallics
Article
Figure 1. ORTEP representations (50% probability level) of the new cationic complexes 3c,e and 4a,b and the neutral analogues 5c,e and 6a,b.
Hydrogen atoms and noncoordinating anions are omitted for clarity, and atom labeling is adjusted for consistency.
a
Table 1. Selected Bond Lengths [Å] and Angles [°] for cationic Cp complexes 3 and Cp* complexes 4
b
b
c
b
b
complex ligand set
3a imi/Cp
3b imi/Cp
3c imi/Cp
3e trz/Cp
3f trz/Cp
3g trz/Cp
4a imi/Cp*
4b imi/Cp*
Fe−Cimi/trz
Fe−CCO
Fe−CCO′
Fe−Ccentroid
CCO−Fe−Cimi/trz
1.969(6)
1.777(5)
1.777(5)
1.731
95.05(19)
95.05(19)
93.4(3)
120.82
1.970(3)
1.774(4)
1.780(3)
1.716
94.2(1)
94.4(2)
92.8(2)
122.2
1.965(6)
1.769(9)
1.768(9)
1.752
93.1(3)
93.7(3)
90.5(4)
121.00
1.976(3)
1.775(3)
1.767(3)
1.718
94.07(12)
92.58(12)
92.36(13)
122.13
1.974(3)
1.765(4)
1.773(4)
1.724
98.82(14)
91.17(14)
92.24(17)
120.46
2.010(6)
1.774(7)
1.767(6)
1.727
99.26(3)
97.54(6)
94.89(7)
121.4(7)
1.979(4)
1.762(5)
1.767(5)
1.743
94.48(19)
94.7(2)
93.1(2)
122.92
1.995(6)
1.759(8)
1.772(8)
1.738
93.6(3)
95.2(3)
91.5(4)
123.50
C
CO′−Fe−Cimi/trz
CCO−Fe−CCO′
Ctrz−Fe−Cpcentroid
N−Ctrz−Fe−Cpcentroid
91.40
91.86
b
92.89
90.97
98.94
98.38
90.75
90.40
a
c
imi = imidazolylidene; trz = triazolylidene. Values obtained from ref 23 (3a and 3b) and ref 27 (3f and 3g). Mixed occupation of anion site with
I (82.3%) or PF6 (17.7%).
Table 2. Selected Bond Lengths [Å] and Angles [°] for Neutral Cp Complexes 5 and Cp* Complexes 6
a
a
a
a
complex ligand set
5a imi/Cp
5b imi/Cp
5c imi/Cp
5d imi/Cp
5e trz/Cp
5g trz/Cp
6a imi/Cp*
6b imi/Cp*
Fe−CNHC
Fe−CCO
Fe−I
Fe−Ccentroid
1.964(3)
1.749(3)
2.6548(3)
1.730
1.972(5)
1.743(6)
2.6597(8)
1.726
1.9652 (19)
1.7470 (19)
2.6428 (3)
1.727
1.980(5)
1.641(9)
2.6445(8)
1.742
1.990(4)
1.777(5)
2.6813(6)
1.744
1.974(3)
1.747(3)
2.6391(4)
1.722
1.970(3)
1.744(3)
2.6582(4)
1.743
1.988(4)
1.736(5)
2.6452(6)
1.743
CCO−Fe−CNHC
CNHC−Fe−I
CCO−Fe−I
CNHC−Fe−Cpcentroid
N−CNHC−Fe−Cpcentroid
97.00(11)
92.35(7)
88.96(9)
123.43
95.2(3)
96.88(15)
85.3(2)
120.54
95.32(8)
92.98 (5)
85.35(6)
124.52
101.2(5)
93.12
96.99(18)
93.02(10)
86.53(13)
121.69
99.48(13)
92.16(7)
89.91(10)
125.15
97.89(13)
92.41(8)
87.91(9)
122.86
97.80(17)
95.46(11)
86.75(15)
122.95
86.5(3)
126.22
107.06
93.88
97.73
134.52
126.89
135.78
105.43
91.98
a
Values obtained from ref 23 (5a and 5b), ref 22 (5d), and ref 27 (5g).
complexes 3f,g proceeded smoothly via deprotonation with
KOtBu,27 and the same methodology was applied to the
triazolium salt 1e to afford the new complex 3e in moderate
yields (Scheme 2). Notably, the iron triazolylidene complex 3f
was also accessible via the decarboxylation route from the new
triazolium carboxylate 2f and [FeCp(CO)2I] in similar yields
as via the free carbene route (39% vs 42%). When irradiated,
the cationic triazolylidene complexes 3e−g transformed into
the neutral complexes 5e−g, which was indicated by a
diagnostic color change from yellow to green as well as
spectroscopic changes similar to those described for the
imidazolylidene analogues, including an upfield shift of the Cp
carbon signals from δC 88 to 80 ppm as well as a ∼20 ppm
downfield shift of the Fe−Ccarbene resonances. All new
imidazolylidene and triazolylidene complexes were stable as
solids in the ambient atmosphere but slowly decomposed in
solution, indicated macroscopically by a gradual color change
to brown and the formation of a precipitate and microscopi-
cally by considerable broadening of the H NMR signals.
1
Remarkably, the synthesis of the Cp* analogues of 3e and 3f
has failed in our hands. Neither the free carbene nor the
decarboxylation route afforded the desired complexes. In an
effort to increase the reactivity, in situ prepared cationic
[FeCp*(CO)2(THF)]OTf was used as an iron precursor42
together with the free carbene from 1f, though only
paramagnetic products were obtained. In contrast, the
triazolium salt 1e with similar steric implications as the
imidazolium salt 1b gave, after deprotonation and reaction
with [FeCp*(CO)2I], the triazolylidene analogue of complex
4b as suggested by the correct relative integral ratio of the
ligand signals and the Cp−Me groups from a crude sample
(Figure S45). Moreover, irradiation of the crude mixture
induced the diagnostic color change to green; however, only
traces of this complex were isolated, and we failed to develop
1540
https://doi.org/10.1021/acs.organomet.1c00200
Organometallics 2021, 40, 1538−1550

Organometallics
pubs.acs.org/Organometallics
Article
Table 3. Vibrational and Electrochemical Data for Cationic and Neutral Fe(II) Complexes 3−6
a
b
Measured in CH2Cl2. Measured in CH2Cl2 using 0.1 M [Bu4N][PF6] as supporting electrolyte, sweep rate of 100 mV s−1, referenced vs SSCE
c
d
using Fc+/Fc as internal standard (E1/2 = +0.46 V, ΔE = 0.1), in parentheses ΔE = Epa − Epc. From ref 22. From ref 27.
synthetic methods to obtain this complex in sufficiently high
quantities. Nonetheless, these experiments indicate that three-
legged piano-stool Fe−Cp* complexes with triazolylidene
complexes should, in principle, be accessible. Notably, Song
and co-workers reported in 2020 a two-legged piano-stool Fe−
Cp* complex with a triazolylidene ligand.31
closer contact in the Cp* complex. Close Cp−H···HNHC
contacts were also identified in solution by Nuclear Over-
hauser effects (NOEs) between the Cp* protons and the HCiPr
protons for 6b. Notably, no such NOE signals were detected in
5b with the same NHC ligand, yet a less bulky Cp instead of
Cp*. Interactions with the Cp ligand only became apparent
when NHCs with bulkier wingtip groups were coordinated.
For example, the Cp protons in complex 5f showed a NOE
with both the mesityl−CH3 and the N−CH2 unit (Figures
S51−S53). These results suggest that steric hindrance of the
Cp* ligand imposes more steric hindrance for bulky NHC
ligands to coordinate to the iron center and rationalize the
failure in preparing the Cp* analogue of complex 3f (vide
supra).
Structural Analyses. Solid state X-ray crystal structures of
representative examples of complexes 3−6 confirmed the
piano-stool geometry and revealed bond lengths and angles in
the expected range (Figure 1, Tables 1 and 2).22,23,27 The Fe−
Ccarbene bonds and the Fe−Cp distances are only marginally
elongated (<0.025 Å) in the Cp* complexes compared to their
Cp analogues. Also, the relative orientation of the carbene
ligand, determined by the N−Ccarbene−Fe−Cpcentroid dihedral
angle θ, does not reveal any significant twists of the carbene in
the cationic complexes 3 and 4 (θ = 91.7° 1.3°) except for
complex 3f and 3g (θ = 98.9° and 98.4°, respectively), which
was attributed to steric constraints between the mesityl groups
and the Cp protons.34 The differences are more noticeable for
the neutral complexes 5 and 6 (θ between 92° and 136°),
suggesting a more flexible rotation. This flexibility may be
facilitated by the smaller I−Fe−CO bond angle as compared to
the CO−Fe−CO bond angle in the cationic analogues.
The steric implications of the Cp* ligand were investigated
by inspecting the shortest distance to NHC wingtip protons in
each structure (Table S1). For the Cp complexes 3 and 5, the
closest Cp−H···HNHC contacts were all between 2.25 and 2.74
Å with 3b displaying the shortest distance and 3c, the longest.
The Cp* complexes featured slightly shorter Cp−CH3···HNHC
distances (2.12−2.35 Å). A comparison of Cp and Cp*
complexes with identical NHC ligand revealed a 0.26 0.12 Å
While the Cp complex 5a with small Me wingtip groups on
the NHC ligand is fluxional due to rotation about the Fe−
CNHC bond and requires low temperature (−20 °C) to reach
1
decoalescence of the resonances in the H NMR spectrum
(ΔG‡ ∼ 59 kJ mol−1),23 the analogous Cp* complex 6a is
more rigid and reveals a slow exchange limit spectrum with two
nonequivalent N−CH3 resonances at δH = 4.04 and 3.86,
respectively (Figure S39).43 In contrast, the signals from the
iBu wingtip groups of complex 5c were broad at room
temperature, though they sharpened upon cooling to −20 °C
(Figure S28, Table S2). From the coalescence temperature of
these signals (Tcoal = 298 K) and the Cimi−H resonance
frequencies, a free energy of activation (ΔG‡) of 58.0 1.0 kJ
mol−1 was calculated for 5c. Complex 5e shows two rotamers
1
in the H NMR spectrum at −25 °C with a free energy of
activation (ΔG‡) of 60.5 1.0 kJ mol−1 (Figure S32, Table
1541
https://doi.org/10.1021/acs.organomet.1c00200
Organometallics 2021, 40, 1538−1550

Organometallics
pubs.acs.org/Organometallics
Article
S3). This higher energy compared to 5c is commensurate with
the larger steric demand of the wingtip groups in 5e.
Donor Properties of the Ligand Sets. To determine the
combined electron donor properties of the NHC ligand and
the Cp(*) moiety, the complexes were analyzed electrochemi-
cally and by IR spectroscopy. IR analysis revealed the
characteristic symmetric and asymmetric bands νs = 2037
12 cm−1 and νas = 1990 13 cm−1 for the cationic bis-carbonyl
complexes 3 and 4 and a single absorption in the 1910−1935
cm−1 range for the neutral monocarbonyl complexes 5 and 6
(Table 3). Triazole-derived NHCs induce generally a 6
1
cm−1 lower energy vibration than imidazolylidenes (cf. 5b vs 5e
and 5d vs 5g) in line with the increased electron donor
properties of triazolylidenes.44 The wingtip group modification
had minor implications for both imidazole- and triazole-
derived carbenes in the cationic systems yet induced some
variation in the neutral complexes (Δνmax = 7 cm−1, cf. 5f vs 5e
and 6a vs 6b). The most substantial shift was noted upon
Figure 2. Normalized cyclic voltammograms of representative
complexes 5a, 5b, 5e, and 6a measured in CH2Cl2, 0.1 M [Bu4N]PF6,
potential in V vs SSCE and referenced to Fc+/Fc (E1/2 = +0.46 V, ΔE
= 0.1 V) or decamethyl ferrocene (E1/2 = +0.11 V, ΔE = 0.1 V), 100
mV s−1 sweep rate.
replacing the Cp ligand for Cp* (Δν = 22
4 cm−1),
h−1, entries 2 and 5; complexes 5d and 5g, TOFmax = 140 vs
230 h−1, entries 4 and 7) with previously reported complex 5g
as the most active complex in this series.27 Third, wingtip
modulations have a marked effect on turnover frequencies,
which increase as follows: Me < iPr, iBu < Mes for the imi
series (entries 1−4) and similarly iPr,iPr < nBu,Mes <
Mes,Mes in the trz series (entries 5−7). The complexes
show varying induction times ranging from 10 up to 90 min,
suggesting that the active catalyst is formed in situ.
Notably, there is no clear correlation between the catalytic
performance and the redox potential as a proxy for the donor
properties of the ligand set (Figure 4a). While there is a good
trend in the Cp series with more donating NHCs increasing
the activity, this trend is not confirmed by the imi/Cp* series,
which should provide much higher activity than that observed
if the electron density at iron is indeed the decisive factor for
catalytic performance.
A more stringent trend was revealed when the catalytic
activity of the complexes was correlated with the steric
influence of the NHC ligand (Figure 4b) as deduced from the
percentage buried volume (%Vbur) calculated with the
SambVca software (Figure S55).45 Accordingly, even a slight
increase of the percentage buried volume leads to a higher
catalyst performance with the mesityl-substituted NHCs (Vbur
= 30−31.2%, TOFmax = 180−230 h−1) performing better than
the analogues with alkyl wingtip groups (Vbur = 25.6−27.4%,
TOFmax = 30−160 h−1). While these steric parameters appear
to be the dominant factor for tailoring catalytic activity, the
electronic factors allow for fine-tuning the activity with trz
systems outperforming the imi analogues (cf. 5b vs 5e and 5d
vs 5g).
In order to probe the steric implications of the Cp*
spectator ligand, different substrates with varying spatial
requirements were evaluated, viz. butyraldehyde as a small
substrate and o-tolualdehyde as a bulky substrate (Table 5,
Figure S56). Except for complexes 5a and 6a with methyl-
substituted NHCs, catalytic activities increased when changing
the substrate from 4-nitrobenzaldehyde to either of these two
aldehydes (Table 5). In addition, turnover frequencies were
generally higher for the bulkier o-tolualdehyde than for
butyraldehyde. Interestingly, the exchange of Cp for Cp*
generally induced an improved catalytic activity toward both
the smaller butyraldehyde (relative to nitrobenzaldehyde) and
the sterically more demanding tolualdehyde (cf. complexes 5b
indicating a major effect of the stronger electron donating Cp*
ligand (cf. 3a,b vs 4a,b,and 5a,b vs 6a,b).
Further insights into the electronic modulation of these
complexes were gained by electrochemical analysis. While the
cationic complexes did not show any oxidation process up to
1.4 V vs SSCE, the neutral complexes 5 and 6 feature a
reversible oxidation in cyclic voltammetry (CV; Figure S54).
The half-wave potentials shift only slightly by <90 mV upon
changing the wingtip group (Table 3). The substitution of the
imidazolylidene scaffold for a triazolylidene lowered the
oxidation potential by 0.065 0.05 V (cf. 5b,d vs 5e,g), and
the exchange of Cp for Cp* had again the largest effect and
shifted the redox potential by 0.27 V (cf. 5a,b vs 6a,b). These
data are in good agreement with the trends deduced from IR
spectroscopy and indicate that the Cp/Cp* modulation is 2−3
times larger than the swap of the NHC from imidazolylidene
to triazolylidene. Wingtip modification is nonlinear, and the
electron-donating character of secondary alkyl groups is
counterbalanced by steric effects, especially with Cp* ligands.
According to both analyses, complex 6b features the most
electron rich iron center with decreasing electron-density along
the series 6b > 6a > 5e > 5g > 5f > 5d > 5b > 5a > 5c. These
effects are displayed in Figure 2 for representative examples
featuring a variable wingtip group, NHC scaffold, and Cp vs
Cp* ligand, respectively.
Catalytic Hydrosilylation of Carbonyl Compounds.
The impact of the broad electronic tunability of these iron
complexes 5 and 6 was evaluated in the reduction of carbonyl
compounds via hydrosilylation.18,27 4-Nitrobenzaldeyde was
employed as a benchmark substrate and PhSiH3, as hydro-
silylation agent; yields were calculated after the Si−O bond
cleavage with Bu4NF (Table 4). Very distinct time−conversion
profiles were observed, which indicate a profound influence of
the ligand set on the catalytic activity (Figure 3). A comparison
of the catalytic performance revealed some general trends.
First, the introduction of a Cp* spectator ligand instead of Cp
is detrimental to catalytic activity and leads to considerably
slower conversion (cf. complexes 5a and 6a, turnover
frequency (TOFmax) = 30 vs 20 h−1, entries 1 and 8) though
this is less pronounced when the carbene contains iPr instead
of methyl wingtip groups (5b vs 6b, entries 2 and 9). Second,
the replacement of the carbene scaffold from imi to trz reduces
the reaction time and enhances the catalytic performance
considerably (cf. complexes 5b and 5e, TOFmax = 50 vs 160
1542
https://doi.org/10.1021/acs.organomet.1c00200
Organometallics 2021, 40, 1538−1550

Organometallics
pubs.acs.org/Organometallics
Article
a
Table 4. Hydrosilylation of 4-Nitro Benzaldehyde by Iron(II) Complexes 5−6
a
General conditions: substrate (0.5 mmol), PhSiH3 (0.6 mmol), [Fe] complex (5 μmol, 1 mol %), C6Me6 as internal standard (50 μmol), 1,2-
b
c
dichloroethane (DCE; 2.5 mL), 60 °C. Conversion determined by 1H NMR spectroscopy as an average of at least two runs. Spectroscopic yield
after alcohol deprotection. From ref 27.
d
spectator ligand do not discriminate bulkier substrates
compared to their Cp analogues with a sterically better
available iron center. The choice of Cp vs Cp* is therefore
substrate dependent and needs to be evaluated for determining
the best performing catalyst system. As shown above for
nitrobenzaldehyde, also for the butyraldehyde and tolualde-
hyde, the TOFmax correlates reasonably well with steric ligand
parameters and much less with the ligand electronics (Figure
S57a,b) though, again, the trz ligands outperform the imi
series.
While there was no steric discrimination observed between
tolualdehyde and butyraldehyde, a steric bias can be deduced,
however, from experiments using very bulky substrates. Thus,
neither mesityl benzaldehyde nor o-tolyl methyl ketone are
Figure 3. Time−conversion profile of the hydrosilylation of 4-
nitrobenzaldehyde by iron(II) complexes 5−6 (conditions as in Table
converted even when using the most potent trz/Cp complex
5f. Likewise, benzylacetate is not hydrosilylated under these
conditions.
Mechanistic Studies. While previous work disclosed a
variety of mechanistically diverse catalytic cycles,27,46−50 we
aimed here to shed some light on the mode of activation of
these carbene iron catalysts in hydrosilylation. Therefore, a set
of stoichiometric experiments was conducted as well as some
tailored modifications of the catalyst precursor in order to
4).
and 6b, TOFmax = 70−80 vs 160−180 h−1; entries 2, 6, 8, and
12). In both the Cp and Cp* analogues, the imidazolylidene
ligand with the iPr wingtip group (5b and 6b) is favorable in
the hydrosilylation catalysis over the methyl analogue (5a and
6a; TOFmax = 10−20 vs 70−80 h−1 (5a vs 5b), 6−20 vs 160−
180 h−1 (6a vs 6b), entries 1, 2, 5−8, 11, and 12). These
results suggest that the sterically hindered catalysts with a Cp*
Figure 4. (a) Plot of oxidation potentials (E1/2) vs TOFmax for the catalytic hydrosilylation of 4-nitrobenzaldehyde; (b) correlation of buried
volume (%Vbur) with TOFmax. Color code: imi/Cp ligand set in blue, trz/Cp ligand set in red, and imi/Cp* ligand set in green.
1543
https://doi.org/10.1021/acs.organomet.1c00200
Organometallics 2021, 40, 1538−1550

Organometallics
pubs.acs.org/Organometallics
Article
a
Table 5. Hydrosilylation of Butyr- and Tolualdehyde by Iron(II) Complexes 5 and 6
a
General conditions: substrate (0.5 mmol), PhSiH3 (0.6 mmol), [Fe] complex (5 μmol, 1 mol %), C6Me6 (50 μmol) as internal standard, DCE
b
c
1
(2.5 mL), 60 °C. Conversion determined by H NMR spectroscopy as an average of at least two runs. Spectroscopic yield determined for
silylated butanol and deprotected 2-methyl-benzylalcohol, respectively.
facilitate the activation process. When the electronic saturation
of the iron complexes 5 and 6 due to their 18 electron
configuration is considered, η5-to-η3 Cp ring slippage51 or the
dissociation of an ancillary ligand constitute plausible
activation pathways. The latter hypothesis is supported by
earlier work that demonstrated that cationic complexes akin to
3 and 4 require UV irradiation to become active hydro-
silylation catalysts.18
Probing Iodide Dissociation. When complex 5f was
dissolved in MeCN, the color gradually darkened and turned
red within 10 min with concomitant formation of a second
species according to 1H NMR spectroscopy (Figure S58). This
change was reversible and was attributed to iodide dissociation
and solvent coordination, a process that may also be
potentially relevant for catalyst activation. To investigate the
relevance of iodide coordination, complex 7c was prepared in
which the iodide was replaced with MeCN as a potentially
better leaving group (Scheme 3). This complex was readily
(Table S9). In an attempt to generate a putative catalytic iron
alkoxide or iron hydride intermediate, complex 7c was used as
a catalyst precursor in the presence of LiOiPr or NaBH4 as
additives.49 Neither of these additives improved the catalytic
activity and, to the contrary, resulted in lower conversions.
1
Also, stoichiometric H NMR experiments did not lead to a
defined complex and in the presence of LiOiPr only revealed
broadening of the signals (Figure S59). Likewise, the addition
of 0.6 equiv of I2 (relative to the iron complex) as a mild
oxidizing agent to access a putative iron(III) species only
lowered the catalytic performance. These data therefore point
to some relevance of iodide coordination for imparting
catalytic activity.
Monitoring Stoichiometric Experiments. The exposure
of complex 5f to 2 equiv of 4-nitrobenzaldehyde in CD2Cl2 at
60 °C for 30 min induced the formation of a minor new set of
aromatic signals at 8.10 and 7.49 ppm of about 5% of the
intensity compared to nitrobenzaldehyde (Figures S60 and
S61). The addition of 4 equiv of PhSiH3 led to the complete
consumption of 4-nitrobenzaldehyde within 10 min (Figure
S60c,d). When a stoichiometric mixture of 4-nitrobenzalde-
hyde and complex 5f in DCE-d4 was heated for 3 h, an almost
full depletion of the aldehyde signals was noted (Figure S62).52
These changes may point to the formation of a paramagnetic
species, such as an iron-alkoxy complex, though these events
are clearly outside the catalytic time regime. Nonetheless, these
data suggest that the substrate is interacting with the complex
but that the resulting species is either not stable enough to be
a
Scheme 3. Synthesis and ORTEP Plot of Complex 7c
a
−
50% probability, all hydrogen atoms and non-coordinating PF6
1
isolated or not detectable by H NMR spectroscopy. Notably,
anions omitted for clarity. Reagents and conditions: (i) AgPF6,
CH2Cl2, room temperature, 1 h; (ii) hν, MeCN, 1 h.
complex 5f on its own is stable in refluxing CD2Cl2 for >2 h
and degrades slowly in DCE-d4 yet without the formation of
any new signals (Figures S63 and S64).
The complementary reaction of complex 5f with 3 equiv of
PhSiH3 at 60 °C in the absence of aldehyde substrate induced
spectral changes only after 20 min with the appearance of a
new set of signals and in particular a weak signal at −14.29
ppm (Figures S65 and S66), suggesting the formation of small
quantities of an iron hydride species (<5%).53,54 The main
species in the spectrum remained 5f, and this was visibly
supported by the bright green color of the solution. The
obtained from the dicarbonyl precursor 3c via AgPF6-mediated
anion exchange followed by UV irradiation in MeCN. The
coordination of MeCN was confirmed by a characteristic
1
singlet at δH = 2.28 in the H NMR spectrum (Figure S46).
Complex 7c catalyzed the hydrosilylation of 4-nitro-
benzaldehyde under standard conditions, though considerably
slower than its iodide analogue 5c. It required 70 h to reach
68% conversion compared to full conversion in 3 h with 5c
1544
https://doi.org/10.1021/acs.organomet.1c00200
Organometallics 2021, 40, 1538−1550

Organometallics
pubs.acs.org/Organometallics
Article
Figure 5. In situ FT-IR spectra recorded during hydrosilylation of 4-nitrobenzaldehyde with 5f as catalyst under standard conditions (Table 4).
Changes over time in 5 min intervals (red at 0 h and blue at 40 min); red arrows indicate the change of signal intensities. Inset: Catalytic profile of
1
two catalytic reactions with conversions determined by H NMR spectroscopy (green circles) and by IR spectroscopy from changes at 1195 and
1711 cm−1 (signals of 4-nitrobenzaldehyde, amber and red traces) and 1082 cm−1 (attributed to the Si−O band in the product, blue trace).
addition of 2 equiv of 4-nitrobenzaldehyde resulted in its full
consumption within 10 min (Figure S65c,d) and the
disappearance of the hydride resonance. Upon addition of an
extra 2 equiv of aldehyde, again some 60% was consumed
within 10 min (Figure S65e), indicating full consumption of
the silane and pointing to an iron species in solution that
remains catalytically active. The presence of signals due to 5f
after these three turnovers revealed that only a small fraction of
the precatalyst was transformed to the active species. Notably,
the iron complex in the presence of only PhSiH3 was stable far
beyond the induction period typically observed in catalytic
runs (up to 25 min for complex 5f, cf. Tables 4 and 5, Figure
S67). Only extended incubation of 5f at 60 °C led to a color
change from green to yellow with a 50% decrease of the signals
of the original iron complex after 16 h (Figure S67).
products (Figure 5). New phenylsilane products are suggested
by the appearance of a broad band around vSiH = 2160 cm−1.57
The plot of the normalized intensity changes over time yields
similar rates as deduced from 1H NMR monitoring of a parallel
reaction (inset of Figure 5).
Catalyst activation was monitored in a set of stoichiometric
experiments and corroborated the conclusions deduced from
NMR spectroscopic analyses. The interaction of complexes 5c
or 5f with 4-nitrobenzaldehyde in a 1:1 molar ratio resulted in
a linear decrease of the carbonyl vibration bands of the
complex and aldehyde at 1935 and 1711 cm−1, respectively,
and the full disappearance within 40 min (Figures S72 and
S73), strongly suggesting an interaction between the substrate
and the catalyst through CO dissociation from the iron
coordination sphere.58 A similar process has been established
for the Shvo catalyst where CO dissociation is more favorable
than η5/η3 ring slippage.59,60
Stoichiometric experiments with complex 5f and PhSiH3 (1
equiv) showed a substantially slower transformation of 5f than
those in the presence of aldehyde under otherwise identical
conditions (20% in 1 h; Figure S75). The addition of 1 equiv
of 4-nitrobenzaldehyde led to a full disappearance after 45 min
of the CO band of the complex and full consumption of the
substrate, together with almost a 50% decrease of the
phenylsilane. The time regime of these experiments suggests
that the iron complex has a much higher affinity for the
aldehyde and indicates that (reversible) CO dissociation and
aldehyde coordination constitute a plausible catalyst activation
pathway. Of course, the currently available data cannot rule out
a Curtin-Hammett scenario with a catalytically irrelevant pre-
equilibrium aldehyde coordination and parallel formation of an
iron hydride as the active species present in too low quantities
to be detected by ReactIR. We note however that aldehyde
coordination and conversion have a strong influence since
earlier work in our group demonstrated that the hydrosilylation
of ketones is hampered unless the catalytically active species is
formed by spiking the reaction with some aldehyde.27
1
Concomitantly, the H NMR resonance of PhSiH3 at 4.15
ppm decreased to ∼5% and new singlets appeared at 5.02 and
5.19 ppm together with downfield shifted aromatic signals,
which were attributed to reshuffled phenylsilane55,56 and
siloxane species; the latter presumably formed due to
fortuitous moisture or oxygen (Figure S68).
The combined results from these NMR studies suggest the
formation of a pre-equilibrium either between the iron
complex and the silane to form an iron hydride as the putative
active species or involving the iron complex and the aldehyde
to form an iron substrate adduct. In either case, there is a
strong preponderance toward the dissociated species, i.e., the
starting materials. Notably, such a pre-equilibrium situation
may rationalize the observed induction periods under catalytic
conditions.
In Situ Reaction Monitoring. The CO functionalities in
both the complex and the aldehyde substrate provided a
diagnostic handle to use ReactIR spectroscopy for investigating
the catalyst activation and performance in situ. A catalytic run
under standard conditions with complex 5f revealed a gradual
decrease of the CO stretch vibration of the 4-nitro-
benzaldehyde at 1711 and 1195 cm−1 and a concomitant
increase of a broad band at 1082 cm−1 attributed to the C−O−
Si unit indicative for the formation of the hydrosilylated
In support of the mechanistic scenario involving ketone
coordination in the catalyst activation step, an additional
1545
https://doi.org/10.1021/acs.organomet.1c00200
Organometallics 2021, 40, 1538−1550

Organometallics
pubs.acs.org/Organometallics
Article
mass spectrometry was carried out with a Thermo Scientific LTQ
Orbitrap XL instrument (ESI-TOF). IR spectra were recorded on a
Jasco 4700 FT-IR instrument in CH2Cl2 solution at 1 cm−1
resolution. Time-resolved online MCT FT-IR spectra were recorded
on a ReactIR 15 Instrument (Mettler Toledo) equipped with a
diamond probe (DiComp, optical range of 3000−650 cm−1). For
online monitoring, the diamond probe was introduced into a 10 mL
Schlenk containing the reaction mixture, and spectra were recorded at
specific times. The experiments were run under nitrogen conditions.
UV irradiation was carried out using a UVP Blak-Ray B-100AP lamp.
Cyclic voltammetry measurements were carried out using a Metrohm
Autolab Model PGSTAT101 potentiostat employing a gastight three-
electrode cell under an argon atmosphere. A platinum disk with 7.0
mm2 surface area was used as the working electrode and polished
before each measurement. The reference electrode was Ag/AgCl; the
counter electrode was Pt foil. Bu4NPF6 (0.1 M) in dry CH2Cl2 was
used as supporting electrolyte with analyte concentrations of
approximately 1 mM. The ferrocenium/ferrocene (Fc+/Fc) redox
couple was used as an internal reference (E1/2 = 0.50 V vs SSCE).61
General Procedure for the Preparation of Complexes 3 and
4. Method A. The imidazolium/triazolium salt (1 equiv) and KOtBu
(1.2 equiv) were suspended in dry THF (10 mL). After 1 h of stirring
at room temperature, the THF was removed under reduced pressure
and the free carbene was extracted in dry toluene (2 × 6 mL); the
resulting suspension was filtered into a dry toluene (3 mL) solution of
[CpFe(CO)2I] or [Cp*Fe(CO)2I] (0.9 equiv). The resulting mixture
was stirred at room temperature under the exclusion of light for 16 h.
The precipitate was collected by filtration, washed with toluene (2 × 5
mL), extracted with CH2Cl2, and dried in vacuo to yield the crude
product as a light-sensitive material, which hampered further
purification.
Method B. The azolium carboxylate 2 and iron precursor were
suspended in dry toluene (6 mL) and stirred at 80 °C for 16 h under
the exclusion of light. After cooling to room temperature, the
precipitate was collected by filtration and purified as described in
Method A.
Synthesis of 3a. According to Method B, complex 3a was afforded
as a yellow powder (65 mg, 43%) by starting from 2a (52 mg, 0.38
mmol) and [CpFe(CO)2I] (114 mg, 0.38 mmol). Analytical data
agree with those reported for this compound.23
Synthesis of 3c. According to Method A, complex 3c was afforded
as a yellow powder (308 mg, 75%) by starting from 1c (334 mg, 1.02
mmol), KOtBu (155 mg, 1.28 mmol), and [CpFe(CO)2I] (259 mg,
0.85 mmol). Single crystals suitable for X-ray diffraction were
obtained by the slow diffusion of Et2O to a CH2Cl2 solution of the
complex.
catalytic experiment was conducted in which complex 6b was
prestirred with 4-nitrobenzaldehyde for 1 h. The addition of
PhSiH3 initiated catalytic turnover immediately without any
activation period, and conversion was complete within 60 min
according to 1H NMR monitoring (Figure S76). In
comparison, without prestirring, an induction time of some
40 min was observed with this catalyst system, indicating that
aldehyde coordination to iron is relevant. In contrast, the
preactivation of complex 6b with PhSiH3 resulted in much
slower conversion, requiring 110 min to reach completion
(Figure S76). These data lend strong support to a catalyst
activation involving CO ligand displacement by the aldehyde
substrate rather than formation of an initial iron hydride
species from the iron complex and the silane.
CONCLUSION
■
This work extends the diversity of piano-stool NHC iron
complexes and includes variation of the NHC scaffold, the
wingtip group, and ancillary ligands as well as modulation of
the Cp ligand with the stronger donating Cp* analogue.
Variability was also implemented on the synthetic level by
introducing decarboxylation, so far unprecedented for iron(II)
precursors, for preparing both imidazolylidene and triazolyli-
dene piano-stool iron complexes. The application of these iron
complexes with diverse ligand sets in the catalytic hydro-
silylation of aldehydes revealed a clear trend in activity that
depends strongly on the NHC scaffold (trz > imi) and revealed
a good correlation with steric parameters as defined by the %
Vbur. Notably, the choice of Cp vs Cp* ligand may be substrate
dependent; for example, the activity increase is opposite for
nitrobenzaldehyde and butyraldehyde. Mechanistic investiga-
tions provide strong support for a catalyst activation that
involves displacement of the CO ligand with the aldehyde
substrate. Such a process rationalizes the critical role of the
substrate in catalyst optimization and also the low activity of
these complexes toward ketones. Moreover, it validates the
need for a high modularity of the ligand sets in these piano-
stool iron(II) complexes as demonstrated in this work. The
option to vary virtually all parameters, from carbene scaffold to
wingtip sterics and electronics, ancillary ligand, Cp substitution
pattern, and counterions around the iron(II) center, will also
offer opportunities for exploring further catalytic applications
beyond hydrosilylation.
1H NMR (400 MHz, CD2Cl2): δ 7.38 (br, 2H, HIm), 5.31 (s, 5H,
Cp), 4.00 (d, J = 7.8 Hz, CH2), 2.22 (sept, J = 7.0 Hz, 2H, CH), 1.02
(d, J = 6.6 Hz, 12H, CH3). 13C{1H} NMR (101 MHz, CD2Cl2): δ
211.81 (CO), 164.47 (CIm−Fe), 125.54 (CIm−H), 87.66 (Cp), 59.20
(CH2), 29.66 (CH), 19.89 (CH3). IR (CH2Cl2, cm−1): 2049, 2002
ν(CO). HRMS: m/z calcd. for C18H25FeIN2O2 [M − I]+, 357.1246;
found, 357.1260. Elemental Analysis calcd. (%) for C18H25FeIN2O2
(484.16): C, 44.65; H, 5.20; N, 5.79; found: C, 44.34; H, 5.15; N,
5.69.
Synthesis of 3e. According to a slightly modified Method A, the
mixture of base and ligand was directly cannulated into the toluene
solution of [CpFe(CO)2I] (161 mg, 0.53 mmol) by starting from 1e
(157 mg, 0.53 mmol) and KOtBu (77 mg, 0.64 mmol). The solvents
were removed under reduced pressure, and the crude was used
directly for the synthesis of 5e. Single yellow crystals suitable for X-ray
diffraction were obtained by slow diffusion of Et2O into a CH2Cl2
solution of the complex.
EXPERIMENTAL SECTION
■
General Comments. Toluene, THF, CH2Cl2, Et2O, and hexane
were dried by passage through solvent purification columns. 1,2-
Dichloroethane was dried over 4 Å molecular sieves and degassed
with argon. All other reagents were commercially available and used
without further purification. Metalation reactions and the purification
of complexes were carried out under an inert nitrogen atmosphere
using standard Schlenk techniques. The synthesis of complexes 3a,
3b, 3d, 3f, 3g, 5a, 5b, 5d, 5f, and 5g22,23,27 have been reported
elsewhere, and the synthesis of all ligands is provided in the
Supporting Information. NMR spectra were measured at 25 °C on
Bruker spectrometers operating at 300 or 400 MHz (1H NMR) and
75 or 101 MHz (13C{1H} NMR), respectively. Chemical shifts (δ in
ppm, coupling constants J in Hz) were referenced to residual solvent
resonances downfield to SiMe4. Assignments were made based on
homo- and heteronuclear shift correlation spectroscopy. The purity of
bulk samples of the complexes has been established by NMR
spectroscopy and by elemental analysis, which was performed at the
University of Bern Microanalytic Laboratory by using a Thermo
Scientific Flash 2000 CHNS-O elemental analyzer. High-resolution
1H NMR (300 MHz, CD2Cl2): δ 5.31 (s, 5H, Cp), 5.11−5.02 (m,
1H, CH), 4.19 (s, 3H, NCH3), 3.62−3.52 (m, 1H, CH), 1.62 (d, J = 6
Hz, 6H, CH3), 1.43 (d, J = 6 Hz, 6H, CH3). 13C{1H} NMR (101
MHz, CD2Cl2): δ 212.57 (CO), 152 (Ctrz−Fe), 87.31 (Cp), 58.30
(NCH), 46.71 (Ctrz), 39.85 (NCH3), 27.58 (CCH), 23.90 (CH3),
20.54 (CH3). IR (CH2Cl2, cm−1): 2042, 1996 ν(CO). HRMS: m/z
calcd. for C16H22FeIN2O [M − I]+, 344.1056; found, 344.1050.
1546
https://doi.org/10.1021/acs.organomet.1c00200
Organometallics 2021, 40, 1538−1550

Organometallics
pubs.acs.org/Organometallics
Article
Elemental Analysis calcd. (%) for C16H22FeIN3O2 (471.01): C, 40.79;
H, 4.71; N, 8.92; found: C, 40.55; H, 4.25; N, 8.90.
Synthesis of 3f. According to Method B, complex 3f was afforded
as a yellow powder (58 mg, 39%) by starting from 2f (66 mg, 0.22
mmol) and [CpFe(CO)2]I (77 mg, 0.25 mmol). Analytical data agree
with those reported for this compound.27
Synthesis of 4a. According to Method A, the crude product was
started from 2a (45 mg, 0.32 mmol) and [Cp*Fe(CO)2I](106 mg,
0.28 mmol), extracted with CH2Cl2, and filtrated through a short pad
of silica; the volatiles were removed by reduced pressure to yield the
complex 4a as a yellow powder (52 mg, 39%). Single crystals suitable
for X-ray diffraction were obtained by slow diffusion of Et2O to a
CH2Cl2 solution of the complex.
1.45 (d, 6.7 Hz, 6H, NCCH3) 1.38, 1.17 (d, 7.3 Hz, 6H, CH3).
13C{1H} NMR (101 CD2Cl2, −25 °C): δ 225.07 (CO), 163.57 (Ctrz−
Fe), 151.25 (Ctrz−C), 79.94 (Cp), 57.89 (NCH), 38.19 (NCH3),
26.84 (CCH), 24.25, 23.32, 20.39, 20.03 (CH3). Minor rotamer
(40%): 1H NMR (400 MHz, CD2Cl2, −25 °C): δ 5.40 (sept, 6.7 Hz,
1H, NCH), 4.71 (sept, 7.3 Hz, 1H, CH), 4.40 (s, 5H, Cp), 4.03 (s,
CH3), 1.55, 1.35 (d, 6.7 Hz, 6H, NCCH3) 1.49−1.45, 1.27 (d, 7.3 Hz,
6H, CH3). 13C{1H} NMR (101 CD2Cl2, − 25 °C): δ 225.17 (CO),
164.52 (Ctrz−Fe), 153.60 (Ctrz−C), 80.24 (Cp), 56.17 (NCH), 38.29
(NCH3), 28.21 (CCH), 23.45, 23.30, 20.37, 19.99 (CH3). IR
(CH2Cl2, cm−1): 1928 ν(CO). HRMS: m/z calcd. for
C15H22FeN3O [M − I]+, 316.1107; found, 316.1099. Elemental
Analysis calcd. (%) C15H22FeIN3O (456.15): C, 40.66; H, 5.00; N,
9.48; found: C, 41.16; H, 5.10; N, 9.52.
Synthesis of 6a. The imidazolium carboxylate 2a (72 mg, 0.51
mmol) and [Cp*Fe(CO)2]I (267 mg, 0.70 mmol) were suspended in
dry toluene (5 mL) and stirred at 80 °C for 16 h under the exclusion
of light. After cooling to room temperature, the precipitate was
collected by filtration and washed with toluene (2 × 3 mL).
According to the general procedure, complex 6a was afforded as a
green powder (39.4 mg, 40%). Single crystals suitable for X-ray
diffraction were obtained by dissolving the product in CH2Cl2 (2 mL)
and layering with dry hexanes.
1H NMR (400 MHz, CD2Cl2): δ 7.42 (s, 2H, HIm), 3.83 (s, 6H,
NCH3), 1.84 (s, 15H, Cp*−CH3). 13C{1H} NMR (101 MHz,
CD2Cl2): δ 214.20 (CO), 172.90 (CIm−Fe), 127.17 (CIm−H), 99.45
(CCp*Ar), 36.86 (NCH3), 10.45 (Cp*−CH3). IR (CH2Cl2, cm−1):
2027, 1977 ν(CO). HRMS: m/z calcd. for C17H23FeIN2O2 [M − I]+,
343.1103; found, 343.1106. Elemental Analysis calcd. (%) for
C17H23FeIN2O2 (470.01): C, 43.43; H, 4.93; N, 5.96; found: C,
42.93; H, 4.68; N, 5.93.
Synthesis of 4b. According to Method A, complex 4b was afforded
as a yellow powder (140 mg, 62%) by starting from 1b (140 mg, 0.47
mmol), KOtBu (67 mg, 0.55 mmol), and [Cp*Fe(CO)2]I (160 mg,
0.43 mmol).
According to Method B, complex 4b was afforded as a yellow
powder (141 mg, 71%) by starting from 2b (75 mg, 0.38 mmol) and
[Cp*Fe(CO)2I] (142 mg, 0.38 mmol). Single crystals suitable for X-
ray diffraction were obtained by slow diffusion of Et2O to a CH2Cl2
solution of the complex.
1H NMR (400 MHz, CD2Cl2): δ 7.08, 6.98 (d, J = 4.0 Hz, 2H,
HIm), 4.04, 3.86 (s, 6H, NCH3), 1.69 (s, 15H, Cp*−CH3). 13C{1H}
NMR (101 MHz, CD2Cl2): δ 226.66 (CO), 191.15 (CIm−Fe),
124.69, 124.47 (CIm−H), 89.51 (CCp*Ar), 43.08 (NCH3), 39.27
(NCH3), 10.65 (Cp*−CH3). IR (CH2Cl2, cm−1): 1914 ν(CO).
HRMS: m/z calcd. for C16H23FeIN2O [M − I]+, 315.1155; found,
315.1154. Elemental Analysis calcd. (%) for C16H23FeIN2O (442.12)
(Et2O) 0.2%: C, 44.16; H, 5.51; N, 6.13; found: C, 43.81; H, 6.05; N,
6.27.
Synthesis of 6b. According to the general procedure, complex 6b
was afforded as a green powder (80 mg, 60%) from 4b (140 mg, 0.27
mmol). Single crystals suitable for X-ray diffraction were obtained by
dissolving the product in CH2Cl2 (2 mL) and layering with dry
hexanes.
1H NMR (CD2Cl2, 400 MHz): δ 7.44 (s, 2H, Im), 4.53 (sept, J =
7.0 Hz, 2H, CH), 1.82 (s, 15H, Cp*−CH3), 1.61, 1.46 (d, J = 6.1 Hz,
12H, CH3). 13C{1H} NMR (CD2Cl2, 101 MHz): δ 214.03 (CO),
169.14 (CIm−Fe), 122.59 (CIm−H), 99.45 (CCp*Ar), 55.15 (CH),
24.91, 24.47, 23.24 (CH3), 10.24 (Cp*−CH3). IR (CH2Cl2, cm−1):
2024, 1977 ν(CO). HRMS: m/z calcd. for C21H31FeIN2O2 [M − I]+,
399.1729; found, 399.1731. Elemental Analysis calcd. (%) for
C21H31FeIN2O2 (526.24): C, 47.93; H, 5.94; N, 5.32; found: C,
47.55; H, 5.83; N, 4.89.
General Procedure for the Preparation of Complexes 5 and
6. Complex 3 or 4 was dissolved in CH2Cl2 and irradiated with a high
intensity UV lamp at 365 nm for 16 h. The resulting green solution
was concentrated to ∼2 mL and layered with dry hexane. After 24 h,
the solution was filtered and evaporated to dryness to yield a dark
green solid.
1H NMR: δ 7.17, 7.07 (d, J = 4.0 Hz, 2H, Im), 5.63−5.42, 5.25−
4.97 (m, 2H, CH), 1.66 (s, 15H, Cp*−CH3), 1.62, 1.48, 1.42, 1.38
(d, J = 6.4 Hz, 12H, CH3). 13C{1H} NMR (101 MHz, CD2Cl2): δ
119.90, 119.35 (CIm−H), 88.99 (CCp*Ar), 53 (CH), 24.89, 24.87,
23.35 (CH3), 10.89 (Cp*−CH3). IR (CH2Cl2, cm−1): 1910 ν(CO).
HRMS: m/z calcd. for C20H31FeIN2O [M − I]+, 371.1780; found,
371.1775. Elemental Analysis calcd. (%) for C20H31FeIN2O (498.08):
C, 48.21; H, 6.27; N, 5.62; found: C, 48.25; H, 6.30; N, 6.07
Synthesis of 7c. The imidazolium salt 1c (334 mg, 1.02 mmol) and
KOtBu (155 mg, 1.28 mmol) were suspended in dry THF (10 mL).
After 1 h of stirring at room temperature, the mixture was added to a
suspension of [CpFe(CO)2]I (159 mg, 0.85 mmol) in dry toluene
(30 mL) and stirred at room temperature for 1 h under the exclusion
of light. The precipitate was collected by filtration, washed with ether
(2 × 10 mL), taken up in CH2Cl2, and filtrated through a short pad of
silica. All volatiles were removed under reduced pressure. The residue
was dissolved in CH2Cl2, and AgPF6 (212 mg, 0.84 mmol) was added.
After 1 h of stirring at room temperature under the exclusion of light,
the mixture was filtrated through a short pad of silica and the solvents
were removed under reduced pressure. The residue was dissolved in
MeCN (15 mL) and irradiated with a high intensity UV lamp at 365
nm for 40 h. The volatiles were removed under reduced pressure.
Single crystals suitable for X-ray diffraction were obtained by
dissolving the product in CH2Cl2 and layering with dry hexanes
(202 mg, 47%).
Synthesis of 5c. According to the general procedure from 3c (308
mg, 0.64 mmol), complex 5c (189 mg, 64%) was afforded. Single
crystals suitable for X-ray diffraction were obtained by dissolving the
crude in CH2Cl2 (2 mL) and layering with dry hexanes.
1H NMR (300 MHz, CD2Cl2, −20 °C): δ 7.20, 7.06 (br, 2H, HIm),
4.90 (dd, J = 13.7, 8.9 Hz, 1H, CH2), 4.43 (s, 5H, Cp−H), 4.34 (dd, J
= 13.6, 7.1 Hz, 1H, CH2), 4.19−4.01 (m, 2H, CH2), 2.39−2.21 (m,
1H, CH), 2.10−2.03 (m, 1H, CH), 1.06 (d, J = 6.6 Hz, 3H, CH3),
1.00 (dd, J = 11.9, 6.8 Hz, 6H, CH3), 0.74 (d, J = 6.6 Hz, 3H, CH3).
13C{1H} NMR (101 MHz, CD2Cl2): δ 224.50 (CO), 184.14
(CIm−Fe), 122.77 (CIm−H), 80.30 (CCp), 60.68, 58.39 (NCH2),
29.51, 29.25 (CH), 19.95, 19.87, 19.72, 19.56 (CH3). IR (CH2Cl2,
cm−1): 1934 ν(CO). HRMS: m/z calcd. for C17H25FeIN2O [M − I]+,
329.1304; found, 329.1311. Elemental Analysis calcd. (%) for
C17H25FeIN2O (456.15): C, 44.76; H, 5.52; N, 6.14; found: C,
44.58; H, 5.37; N, 6.25.
Synthesis of 5e. The general procedure starting from 3e afforded
complex 5e (50 mg, 21%, overall yield from 1e). Single crystals
suitable for X-ray diffraction were obtained by dissolving the crude in
CH2Cl2 (2 mL) and layering with dry hexanes.
1H NMR (400 MHz, CD2Cl2): δ 7.29 (s, 2H, HIm), 4.71 (s, 5H,
Cp), 4.16−4.11 (m, 2H, CH2), 3.96−3.91 (m, 2H, CH2), 2.28 (s, 3H,
CH3CN), 2.18 (m, 2H, CH), 1.02, 0.96 (d, J = 6.6 Hz, 12H, CH3).
13C{1H} NMR (101 MHz, CD2Cl2): δ 220.05 (CO), 176.56 (CIm−
Fe), 134.61 (CNCH3), 124.24 (CIm−H), 82.59 (Cp), 58.43 (CH2),
29.73 (CH), 20.02 (CH3), 5.16 (CH3CN). IR (CH2Cl2, cm−1): 1980
ν(CO). HRMS: m/z calcd. for C19H28FeN3O [M − PF6]+, 370.1576;
found, 370.1575. Elemental Analysis calcd. (%) for C19H28F6FeN3OP
1H NMR (400 MHz, toluene-d, 65 °C): δ 6.42−6.08, 4.52−4.27
(broad, 2H, CH), 4.20 (s, 5H, Cp), 3.11 (s, CH3), 1.53−1.35, 1.08−
0.99, 0.95−0.83 (broad, 12H, CH3). Major rotamer (60%): 1H NMR
(400 MHz, CD2Cl2, − 25 °C): δ 6.36 (sept, 6.7 Hz, 1H, NCH), 3.96
(sept, 7.3 Hz, 1H, CH), 4.39 (s, 5H, Cp), 4.01 (s, CH3), 1.65, 1.49−
1547
https://doi.org/10.1021/acs.organomet.1c00200
Organometallics 2021, 40, 1538−1550

Organometallics
pubs.acs.org/Organometallics
Article
(515.26): C, 44.29; H, 5.48; N, 8.16; found: C, 44.33; H, 5.66; N,
8.11.
Authors
Pamela V. S. Nylund − Department of Chemistry &
Typical Procedure for Hydrosilylation Catalysis. A solution of
4-nitrobenzaldehyde (0.5 mmol), phenylsilane (74 μL; 0.6 mmol),
and hexamethylbenzene (8.1 mg; 0.05 mmol) or 1,3,5-trimethox-
ybenzene (8.4 mg; 0.05 mmol) in 1,2-dichloroethane (2.0 mL) was
stirred at 60 °C for 10 min under an N2 atmosphere. The iron
complex was added from a stock solution (0.5 mL, 0.01 M, 5 μmol),
and aliquots were taken at specific times, diluted with CDCl3, and
Biochemistry, University of Bern, 3012 Bern, Switzerland
Nathalie C. Ségaud − Department of Chemistry &
Biochemistry, University of Bern, 3012 Bern, Switzerland;
orcid.org/0000-0002-9221-1416
Complete contact information is available at:
https://pubs.acs.org/10.1021/acs.organomet.1c00200
1
analyzed by H NMR spectroscopy.
Crystallographic Details. All measurements were made on an
Oxford Diffraction SuperNova area-detector diffractometer62 using
mirror optics, monochromated Mo Kα radiation (λ = 0.71073 Å), and
Al filtering.63 The unit cell constants and an orientation matrix for
data collection were obtained from a least-squares refinement of the
setting angles of reflections in the range of 2.0° < θ < 27.9°. A total of
728 frames were collected using ω scans with 8 + 8 s exposure time, a
rotation angle of 1.0° per frame, a crystal−detector distance of 65.0
mm, and T = 173(2) K. Data reduction was performed using the
CrysAlisPro62 program. The intensities were corrected for Lorentz
and polarization effects, and a numerical absorption correction based
on Gaussian integration over a multifaceted crystal model was applied.
Data collection and refinement parameters are presented in the
Supporting Information. The structure was solved by direct methods
using SHELXT,64 which revealed the positions of the non-hydrogen
atoms of the title compound. All non-hydrogen atoms were refined
anisotropically. All H atoms were placed in geometrically calculated
positions and refined using a riding model where each H atom was
assigned a fixed isotropic displacement parameter with a value equal
to 1.2 Uequiv of its parent atom (1.5 Uequiv for methyl groups).
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank the European Research Council (ERC 615653) for
generous financial support of our work in this area and the
group at Chemical Crystallography of the University of Bern
for X-ray analyses.
REFERENCES
■
(1) Bauer, I.; Knölker, H. J. Iron Catalysis in Organic Synthesis.
Chem. Rev. 2015, 115, 3170−3387.
(2) Bolm, C.; Legros, J.; Le Paih, J.; Zani, L. Iron-Catalyzed
Reactions in Organic Synthesis. Chem. Rev. 2004, 104, 6217−6254.
(3) Morris, R. H. Asymmetric Hydrogenation, Transfer Hydro-
genation and Hydrosilylation of Ketones Catalyzed by Iron
Complexes. Chem. Soc. Rev. 2009, 38, 2282−2291.
(4) Zhang, M.; Zhang, A. Iron-Catalyzed Hydrosilylation Reactions.
Appl. Organomet. Chem. 2010, 24, 751−757.
Refinement of the structure was carried out on F2 using full-matrix
2
(5) Du, X.; Huang, Z. Advances in Base-Metal-Catalyzed Alkene
Hydrosilylation. ACS Catal. 2017, 7, 1227−1243.
least-squares procedures, which minimized the function Σw(Fo
−
Fc2)2. The weighting scheme was based on counting statistics and
included a factor to downweigh the intense reflections. All
calculations were performed using the SHELXL-2014/765 program
in OLEX2.66 Further crystallographic details are compiled in Tables
S2−S6. Crystallographic data for the structures of all compounds
reported in this paper have been deposited with the Cambridge
Crystallographic Data Centre (CCDC) as supplementary publication
numbers 2072827 (3c), 2072824 (3e), 2072829 (4a), 2072831 (4b),
2072826 (5c), 2072828 (5e), 2072825 (6a), 2072832 (6b), and
2072830 (7c).
(6) Lopes, R.; Royo, B. Iron N-Heterocyclic Carbenes in Reduction
Reactions. Isr. J. Chem. 2017, 57, 1151−1159.
(7) Wei, D.; Darcel, C. Iron Catalysis in Reduction and
Hydrometalation Reactions. Chem. Rev. 2019, 119, 2550−2610.
́
(8) Raya-Barón, A.; Ona-Burgos, P.; Fernández, I. Iron-Catalyzed
̃
Homogeneous Hydrosilylation of Ketones and Aldehydes: Advances
and Mechanistic Perspective. ACS Catal. 2019, 9, 5400−5417.
(9) Yun, J.; Kim, D.; Yun, H. A New Alternative to Stryker’s Reagent
in Hydrosilylation: Synthesis, Structure, and Reactivity of a Well-
Defined Carbene-Copper(Ii) Acetate Complex. Chem. Commun.
2005, 942, 5181.
ASSOCIATED CONTENT
■
(10) Lopes, R.; Pereira, M. M.; Royo, B. Selective Reduction of
Nitroarenes with Silanes Catalyzed by Nickel N-Heterocyclic Carbene
Complexes. ChemCatChem 2017, 9, 3073−3077.
(11) Hofmann, R. J.; Vlatkovic, M.; Wiesbrock, F. Fifty Years of
Hydrosilylation in Polymer Science: A Review of Current Trends of
Low-Cost Transition-Metal and Metal-Free Catalysts, Non-Thermally
Triggered Hydrosilylation Reactions, and Industrial Applications.
Polymers 2017, 9, 534.
(12) Bart, S. C.; Lobkovsky, E.; Chirik, P. J. Preparation and
Molecular and Electronic Structures of Iron(0) Dinitrogen and Silane
Complexes and Their Application to Catalytic Hydrogenation and
Hydrosilation. J. Am. Chem. Soc. 2004, 126, 13794−13807.
(13) Wu, J. Y.; Stanzl, B. N.; Ritter, T. A Strategy for the Synthesis of
Well-Defined Iron Catalysts and Application to Regioselective Diene
Hydrosilylation. J. Am. Chem. Soc. 2010, 132, 13214−13216.
(14) Sanagawa, A.; Nagashima, H. Cobalt(0) and Iron(0)
Isocyanides as Catalysts for Alkene Hydrosilylation with Hydro-
siloxanes. Organometallics 2018, 37, 2859−2871.
sı
* Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acs.organomet.1c00200.
Experimental procedures for ligands, NMR spectra, free
energy calculations, cyclic voltammetry, crystallographic
details, buried volume calculations, and catalytic and
mechanistic details (PDF)
Accession Codes
CCDC 2072824−2072832 contain the supplementary crys-
tallographic data for this paper. These data can be obtained
free of charge via www.ccdc.cam.ac.uk/data_request/cif, or by
emailing data_request@ccdc.cam.ac.uk, or by contacting The
Cambridge Crystallographic Data Centre, 12 Union Road,
Cambridge CB2 1EZ, UK; fax: +44 1223 336033.
(15) Hu, M. Y.; He, Q.; Fan, S. J.; Wang, Z. C.; Liu, L. Y.; Mu, Y. J.;
Peng, Q.; Zhu, S. F. Ligands with 1,10-Phenanthroline Scaffold for
Highly Regioselective Iron-Catalyzed Alkene Hydrosilylation. Nat.
Commun. 2018, 9, 1−12.
(16) Zhang, M.; Zhang, J.; Ni, X.; Shen, Z. Bis(Phenolate) N-
Heterocyclic Carbene Rare Earth Metal Complexes: Synthesis,
Characterization and Applications in the Polymerization of n-Hexyl
Isocyanate. RSC Adv. 2015, 5, 83295−83303.
AUTHOR INFORMATION
■
Corresponding Author
Martin Albrecht − Department of Chemistry & Biochemistry,
University of Bern, 3012 Bern, Switzerland; orcid.org/
0000-0001-7403-2329; Email: martin.albrecht@
dcb.unibe.ch
1548
https://doi.org/10.1021/acs.organomet.1c00200
Organometallics 2021, 40, 1538−1550

Organometallics
pubs.acs.org/Organometallics
Article
(17) Kandepi, V. V. K. M.; Cardoso, J. M. S.; Peris, E.; Royo, B.
Iron(II) Complexes Bearing Chelating Cyclopentadienyl-N-Hetero-
cyclic Carbene Ligands as Catalysts for Hydrosilylation and Hydrogen
Transfer Reactions. Organometallics 2010, 29, 2777−2782.
(18) Jiang, F.; Bézier, D.; Sortais, J. B.; Darcel, C. N-Heterocyclic
Carbene Piano-Stool Iron Complexes as Efficient Catalysts for
Hydrosilylation of Carbonyl Derivatives. Adv. Synth. Catal. 2011,
353, 239−244.
Precursors to N-Heterocyclic Carbene Complexes of Rh, Ru, Ir, and
Pd. J. Am. Chem. Soc. 2005, 127, 17624−17625.
(36) Wyer, E.; Gucciardo, G.; Leigh, V.; Muller-Bunz, H.; Albrecht,
̈
M. Comparison of Carbene and Imidazole Bonding to a Copper(I)
Center. J. Organomet. Chem. 2011, 696, 2882−2885.
(37) Albrecht, M.; Maji, P.; Häusl, C.; Monney, A.; Muller-Bunz, H.
̈
N-Heterocyclic Carbene Bonding to Cobalt Porphyrin Complexes.
Inorg. Chim. Acta 2012, 380, 90−95.
(19) Bézier, D.; Jiang, F.; Roisnel, T.; Sortais, J. B.; Darcel, C.
Cyclopentadienyl-NHC Iron Complexes for Solvent-Free Catalytic
Hydrosilylation of Aldehydes and Ketones. Eur. J. Inorg. Chem. 2012,
2012, 1333−1337.
(38) Ségaud, N.; McMaster, J.; Van Koten, G.; Albrecht, M.
Imidazolylidene Cu(II) Complexes: Synthesis Using Imidazolium
Carboxylate Precursors and Structure Rearrangement Pathways. Inorg.
Chem. 2019, 58, 16047−16058.
̌
(20) Demir, S.; Gökcȩ , Y.; Kaloglu, N.; Sortais, J. B.; Darcel, C.;
(39) Shapovalov, S. S.; Tikhonova, O. G.; Skabitskii, I. V.; Kolos, A.
V.; Sakharov, S. G.; Torubaev, Y. V. Oxidation of Iron Complex with
NHC Ligand with Molecular Iodine. Russ. J. Inorg. Chem. 2019, 64,
1418−1423.
(40) Shapovalov, S. S.; Tikhonova, O. G.; Grigor’eva, M. O.;
Skabitskii, I. V.; Simonenko, N. P. Metal Complexes with the N-
Heterocyclic Ligand: Synthesis, Structures, and Thermal Decom-
position. Russ. J. Coord. Chem. 2019, 45, 706−711.
(41) Naumann, S.; Schmidt, F. G.; Schowner, R.; Frey, W.;
Buchmeiser, M. R. Polymerization of Methyl Methacrylate by Latent
Pre-Catalysts Based on CO2-Protected N-Heterocyclic Carbenes.
Polym. Chem. 2013, 4, 2731.
̈
Ozdemir, I. Synthesis of New Iron-NHC Complexes as Catalysts for
Hydrosilylation Reactions. Appl. Organomet. Chem. 2013, 27, 459−
464.
(21) Lopes, R.; Cardoso, J. M. S.; Postigo, L.; Royo, B. Reduction of
Ketones with Silanes Catalysed by a Cyclopentadienyl- Functionalised
N-Heterocyclic Iron Complex. Catal. Lett. 2013, 143, 1061−1066.
(22) Buchgraber, P.; Toupet, L.; Guerchais, V. Syntheses, Properties,
and X-Ray Crystal Structures of Piano-Stool Iron Complexes Bearing
an N-Heterocyclic Carbene Ligand. Organometallics 2003, 22, 5144−
5147.
(23) Mercs, L.; Labat, G.; Neels, A.; Ehlers, A.; Albrecht, M. Piano-
Stool Iron(II) Complexes as Probes for the Bonding of N-
Heterocyclic Carbenes: Indications for π-Acceptor Ability. Organo-
metallics 2006, 25, 5648−5656.
(24) Guisado-Barrios, G.; Bouffard, J.; Donnadieu, B.; Bertrand, G.
Crystalline 1H-l,2,3-Triazol-5-Ylidenes: New Stable Mesoionic
Carbenes (MICs). Angew. Chem., Int. Ed. 2010, 49, 4759−4762.
(25) Bouffard, J.; Keitz, B. K.; Tonner, R.; Guisado-Barrios, G.;
Frenking, G.; Grubbs, R. H.; Bertrand, G. Synthesis of Highly Stable
1,3-Diaryl-1 H −1,2,3-Triazol-5-Ylidenes and Their Applications in
Ruthenium-Catalyzed Olefin Metathesis. Organometallics 2011, 30,
2617−2627.
(26) Trnka, T. M.; Morgan, J. P.; Sanford, M. S.; Wilhelm, T. E.;
Scholl, M.; Choi, T. L.; Ding, S.; Day, M. W.; Grubbs, R. H. Synthesis
and Activity of Ruthenium Alkylidene Complexes Coordinated with
Phosphine and N-Heterocyclic Carbene Ligands. J. Am. Chem. Soc.
2003, 125, 2546−2558.
(27) Johnson, C.; Albrecht, M. Triazolylidene Iron(II) Piano-Stool
Complexes: Synthesis and Catalytic Hydrosilylation of Carbonyl
Compounds. Organometallics 2017, 36, 2902−2913.
(28) Ohki, Y.; Hatanaka, T.; Tatsumi, K. J. Am. Chem. Soc. 2008, 130
(7), 17174−17186.
(29) Liang, Q.; Osten, K. M.; Song, D. Iron-Catalyzed Gem-Specific
Dimerization of Terminal Alkynes. Angew. Chem., Int. Ed. 2017, 56,
6317−6320.
(30) Liang, Q.; Sheng, K.; Salmon, A.; Zhou, V. Y.; Song, D. Active
Iron(II) Catalysts toward Gem-Specific Dimerization of Terminal
Alkynes. ACS Catal. 2019, 9, 810−818.
(42) Akita, M.; Terada, M.; Tanaka, M.; Morooka, Y. Some
Additional Aspects of Versatile Starting Compounds for Cationic
Organoiron Complexes: Molecular Structure of the Aqua Complex
[(H5-C5Me4Et)Fe(CO)2(OH2)]BF4 and Solution Behavior of the
THF Complex [(H5HC5R5)Fe(CO)2(THF)]BF4. J. Organomet.
Chem. 1996, 510, 255−261.
(43) Gunther, H. NMR-Spektroskopie; Georg Thieme Verlag
̈
Stuttgart, 1973.
(44) Mathew, P.; Neels, A.; Albrecht, M. 1,2,3-Triazolylidenes as
Versatile Abnormal Carbene Ligands for Late Transition Metals. J.
Am. Chem. Soc. 2008, 130 (130), 13534−13535.
(45) Falivene, L.; Cao, Z.; Petta, A.; Serra, L.; Poater, A.; Oliva, R.;
Scarano, V.; Cavallo, L. Towards the Online Computer-Aided Design
of Catalytic Pockets. Nat. Chem. 2019, 11, 872−879.
(46) Gutsulyak, D. V.; Kuzmina, L. G.; Howard, J. A. K.;
Vyboishchikov, S. F.; Nikonov, G. I. Cp(Pri2MeP)FeH2SiR3:
Nonclassical Iron Silyl Dihydride. J. Am. Chem. Soc. 2008, 130,
3732−3733.
(47) Lam, Y. C.; Nielsen, R. J.; Goddard, W. A.; Dash, A. K. The
Mechanism for Catalytic Hydrosilylation by Bis(Imino)Pyridine Iron
Olefin Complexes Supported by Broken Symmetry Density Func-
tional Theory. Dalt. Trans. 2017, 46, 12507−12515.
(48) Smith, P. W.; Dong, Y.; Tilley, T. D. Efficient and Selective
Alkene Hydrosilation Promoted by Weak, Double Si-H Activation at
an Iron Center. Chem. Sci. 2020, 11, 7070−7075.
(49) Bleith, T.; Gade, L. H. Mechanism of the Iron(II)-Catalyzed
Hydrosilylation of Ketones: Activation of Iron Carboxylate
Precatalysts and Reaction Pathways of the Active Catalyst. J. Am.
Chem. Soc. 2016, 138, 4972−4983.
(31) Liang, Q.; Hayashi, K.; Rabeda, K.; Jimenez-Santiago, J. L.;
Song, D. Piano-Stool Iron Complexes as Precatalysts for Gem-Specific
Dimerization of Terminal Alkynes. Organometallics 2020, 39, 2320−
2326.
(50) Metsänen, T. T.; Gallego, D.; Szilvási, T.; Driess, M.; Oestreich,
M. Peripheral Mechanism of a Carbonyl Hydrosilylation Catalysed by
an SiNSi Iron Pincer Complex. Chem. Sci. 2015, 6, 7143−7149.
(51) O'Connor, J. M.; Casey, C. P. Ring-Slippage Chemistry of
Transition-Metal Cyclopentadienyl and Indenyi Complexes. Chem.
Rev. 1987, 87, 307−318.
(32) Riener, K.; Haslinger, S.; Raba, A.; Högerl, M. P.; Cokoja, M.;
Herrmann, W. A.; Kuhn, F. E. Chemistry of Iron N -Heterocyclic
̈
Carbene Complexes: Syntheses, Structures, Reactivities, and Catalytic
Applications. Chem. Rev. 2014, 114, 5215−5272.
(33) Royo, B. Cyclopentadienyl-Functionalized n-Heterocyclic
Carbene Complexes of Iron and Nickel: Catalysts for Reductions.
In AdvancesAdvances in Organometallic Chemistry and Catalysis;
Pombiero Armando, J. L., Ed.; John Wiley & Sons Inc., 2014;
DOI: 10.1016/s1351-4180(14)70389-7.
(52) No product was observed by 1H NMR, nor was any
precipitation observed. No precipitation was observed in the NMR
tube, nor were any new signals observed, indicative of a product
formation.
(53) Hamon, J.-R.; Hamon, P.; Toupet, L.; Costuas, K.; Saillard, J.-
Y. Classical and Non-Classical Iron Hydrides: Synthesis, NMR
Characterisation, Theoretical Investigation and X-Ray Crystal
Structure of the Iron(IV) Dihydride [Cp*Fe(Dppe)(H)2] +BF4-.
C. R. Chim. 2002, 5, 89−98.
(34) Johnson, C.; Albrecht, M. Piano-Stool N-Heterocyclic Carbene
Iron Complexes: Synthesis, Reactivity and Catalytic Applications.
Coord. Chem. Rev. 2017, 352, 1−14.
(35) Voutchkova, A. M.; Appelhans, L. N.; Chianese, A. R.;
Crabtree, R. H. Disubstituted Imidazolium-2-Carboxylates as Efficient
1549
https://doi.org/10.1021/acs.organomet.1c00200
Organometallics 2021, 40, 1538−1550

Organometallics
pubs.acs.org/Organometallics
Article
(54) Li, H.; Misal Castro, L. C.; Zheng, J.; Roisnel, T.; Dorcet, V.;
Sortais, J. B.; Darcel, C. Selective Reduction of Esters to Aldehydes
under the Catalysis of Well-Defined NHC-Iron Complexes. Angew.
Chem., Int. Ed. 2013, 52, 8045−8049.
(55) Becker, B.; Corriu, R. J. P.; Guérin, C.; Henner, B. J. L.
Hypervalent Silicon Hydrides: Evidence for Their Intermediacy in the
Exchange Reactions of Di- and Tri-Hydrogenosilanes Catalysed by
Hydrides (NaH, KH and LiAlH4). J. Organomet. Chem. 1989, 369,
147−154.
(56) Docherty, J. H.; Dominey, A. P.; Thomas, S. P. Nucleophile
Induced Ligand Rearrangement Reactions of Alkoxy- and Arylsilanes.
Tetrahedron 2019, 75, 3330−3335.
(57) Bellamy, L. J. The Infra-Red Spectra of Complex Molecules, 3rd
ed.; Chapman and Hall Ltd.: London, 1975.
1
(58) Analysis of this reaction mixture by H NMR spectroscopy
revealed the same changes in chemical shifts as observed by 1H NMR
spectroscopic reaction monitoring (Figure S74 cf. Figure S63).
(59) Warner, M. C.; Verho, O.; Bäckvall, J. E. CO Dissociation
Mechanism in Racemization of Alcohols by a Cyclopentadienyl
Ruthenium Dicarbonyl Catalyst. J. Am. Chem. Soc. 2011, 133, 2820−
2823.
(60) Stewart, B.; Nyhlen, J.; Martín-Matute, B.; Bäckvall, J. E.;
Privalov, T. A Computational Study of the CO Dissociation in
Cyclopentadienyl Ruthenium Complexes Relevant to the Racemiza-
tion of Alcohols. Dalt. Trans. 2013, 42, 927−934.
(61) Connelly, N. G.; Geiger, W. E. Chemical Redox Agents for
Organometallic Chemistry. Chem. Rev. 1996, 96, 877−910.
(62) Oxford Diffraction Ltd. CrysAlisPro (Version 1.171.40.39a);
Oxford Diffraction Ltd., Yarnton, Oxfordshire, UK, 2018; pp 279−
288.
(63) Macchi, P.; Bu, H.; Chimpri, A. S.; et al. Low-Energy
Contamination of Mo Microsource X-Ray Radiation: Analysis and
Solution of the Problem. J. Appl. Crystallogr. 2011, 44, 763−771.
(64) Sheldrick, G. M. Research Papers SHELXT - Integrated Space-
Group and Crystal- Structure Determination Research Papers. Acta
Crystallogr., Sect. A: Found. Adv. 2015, 71, 3−8.
(65) Sheldrick, G. M. Crystal Structure Refinement with SHELXL.
Acta Crystallogr., Sect. C: Struct. Chem. 2015, 71, 3−8.
(66) Dolomanov, O. V.; Bourhis, L. J.; Gildea, R. J.; Howard, J. A.
K.; Puschmann, H. OLEX2: A Complete Structure Solution,
Refinement and Analysis Program. J. Appl. Crystallogr. 2009, 42,
339−341.
1550
https://doi.org/10.1021/acs.organomet.1c00200
Organometallics 2021, 40, 1538−1550