Synthesis and electrochemical study of iron, chromium and tungsten aminocarbenes: Role of ligand structure and central metal nature

Article abstract of DOI:10.1016/j.electacta.2010.02.057

Several series of Fischer-type aminocarbene complexes with central Fe, Cr or W atoms and with various carbene substitution were synthesized and electrochemically investigated by dc-polarography and cyclic voltammetry. The shifts and changes of reduction a

Full text of DOI:10.1016/j.electacta.2010.02.057

Electrochimica Acta 55 (2010) 8341–8351
Contents lists available at ScienceDirect
Electrochimica Acta
journal homepage: www.elsevier.com/locate/electacta
Synthesis and electrochemical study of iron, chromium and tungsten
aminocarbenes: Role of ligand structure and central metal nature
Irena Hoskovcováa,1, Jana Rohácˇováb, Dalimil Dvorˇákb, Tomásˇ Tobrmanb, Stanislav Zálisˇ c,
Radka Zveˇrˇinováa,c, Jirˇí Ludvíkc,∗,1
a Department of Inorganic Chemistry, Institute of Chemical Technology, Technická 5, 166 28 Prague 6, Czech Republic
b Department of Organic Chemistry, Institute of Chemical Technology, Technická 5, 166 28 Prague 6, Czech Republic
c J. Heyrovsky´ Institute of Physical Chemistry, Academy of Science of the Czech Republic, Dolejsˇkova 3, 182 23 Prague 8, Czech Republic
a r t i c l e i n f o
a b s t r a c t
Article history:
Several series of Fischer-type aminocarbene complexes with central Fe, Cr or W atoms and with vari-
ous carbene substitution were synthesized and electrochemically investigated by dc-polarography and
cyclic voltammetry. The shifts and changes of reduction and oxidation potentials were evaluated using
the linear free energy relationship (LFER) approach with respect to (a) the type of coordination, (b) the
substitution on the carbene ligand and (c) the nature of the central metal atom. The analysis of measured
data confirms that the reduction center is localized on the carbene moiety and is strongly influenced
by both electronic and sterical properties of its substituents. The oxidation proceeds on the metal and
depends mainly on its nature and on the ␲-acidity of the ligands. Electrochemistry thus represents an
important experimental approach to the description and understanding of the molecular electronic struc-
ture and redox properties. Experimental results are supported by DFT calculation of HOMO and LUMO
orbitals shape and composition.
Received 30 November 2009
Received in revised form 14 February 2010
Accepted 18 February 2010
Available online 26 February 2010
Keywords:
Fischer aminocarbene complexes
Voltammetry–polarography
HOMO–LUMO displacement
LFER
Structurally dependent redox properties
© 2010 Elsevier Ltd. All rights reserved.
1. Introduction
between structure and chemical properties. This investigation is
the main goal of the presented contribution.
Correlation between composition of metal coordination sphere
and redox behaviour of the complex has been studied for a long
time. For organometallic compounds, a theory was built [1] in
which ligand constants PL are used for characterizing influence
of certain ligand type on electrochemical properties of the whole
molecule. In addition to this, the oxidation and reduction potentials,
their sequences and differences can be influenced also by substi-
tution on the organic ligands themselves. Such an influence can be
expressed quantitatively and correlated with electrochemical data
by means of various ꢀ-constants of Hammett type.
Fischer aminocarbene complexes are widely used in organic
synthesis, both as reactants and catalysts, where the substitution
on the organic carbene ligand plays an important role. Therefore,
various new structural types are being synthesized and tested [2].
Recently, a possibility of using Fischer carbenes as electrochemical
probes for biosensors has been shown [3].
In our previous paper [4] we reported electrochemical prop-
erties of the first series of five chromium non-chelate and four
chelate complexes of type I and II, respectively. For comparison,
one tungsten non-chelate complex III and one chromium chelate
complex VII bearing on the carbene carbon a methyl group instead
of the substituted phenyl were synthesized and examined (Fig. 1).
Using the linear free energy relationship (LFER) approach [5] we
demonstrated experimentally that the reduction takes place on the
carbene carbon atom and the reduction potential is sensitive to its
substitution. A remarkable influence of sterical hindrance for the
type II was observed and explained in agreement with NMR study
of the hindered rotation of the aromatic ring attached to the car-
bene carbon atom [6]. Based on analysis of experimental data it was
also concluded that the oxidation center of the molecule should be
located prevalently on the central metal atom with strong sensitiv-
ity to the nature of the metal and the number of CO ligands bonded
to it but with a negligible influence of the carbene substitution.
In the present study, we turned our attention to the Fischer
aminocarbenes with iron as a central metal atom and to their com-
parison with previously studied chromium complexes. Due to its
different electron structure and coordination number five, some
differences in behaviour of iron compounds can be expected. Two
new analogous series of five iron non-chelate (V) and five chelate
Aimed design of new carbene systems requires deeper under-
standing of their redox abilities and knowledge of the relationship
∗
Corresponding author. Tel.: +420 266053217.
E-mail addresses: jiri.ludvik@jh-inst.cas.cz, ludvik@jh-inst.cas.cz (J. Ludvík).
ISE members.
1
0013-4686/$ – see front matter © 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.electacta.2010.02.057

8342
I. Hoskovcová et al. / Electrochimica Acta 55 (2010) 8341–8351
(VI) complexes were synthesized and examined in order to (1) char-
acterize electrochemically the new compounds; (2) compare the
electronic and redox properties of chromium and iron analogues;
(3) compare the electronic and redox properties of chelated and
non-chelated compounds; (4) correlate the conclusions derived
from the analysis of experimental data with quantum chemical
calculations.
In addition to this, in the frame of this contribution sev-
eral other aminocarbene complexes were synthesized in order to
extend and complete the already existing series. They are: tungsten
chelate IV (analogous to the already synthesized non-chelate III),
chromium non-chelates IXa, c, e with a methyl in ortho-position,
two missing compounds in series II and VII – chromium non-
chelates IIb and VIIc and a chelate analogue to VIIb – compound
VIIIb. Such variety of compounds gives good possibility to refine
the previous conclusions and, eventually, formulate more general
rules. The overview of investigated aminocarbenes is presented in
Table 1.
Merck. N,N-Dimethylbenzamide, N,N-dimethylformamide and
N,N-dimethylacetamide were obtained from Aldrich, 4-substituted
N,N-dimethylbenzamides, N,N-dimethyl-2-methylbenzamide and
N,N-dimethyl-2-methyl-4-methoxybenzamide were obtained by
the reaction of the corresponding acylchloride with excess of
N,N-dimethylamine in diethylether. N,N-Diallylbenzamide [12] and
4-substituted N,N-diallylbenzamides [6] were prepared according
to the reported procedures. Melting points were determined on a
Kofler block and are uncorrected. 1H and 13C NMR spectra were
recorded on a Varian Gemini 300 spectrometer (1H at 300 MHz,
13C at 75.4) in CDCl3.
2.3.1. N,N-Dimethyl-2-methyl-4-(trifluoromethyl)benzamide
2-Methyl-4-(trifluoromethyl)benzoic acid was prepared using
the methodology for the direct ortho-lithiation of unprotected ben-
zoic acids [13]. Reaction started from 4-(trifluoromethyl)benzoic
acid (0.32 g, 2 mmol), sec-butyllithium (4.4 mmol), TMEDA (0.66 ml,
4.4 mmol) and iodomethane (0.5 ml, 8 mmol) afforded 280 mg
(69%) of crude 2-methyl-4-(trifluoromethyl)benzoic acid as a white
solid which was recrystallized from heptane.
2. Experimental
1H NMR (CDCl3): 2.71 (s, 3H, CH3), 7.55 (br s, 2H, Ar–H), 8.14 (d,
J = 8.79 Hz, 1H, Ar–H).
2.1. Electrochemistry
To
a solution of 2-methyl-4-(trifluoromethyl)benzoic acid
All electrochemical measurements were carried out in a 10 ml
two-compartment cell of Metrohm type, in non-aqueous dimethyl-
formamide (DMF) purified by azeotropic distillation with 5% water
and 5% benzene followed by vacuum fractionation [4]. The solu-
tion was deaerated by a stream of argon. Tetrabutylammonium
hexafluorophosphate (Bu4N PF6) p.a. Fluka recrystallized from
aqueous ethanol was used as supporting electrolyte in concentra-
tion 0.1 mol/l. The sample solution was prepared always directly
in the cell by dissolving the known amount of solid compound;
sample concentration was about 3 × 10−4 mol/l.
(0.091 g, 0.45 mmol) and DCC (0.103 g, 0.5 mmol) in 5 ml of THF
1 M solution of dimethylamine in Et2O (1.4 ml, 1.4 mmol) of was
added. Resulting mixture was stirred 17 h at ambient temper-
ature, then acetic acid (2 ml) was added and the mixture was
stirred 30 min at ambient temperature. Evaporation of the sol-
vents followed by flash chromatography on silica in hexane–ethyl
acetate (1:2) gave 80 mg (81%) of desired N,N-dimethyl-2-methyl-
4-(trifluoromethyl)benzamide as yellow oil.
1H NMR: 2.35 (s, 3H, CH3), 2.83 (s, 3H, NCH3), 3.15 (s, 3H, NCH3),
7.30 (d, J = 8.79 Hz, 1H, Ar–H), 7.48 (br s, 2H, Ar–H).
For electrochemical experiments a computer-controlled Eco-
Tribo polarograph (PC-ETP, ECO-TREND PLUS Prague, Czech
Republic) or a PA3 potentiostat (Laboratorní prˇístroje) was used
with a std. Ag/Ag+ or SCE reference electrode (all data are converted
vs. SCE reference electrode) and a platinum sheet as an auxiliary
electrode. Reduction was studied by dc-polarography on dropping
mercury electrode (DME). For cyclic voltammetry (CV), a freshly
polished stationary platinum disk (diameter 0.5 mm) electrode was
used for both oxidation and reduction, scan rate ranging from
100 to 800 mV/s. The accuracy of the potential measurement was
0.005 V. Under given condition, it was possible to determine both
Ered and Eox for all the substances: Ered = E1/2 from dc-polarography,
Eox(red) = 1/2(Epa + Epc) from CV measurement.
2.4. Preparation of aminocarbene complexes
Chromium, tungsten and iron aminocarbene complexes Ib,d [4],
Ic [14], IIa,b,d,e [6], IIc [15], Vc [16], VIc [17], VIIb,c [18] and VIIIb
[19] were prepared according to the reported procedures.
2.4.1. General procedure for the preparation of chromium and
tungsten aminocarbene complexes Ia,e, IIIc, IVc and IX [14]
The solution of sodium naphthalenide prepared from sodium
(0.43 g, 18.5 mmol), naphthalene (2.37 g, 18.5 mmol) in THF (50 ml)
was slowly added under argon atmosphere to the stirred suspen-
sion of Cr(CO)6 (1.54 g, 7 mmol) or W(CO)6 (2.46 g, 7 mmol) in THF
(40 ml) at −78 ◦C. The mixture was allowed to warm to 0 ◦C and kept
at this temperature until all solid carbonyl dissolved (0.5 h). After
cooling to −78 ◦C a solution of the corresponding amide (7 mmol)
in THF (10 ml) was added through a double-ended needle. The
mixture was stirred at −78 ◦C for 30 min and then 30 min at 0 ◦C.
After cooling to −78 ◦C, Me3SiCl (1.8 ml; 14 mmol) was added via
a syringe. The solution was stirred at −78 ◦C for 30 min, then the
cooling bath was removed, the mixture was allowed to warm to
0 ◦C and neutral alumina (8 g) was added. The solvent was removed
under reduced pressure on a rotatory evaporator (bath temperature
<30 ◦C) and the residue was dried for several hours under high vac-
uum to remove all the THF. Hexane (50 ml) was then added and the
mixture was stirred vigorously for several minutes under argon
atmosphere. The formed suspension was then transferred on top
of a column filled with 50 g of silica. Naphthalene was eluted with
pure hexane and further elution with a hexane–CH2Cl2 mixture
(5:1–2:1) gave the product.
2.2. DFT calculations
Ground state electronic structure calculations on carbene Cr and
Fe complexes Ic, IIc, Vc, VIc and VIIc have been done by density
functional theory (DFT) methods using Gaussian 03 [7] program
package. B3LYP hybrid functional [8–10] was used together with
6–31G* polarized double-␨ basis sets [11] for H, C, Fe and Cr atoms.
Geometry of all complexes was optimized without any symme-
try constrains, vibrational analysis was used for characterization of
energy minima. MO analysis was done at optimized structures, MO
plots were generated by GaussView software.
2.3. Chemicals for carbene complexes
All experiments were carried out under argon. Tetrahydrofu-
ran was distilled from benzophenone ketyl under argon prior to
use. Chromium hexacarbonyl, tungsten hexacarbonyl, iron pen-
tacarbonyl and Me3SiCl were purchased from Aldrich and were
used without purification. Silica and alumina were obtained from

I. Hoskovcová et al. / Electrochimica Acta 55 (2010) 8341–8351
8343
Table 1
Overview of the studied compounds and Hammett’s ꢁ-values of individual series.
Structure
R
Number
Synthesis [lit.]
Ered slope analysis
ꢁ = 0.344, R2 = 0.957
Eox slope analysis
–OCH3
–CH3
–H
–Cl
–CF3
Ia
Ib
Ic
Id
Ie
This work
[4]
[14]
[4]
This work
ꢁ = 0.082, R2 = 0.908
–OCH3
–CH3
–H
–Cl
–CF3
IIa
IIb
IIc
IId
IIe
[6]
[6]
[17]
[6]
[6]
ꢁ = 0.224, R2 = 0.925
ꢁ = 0.070, R2 = 0.981
–H
IIIc
This work
–
–
–H
IVc
This work
–
–
–OCH3
–CH3
–H
–Cl
–CF3
Va
Vb
Vc
Vd
Ve
This work
This work
[16]
This work
This work
ꢁ = 0.334,
R
ꢁ = 0.069,
2 = 0.976
R
2 = 0.968
–OCH3
–CH3
–H
–Cl
–CF3
VIa
VIb
VIc
VId
VIe
This work
This work
[17]
This work
This work
ꢁ = 0.255,
2 = 0.857
ꢁ = 0.044,
R
2 = 0.708
R
–CH3
–H
–C6H5
VIIb
VIIc
VIIf = Ic
[14]
This work
[14]
ꢁ = 1.990,
R
ꢁ = 0.332,
R
2 = 0.882
2 = 0.903
–CH3
–C6H5
VIIIb
VIIIf = IIc
[19]
[17]
–
–

8344
I. Hoskovcová et al. / Electrochimica Acta 55 (2010) 8341–8351
Table 1 (Continued)
Structure
R
Number
Synthesis [lit.]
Ered slope analysis
ꢁ = 0.304,
Eox slope analysis
ꢁ = 0.092,
–OCH3
–H
–CF3
IXa
IXc
IXe
This work
This work
This work
R
2 = 0.958
R
2 = 0.999
Data correlated to ꢀi (see the text, Section 3.1).
2.4.2. Pentacarbonyl[(N,N-dimethylamino)
(4-methoxyphenyl)methylene]chromium(0) (Ia)
Elution with hexane–CH2Cl2 mixture (3:1) afforded Ia in 81%
yield as yellow crystals. M.p. 44–45 ◦C, Ref. [15]: 47 ◦C. 1H NMR: ı
3.04 (s, 3H; anti-NCH3), 3.81 (s, 3H; syn-NCH3), 3.98 (s, 3H; OCH3),
6.64 (d, J = 8.8 Hz, 2H; Ar–H), 6.91 (d, J = 8.5 Hz, 2H; Ar–H).
6.81 (brs, 2H, o-Ph–H), 7.20 (m, 1H, Ph–H), 7.37 (m, 2H, Ph–H);
13C NMR: ı 263.6 (W C, JWC = 85 Hz), 210.7 (JWC = 77 Hz, CO),
210.2 (JWC = 64 Hz, CO), 203.0 (JWC = 64 Hz, CO), 202.8 (JWC = 62 Hz,
CO), 149.2 (C–Ph), 131.2 (–CH ), 128.3 (CH–Ph), 126.8 (CH–Ph),
120.3 (CH–Ph), 119.9( CH2), 70.1 (chelated –CH ), 62.8 (chelated
NCH2), 59.1 (chelated CH2), 56.1 (NCH2); IR (CHCl3) ꢃ 2024 (s),
1926 (vs), 1904 (vs), 1516 (w) cm−1. Anal. Calcd for C17H15NO4W:
C, 42.44; H, 3.14; N, 2.91. Found: C, 42.36; H, 3.09; N, 2.78.
2.4.3. Pentacarbonyl[(N,N-dimethylamino)
(4-trifluoromethylphenyl)methylene]chromium(0) (Ie)
Mixture of hexane–CH2Cl2 (3:1) was used for the elution. Yellow
crystals, yield 66%, M.p. 89–91 ◦C, Ref. [15]: 90 ◦C. 1H NMR: ı 3.02
(s, 3H; anti-NCH3), 3.98 (s, 3H; syn-NCH3), 6.78 (d, J = 8.0 Hz, 2H;
Ar–H), 7.61 (d, J = 8.0 Hz, 2H; Ar–H).
2.4.6. Pentacarbonyl[(N,N-dimethylamino)(2-methyl-4-
methoxyphenyl)methylene]chromium(0)(IXa)
General procedure starting from 0.56 g (3 mmol) of 2-
methyl-4-methoxy-N,N-dimethylbenzamide afforded elution with
hexane–CH2Cl2 mixture (5:1) IXa (0.34 g, 31%) in the form of yel-
low crystals. M.p. 79–82 ◦C; 1H NMR: ı 2.03 (s, 3H; ArCH3), 3.02
(s, 3H; anti-NCH3), 3.80 (s, 3H; syn-NCH3), 3.99 (s, 3H; OCH3), 6.64
(d, J = 8.5 Hz, 1H; Ar–H), 6.69 (s, 1H; Ar–H), 6.78 (d, J = 10.9 Hz, 1H;
Ar–H). 13C NMR: ı 277.7 (C Cr), 223.8 (CO), 217.3 (CO), 157.8
(C–Ar), 145.7 (C–Ar), 127.4 (C–Ar), 121.1 (CH–Ar), 115.9 (CH–Ar),
111.6 (CH–Ar), 55.7 (CH3–O), 51.1 (CH3–N), 45.1 (CH3–N), 19.1
2.4.4. Pentacarbonyl[(N,N-dimethylamino)
phenylmethylene]tungsten(0) (IIIc)
Elution with hexane–CH2Cl2 mixture (5:1) gave IIIc in 62% yield
as pale yellow crystals. M.p. 99 ◦C, Ref. [18]: 93–94 ◦C. 1H NMR: ı
3.05 (s, 3H; NCH3), 3.94 (s, 3H; NCH3), 6.76 (d, J = 8.5 Hz, 2H; Ph–H),
7.16 (t, J = 15.0 Hz, 1H; Ph–H), 7.39 (t, J = 15.5 Hz, 2H; Ph–H).
(CH3–Ar). IR (KBr): ꢃ 2052 (m), 1891 (s) cm−1
.
2.4.5. Tetracarbonyl[(ꢂ2-N-allyl-N-allylamino)
phenylmethylene]tungsten(0) (IVc)
2.4.7. Pentacarbonyl[(N,N-dimethylamino)(2-methylphenyl)
The above method starting from 1.01 g (5 mmol) of N,N-
diallylbenzamide afforded after elution with hexane–CH2Cl2
mixture (5:1) crystalline pentacarbonyl[(N,N-diallylamino)
phenylmethylene]tungsten(0) 0.9 g followed with desired
IVc (0.081 g). Pentacarbonyl[(N,N-diallylamino)phenylmethylene]
tungsten(0) was dissolved in toluene (10 ml) and the solution was
refluxed under argon for 4 h. Chromatography on silica (30 g) in
hexane–CH2Cl2 5:1 gave another IVc (0.8 g, 33%) as yellow oil.
methylene]chromium(0) (IXc)
The
above
procedure
starting
from
2-methyl-N,N-
dimethylbenzamide (0.82 g, 5 mmol) furnished IXc (0.93 g,
55%) as yellow crystals.
M.p. >91 ◦C (decomp.); 1H NMR: ı 2.07 (s, 3H, Ph−CH3), 3.04
(s, 3H, N−CH3), 4.01 (s, 3H, N−CH3), 6.71 (d, J = 7.7 Hz, 1H, Ph–H),
7.05−7.17 (m, 2H, Ph–H), 7.22 (t, J = 7.4 Hz, 1H, Ph–H). 13C NMR: ı
276.5 (C Cr), 223.2 (CO), 216.8 (CO), 151.6 (C−Ph), 130.6 (CH−Ph),
126.1 (CH−Ph), 126.0 (CH−Ph), 125.4 (C−Ph), 119.5 (CH−Ph), 51.0
(CH3−N), 45.2 (CH3−N), 18.8 (CH3−Ph).
2.4.5.1. Pentacarbonyl[(N,N-diallylamino)(phenyl)methylene]
tungsten(0). Yellow solid; M.p. 49–50 ◦C; 1H NMR: ı 3.97 (d,
J = 5.5 Hz, 2H, NCH2), 4.88 (d, J = 5.5 Hz, 2H, NCH2), 5.13 (dd, J = 1.1,
17.0 Hz, 1H, = CHH), 5.30 (dd, J = 1.1, 10.4 Hz, 1H, CHH), 5.41 (dd,
J = 1.1, 17.0 Hz, 1H, CHH), 5.49 (dd, J = 1.1, 10.4 Hz, 1H, CHH),
5.59 (m, 1H, CH–), 6.02 (m, 1H, CH–), 6.79 (m, 2H, Ph–H), 7.15
(m, 1H, Ph–H), 7.38 (m, 2H, Ph–H); 13C NMR: ı 259.9 (W C),
204.2 (CO), 198.2 (CO), 153.0 (C–Ph), 131.4 (–CH ), 131.0 (–CH ),
128.2 (CH–Ph), 126.3 (CH–Ph), 120.3 ( CH2), 119.8 ( CH2), 119.1
(CH–Ph), 64.1 (NCH2), 55.2 (NCH2); IR (CHCl3) ꢃ 2063 (w), 1980
(w), 1929 (vs) cm−1. Anal. Calcd for C18H15NO5W: C, 42.46; H,
2.97; N, 2.75. Found: C, 42.02; H, 2.80; N, 2.69.
IR (CHCl3): ꢃ 2055, 1972, 1930, 1533, 1400 cm−1. Anal. Calcd for
C15H13NO5Cr: C, 53.10; H, 3.86; N, 4.13. Found: C, 53.37; H, 3.94;
N, 4.11.
2.4.8. Pentacarbonyl[(N,N-dimethylamino)(2-methyl-4-
trifluoromethylphenyl)methylene] chromium(0) (IXe)
General procedure starting from 0.206 g (0.89 mmol) of
N,N-dimethyl-2-methyl-4-(trifluoromethyl) benzamide 0.206 g
(0.89 mmol) afforded after chromatography in hexane–CH2Cl2
mixture (5:1) IXe (0.07 g, 19%) as yellow solid.
1H NMR: 2.13 (s, 3H, CH3), 3.06 (s, 3H, NCH3), 4.04 (s, 3H, NCH3),
6.83 (d, J = 8.2 Hz, 1H, Ar–H), 7.44 (s, 1H, Ar–H), 7.51 (d, J = 8.8 Hz,
1H, Ar–H). 13C NMR: 18.6 (CH3), 45.5 (NCH3), 51.08 (NCH3)120.1
(CAr), 123.2 (CAr), 123.9 (q, J = 271 Hz, CF3), 126.6 (CAr), 127.8 (CAr),
127.9 (q, J = 32 Hz, C–CF3), 154.0 (CAr), 216.8 (CO), 223.1 (CO), 275.2
(C Cr).
2.4.5.2. Tetracarbonyl[(ꢂ2-N-allyl-N-allylamino)(phenyl)methylene]
tungsten(0) (IVc). Yellow oil; 1H NMR: ı 3.44 (d, J = 8.2 Hz, 1H,
chelated CHH), 3.46 (d, J = 13.2 Hz, 1H, chelated CHH), 3.90 (m,
1H, chelated NCH2 and 2H, NCH2), 4.56 (m, 1H,chelated CH–), 4.77
(dd, J = 4.9, 13.7 Hz, 1H, chelated NCH2), 5.20 (dd, J = 1.1, 17.6 Hz,
1H, CHH), 5.30 (dd, J = 1.1, 10.4 Hz, 1H, CHH), 5.67 (m, 1H, CH–),
For the preparation of iron aminocarbene complexes Va, Vb, Vd,
Ve and VIa, the same general procedure described for the prepara-

I. Hoskovcová et al. / Electrochimica Acta 55 (2010) 8341–8351
8345
2.4.14. Tricarbonyl[(ꢂ2-N-allyl-N-allylamino)
(4-methylphenyl)methylene]iron(0) (VIb)
General procedure starting from 0.86 g (4.0 mmol) of 4-methyl-
N,N-diallylbenzamide gave after elution with hexane–CH2Cl2
mixture (2:1) 0.44 g (32%) of VIb as a yellow oil.
tion of the chromium and tungsten aminocarbene complexes was
used. After deposition on alumina and evaporation of the solvents
the product was isolated by chromatography on neutral alumina
(100 g). Naphthalene was eluted with pure hexane and further elu-
tion with a hexane–CH2Cl2 mixture (5:1–2:1) gave the product.
1H NMR (300 MHz, CDCl3): ı 1.56 (d, J = 10.5 Hz, 1H, CH2),
2.04 (d, J = 7.8 Hz, 1H, CH2), 2.36 (s, 3H, CH3), 3.20 (m, 1H,
–CH ), 3.76 (dd, J = 6.0 Hz, J = 15.0 Hz, 1H, NCH2), 3.88 (dd, J = 6.0 Hz,
J = 15.0 Hz, 1H, NCH2), 4.17 (dd, J = 5.1 Hz, J = 14.4 Hz, 1H, NCH2), 4.35
(d, J = 14.4 Hz, 1H, NCH2), 5.17 (d, J = 17.1 Hz, 1H, CH2), 5.26 (d,
J = 10.5 Hz, 1H, CH2), 5.55 (m, 1H, –CH ), 6.78 (brs, 2H, ArH), 7.18
(d, J = 8.1 Hz, 2H, ArH).
2.4.9. Tetracarbonyl[(N,N-dimethylamino)(4-
methoxyphenyl)methylene]iron(0) (Va)
General procedure starting from 0.896 g (5 mmol) of 4-
methoxy-N,N-dimethylbenzamide gave after elution with
hexane–CH2Cl2 mixture (3:1) 0.33 g (20%) of Va as bright-yellow
crystals.
M.p. 80–85 ◦C; 1H NMR: ı 3.10 (s, 3H; anti-NCH3), 3.82 (s, 3H;
syn-NCH3), 3.99 (s, 3H; OCH3), 6.78 (d, J = 8.5 Hz, 2H; Ar–H), 6.90
(d, J = 8.8 Hz, 2H; Ar–H). 13C NMR: ı 260.4 (C Cr), 215.0 (CO), 158.3
(C–Ar), 146.5 (C–Ar), 122.0 (CH–Ar), 113.9 (CH–Ar), 55.4 (CH3–O),
50.5 (CH3–N), 45.9 (CH3–N). IR (CHCl3): ꢃ 2041 (m), 1966 (w), 1926
(s) cm−1. Anal. Calcd for C14H13FeNO5: C, 50.78; H, 3.96; N, 4.23.
Found: C, 50.69; H, 3.98; N, 4.25.
2.4.15. Tricarbonyl[(ꢂ2-N-allyl-N-allylamino)(4-chlorophenyl)
methylene]iron(0) (VId)
General procedure starting from 0.95 g (4.0 mmol) of 4-chloro-
N,N-diallylbenzamide gave after elution with hexane–CH2Cl2
mixture (4:1) 0.60 g (42%) of VIb as a yellow oil.
1H NMR (300 MHz, CDCl3): ı 1.48 (d, J = 10.5 Hz, 1H, CH2),
2.04 (d, J = 9.0 Hz, 1H, CH2), 3.20 (m, 1H, –CH ), 3.74 (dd, J = 6.0 Hz,
J = 15.0 Hz, 1H, NCH2), 3.84 (dd, J = 6.0 Hz, J = 15.0 Hz, 1H, NCH2), 4.16
(dd, J = 4.8 Hz, J = 14.4 Hz, 1H, CH2), 4.33 (d, J = 14.4 Hz, 1H, CH2), 5.15
(d, J = 16.8 Hz, 1H, CH2), 5.26 (d, J = 10.2 Hz, 1H, CH2), 5.55 (m, 1H,
–CH ), 6.80 (brs, 2H, ArH), 7.33 (d, J = 8.1 Hz, 2H, ArH).
2.4.10. Tetracarbonyl[(N,N-dimethylamino)
(4-methylphenyl)methylene]iron(0) (Vb)
Elution with hexane–CH2Cl2 mixture (5:1) furnished Vb (0.76 g,
48%) as orange-yellow crystals.
M.p. >80 ◦C (decomp.); 1H NMR: ı 2.35 (s, 3H; ArCH3), 3.09 (s,
3H; anti-NCH3), 3.99 (s, 3H; syn-NCH3), 6.73 (d, J = 7.9 Hz, 2H; Ar–H),
7.18 (d, J = 7.9 Hz, 2H; Ar–H). IR (CHCl3): ꢃ 2041 (m), 1966 (w), 1925
(s) cm−1. Anal. Calcd for C14H13FeNO4: C, 53.36; H, 4.16; N, 4.45.
Found: C, 53.69; H, 3.99; N, 4.39.
3. Results and discussion
3.1. General approach
The presented new iron and tungsten complexes together
with the previously reported series of aminocarbene complexes
of chromium [4] form a “multidimensional” set of analogous
compounds (cf. Table 1). This extension enabled more complex
evaluation of structural influences in various types of Fischer
aminocarbene complexes.
2.4.11. Tetracarbonyl[(N,N-dimethylamino)
(4-chlorophenyl)methylene]iron(0) (Vd)
Elution with hexane–CH2Cl2 mixture (4:1) furnished Vd (0.60 g,
36%) as yellow crystals. M.p. >110 ◦C (decomp.); 1H NMR: ı 3.11 (s,
3H; anti-NCH3), 4.00 (s, 3H; syn-NCH3), 6.79 (d, J = 8.2 Hz, 2H; Ar–H),
7.36 (d, J = 8.5 Hz, 2H; Ar–H). 13C NMR: ı 259.5 (C Cr), 214.6 (CO),
151.2 (C–Ar), 132.6 (C–Ar), 129.0 (CH-Ar), 121.8 (CH-Ar), 50.6 (CH3-
(1) The series I and II were correlated with analogous tungsten
complexes III and IV to evaluate parallels within Group 6 metals
behaviour.
N), 46.2 (CH3–N). IR (CHCl3): ꢃ 2044 (m), 1969 (w), 1930 (s) cm−1
.
Anal. Calcd for C13H10ClFeNO4: C, 46.54; H, 3.00; N, 4.17. Found: C,
46.32; H, 2.91; N, 4.19.
(2) The series I and II were correlated with analogous iron com-
plexes V and VI to compare complexes with the same ligands
but different central atom and different coordination number.
(3) The effect of the presence/absence of the chelate ring will be
followed using comparison between the non-chelates I, V and
VIIb on one side and corresponding chelates II, VI and VIIIb on
the other side.
2.4.12. Tetracarbonyl[(N,N-dimethylamino)
(4-trifluoromethylphenyl)methylene]iron(0) (Ve)
Elution with hexane–CH2Cl2 mixture (3:1) furnished Ve (0.76 g,
41%) as bright-yellow crystals.
M.p. 98–100 ◦C; 1H NMR: ı 3.12 (s, 3H; anti-NCH3), 4.03 (s,
3H; syn-NCH3), 6.96 (d, J = 8.2 Hz, 2H; Ar–H), 7.65 (d, J = 7.9 Hz, 2H;
Ar–H). 13C NMR: ı 259.1 (C Cr), 214,4 (CO), 155.4 (C–Ar), 129.4
(C–Ar), 126.0 (CH–Ar), 120.7 (CH–Ar), 50.5 (CH3–N), 46.4 (CH3–N).
IR (CHCl3): ꢃ 2045 (m), 1971 (w), 1932 (s) cm−1. Anal. Calcd for
C14H10F3FeNO4: C, 45.56; H, 2.73; N, 3.80. Found: C, 45.58; H, 2.59;
N, 3.60.
(4) The ꢁ-value of the small series of the ortho-phenylsubstituted
chromium non-chelates IXa, c, e will be compared with those
2.4.13. Tricarbonyl[(ꢂ2-N-allyl-N-allylamino)
(4-methoxyphenyl)methylene]iron(0) (VIa)
General procedure starting from 0.81 g (3.5 mmol) of 4-
methoxy-N,N-diallylbenzamide gave after elution with hexane–
CH2Cl2 mixture (4:1) 0.33 g (26%) of VIa as a brown oil.
1H NMR: ı 1.48 (d, J = 10.4 Hz, 1H, chelated CH2), 2.01 (d,
J = 7.7 Hz, 1H, chelated CH2), 3.18 (m, 1H, chelated –CH ), 3.75
(dd, J = 15.1, 6.0 Hz, 1H, NCH2), 3.81 (s, 1H, OCH3), 3.89 (dd, J = 15.1,
5.8 Hz, 1H, NCH2), 4.14 (dd, J = 14.3, 5.2 Hz, 1H, chelated NCH2), 4.35
(dd, J = 14.0, 0.8 Hz, 1H, chelated NCH2), 5.15 (dd, J = 17.0, 1.0 Hz, 1H,
CH2), 5.24 (dd, J = 10.3, 1.0 Hz, 1H, CH2), 5.52–5.66 (m, 1H, –CH ),
5.75–5.95 (m, 4H, H–Ar).
Fig. 1. Electrochemical reduction of Cr and W complexes.

8346
I. Hoskovcová et al. / Electrochimica Acta 55 (2010) 8341–8351
of the series I and II to follow the correlation between the ease
of phenyl ring rotation and influence of its p-substituent on the
reduction potential.
additional methyl group in the o-position on the phenyl should
also sterically hinder the rotation. In comparison with the anal-
ogous series I, lower value of the reaction constant was really
found. Moreover, it follows from NMR results that for the series
IX the rotation along the carbene–phenyl C–C bond is blocked and
the ortho-methylated derivatives are split into two isolable enan-
tiomeres [20] (cf. Table 1; Fig. 1).
(5) The non-chelate compounds VII substituted directly at the car-
bene carbon represent another small series where the influence
of substitution is to be followed.
According to expectation, the reduction potentials of com-
pounds IXa,c,e are in all cases shifted more negatively when
comparing with the analogous non-o-methylated compounds Ia,c,e
due to the electron pushing effect of the additional methyl sub-
stituent.
First, the complete set of electrochemical data was measured
using dc-polarography and cyclic voltammetry on Pt and Hg
stationary electrodes and fundamental reduction and oxidation
properties of all studied compounds were characterized. The reduc-
tion proceeds as a two-electron irreversible process around −2 V on
both electrode materials, the oxidation is manifested as a one elec-
tron, reversible couple at the cyclic voltammogram on platinum
electrode [4].
Slope ꢁ of the Hammett equation gives us a good possibility of
experimental observation and evaluation of the electronic commu-
nication (depending on the extent of electron delocalization and the
distance) between the substituent and the reaction center.
Similarly to the series I and II, in the case of series VII and VIII the
difference in potentials between the chelated and analogous non-
chelated derivatives is again very small. The dramatic difference,
however, was found in the reaction constants ꢁ: in the series VII and
VIII where the substituent is bonded directly (=closer) to the car-
bene carbon atom, the Hammett’s slope is about 5–6 times higher
than in the series I, II or IX. This finding shows that the direct sub-
stitution of the reduction center has here an extreme influence on
the reduction potential of the whole molecule. This result should be
very valuable namely for the design of new generations of carbene
complexes and for “tailoring” of their properties.
The investigation of the couple of tungsten complexes – the
non-chelate IIIc (analogue to Ic) and the corresponding chelate
IVc (analogue to IIc) offers a comparison within the Group 6 met-
als. The tungsten complexes are reduced slightly (about 30–50 mV)
less negatively (more easily) than the analogous chromium ones.
This fact is in accordance with the higher polarizability of tung-
sten compared with chromium. When comparing the tungsten
chelate and non-chelate compounds, their reduction potentials
are also very close to each other like in the case of chromium
complexes.
According to the LFER approach [5], the value of the slope ꢁ
(reaction constant) in the equation:
Eox (red) = ꢁ · ꢀp(ꢀi) + c
reflects the extent of electronic communication between the sub-
stituent and the reduction (oxidation) center, whereas the constant
ꢀp characterizes the sum of induction and resonance effects of the
substituent in the p-position on phenyl ring; eventually, for series
VII and VIII only, the constant ꢀi characterizes the induction effect
of the directly bonded substituent. From the linearity of the correla-
tion Ered (or Eox) of all involved derivatives vs. the corresponding ꢀp
(ꢀi) values follows, that in the frame of the studied series the same
reduction (oxidation) mechanism occurs, the reduction (oxidation)
centers remain unchanged and the structure of all starting deriva-
tives as well as intermediates is analogous. Any potential value
non-fitting the linear dependence points to a different behaviour
of the respective molecule, which means different localization of
redox center, or different electron displacement and thus different
mechanism.
On the other hand, the ꢁ-values reflect the extent of the
intramolecular interaction within the molecule, i.e. electronic “con-
ductivity” of the molecular bridge. Every difference in ꢁ-values
requires explanation on the basis of the structure and electron
delocalization.
It is possible to conclude (e.g. for the eventual design of new
analogues and prediction of their properties) that the reduction
potential depends strongly on the substitution of the carbene car-
bon, in the case of the free rotation along the carbene–phenyl C–C
bond, the substitution in the para-position on the phenyl ring plays
a role, too. On the other hand, the reduction potentials are not sub-
stantially affected by the presence of chelating allylic substituent,
hence an influence of the nitrogen substitution and the number of
CO ligands is negligible. Also the change of the central metal atom
(within the Group 6) plays only a minor role.
3.2. Reduction
3.2.1. Reduction of Cr and W complexes
There are five general types of chromium complexes to be
discussed: the non-chelates with p-substituted phenyl – type I
[(CO)5Cr C{N(CH3)2}(C6H4-p-R)], the analogous chelates – type
II [(CO)4Cr C{N(C3H5)2}(C6H4-p-R)], the directly substituted non-
chelates – type VII [(CO)5Cr C{N(CH3)2}R], the analogous chelate
VIIIb and the non-chelates with o-Me-p-substituted phenyl – type
IX [(CO)5Cr C{N(CH3)2}(C6H3-o-CH3-p-R)]. All of them have coor-
dination number six. In molecules of the type II and VIII one of the
allylic groups is coordinated to the central metal atom, the chelate
ring is formed and, hence, the number of CO ligands is lower by one
compared with the types I and VII.
3.2.1.1. Comparison of aminocarbene and alkoxycarbene complexes.
The above measurements permit also a comparison of two types of
Fischer carbene complexes with different stabilizing heteroatom:
the aminocarbene (data from Fig. 1; Table 1) and alkoxycarbene
complexes of Cr and W (data published in [21]), whose reduction
potentials were measured by CV under similar conditions (in non-
aqueous acetonitrile, vs. SCE).
Difference between reduction potentials of series I and II
(chromium non-chelates and chelates, cf. Fig. 1) is very small. On
the other hand, there is a remarkable feature concerning their
slopes ꢁ: the slope of the non-chelated carbenes is higher than
that of chelates and both lines cross at the ꢀp value equal to zero.
This effect was attributed to the lower electronic communication
between the para-phenyl substituent and the carbene reduction
center in chelates, caused by sterically hindered rotation along the
carbene–phenyl C–C bond.
Replacement of dimethylaminogroup by methoxygroup facil-
itates reduction of the compounds very strongly – the reduction
potentials are shifted positively by about 600 mV as the carbene
carbon atom becomes more positively charged when bonded to
a more electronegative atom of oxygen. Coming from chromium
methoxycarbenes to tungsten ones, reduction becomes easier by
several tens of mV, like in the case of aminocarbenes. Some new
alkoxycarbenes consisting of four chromium and/or tungsten units
attached to macrocyclic ligands were synthesized and their CV was
measured [22]. Also here, the reduction potential was found to be
some 60–70 mV more positive for tungsten compounds compar-
To support this explanation, the series IX was synthesized and
electrochemically investigated. In this non-chelated series, the

I. Hoskovcová et al. / Electrochimica Acta 55 (2010) 8341–8351
8347
Fig. 2. Comparison of reduction potentials of the Fe chelates and non-chelates with
those of the analogous Cr derivatives.
Fig. 3. Electrochemical oxidation of Cr and W complexes.
ing with their chromium analogues in accordance with the higher
polarizability of tungsten.
of chromium chelates (series I) as well as non-chelates (series II)
depend only very slightly on the carbene carbon substitution. In the
case of the directly substituted chromium carbenes VII and VIII the
ꢁ-value is about 5 times higher, but still by about one order of mag-
nitude lower then the ꢁ-value of their reduction plot (ꢁred = 1.99;
ꢁox = 0.33).
On the other hand, it is remarkable that the chelates (chromium
series II and tungsten complex IVc) are oxidized 300–400 mV less
positively (much more easily) than the corresponding non-chelates
(I and III). The reason is that the oxidation potential depends on
the number of carbonyl ligands. Since CO groups have strong ␲-
acceptor properties, they lower the electron density on the metal
atom and make its oxidation more difficult. In the chelates, one
electron withdrawing carbonyl is replaced by one allyl ligand with-
out pronounced ␲-acceptor properties; therefore, the oxidation of
chelates is significantly easier.
Similarly, the influence of the substituent in para-position
of the phenyl ring attached to the carbene carbon on reduc-
tion potentials is comparable: For [(CO)5Cr C(OCH3)(C6H5)]
complex I*c (analogous to Ic), E1/2 = −1.34 V was found, for
[(CO)5Cr C(OCH3)(C6H4-p-CH3)]
complex
I*b
(analogous
to Ib), E1/2 = −1.37 V. This difference of 30 mV corresponds
well with the 45 mV difference between Ic and Ib reduction
potentials.
3.2.2. Reduction of iron compounds
Two series of new iron aminocarbene complexes were syn-
thesized: the non-chelates with p-substituted phenyl – type
V
[(CO)4Fe C{N(CH3)2}(C6H4-p-R)], which is similar to the
chromium series I, and the analogous chelates type VI
–
Accordingly, the oxidation of Type I and Type VII complexes,
which have the same central atom and the same number of CO
ligands, proceeds at mutually close Eox values. The Eox values found
for both Type II and Type VIII chelated compounds are also close
one to another but again, both of them are lower by about 300 mV
comparing with I and VII.
[(CO)3Fe C{N(C3H5)2}(C6H4-p-R)], which corresponds to the
chromium series II. All iron complexes have the coordination num-
ber five. In molecules of the type VI, one of the allylic groups is
coordinated to the central metal atom, the chelate ring is formed
and, hence, the number of CO ligands is lower by one (like in the
chromium series II).
Fig. 3 shows another important feature: the oxidation potentials
depend strongly on the central metal atom: Tungsten complexes
are oxidized about 200 mV more positively (oxidation is more diffi-
cult) than their chromium analogues. This fact is in accordance with
the easier reduction of the tungsten complexes and is related again
to the higher polarizability of tungsten compared with chromium.
This is in accordance with the fact that the oxidation center of the
whole aminocarbene complex is the metal atom.
The above-described relationships between the structure and
redox properties of aminocarbenes seem to be more general.
Recently, a work has been published [3] dealing a.o. with electro-
chemical oxidation of some tungsten and chromium aminocarbene
complexes. Even if the structure of compounds, the electrochem-
ical conditions and the aim of the paper were different, their data
fit well with our findings mentioned above:
The iron aminocarbene complexes were investigated electro-
chemically in order to characterize the new compounds and to
compare their redox and other properties with the above inves-
tigated chromium and tungsten analogues of the type I, II, III and
IV (Fig. 2).
From the linearity of the E1/2 vs. ꢀp dependence it is evident
that the reduction mechanism of all studied iron derivatives is the
same, analogously to the chromium compounds. Even in this case,
the reduction center is located at the carbene moiety since the
reduction potential is substantially affected by the substitution on
the phenyl ring according to the LFER and the slopes (ꢁ-values)
are approximately the same for series V and I. As in the case of
chromium compounds, the slope of chelated iron aminocarbenes
VI is lower than that for the non-chelated ones (V). In contrast
to the chromium analogues, however, the reduction of chelated
iron aminocarbene complexes (type VI) is by 120 mV more diffi-
cult than that of non-chelated type V. This fact will be discussed
later together with the general comparison of iron and chromium
analogues.
(1) The
compounds
[(CO)5Cr C(H)N(CH2Ph)2]
and
[(CO)5Cr C(CH3)N(CH2Ph)2] are analogous to our VIIc and VIIb
and their respective Eox + 0.419 V and + 0.401 V vs. Fc+/Fc corre-
spond exactly to Eox (VIIc) = +0.819 V and Eox (VIIb) = +0.800 V
vs. SCE.
3.3. Oxidation
(2) For chromium complexes bearing a side chain with chelating
group, decrease of the number of CO groups by one leads to
lowering of the oxidation potential by 0.33 V (cf. approx. 0.35 V
difference between Type I and II.)
3.3.1. Oxidation of Cr and W complexes
According to the Hammett plots shown in Fig. 3, the slopes of the
Eox = ꢁ · ꢀp dependence are very low, hence the oxidation potentials

8348
I. Hoskovcová et al. / Electrochimica Acta 55 (2010) 8341–8351
The explanation should start with a deeper consideration
involving structural aspects and electronic properties of Fischer
carbene complexes as well as coordination abilities of ligands. The
main feature of Fischer carbene complexes is that the electron
withdrawing CO ligands lower the electron density at the central
metal atom. The formally double bond between the carbene carbon
and metal is polarized towards the metal atom, charging the car-
bene carbon positively (therefore the Fischer carbenes are called
“electrophilic”). The metal is now more electron rich, whereas the
electron-poor carbene is stabilized by the non-bonding electron
pair from the donating heteroatom (nitrogen or oxygen), giving rise
to a higher bond order between the carbene carbon and the group
X, according to the mesomeric expectations (Scheme 1).
Both divergences mentioned above originate in the different
properties of the central metal atom. Chromium has coordination
number six, iron has coordination number five and two more d-
electrons. Size of the both metals is about the same (single bonded
metallic radii of Cr and Fe are 119 pm and 117 pm, respectively).
It follows that in the coordination sphere of chromium some
intramolecular repulsion among ligands exists and hence the
metal–ligand bond lengths are expected to be longer that in the 5-
membered coordination sphere of iron. This premise is supported
by the structural data published in [23] for a binuclear aminocar-
bene complex (CO)5Cr C(NMe2)–C6H4–(NMe2)C Fe(CO)4. In this
compound, an average distance Cr–CO was 188 pm, Fe–CO 178 pm;
Fig. 4. Comparison of the oxidation potentials of the Fe chelates and non-chelates
with those of the analogous Cr derivatives.
(3) Two couples of analogous chromium and tungsten complexes
were measured. Like in our work, oxidation of tungsten deriva-
tives is always more difficult, ꢄE (Cr–W) was 0.11 V and 0.09 V,
while we have found ꢄE (Ic–IIIc) 0.18 V.
length of the metal–carbene carbon bond was 215 pm for Cr
C
3.3.2. Oxidation of iron compounds
and 200 pm for Fe C. This metal–carbene carbon bond variation
is supported by DFT calculations. Cr C and Fe C bond lengths
of 213.4 pm and 198.5 pm were calculated for the non-chelate
complexes Ic and Vc, respectively. Calculations on the chelate com-
plexes IIc and VIc gave Cr C and Fe C bond lengths of 208.6 pm
and 195.8 pm, respectively.
Thus in chromium complexes the distances Cr-ligand are rel-
atively long, an overlap of ␲-orbitals is not deep enough and the
back donation Cr → C is negligible. Therefore,
The oxidation potentials of iron complexes (series V and VI)
depend only very slightly on the substituent R, the slopes of the
Eox = ꢁ · ꢀp dependence are very low and the ꢁ-values are very close
to those of analogous chromium complexes I and II (Fig. 4).
The oxidation potentials of chelated (series VI) as well as non-
chelated (series V) derivatives are less positive than in the case
of chromium or tungsten complexes. This finding reflects the fact
that iron is more electron-rich metal with higher number of valence
electrons and, in addition to this, the number of carbonyl ligands is
lower by one.
What is, however, surprising, is the fact that the oxidation of iron
chelates and non-chelates proceeds in the same potential region.
Moreover, the non-chelates are oxidized even slightly more easily
than the chelates, although the chelates bear one CO ligand less
(in chromium or tungsten complexes, the difference between the
chelate and non-chelate oxidation potentials is very high: about
350 mV, cf. Fig. 4).
(a) the reduction center (most probably the C N double bond) is
affected neither by the number of CO ligands at the chromium
atom nor by the coordinated allyl group. Hence, the reduction
of chromium chelates and non-chelates proceeds in the same
region of potentials;
(b) the oxidation center (chromium atom) is isolated, therefore
the lower number of CO ligands as well as the replacement of
one CO ligand by the allyl influences substantially the oxida-
tion potential. Hence, the oxidation of chromium chelates and
non-chelates differ by about 350 mV.
These significant differences in redox properties between anal-
ogous chromium (or tungsten) and iron aminocarbene complexes
will be dicussed in the following section.
In iron complexes the metal–ligand distances are shorter and
the orbital overlap between them is more important. Under these
circumstances, importance of the overlap of ␲-orbitals grows up
and increased back donation is observed not only to CO ligands
(chromium carbenes I show ꢃ(CO) at 2054 cm−1, whereas for iron
carbenes V the vibration is shifted down to 2041 cm−1) but also to
carbene carbon and to the ␲-antibonding orbitals of the allyl group.
Simultaneously, better ꢀ-donation from allyl to iron occurs, making
thus the electron density at the metal even higher. Therefore
3.4. Difference Cr–Fe
Let us summarize the observed differences.
The reduction of chelated iron aminocarbene complexes (type
VI) is by 120 mV more difficult than that of non-chelated type
V, whereas the chromium chelates (II) and non-chelates (I) are
reduced in the same potential region (cf. Fig. 2).
The oxidation of iron chelates VI and non-chelates V proceeds in
the same potential region, moreover, the chelates are oxidized even
by tens mV more difficultly than the non-chelates, whereas the
chromium chelates II are oxidized much more easily (by 350 mV)
than chromium non-chelates I (cf. Fig. 4).
(a) due to the Fe → C back donation, the reduction center (carbene
carbon or the C N bond) is influenced. Hence, the reduction
potentials of the iron chelate derivatives (VI) are slightly more
Scheme 1. X = NR2, O-Alk.

I. Hoskovcová et al. / Electrochimica Acta 55 (2010) 8341–8351
8349
Fig. 5. The visualization of FMOs for chromium and iron complexes:.
Table 2
The contributions (%) of individual fragments to FMOs.
a
M
COtotal
Ccarb
N
R1
R2
Ph
LUMO
HOMO
5
56
16
36
51
2
17
0
3
0
3
0
6
5
Ic
LUMO
HOMO
5
59
9
30
42
2
14
0
4
5
0
20
2
IIc
7b
LUMO
HOMO
9
54
5
40
54
2
18
0
3
2
3
0
8
2
Vc
LUMO
HOMO
8
43
4
17
37
1
13
2
4
5
0
28
2
VIc
35b
R1 = R2 = CH3 for Ic and Vc.
R1 = R2 = allyl for IIc and VIc.
a
The substituent on N closer to the metal atom.
Chelated allyl.
b

8350
I. Hoskovcová et al. / Electrochimica Acta 55 (2010) 8341–8351
negative (by 120 mV) than the non-chelates (V), where the
the structure and redox properties in the Fischer aminocarbene
complexes and to find general rules enabling reliable prediction
of behavior of eventual further generations of carbene catalysts.
Experimental data were treated by the LFER approach in order to
evaluate extent of the electronic communication in the molecules
and to compare relative significance of various effects.
It was found that the reduction potential is very strongly influ-
enced by the direct substituents on the carbene carbon (series VII,
VIII). In the case of the aminocarbenes with aryl substituents (types
I, II, II, IV, V, VI and IX), the substitution on this phenyl ring in
para-position influences reduction potentials significantly.
The values of the oxidation potentials depend on the nature of
the central atom and on the number of ␲-acidic ligands in its coordi-
nation sphere and they are nearly insensitive to the carbene carbon
substitution. For Cr and W compounds, a straightforward correla-
tion between number of CO ligands and oxidation potential holds:
loss of one carbonyl group makes the oxidation by approx. 0.35 V
easier – less positive.
higher number of CO ligands makes the reduction easier.
(b) considering the oxidation center, two contradictional effects
can be followed: The presence of the donating allyl in chelates
together with the absence of one CO ligand should make the
oxidation easier; however, the strong Fe → C back donation
(proved by the mentioned reduction behaviour) is able to com-
pensate partially the above effect. Moreover, allylic group itself,
situated in very proximity of iron, can act as a ␲-acidic group
and take some electron density off. An additional support for
this explanation is given by 1H NMR spectra of the complexes:
chemical shift ı of the chelated –CH CH2 group is 3.22 for
chromium chelates (II), whereas for iron chelates it is only
1.52, due to a partial shielding by increased electron density.
As a result, the oxidation potentials of iron chelates and non-
chelates are very similar. The attempts of interpretation of
experimental data should be, however, supported by quantum
chemical calculations.
The difference between electronic structure of iron and
chromium aminocarbenes is reflected namely in the role of chelat-
ing allyl and in the extent of the metal-carbene or metal-allyl back
donation. The complexes of the Group 6 metals (Cr and W) have
their oxidation and reduction centers separated, presence of the
allyl plays a negligible role and this back donation is small, too.
In the pentacoordinated iron complexes, however, the bond
length metal–ligand is shorter than in the hexacoordinated Cr or
W molecules. Hence, better orbital overlap causes increasing of
some ␲-acidic character of the allylic and carbene ligands. Conse-
quently, reduction of iron chelates is more difficult then reduction
of non-chelates (carbene moiety becomes more electron rich), and
the more easy oxidation of iron chelates shows the ability of the
allyl and carbene groups to compensate the missing carbonyl lig-
and. The existence of Fe → allyl back donation is supported by 1H
NMR data. It is possible to consider the ␲-acidity of allylic and car-
bene ligands to be not a constant property of the ligands but rather
a property displayed with respect to a given metal.
3.5. DFT calculations
3.5.1. Localization of HOMO and LUMO. The role of the allylic
chelating ligand
Quantum chemical calculations were performed in order to
characterize redox orbitals and to prove and refine the interpreta-
tion of experiments. Fig. 5 depicts the displacement of HOMO and
LUMO orbitals and Table 2 provides the percentage of composing
fragments in these frontier orbitals (FMOs).
From Fig. 5 and Table 2 it is evident that the oxidation cen-
ter (HOMO orbital) in chromium and iron non-chelate and chelate
complexes is preferably localized on the metal center and on car-
bonyl ligands. On the other hand, in the chelates, the allyl ligand
is always involved in the HOMO orbital. Whereas in chromium
complexes the participation of allyl is rather small (7%), in iron
derivatives allyl participates substantially (35%), with pronounced
␲-acidic behaviour, as we have discussed based on experimental
results. Consequently, the metal contribution (43%) to HOMO is
lower than in chromium complexes.
Fig. 5 and Table 2 confirm that the center of reduction (LUMO
orbital) in chromium and iron non-chelate complexes (Ic and Vc,
respectively) is localized prevalently on the carbon and nitrogen
atoms. This fact demonstrates that among the mesomeric forms
of Fischer electrophilic aminocarbene complexes (Scheme 1) the
imine form with the formal C N double bond prevails and is
reduced. It is important to note that in non-chelates the phenyl
ring attached to the carbene carbon is involved neither in LUMO,
nor in HOMO. On the other hand, in the chelates, namely with
iron (VIc), the LUMO orbital is more spread over the molecule with
contributing ␲ orbitals of the phenyl ring.
As a support for the above discussion and for the interpretation
of the electrochemical data and their LFER treatment DFT calcula-
tions were utilized.
Acknowledgement
This contribution was supported by the grant of Grant Agency
of the Academy of Sciences of the Czech Republic (GAAVCR), no.:
IAA400400813.
References
[1] J. Chatt, C.T. Kan, G.J. Leigh, C.J. Pickett, D.R. Stanley, J. Chem. Soc., Dalton (1980)
2032.
Calculations also indicate that the reduction of the chelates
should be more sensitive to substitution on the phenyl group than
in the case of the non-chelates.
[2] F. Zaragosa-Dörwald, Metal Carbenes in Organic Synthesis, Wiley-VCH, 1999.
[3] C. Baldoli, P. Cerea, L. Falciola, C. Giannini, E. Licandro, S. Maiorana, P. Mussini,
D. Perdicchia, J. Organomet. Chem. 690 (2005) 5777.
[4] I. Hoskovcová, J. Rohácˇová, L. Meca, T. Tobrman, D. Dvorˇák, J. Ludvík, Elec-
trochim. Acta 50 (2005) 4911.
[5] P. Zuman, Substituent Effects in Organic Polarography, Plenum Press, New York,
4. Conclusions
1967.
ˇ
[6] T. Tobrman, L. Meca, H. Dvorˇáková, J. Cerny´, D. Dvorˇák, Organometallics 25
(2006) 5540.
For purposes of this study, 30 different aminocarbenes were
synthesized and their reduction and oxidation potentials were
determined in aprotic media. Based on such a large choice of
carbene complexes it was possible to correlate the measured elec-
trochemical data with single structural aspects and to compare
mutually the results for various homologous series of compounds
(Cr–W–Fe as central atoms, chelated or non-chelated complexes, p-
phenyl ring substitution or direct carbene carbon substitution). We
looked for analogies and discrepancies in order to explain the rea-
sons for observed differences, to describe the relationship between
[7] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheese-
man, J.A. Montgomery Jr., T. Vreven, K.N. Kudin, J.C. Burant, J.M. Millam, S.S.
Iyengar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega, G.A.
Petersson, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa,
M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X. Li, J.E. Knox, H.P.
Hratchian, J.B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R.E. Strat-
mann, O. Yazyev, A.J. Austin, R. Cammi, C. Pomelli, J.W. Ochterski, P.Y. Ayala, K.
Morokuma, G.A. Voth, P. Salvador, J.J. Dannenberg, V.G. Zakrzewski, S. Dapprich,
A.D. Daniels, M.C. Strain, O. Farkas, D.K. Malick, A.D. Rabuck, K. Raghavachari,
J.B. Foresman, J.V. Ortiz, Q. Cui, A.G. Baboul, S. Clifford, J. Cioslowski, B.B. Ste-
fanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R.L. Martin, D.J. Fox, T. Keith,
M.A. Al-Laham, C.Y. Peng, A. Nanayakkara, M. Challacombe, P.M.W. Gill, B. John-

I. Hoskovcová et al. / Electrochimica Acta 55 (2010) 8341–8351
8351
son, W. Chen, M.W. Wong, C. Gonzalez, J.A. Pople, Gaussian 03, Revision C.02,
[16] D. Dvorˇák, Organometallics 14 (1995) 570.
Gaussian, Inc., Wallingford, CT, 2004.
[17] L. Vyklicky´, H. Dvorˇáková, D. Dvorˇák, Organometallics 20 (2001) 5419.
[8] A.D. Becke, J. Chem. Phys. 98 (1993) 5648.
[9] A.D. Becke, Phys. Rev. A 38 (1988) 3098.
[18] E.O. Fischer, K.R. Schmid, W. Kalbfus, C.G. Kreiter, Chem. Ber. 106 (1973)
3893.
[10] P.C. Hariharan, J.A. Pople, Theo. Chim. Acta 28 (1973) 213.
[11] V.A. Rassolov, J.A. Pople, M.A. Ratner, T.L. Windus, J. Chem. Phys. 109 (1998)
1223.
[12] N.O. Brace, J. Org. Chem. 36 (1971) 3187.
[13] B. Bennetau, J. Mortier, J. Moyroud, J.-L. Guesnet, J. Chem. Soc., Perkin Trans. 1
(1995) 1265.
[19] A. Parlier, R. Yefsah, M. Rudler, H. Rudler, J. Organomet. Chem. 381 (1990) 191.
[20] L. Meca, Thesis, Institute of Chemical Technology, Prague, 2004.
[21] C.P. Casey, L.D. Albin, M.C. Saeman, D.H. Evans, J. Organomet. Chem. 155 (1978)
C37.
[22] I. Fernández, M.J. Manchen˜o, M. Gómez-Gallego, M.A. Sierra, Org. Lett. 5 (2003)
1237.
[14] R. Imwinkelried, L.S. Hegedus, Organometallics 7 (1988) 702.
[15] R. Hafner, L.S. Hegedus, G. De Weck, B. Hawkins, K.H. Dötz, J. Am. Chem. Soc.
110 (1988) 8413.
[23] M. Havránek, M. Husˇák, D. Dvorˇák, Organometallics 14 (1995) 5024.