Synthesis and characterization of redox-active mononuclear Fe(κ2-dppe)(η5-C5Me 5)-Terminated π-conjugated wires

Article

pubs.acs.org/Organometallics

Synthesis and Characterization of Redox-Active Mononuclear

Fe(κ²‑dppe)(η⁵‑C₅Me₅)‑Terminated π‑Conjugated Wires

Katy Green,^†Nicolas Gauthier,^†Hiba Sahnoune,^†Gilles Argouarch,^†Loic Toupet,^‡Karine Costuas,^†

Arnaud Bondon,^§Bruno Fabre,*^,†Jean-Francois Halet,*^,†and Frederic Paul*^,†

̧

́

^†Institut des Sciences Chimiques de Rennes, UMR 6226 CNRS, Universite

France

́

de Rennes 1, Campus de Beaulieu, 35042 Rennes Cedex,

^‡Institut de Physique de Rennes, UMR 6251 CNRS, Universite

́

de Rennes 1, Campus de Beaulieu, 35042 Rennes Cedex, France

^§RMN-ILP, UMR 6026 CNRS, Universite

France

́

de Rennes 1, IFR 140, PRISM, CS 34317, Campus de Villejean, 35043 Rennes Cedex,

S

* Supporting Information

ABSTRACT: Several new redox-active Fe(κ²-dppe)(η⁵-C₅Me₅) arylace-

tylide complexes featuring pendant ethynyl (Fe(κ²-dppe)(η⁵-C₅Me₅)-

[{CC(1,4-C₆H₄)}_nCCH] (1b−d; n = 1−3), Fe(κ²-dppe)(η⁵-

C₅Me₅)[CC(1,3-C₆H₄)CCH] (2)) or ethenyl (Fe(κ²-dppe)(η⁵-

C₅Me₅)[CC(1,4-C₆H₄)CHCH₂] (3)) groups have been synthesized

and characterized under their Fe(II) and Fe(III) states. In contrast to the

known ethynyl Fe(III) complex [Fe(κ²-dppe)(η⁵-C₅Me₅)(CCH)][PF₆]

(1a[PF₆]), most of the new Fe(III) derivatives turned out to be kinetically

stable in solution. A consistent picture of the electronic structure of the

latter complexes in both redox states emerged from experimental data and

DFT calculations. This study revealed that beyond the ﬁrst 1,4-phenylene ring, modiﬁcation or extension of the carbon-rich

linker using (4-phenylene)ethynylene spacers will have only a minor inﬂuence on their electronic properties in their ground state,

while still maintaining some (weak) electronic interaction along the carbon-rich backbone.

process, we decided to study the electron-transfer between the

“Fe(κ²-dppe)(η⁵-C₅Me₅)” end group (dppe = 1,2-bis-

(diphenylphosphino)ethane) and the underlying Si surface

when anchored with various types of spacers. To this aim, the

series of mononuclear complexes Fe(κ²-dppe)(η⁵-C₅Me₅)

[{CC(1,4-C₆H₄)}_nCCH] (1a−d; n = 0−3), Fe(κ²-dppe)-

(η⁵-C₅Me₅)[CC(1,3-C₆H₄)CCH] (2), and Fe(κ²-dppe)-

(η⁵-C₅Me₅)[CC(1,4-C₆H₄)CHCH₂] (3) featuring (poly)-

phenyleneethynylene ligands of various lengths and topologies

terminated by linking end groups was targeted (Scheme 1).

Indeed, as previously communicated,¹⁷1b−d will lead to the

corresponding monolayers when reacted with Si−H surfaces

using the UV-activated hydrosilylation reaction initially

developed by Chidsey and co-workers for phenylacetylene or

styrene,¹⁸while 2 and 3 should react similarly. Then, in order

to be able to optimize this photochemical grafting step, we have

also conducted a thorough experimental and theoretical study

on these derivatives in order to see how a change in the

connectivity of the terminal group on the aryl ring (meta vs

para), a change in the terminal group nature (ethynyl vs

ethenyl), or simply the extension of the 1,4-phenylene

ethynylene spacer would aﬀect the terminal group involved in

the photochemical grafting reaction. In particular, extension of

INTRODUCTION

■

Functional molecular materials are playing an ever-increasing

role in the fabrication of smart integrated devices that can

perform inter alia logic operations.¹⁻³In this respect, the

modiﬁcation of conducting surfaces at the molecular level with

redox-active “building blocks” constitutes a powerful approach,

particularly when the goal is to obtain integrated systems

devoted to information storage or transfer.⁴⁻⁶In parallel with

these studies, several families of redox-active carbon-rich group

8 σ-alkynyl complexes have been identiﬁed as interesting

candidates for reversible charge storage at the molecular level^7,8

and were shown to present promising prospects for the

elaboration of molecular-based electro-switchable devices.^9,10

Semiconducting surfaces, such as doped silicon, are attractive

for the elaboration of molecular-scale devices, since “π-

conjugation” between the redox tag and the surface can in

principle be maintained when the grafting is conducted from an

alkyne-terminated molecular precursor and oxide-free hydro-

gen-terminated silicon (Si−H) surfaces.^11,12To our knowledge,

while such silicon-based interfaces derivatized with various

types of redox-active centers, such as ferrocene¹³⁻¹⁵and metal-

complexed porphyrins,^1,2,4,16have recently attracted sustained

attention, the interfacial electron-transfer kinetics have not yet

been thoroughly investigated with metal alkynyl-based

monolayers.^5,6,14,17In order to learn how the nature and

length of the spacer inﬂuence the interfacial electron-exchange

Received: June 5, 2013

Published: July 31, 2013

4366

dx.doi.org/10.1021/om400515g | Organometallics 2013, 32, 4366−4381

Organometallics

Article

Scheme 1. Targeted Organoiron(II) Complexes: 1a−d, 2, and 3 ([Fe] = Fe(κ²-dppe)(η⁵-C₅Me₅))

Scheme 2. Synthesis of 1b−d and 2

the carbon-rich linker will result in a narrowing of the gap

between the occupied and empty molecular orbitals centered

on the (poly)phenylene ethynylene chain, and this phenom-

enon might signiﬁcantly aﬀect both the reactivity and the

stability of the terminal functional group. Their electron-

transfer properties within the monolayers should also be

modiﬁed. Finally, the kinetic stability of the corresponding

Fe(III) complexes might turn out to be an important issue,

given that oxidation reactions might take place at the Si−H

interface during the grafting process^18balong with the fact that

the shorter of these compounds (1a[PF₆]) is known to

spontaneously dimerize in solution.¹⁹

We therefore report in this paper (i) the synthesis of the

phenylene ethynylene mononuclear Fe(II) precursors 1b−d, 2,

and 3, featuring a pendant ethynyl or ethenyl group, (ii) the

characterization of their corresponding Fe(III) radical cations

(1b−d[PF₆], 2[PF₆], 3[PF₆]), and (iii) DFT computations on

several of these compounds to better understand their reactivity

toward the photochemical grafting reaction. The preparation

and characterization of the corresponding hybrid molecule/Si

interfaces as well as the measurements of the interfacial

electron-transfer kinetics will be reported in a subsequent

paper.²⁰

steps via vinylidene intermediates that were deprotonated

immediately after isolation with either t-BuOK or 1,8-

diazabicyclo[5.4.0]undec-7-ene (DBU). Deprotonation of

trimethylsilyl-protected vinylidene complexes with t-BuOK

resulted in the simultaneous removal of the silyl protective

group, whereas the use of DBU allowed for the clean isolation

of the silyl-terminated compounds (5a−c). The trimethylsilyl

protective group in 5a−c was subsequently removed with

potassium carbonate following standard procedures.²⁷Com-

plexes 1b and 1c were also accessed from the triisopropylsilyl-

protected complexes (5′b and 5′c) after desilylation using tetra-

n-butylammonium ﬂuoride (TBAF). The latter approach

turned out to be less productive than the former one and

requires harsher deprotection conditions. In line with a

previous study,²²trimethylsilyl complexes (5a−c) appear to

be better precursors of the corresponding ethynyl derivatives

than triisopropyl derivatives (5′a,b).

The synthesis of the ethenyl complex 3 was achieved from

the known organoiron Fe(II) aldehyde complex precursor 7²⁸

in one step, by reaction with a phosphonium salt in a Wittig-

type reaction (Scheme 3).²⁹

Scheme 3. Synthesis of 3

RESULTS

■

Synthesis and Characterization of Fe(II) Complexes

with Pendant Ethynyl and Ethenyl Groups. Among the

ethynyl-terminated mononuclear organoiron complexes that

were targeted for the grafting on Si−H surfaces only 1a^19,21and

1b²²had been reported. The new compounds 1c−d, 2, and 3

were obtained by similar reactions, starting from the

corresponding ethynyl precursors and the iron(II) chloride

complex 4,²³in two steps, following a well-established protocol

(Scheme 2).²⁴Several of the alkynes required for these

syntheses have been previously reported.^25,26All were newly

synthesized (Supporting Information) and reacted with the

Fe(II) chloride precursor 4. The synthesis proceeds in two

The new Fe(II) complexes were fully characterized by MS,

IR (in solution and solid state), and NMR spectroscopies. In

addition, the solid-state structures of 1b,c, 2, and 3 and of the

silyl precursors 5a and 6 were obtained by X-ray diﬀraction

(vide infra). ³¹P{¹H} NMR constitutes a good handle to probe

the purity of the various products by observation of a unique

singlet near 101 ppm for the chemically equivalent phosphine

atoms of the dppe ligand of each compound. However this

signal appears to be poorly sensitive to bridge extension in 1b−

4367

dx.doi.org/10.1021/om400515g | Organometallics 2013, 32, 4366−4381

Organometallics

Article

d. ¹H NMR is also poorly informative on the chemical

modiﬁcations taking place on the bridge within 1a−d, apart

from evidencing the increasing number of aromatic signals. In

contrast, ¹³C NMR allows for the clear detection of the

additional alkyne signals (near 90 ppm) when proceeding from

1a to 1d, while the α-alkynyl carbon atom comes out each time

around 146 ppm for all of these compounds.^22,30,31

Table 2. Electrochemical Data for Selected Fe(κ²-dppe)(η⁵-

C₅Me₅)(R) Complexes

ab

,

bc

,

compd

R

E° [ΔE_p] (V)

d

1b

1c

1d

2

CC(1,4-C₆H₄)CCH

{CC(1,4-C₆H₄)}₂CCH

{CC(1,4-C₆H₄)}₃CCH

CC(1,3-C₆H₄)CCH

CC(1,4-C₆H₄)CHCH₂

CC(1,4-C₆H₄)CCTMS

CC(1,3-C₆H₄)CCTMS

CCPh

−0.11 [0.08]

−0.12 [0.08]

−0.13 [0.09]

−0.16 [0.07]

−0.12 [0.08]

−0.14 [0.09]

−0.15 [0.08]

−0.11 [0.08]

−0.11 [0.07]

−0.11 [0.06]

−0.12 [0.07]

−0.15 [0.07]

The IR spectra clearly reveal an increasing number of ν_CC

3

modes upon bridge extension between 1b and 1c (Table 1) or

e

f

5b

6

Table 1. Experimental (in CH₂Cl₂) and Computed (in

Parentheses) IR ν_CCFrequencies (cm⁻¹) for Selected

Fe(κ²-dppe)(η⁵-C₅Me₅)(R) Complexes

g

8

−0.14 [0.08]

a

Conditions: CH₂Cl₂solvent, 0.1 M [Bu₄N][PF₆] supporting

b

electrolyte, 20 °C, Pt electrode, 0.1 V s⁻¹. The ratio between the

a

Fe(II)

Fe(III)

Δν_CC

anodic and cathodic peak current intensities was unity in each case

c

(I_pa/I_pc= 1). Conditions: CH₃CN solvent, 0.1 M [Bu₄N][ClO₄]

compd

R

ν_CC

Fe(II)/Fe(III)

d

supporting electrolyte, 20 °C, Pt electrode, 0.1 V s⁻¹. Potential

b

1a

1b

CCH

1910

e

f

diﬀerence between the anodic and cathodic peaks. From ref 22. From

ref 35. See ref 30.

cd

,

CC(1,4-C₆H₄)

CCH

2049/2030

(2066)

1994 (1996)

−36 (−70)

g

e

2100 (2125 )

2100 (2136)

1992 (2011)

2105 (2137)

2210 (2208)

1988 (2009)

2105 (2133)

0 (+11)

1c

{CC(1,4-

2043 (2067)

−51 (−56)

−2 (+5)

+4 (−15)

−55 (3)

−2 (+21)

+5 (−28)

e

1b−d and 3 and at higher potential for 2. In line with previous

observations,^22,30this trend correlates well with the electronic

eﬀect of the substituent on the ﬁrst ring. Indeed, based on the

redox potentials, we ﬁnd the following order: p-CCH > p-

CCSiMe₃(∼p-CCAr) > m-CCH > m-CCSiMe₃> H

> p-CHCH₂, which qualitatively ﬁts with the electron-

withdrawing character of the latter, according to the Hammett

electronic substituent parameter scale (when available).³⁴

Notably, these values also indicate that beyond the second

ring further extension of the arylethynyl ligand does not aﬀect

signiﬁcantly the metal-centered oxidation redox potential, since

1c and 1d display the same value within experimental

uncertainties.

Synthesis and Characterization of Corresponding

Fe(III) Complexes. In accordance with the previous cyclic

voltammetry (CV) studies, the corresponding Fe(III) com-

plexes could be synthesized and isolated after chemical

oxidation using ferrocenium hexaﬂuorophosphate (Scheme

4). The completion of the reaction was apparent from the

combined CV and IR spectra of the various products isolated.

Thus, the CVs of 1b,c[PF₆], 2[PF₆], and 3[PF₆] remain

identical to that of the starting Fe(II) complexes, indicating the

absence of decomposition or formation of new electroactive

side-products during the measurement time, while the IR

showed a clear shift toward lower wavenumbers of the

diagnostic ν_CCFestretching mode (Table 1), as expected

from previous investigations.^33,35Note that 1d[PF₆] decom-

poses in solution at ambient temperature. This Fe(III) species

could therefore not be isolated in a pure state and was only

transiently characterized when generated in solution, exhibiting

spectroscopic signatures consistent with these obtained for the

shorter analogues 1b,c[PF₆]. The ν_CCFeshift reveals a

weakening of the alkynyl bond order upon oxidation. However,

its value for 1b, 2, and 3 is not simple to determine with

accuracy, due to the existence of Fermi coupling.^32,33Overall,

the shift amounts to ca. 50 cm⁻¹for 1c and 1d, which

corresponds to that observed for the phenylalkynyl complex 8,

and seems poorly aﬀected when going from 1b[PF₆] to

1d[PF₆]. Relatively, the energies of the other ν_CCmodes of

these complexes and of the ν_CHCH2mode in 3 are almost not

aﬀected by oxidation, further suggesting that the electronic hole

is mainly metal-centered in these complexes and not

delocalized beyond the ﬁrst phenyl ring of the bridge. Thus,

C₆H₄)}₂CCH

2107 (2132 )

2206 (2220)

2043 (2112)

1d

{CC(1,4-

e

C₆H₄)}₃CCH

2107 (2117 )

2205

2210

(2200/2221)

e

(2249/2267 )

f

2

CC(1,3-C₆H₄)

CCH

2058 /

2009 (2006)

−33 (−65)

2042 (2071)

2105 (2135)

2109 (2148)

1990 (1995)

+4 (+13)

d

3

8

CC(1,4-C₆H₄)

CHCH₂

2055/2035

(2066)

−45 (−71)

g

b

CCPh

2053

2021/1988

−32

a

−

Minimum Fe(II) vs Fe(III) ν_CCdiﬀerence ( 4 cm⁻¹; PF₆

counterion). Similar in CH₂Cl₂to that in KBr pellets.¹⁹See also

b

c

d

ref 22. Two bands were observed for the Fe(III) parents, presumably

due to Fermi coupling.³³Very weak. Shoulder. See ref 24.

e

f

g

5a and 5b series. In all complexes, except for 1a, the ν_CC

mode corresponding to the alkynyl bonds ligated to the Fe(II)

center is observed around 2040 cm⁻¹, while those correspond-

ing to the terminal alkyne bond appear as very weak

absorptions near 2105 or 2155 cm⁻¹, depending on whether

the alkyne is ligated to H or SiMe_3,respectively. The additional

mode corresponding to the central alkyne in 1c or 5b is

observed near 2210 cm⁻¹.³²Interestingly, further progression

from 1c to 1d or 5b to 5c does not lead to any signiﬁcant

change in the ν_CCrange. Most likely, the additional ν_CC

mode expected is superimposed to the internal mode near 2210

cm⁻¹, as suggested by the DFT calculations (vide infra).

Congruent with this hypothesis, a shoulder is observed on this

absorption peak near 2210 cm⁻¹in the Raman spectrum, which

could correspond to the expected additional mode (see

Supporting Information). Finally, for 3, the additional

stretching mode observed near 1620 cm⁻¹and not observed

for the other Fe(II) complexes is tentatively attributed to the

ν_CHCH2mode.

Cyclic Voltammetry. Cyclic voltammograms of the

mononuclear compounds 1b−d, 2, and 3 were done in

dichloromethane medium. A single reversible one-electron

system was observed for all complexes (Table 2), correspond-

ing to the metal-centered oxidation Fe(II)/Fe(III).⁸Compared

to the phenylethynyl complex Fe(κ²-dppe)(η⁵-C₅Me₅)(C

CPh) (8), the oxidation process occurs at lower potential for

4368

dx.doi.org/10.1021/om400515g | Organometallics 2013, 32, 4366−4381

Organometallics

Article

Scheme 4. Synthesis of 1b−d[PF₆], 2[PF₆], and 3[PF₆]

Table 3. Experimental and Computed (in Parentheses) ESR Spectroscopic Data for Selected [Fe(κ²-dppe)(η⁵-C₅Me₅)(R)][PF₆]

ab

,

Complexes

compd

R

g₁

g₂

g₃

Δg

⟨g⟩

1b[PF₆]

1c[PF₆]

CC(1,4-C₆H₄)CCH

{CC(1,4-C₆H₄)}₂CCH

{CC(1,4-C₆H₄)}₃CCH

CC(1,3-C₆H₄)CCH

CC(1,4-C₆H₄)CHCH₂

CC(1,4-C₆H₄)CCTMS

{CC(1,4-C₆H₄)}₂CCTMS

{CC(1,4-C₆H₄)}₃CCTMS

CC(1,4-C₆H₄)CCTIPS

{CC(1,4-C₆H₄)}₂CCTIPS

CC(1,3-C₆H₄)CCTMS

CCPh

1.972 (1.978)

1.977 (1.985)

1.975 (1.982)

1.979 (1.971)

1.978 (1.976)

1.975

2.031 (2.032)

2.033 (2.026)

2.032 (2.022)

2.034 (2.029)

2.030 (2.028)

2.032

2.481 (2.259)

2.472 (2.197)

2.459 (2.214)

2.473 (2.305)

2.428 (2.263)

2.478

0.509 (0.281)

0.495 (0.212)

0.484 (0.232)

0.494 (0.334)

0.450 (0.287)

0.503

2.161 (2.090)

2.160 (2.069)

2.155 (2.073)

2.162 (2.102)

2.145 (2.089)

2.161

c

1d[PF₆]

2[PF₆]

3[PF₆]

5a[PF₆]

5b[PF₆]

5c[PF₆]

5′a[PF₆]

5′c[PF₆]

6[PF₆]

1.974

2.032

2.476

0.502

2.160

1.974

2.031

2.482

0.508

2.162

1.975

2.028

2.455

0.480

2.152

1.976

2.034

2.481

0.505

2.163

1.978

2.035

2.466

0.488

2.159

d

8[PF₆]

1.975

2.033

2.464

0.489

d

2.157

a

b

c

At ca. 70 K in CH₂Cl₂/1,2-C₂H₄Cl₂(1:1) glass. g values 0.005. Not pure: other minor species also present. See ref 35.

for the ethynyl-terminated complexes, oxidation leads to a

only a marginal eﬀect on the radical delocalization. As

previously shown with other such arylalkynyl compounds,³⁵

the ESR anisotropy (Δg) is often indicative of the electron-

attracting inﬂuence of the substituent on the ring. Presently,

comparison between the Δg values obtained for 1b[PF₆] and

2[PF₆] or for 5b[PF₆] and 6[PF₆] suggests that an alkyne

group is less electron withdrawing when located in the meta

position than in the para position on the ﬁrst aryl ring. Actually,

the values found for the meta derivatives compare within

experimental uncertainties with that obtained for the

compound 8[PF₆], possessing an unsubstituted phenyl

ring.^35,36Likewise, comparison between the anisotropy of

1b[PF₆] and 3[PF₆] suggests that the vinyl group is a slightly

more electron-releasing group than the ethynyl group and even

ν

mode at slightly lower energies for 1b[PF₆] (Δν = 19

C−H

cm⁻¹) and 1c[PF₆] (Δν = 26 cm⁻¹) than for the Fe(II) parent

compounds, but not for 1d[PF₆] and 2[PF₆]. Notably, the

ν_CCHmode is signiﬁcantly less intense than the internal and

alkynyl ν_CCstretches in compounds 1b−d and 3.

The strong metallo-centered character of these Fe(III)

radicals is also evidenced by their rhombic electron spin

resonance (ESR) signatures at ca. 70 K, which exhibit only

marginal changes in anisotropy (Δg) or mean g value (⟨g⟩)

between 1b−d[PF₆], 2[PF₆], and 3[PF₆] (Table 3). Thus,

among the Fe(III) para-ethynyl/ethenyl complexes (1b−

d[PF₆] and 3[PF₆]) or silyl-protected precursors (5a−

c[PF₆]), extension of the conjugation pathway has apparently

4369

dx.doi.org/10.1021/om400515g | Organometallics 2013, 32, 4366−4381

Organometallics

Article

Figure 1. ¹H NMR spectra of 2[PF₆] (a) and 1b[PF₆] (b) in CD₂Cl₂at 20 °C with proposed assignment according to Chart 1 for selected protons.

1

Table 4. Observed H NMR Shifts (δ 0.1 ppm) Recorded for Selected Protons of the [Fe(κ²-dppe)(η⁵-C₅Me₅)(R)][PF₆]

a

Complexes at 20 °C in CD₂Cl₂

R

dppe

C₅Me₅

H_Cp*

compd

R

H₁

H₂

H₃

H₄

H₅

H₆

H₇

H_Ar

CH₂

1b[PF₆]

1c[PF₆]

1d[PF₆]

2[PF₆]

CC(1,4-C₆H₄)CCH

{CC(1,4-C₆H₄)}₂CCH

{CC(1,4-C₆H₄)}₃CCH

CC(1,3-C₆H₄)CCH

CC(1,4-C₆H₄)CHCH₂

CC(C₆H₅)

−20.1

0.9

30.4

31.3

−39.6

−41.8

−42.4

27.8

7.9−1.3

7.9−1.7

7.9−1.6

10.0−2.0

−2.8

−2.7

−2.9

−2.8

−10.3

−10.5

−10.3

b

9.3

9.5

3.6

c

e

bc

,

d

c

2.5

31.6

n.a.

8.9

−36.7

−22.3

29.2

−38.2

32.5

3[PF₆]

−24.6

−41.7

28.1

−46.6

e

8[PF₆]

−41.7

7.9−1.8

c

−10.5

d

a

b

Proposed assignment according to Chart 1 (CHDCl₂taken at 5.35 ppm). Partly hidden behind another signal. Tentative assignment. Not

assigned. See ref 38.

e

Chart 1. Labeling of Selected Protons of 1b−d[PF₆], 2[PF₆], 3[PF₆], and 8[PF₆] ([Fe] = Fe(κ²-dppe)(η⁵-C₅Me₅))

4370

dx.doi.org/10.1021/om400515g | Organometallics 2013, 32, 4366−4381

Organometallics

Article

Figure 2. Thermal ellipsoid plots (ORTEP) of the two ethynyl complexes 1b (a) and 1c (b) at the 50% probability level. Hydrogen atoms and

solvent molecules have been omitted for clarity.

Figure 3. Thermal ellipsoid plots (ORTEP) of the Fe(II) and Fe(III) ethynyl complexes 2 (a) and 2[PF₆] (b) at the 50% probability level.

Hydrogen atoms have been omitted for clarity.

slightly more than a hydrogen atom when compared to the Δg

obtained for 8[PF₆].

These Fe(III) compounds were also characterized by H

substitution on the aryl ring and is actually also supported by

the DFT-computed carbon-based spin densities (see below).⁴¹

The protons of the second aromatic ring in 1c[PF₆] are much

less shifted than those of the ﬁrst ring and apparently even less

than for the third ring in 1d[PF₆], which is consistent with no

signiﬁcant spin density being delocalized beyond the second

ring in these compounds. This statement can also be deduced

from the ethynyl proton shifts which appear at 20.1 ppm for

1b[PF₆] and at 0.9 ppm for 1c[PF₆]. These observations are in

line with the dominantly metal-centered nature of these

radicals, a statement already inferred from ESR experiments,

but also apparent from IR measurements. Finally, the ethenyl

protons H₁and H₂in 3[PF₆] are slightly more shielded than

the ethynyl one (H₁) in 1b[PF₆], in line with the slightly higher

spin density found on the former substituent by DFT (see

later).

Crystallography. Figure 2 depicts the solid-state structures

of Fe(II) complexes 1b and 1c. The structures of the Fe(II)

and Fe(III) meta-substituted compounds 2 and 2[PF₆] and of

the ethenyl-terminated Fe(II) complex 3 are shown in Figures

3 and 4, respectively. Those of the triisopropylsilyl-protected

precursor of 1b (5′a) and of the trimethylsilyl-protected

precursor of 2 (6) are given in the Supporting Information.

Compounds 1b and 5′c have been characterized previously, but

their solid-state structures have not been reported,²²while the

structure of 1c has been previously reported in a

communication.¹⁷In the present case, the structural data for

1

NMR (Figure 1). All, except 1d[PF₆], for which not all signals

could be identiﬁed due to its kinetic instability in solution, gave

rise to well-resolved spectra characteristic of a single Fe(III)

metallo-centered radical.^8,37The shift of their protons was

identiﬁed by analogy with those of related complexes, by

integration, and whenever possible by polarization transfer

experiments. Most often, this led to an unambiguous

assignment of all the observed signals.³⁸The NMR shifts of

the dppe and C₅Me₅protons were very constant from one

complex to the other among 1b,c[PF₆], 2[PF₆], and 3[PF₆],

most of the changes taking place for the protons of the

1

arylethynyl ligand. The shifts of these H NMR signals (Table

4), which mostly depend on the so-called contact contribution,

reﬂect the changes in spin delocalization along the backbone of

the unsaturated carbon-rich ligand.^39,40First, based on the

DFT-computed spin densities (vide infra), their sign reﬂects the

spin alternation of the vicinal carbon atoms, in line with a spin

delocalization/polarization taking place mostly in the π-

manifold. This is plainly illustrated by the shift of the terminal

ethynyl proton, which was identiﬁed by ¹H−¹³C HSQC

correlation experiments for 1b[PF₆] (see Supporting Informa-

tion). The latter comes out at −20.1 ppm. For 2[PF₆], it is

found at the positive value 8.9 ppm. Such a reversal of the sign

of the observed shift can be expected based on the meta vs para

4371

dx.doi.org/10.1021/om400515g | Organometallics 2013, 32, 4366−4381

Organometallics

Article

Packing is certainly also at the origin of the unusual C45−

C36−Si1 angle of 168.2° (signiﬁcantly diﬀerent from 180°)

observed for 5′a. For the ethenyl complex 3, the coplanarity of

the ethenyl group and the adjacent phenyl ring, which allows

maximizing their π overlap, can apparently be accommodated

by packing forces.

UV−Visible Absorption Spectroscopy. The UV−vis

spectra of these compounds were recorded in their various

redox states (Table 5). In line with previous studies on the

electrochromism of related iron alkynyl complexes,^32,47Fe(II)

derivatives exhibit an orange color resulting from the presence

of an allowed MLCT (ε ≥ 8000 M⁻¹cm⁻¹) absorption in the

blue edge of the visible spectrum,³⁰while Fe(III) complexes are

dark brown-colored due to the apparence of an LMCT

absorption at lower energy, near 700 nm (ε ≤ 7000 M⁻¹cm⁻¹),

resulting from the reversal of polarity taking place upon

oxidation (Figure 5a,b).³⁵In the 1b-d/3 series, but also in 1b−

d[PF₆]/3[PF₆], the extension of the π-system on the alkynyl

ligand induces a slight bathochromic shift of these transitions

relative to those of the phenylethynyl complexes 8/8[PF₆] (see

also Computational Details).³⁵A similar feature is also

observed in the Fe(II) trimethylsilyl-protected complexes 5a−

c (Supporting Information). As expected, the introduction of

an ethynyl group in a nonconjugated (meta) position of the

arylethynyl ligand has a weaker eﬀect than in para position. The

hypsochromic shift of the MLCT of 2 relative to 1b can be

attributed to the loss of conjugation between the organoiron

center and the terminal ethynyl group located in meta position

in 2. Notably, compared to 1b, a slightly lower bathochromic

shift is observed for 3, which has an ethenyl group appended in

para position, in line with the better electron-releasing

capability of the vinyl substituent compared to the ethynyl

one, also revealed by electrochemistry.

Figure 4. Thermal ellipsoid plots (ORTEP) of the ethenyl complex 3

at the 50% probability level. Hydrogen atoms have been omitted for

clarity.

1b,c, 2, and 3 provide us with a mean to evaluate their spatial

extension, which constitutes important information for

estimating the surface area occupied by these molecules once

grafted on Si−H surfaces. All these compounds crystallize in

monoclinic or triclinic systems in centrosymmetric space

groups (Supporting Information, Table S1). Both distance

and angle values are expected for Fe(II) piano-stool complexes

(Supporting Information, Table S2),^{22,24,28,30,35,42}while a slight

but characteristic lengthening of the Fe−C and Fe−P bonds

and a concomitant shortening of the Fe−C37 bond are

observed for the Fe(III) complex 2[PF₆].^35,38,41,43Among the

Fe(II) species, the terminal ethynyl bonds in 1b, 1c, and 2 are

always slightly shorter (ca. 1.18 Å) than the internal ones (ca.

1.22 Å), while the ethenyl bond in 3 (1.312(3) Å) is

signiﬁcantly longer and close to typical values encountered

for organic molecules (1.321 Å).⁴⁴In contrast to related alkynyl

complexes featuring extended ligands,⁴⁶these compounds

roughly preserve their “rigid rod structure” in the solid state,

with only a slight S-shaped deviation from linearity taking place

for 1c. This characteristic deviation⁴⁶as well as the non-

coplanarity between the two phenyl rings of the arylalkynyl

ligand (torsion angle of 36.1° between them) is most likely

induced by the packing rather than by intramolecular eﬀects.³⁰

Likewise, for the Fe(III) complexes, the LMCT band of

3[PF₆] shows up at slightly lower energies than for 1b[PF₆]

and 2[PF₆], in line with a little more pronounced electron-

releasing character for the para-vinyl group relative to para- and

meta-ethynyl substituents, while marginal diﬀerences between

LMCT bands are stated between 1c[PF₆] and 1d[PF₆].³⁵In

addition, for the Fe(III) radical cations (Figure 5b), forbidden

a

Table 5. UV−Vis Spectroscopy Data for Selected [Fe(dppe)(η⁵-C₅Me₅)(R)]ⁿ⁺Complexes in CH₂Cl₂(n = 0, 1)

compd

1a

R

absorptions in nm [10⁻³ε in M⁻¹cm⁻¹

264 [15.1, sh], 402 [1.8, sh], 500 [0.6, sh]

268 [37.2], 412 [16.8]

]

ref

CCH

this work

1b

CC(1,4-C₆H₄)CCH

1b[PF₆]

1c

266 [29.9], 331 [16.6, sh], 428 [5.1, sh], 495 [2.6, sh], 593 [1.4], 704 [1.7], 1863 [0.10]

270 [37.2], 318 [32.2], 436 [19.7]

{CC(1,4-C₆H₄)}₂C

CH

1c[PF₆]

328 [38.1, sh], 336 [37.2, sh], 407 [8.4, sh], 500 [2.6, sh], 626 [1.3], 735 [2.6], 1864 [0.15]

337 [47.1], 476 [9.2]

this work

1d

{CC(1,4-C₆H₄)}₃C

CH

a

1d[PF₆]

338 [69.7, sh], 358 [56.2, sh], 442 [6.8, sh], 738 [3.6], 1864 [0.12]

358 [8.6], 398 [sh, 6.1]

this work

2

CC(1,3-C₆H₄)CCH

2[PF₆]

275 [30.7, sh], 384 [4.0, sh], 571 [2.4], 655 [2.8], 1864 [0.12]

274 [24.3], 406 [13.1]

3

CC(1,4-C₆H₄)CH

CH₂

3[PF₆]

8

268 [36.3], 352 [9.8, sh], 412 [3.6, sh], 472 [2.5], 612 [1.5, sh], 726 [2.9], 1827 [0.10]

277 [sh, 14.5], 350 [13.6]

this work

30

CCPh

8[PF₆]

261 [32.6, sh], 280 [27.4, sh], 301 [18.8, sh], 342 [5.9, sh], 379 [3.6, sh], 575 [2.3, sh], 662 [3.1], 1846 35

[0.09]

a

Data should be analyzed with caution due to the instability established for this particular Fe(III) compound in solution; sh = shoulder.

4372

dx.doi.org/10.1021/om400515g | Organometallics 2013, 32, 4366−4381

Organometallics

Article

Figure 5. UV−vis spectra of 1b−d, 2, and 3 (a) and 1b−d[PF₆], 2[PF₆], and 3[PF₆] (b) complexes in CH₂Cl₂. Scaled-up visible region for the

Fe(III) complexes (c).

Figure 6. DFT molecular orbital diagrams of the [Fe(κ²-dppe)(η⁵-C₅Me₅){CC(1,4-C₆H₄)}_nCCH] para complexes (n = 1 (1b), 2 (1c), 3

(1d)), [Fe(κ²-dppe)(η⁵-C₅Me₅){CC(1,3-C₆H₄)}CCH] meta complex (2), and [Fe(κ²-dppe)(η⁵-C₅Me₅){CC(1,3-C₆H₄)}CHCH₂] (3).

The Fe (left)/carbon-chain (right) percentage contributions are given in italics.

bands at lower energy (ε ≈ 100 M⁻¹cm⁻¹) corresponding to

forbidden d−d excitations are also observed near 1900 nm.³⁵

DFT Calculations. Density-functional theory (DFT)

calculations were carried out on compounds 1b−d, 2, and 3

and their radical cations in order to analyze and compare their

electronic structure and the ensuing physical properties (see the

Experimental Section for computational details).⁴⁸Their

structural arrangements were ﬁrst optimized and compared

(Supporting Information, Table S3). Computed data match

reasonably well with the available experimental values

(Supporting Information, Table S4), with the largest bond

length deviations found for the Fe−Cp* (centroid) (<0.05 Å)

and Fe−P distances (<0.03 Å). As often observed for this kind

of acetylide metal complex, the computed CC distances are

slightly overestimated by 0.02−0.03 Å on average, with respect

to the experimental ones, in 1b and 1c. As expected, the

extension of the unsaturated bridge (CCPh)_nfrom n = 1

(1b) to 3 (1d) hardly aﬀects the Fe−P(dppe) and Fe−Cp*

(centroid) bond lengths. Similar to experiment, the computed

inner CC distances (close to the metal center) are ca. 0.03 Å

longer than the outer ones. Computed bond distances for 1d,

4373

dx.doi.org/10.1021/om400515g | Organometallics 2013, 32, 4366−4381

Organometallics

Article

not measured experimentally, are reasonably similar to those

computed for 1b and 1c.

(vide supra). This HOMO is part of the “t_2g” set expected for

these 18-electron pseudo-octahedral complexes (Figure 7).

Being strongly metallic in character, only very small energy

stabilization occurs upon elongation of the carbon chain.

Although the Koopmans approximation⁴⁹is not valid within

the DFT approach, the changes of the ﬁrst adiabatic ionization

potentials (IP) in the series are susceptible to be correlated to

the energies of the HOMOs (which are close in energy, vide

supra) if the electronic and geometric reorganizations upon

oxidation are moderate. Indeed, IPs computed without or with

solvent (dichloromethane) eﬀect are almost equal (ca. 5.25 and

4.65 eV, respectively). This is in agreement with the

experimental oxidation potentials of species 1b−d, which are

comparable.

The LUMO of 1b, 2, and 3 and the LUMO+1 of 1c and 1d

are strongly localized on the Fe(κ²-dppe)(η⁵-C₅Me₅) moiety.

On the other hand, the LUMO of 1c and 1d is mainly localized

on the carbon chain, speciﬁcally on the outer part (Figure 7).

Indeed, the antibonding character of these π*-orbitals

diminishes with the lengthening of the carbon chain inducing

a gradual energy stabilization. Consequently, the HOMO−

LUMO energy gap decreases from 1b (1.79 eV) to 1c (1.52

eV) to 1d (1.18 eV). As expected, the HOMO−LUMO gap of

2 (1.78 eV) and 3 (1.77 eV) is close to that of 1b.

Upon oxidation, a substantial lengthening of the Fe−P bond

lengths (ca. 0.06 Å) and, to a lesser extent, of the Fe−Cp*

(centroid) along with a slight Fe−CC (ca. 0.03 Å)

contraction is noted. A slight lengthening of the inner CC

distances is computed. Surprisingly enough, no geometrical

change is noted for the outer CC upon oxidation. These

structural modiﬁcations are in line with those previously

observed for related systems.^35,38,41They can easily be

understood by examining the HOMOs of those systems (see

the molecular orbital diagrams in Figure 6). Indeed, the

HOMO of 1b−d, 2, and 3 is π in character, heavily localized on

the Fe center and on the adjacent ethynyl group and, to a lesser

extent, on the next phenyl ring (Figure 7). It is, in turn, mainly

Fe−C37 antibonding and C37−C38 bonding in character.

Consequently, oxidation of these systems leads to some

shortening of Fe−C37 and a slight lengthening of C37−C38

The energies of the vibrational frequencies of the ethynyl

CC vibrators were computed for these compounds and their

radical cations. Overall, they compare very well with the

experimental values (Table 1). As experimentally observed for

all the neutral compounds, the ethynyl bond neighboring the

Fe center vibrates at somewhat smaller energy than the outer

ones. No change in energy is observed for this vibrator upon

chain lengthening. However, the vibrational energy decreases

by ca. 50 cm⁻¹upon oxidation (Table 1). Comparably, the

energy of the other ethynyl groups remains nearly unchanged at

ca. 2100−2200 cm⁻¹. This is in agreement with the nodal

properties of the HOMOs, partially depopulated upon

oxidation, which are hardly localized on these CC groups

(vide supra).

Mulliken atomic spin densities were computed for the

monocationic species 1b−d⁺, 2⁺, and 3⁺to evaluate the

delocalization of the unpaired electron in these complexes

(Supporting Information, Table S5). Results are graphically

represented in Figure 8. The spin density distribution tracks

close to the HOMO spatial distribution of the neutral

complexes, which is, indeed, partially depopulated. It is mostly

localized on the iron and adjacent ethynyl group for each

system (Figure 7). Indeed, for all complexes, most of the spin

density is found on the metal center (ca. 0.70 electron) and to a

lesser extent on the inner CC group (ca. 0.35 electron). A

slight spin-density decrease on the metal is noticed upon chain

lengthening for the beneﬁt of the adjacent ethynyl group of the

chain itself. It is noteworthy that the spin density is somewhat

more distributed over the whole backbone for 1b⁺, 2⁺, and 3⁺.

It appears that regardless of the redox state (n = 0 or 1) of

1b−dⁿ⁺, 2ⁿ⁺, and 3ⁿ⁺, their MOs which are heavily localized on

the terminal alkyne or alkene moieties will strongly resemble

those observed for the purely organic molecules in which the

organometallic [Fe(κ²-dppe)(η⁵-C₅Me₅)(CC)]ⁿ⁺capping

groups have been replaced by a hydrogen or by a moderately

electron-releasing substituent (such as a methoxy group),

especially when the phenylene-ethynyl chain is extended. Thus,

based on the assumption that the photochemical initiation

previously used primarily involves homolysis of surface Si−H

Figure 7. Plots of the HOMO and LUMO for the complexes 1b−d, 2,

and 3. Contour values are 0.03 (e/Bohr³)^1/2. The energy is given in

parentheses.

4374

dx.doi.org/10.1021/om400515g | Organometallics 2013, 32, 4366−4381

Organometallics

Article

DISCUSSION

■

Both the spectroscopic results and the DFT calculations

conﬁrm that the isolated ethynyl- or ethenyl-terminated

compounds 1b−d, 2, and 3 exhibit the classical features of

Fe(II) arylalkynyl complexes,³⁰with a dominantly metal-

centered HOMO. Spectroscopy (NMR and IR) and

computations on 1b−d reveal that the terminal alkyne (or

alkene) group in these organometallic complexes is not very

diﬀerent from those connected to purely organic aromatic

scaﬀolds, especially for compounds featuring extended phenyl-

ene-ethynylene spacers, due to the strong localization of the

HOMO on the Fe(II) fragment. For instance, based on its

characteristic ν_CCstretching mode, the terminal alkynyl group

appears not to be signiﬁcantly aﬀected in solution by the

structural changes within 1b,c or 2. Changes observed

experimentally are much less than those computed and remain

close to the experimental uncertainty. Also, the Fe(II/III)

oxidation potential appears to be only weakly dependent on the

substitution pattern of the closest phenylene ring of the carbon-

rich spacer, in line with an even weaker interaction with the

terminal group in the longer compounds. However, the

characteristic bathochromic shifts of their MLCT transitions

in the UV−visible range reveal that some electronic

communication is nevertheless present along the metal and

phenylene ethynylene backbone in these Fe(II) compounds.

Upon oxidation, the ν_CCstretching mode of the terminal

alkyne appears also to not be signiﬁcantly aﬀected. ESR reveals

a dominantly metal-centered rhombic radical, the anisotropy of

which is primarily determined by the substitution pattern of the

closest phenylene ring of the carbon-rich spacer. In this respect,

Figure 8. Plots of the spatial distribution of the spin density for

compounds 1b−d⁺, 2⁺, and 3⁺. Isocontour values: 0.003 e/bohr³.

bonds,^18bthe reactivity of 1b−d, 2, and 3 should resemble that

of the purely organic analogues, which is indeed observed.¹⁷

EPR properties of the optimized 1b−d⁺, 2⁺, and 3⁺

complexes were also computed. The resulting g-tensor

components are given in Table 3 for a comparison with

experimental values. The agreement between the computed and

experimental anisotropies Δg is moderately satisfactory, with

the computed ones being somewhat smaller than the

experimental ones. However, these computations were done

on a single conformation,^38,50while the eﬀect of the

surrounding medium was not taken into consideration. With

those restrictions in mind, the overall trend can be considered

to be comparable to that experimentally observed and also in

agreement with the computed spatial distribution of the spin

density. Extension of the carbon chain hardly aﬀects the spin

density distribution of these monocationic species. This

contrasts with metal-polyynyl complexes, which show much

more pronounced delocalization of the spin density over the

carbon chain.⁵⁰

Finally, the TD-DFT-calculated energies of the lowest

allowed electronic excitations of the neutral complexes 1b−d

were also computed (Supporting Information, Tables S6 and

S7). As observed experimentally, a substantial red shift of the

lowest excitation energies is noted upon lengthening of the

carbon chain (418, 486, and 527 nm for 1b, 1c, and 1d,

respectively). These excitations mostly involve HOMO−

LUMO transitions. Accordingly, with the energy of the

HOMO hardly changing and that of the LUMO decreasing

upon chain lengthening, some bathochromic shift is observed.

The ﬁrst lowest excitation energies were also computed for the

cationic complexes 1b⁺−d⁺. Interestingly, very weak excitations

are found in the NIR region (1825, 1833, and 1897 nm for 1b⁺,

1c⁺, and 1d⁺, respectively), which compare very well with the

very low-energy bands of ca. 1865 nm measured experimen-

tally. As expected, they involve metal d−d transitions. Very

weak electronic excitations are computed around 600 nm for all

the cations (see Table S7). They might correspond to weak

bands experimentally observed somewhat red-shifted around

700 nm (see above). They mostly involve ligand-to-metal and/

or metal-ligand-to-metal transitions.

1

the H NMR of the Fe(III) complexes is more informative. It

reveals that while sizable spin delocalization takes place on the

ﬁrst phenylene unit and also on the second C₂unit, this

phenomenon drastically drops when further progressing along

the carbon-rich chain. The magnitude (and sign) of the spin

present on the second C₂unit clearly depends on the topology

of the ﬁrst ring (para > meta) and on its nature (ethenyl >

ethynyl). However, on the basis of the relative isotropic shifts of

the protons of these units, it remains lower than in the ﬁrst

phenylene ring. The unpaired spin present on the terminal C₂

unit of all compounds should therefore not exceed a few

percent, in line with the available DFT computations.³⁸Thus,

in contrast to their shorter homologue 1a⁺, which presents a

much larger spin density on the β-ethynyl carbon and promptly

dimerizes upon oxidation,^19,511b,c⁺are kinetically much more

stable in solution. The gain in kinetic stability observed for

1b,c[PF₆], 2[PF₆], and 3[PF₆] over 1a[PF₆] certainly results

from the presence of an aryl ring in the β-position to the metal

which sterically shields the reactive β-carbon atom, thereby

preventing coupling at this position and inhibiting further spin

delocalization along the carbon-rich ligand.⁵¹Note that in spite

of a very small spin density on the terminal alkynyl unit,

1d[PF₆] slowly undergoes degradation in solution, even in the

absence of oxygen.⁵²In this case, a larger portion of the 1,4-

phenylene-ethnylene chain is exposed to the reaction medium.

The decomposition stated for 1d[PF₆] might therefore be

attributed to the occurrence of a radical coupling process taking

place at other positions. Thus, except for 1d, the stability of

1b−d, 2, or 3 should not be a determining issue during the

grafting reaction, even if partial oxidation takes place on the Si−

H surface or within the monolayer,^18bprovided the reaction is

conducted under an inert atmosphere.

4375

dx.doi.org/10.1021/om400515g | Organometallics 2013, 32, 4366−4381

Organometallics

Article

As expected from previous investigations on arylalkynyl

Fe(III) complexes,³⁵DFT conﬁrmed that the SOMO was

largely metal-centered in these compounds, even when the aryl

alkynyl ligand is extended. Experimentally, as discussed above,

this translates to a quasi-invariance of the ESR signature (Table

3) and of the Fe(III/II) oxidation potentials (Table 2) of these

compounds when featuring a carbon-rich ligand with two aryl

rings or more, regardless of the exact nature of the terminal

groups on the second ring (CC−Ar, CC−TMS, CC−

TIPS, or CC−H).⁵³This statement is also substantiated by

the very slight changes in paramagnetic ¹H NMR shifts

observed for the protons of the ﬁrst phenyl ring upon chain

extension, which are much more sensitive probes to changes in

spin density than is ESR anisotropy.³⁸However, a closer

examination of these signals reveals very slight upﬁeld and

downﬁeld shifts of the ortho and meta hydrogen atoms,

respectively, when proceeding from 1b to 1d, in line with a

concomitant slight enhancement of the spin delocalized from

the metal toward the (poly)arylalkynyl ligand. The latter, also

found in DFT computations, might be at the origin of the

increased kinetic instability of these radicals upon extension of

the carbon-rich linker,^51,54as experimentally stated for 1d[PF₆]

in solution. In this connection, the very slight bathochromic

shift experienced by the LMCT transition of 1c,d upon chain

elongation further indicates that only a very weak electronic

interaction takes place between the metallic end group and

phenylethynyl linker for the extended compounds in the

Fe(III) state.

possibly at the origin of its increased reactivity in solution. In all

the other cases, any adventitious oxidation of the Fe(II)

precursor during the photochemical grafting reaction should

not be detrimental to the monolayer formation, provided the

reaction is performed under an inert atmosphere.⁵²In

accordance with these statements, we will see in a subsequent

paper that these functional compounds can be conveniently

grafted on Si−H surfaces by using a photochemical protocol

operating under rather “mild” conditions.²⁰In addition, we will

show that the resulting monolayers exhibit remarkable

electronic exchange rates through their covalent interfacial

Si−C bond compared to related organometallic silicon-based

interfaces.

EXPERIMENTAL SECTION

■

General Procedures. All reactions and workup procedures were

carried out under dry, high-purity argon using standard Schlenk

techniques. All solvents were freshly distilled and purged with argon

before use. Infrared spectra were obtained using a Bruker IFS28 FT-IR

spectrometer (400−4000 cm⁻¹) or using a Bruker Optics Vertex 70

FT-IR spectrometer in the transmission mode (100 scans, 2 cm⁻¹

resolution and automatic gain) using a DTGS detector. Raman spectra

of the solid samples were obtained by diﬀuse scattering on the same

apparatus and recorded in the 100−3500 cm⁻¹range (Stokes

emission) with a laser excitation source at 1064 nm (25 mW) and a

quartz separator with an FRA 106 detector. NMR spectra were

acquired at 298 K on a Bruker DPX200, a Bruker Ascend 400 MHz

NMR, or a Bruker AVANCE 500, with a 5 mm broadband observe

probe equipped with a z-gradient coil. Experimental details regarding

measurements on paramagnetic Fe(III) complexes can be found in

previous contributions.^38,41,55Chemical shifts are given in ppm and

CONCLUSION

■

referenced to the residual nondeuterated solvent signal for ¹H and ¹³

C

We have reported here the synthesis and characterization of

new redox-active Fe(II) arylacetylide complexes 1b−d and 2

featuring a pendant ethynyl group, and 3 with a pendant

ethenyl group, as well. As discrete species, the experimental

data and the DFT calculations reveal that beyond the ﬁrst 1,4-

phenylene ring further extension of the carbon-rich linker

induces only minor changes in their electronic properties in

both redox states. Therefore, as previously observed,²⁰the

terminal ethynyl or ethenyl group of these compounds should

react similarly to that of a purely organic aromatic styryl or

phenylethynyl derivative, and, at least for 1b−d and 2, only

minor diﬀerences in reactivity can be anticipated between them

during the photochemical grafting reaction. The UV−vis

absorption spectra nevertheless reveal the existence of some

electronic interaction along the (poly)phenylene-ethynylene

backbone, by showing a clear bathochromic shift of the MLCT

transition upon chain lengthening. Accordingly, this kind of

conjugated carbon-rich spacer should facilitate the electronic

interaction between the terminal metallic Fe(II) center and the

silicon surface subsequent to grafting. We have also isolated and

characterized each of these electron-rich alkynyl complexes

under their cationic Fe(III) state, and roughly similar

conclusions can be drawn regarding the electronic structure

of the terminal ethynyl or ethenyl group and regarding the

electronic communication through the phenylethynyl linker.

Notably, all these radical cations except 1d[PF₆] turned out

to be kinetically stable in solution during periods exceeding

those previously used for performing the grafting reaction. For

the Fe(III) species, this remarkable stability is in line with the

dominant metal-centered nature of these organometallic

radicals. In this respect, the longest derivative (1d[PF₆])

among them presents however a slightly higher delocalization

of the electronic vacancy on the carbon-rich ligand, which is

and external H₃PO₄(0.0 ppm) for ³¹P NMR spectra. Cyclic

voltammograms were recorded in dry CH₂Cl₂solutions (containing

0.10 M [n-Bu₄N][PF₆], purged with argon, and maintained under an

argon atmosphere) using an EG&G-PAR model 263 potentiostat/

galvanostat or using an Autolab electrochemical analyzer (PGSTAT 30

potentiostat/galvanostat from Eco Chemie B.V.) equipped with the

GPES software in a homemade three-electrode Teﬂon cell. The

working electrode was a Pt disk, the counter electrode a Pt wire, and

0/+

the reference electrode a saturated calomel electrode. The FeCp₂

couple (E_1/2: 0.46 V, ΔE_p= 0.09 V; I_pa/I_pc= 1) was used as an internal

calibrant for the potential measurements.⁵⁶Near-IR and UV−visible

spectra were recorded as CH₂Cl₂solutions, using a 1 cm long quartz

cell on a Cary 5000 spectrometer. EPR spectra were recorded on a

Bruker EMX-8/2.7 (X-band) spectrometer, at 77 K (liquid nitrogen).

Elemental analysis and high-resolution mass spectra (ESI on

Micromass MS/MS ZABSpec TOF spectrometer) were performed

at the “Centre Regional de Mesures Physiques de l′Ouest” (CRMPO),

Universite

́

de Rennes 1. The complex Fe(κ²-dppe)(η⁵-C₅Me₅)(Cl) (4)

was obtained following published procedures.²³Other chemicals were

purchased from commercial suppliers and used as received.

Synthesis of Fe(κ²-dppe)(η⁵-C₅Me₅){CC(1,4-C₆H₄)CCSiMe₃}

(5a). A solution of Fe(κ²-dppe)(η⁵-C₅Me₅)(Cl) (4; 509 mg, 0.81

mmol), ((4-ethynylphenyl)ethynyl)trimethylsilane (178 mg, 0.90

mmol), and KPF₆(235 mg, 1.28 mmol) in THF (20 mL) and

MeOH (20 mL) was stirred for 16 h. The solvent was removed in

vacuo. The brown solid was extracted with CH₂Cl₂(10 mL), n-pentane

(50 mL) was added to the stirred extract, and the crude vinylidene

complex 5a-vin[PF₆] was collected as a brown precipitate. ³¹P{¹H}

NMR (CDCl₃, 81 MHz): δ 87.6 (s, 2P, dppe), −143.0 (sept, 1P,

−

PF₆). ¹H NMR (CDCl₃, 200 MHz): δ 7.85−7.00 (m, 22H, H_Ar/dppe),

6.23 (m, 2H, H_Ar) 5.06 (m, 1H, FeCCH), 3.16 (m, 2H,

CH_2/dppe), 2.56 (m, 2H, CH_2/dppe), 1.62 (s, 15H, H_Cp*), 0.28 (s, 9H,

H_SiMe3). This crude vinylidene complex was dissolved in THF (20

mL), DBU (0.2 mL, 1.32 mmol) was added, and the mixture was

stirred for 10 min. The solvent was removed in vacuo, yielding the

crude product as a red solid. The product was puriﬁed on a short

4376

dx.doi.org/10.1021/om400515g | Organometallics 2013, 32, 4366−4381

Organometallics

Article

column of deactivated silica under an inert atmosphere, eluting with

toluene. The solvent was removed and the orange solid washed with n-

pentane, aﬀording 5a as an orange solid in 84% yield (536 mg, 0.68

mmol). The identity of 5a was conﬁrmed by comparison of

15H, H_Cp*), 0.24 (m, 9H, H_SiMe3). ¹³C{¹H} NMR (C₆D₆, 50 MHz): δ

2

146.7 (t, J_P,C= 39 Hz, FeCC), 139.9−128.5 (C_Ar, CH_Ar), 124.9

(C_Ar), 122.7 (C_Ar), 121.7 (FeCC), 117.1 (C_Ar), 105.9, 95.9, 94.0,

89.9 (CC), 88.1 (C₅(CH₃)₅), 31.1 (m, CH₂), 10.4 (C₅(CH₃)₅), 0.1

(Si(CH₃)₃)). UV−vis (CH₂Cl₂, λ_max/nm [ε/10³M⁻¹cm⁻¹]): 316

[40.3], 460 [20.6].

spectroscopic data with literature values.²²UV−vis (CH₂Cl₂, λ_max

nm [ε/10³M⁻¹cm⁻¹]): 269 [38.3], 415 [17.1].

/

Synthesis of Fe(κ²-dppe)(η⁵-C₅Me₅){CC(1,4-C₆H₄)CCSi(ⁱPr)₃}

(5′a). A solution of Fe(κ²-dppe)(η⁵-C₅Me₅)(Cl) (4; 1.90 g, 3.04

mmol), ((4-ethynylphenyl)ethynyl)triisopropylsilane (859 mg, 3.04

mmol), and NH₄PF₆(592 mg, 3.632 mmol) in MeOH (40 mL) was

stirred for 12 h. The solvent was removed in vacuo. The brown solid

was extracted with CH₂Cl₂(2 × 30 mL), and diethyl ether (40 mL)

was added to the stirred extract after concentration to ca. 5 mL. The

crude vinylidene complex 5′a-vin[PF₆] (2.80 g) was collected as a

brown precipitate. FT-IR (KBr, cm⁻¹): ν 2146 (m, CC), 1620 (s,

Synthesis of Fe(κ²-dppe)(η⁵-C₅Me₅){[CC(1,4-C₆H₄)]₂CCSi(ⁱPr)₃}

(5′b). A solution of ((4-((4-ethynylphenyl)ethynyl)phenyl)ethynyl)-

triisopropylsilane (200 mg, 0.522 mmol) in THF (15 mL) was added

to a mixture of Fe(κ²-dppe)(η⁵-C₅Me₅)(Cl) (4; 271 mg, 0.435 mmol)

and NH₄PF₆(106 mg, 0.652 mmol) in MeOH (30 mL) and was

stirred for 24 h. After evaporation of the solvent, the orange solid was

extracted with CH₂Cl₂and ﬁltrated. Concentration of the extract and

precipitation by excess diethyl ether (20 mL) allowed the isolation of

the crude vinylidene complex 5′b-vin[PF₆] as an orange solid in 78%

yield (380 mg, 0.340 mmol). FT-IR (KBr, cm⁻¹): ν 2208 (w, CC),

−

FeCCH), 838 (vs, PF₆). This crude vinylidene complex was

−

2149 (vs, CC), 1618 (FeCCH), 824(PF₆). ³¹P{¹H} NMR

dissolved in THF (50 mL), KOt-Bu (618 mg, 5.507 mmol) was added,

and the mixture was stirred for 60 min. The solvent was removed in

vacuo, yielding the crude product as a red solid. The solvent was

removed and the orange solid washed with n-pentane, aﬀording 5′a as

an orange solid in 88% yield (2.10 g, 2.411 mmol). Anal. Calcd for

C₅₅H₆₄P₂SiFe: C, 75.84; H, 7.41. Found: C, 75.81; H, 7.44. The

identity of 5′a was also conﬁrmed by comparison of spectroscopic data

with literature values.²²Crystals of this complex could be grown by

slow evaporation of a diethyl ether solution of the complex. UV−vis

(CH₂Cl₂, λ_max/nm [ε/10³M⁻¹cm⁻¹]): 280 [35.6], 416 [21.0].

Synthesis of Fe(κ²-dppe)(η⁵-C₅Me₅){CC(1,4-C₆H₄)CCH} (1b).

From 5a: A solution of 5a (512 mg, 0.65 mmol) and K₂CO₃(116 mg,

0.84 mmol) in THF (20 mL) and MeOH (20 mL) was stirred for 16

h, and the solvent was removed in vacuo. The product was extracted

with toluene, the solvent was removed, and the orange solid was

washed twice with n-pentane, aﬀording 1b as an orange solid in 85%

yield (394 mg, 0.55 mmol). The identity of 1b was conﬁrmed by

comparison of spectroscopic data with literature values.²²Crystals of

the complex could also be grown by slow evaporation of a benzene

solution of the complex. From 5′a: The complex 5′a (392 mg, 0.450

mmol) was solubilized in THF (30 mL), and 0.9 mL (0.900 mmol) of

a 1 M commercial solution of tetrabutylammonium ﬂuoride (TBAF)

in THF was syringed into the reaction medium, which was stirred for

24 h before the solvent was removed in vacuo. The product was

extracted with toluene, the solvent was concentrated, and the orange

solid was precipitated with n-pentane and washed twice with n-

pentane, aﬀording 1b as an orange solid in 51% yield (165 mg, 0.231

mmol). Raman (neat, cm⁻¹): ν 2097(m, CCH), 2047, 2031(s,

FeCC).

−

1

(CDCl₃, 81 MHz): δ 87.6 (s, 2P, dppe), −143.0 (sept, 1P, PF₆). H

NMR (CDCl₃, 200 MHz): δ 7.91−6.80 (m, 26H, H_Ar/dppe), 6.3 (m,

2H, H_Ar) 5.09 (m, 1H, FeCCH), 3.11 (m, 2H, CH_2/dppe), 2.55

(m, 2H, CH_2/dppe), 1.64 (s, 15H, H_Cp*), 1.20 (s, 3H, Si[CH(CH₃)₂]),

1.18 (s, 18H, Si[CH(CH₃)₂]). The vinylidene complex (360 mg, 0.323

mmol) and t-BuOK (72 mg, 0.644 mmol) were dissolved in THF (20

mL) and stirred for 2 h. After evacuation of the solvent, the orange

solid was extracted with toluene (20 mL) and ﬁltrated. The solvent

was evaporated and the orange solid was washed twice with n-pentane,

aﬀording 5′b as an orange solid in 71% yield (323 mg, 0.229 mmol).

Anal. Calcd for C₆₃H₄₆₈P₂SiFe: C, 77.92; H, 7.06. Found: C, 77.72; H,

7.18. HRMS: m/z 970.4230 [M]⁺, m/z calcd for [C₆₃H₆₈P₂Si⁵⁶Fe]

970.3914. FT-IR (KBr, cm⁻¹): ν 2205 (w, CC), 2151 (w, CC),

2043 (vs, CCFe). ³¹P{¹H} NMR (C₆D₆, 81 MHz): δ 101.2 (s). ¹H

3

NMR (C₆D₆, 200 MHz): δ 7.96 (m, 4H, H_dppe), 7.53(d, J_H,H= 7.8

Hz, 2H, H_Ar), 7.35−7.04 (m, 22H, H_Ar/dppe), 2.61 (m, 2H, CH_2/dppe),

1.83 (m, 2H, CH_2/dppe), 1.53 (s, 15H, H_Cp*), 1.21 (m, 21H,

2

Si[CH(CH₃)₂]). ¹³C{¹H} NMR (C₆D₆, 125 MHz): δ 145.4 (t, J_P,C

= 38 Hz, FeCC,), 141.5−128.5 (C_Ar, C_ArH), 122.5 (FeCC), 117.8

(C_Ar), 108.7, 94.7, 92.9, 90.7 (CC), 88.9 (C₅(CH₃)₅), 31.8 (m,

CH₂), 19.7 (Si[CH(CH₃)₂]), 12.5 (Si[CH(CH₃)₂]), 11.2 (C₅(CH₃)₅).

UV−vis (CH₂Cl₂, λ_max/nm [ε/10³M⁻¹cm⁻¹]): 318 [30.3], 456

[17.9]. CV (CH₂Cl₂, 0.1 M [NBu₄][PF₆], 20 °C, 0.1 V s⁻¹) E° in V vs

SCE(ΔE_pin V, i_pa/i_pc): −0.12 (0.074, 1.0).

Synthesis of Fe(κ²-dppe)(η⁵-C₅Me₅){[CC(1,4-C₆H₄)]₂CCH}

(1c). From 5b: A solution of 5b (620 mg, 0.74 mmol), and K₂CO₃

(142 mg, 1.03 mmol) in THF (40 mL) and MeOH (30 mL) was

stirred for 16 h, and the solvent was removed in vacuo. The product

was extracted with toluene, the solvent was removed, and the orange

solid was washed twice with n-pentane, aﬀording 1c as an orange solid

in 86% yield (490 mg, 0.64 mmol). From 5′b: TBAF (1 M solution in

THF, 0.041 mL, 0.041 mmol) was added to a solution of 5′b (200 mg,

0.205 mmol) in THF (30 mL), and the reaction mixture was stirred

for 48 h. After evacuation of the solvent, the orange solid was extracted

with toluene (20 mL) and ﬁltrated. The solvent was evaporated under

reduced pressure, and the orange solid was washed twice with n-

pentane, aﬀording 1c as an orange solid in 43% yield (250 mg, 0.092

mmol). Crystals of the complex were grown by slow diﬀusion of n-

pentane vapors into a toluene solution of the complex. Anal. Calcd for

C₅₄H₄₈P₂Fe: C, 79.60; H, 5.94. Found: C, 79.32; H, 6.23. HRMS: m/z

814.2590 [M]⁺, m/z calcd for [C₅₄H₄₈P₂⁵⁶Fe] 814.2581. FT-IR (KBr,

cm⁻¹): ν 3280 (w, C−H), 2206 (w, CC), 2106 (vw, CC),

2045 (vs, CCFe). Raman (neat, cm⁻¹): ν 2206 (s, CC), 2107 (w,

CC), 2042 (m, CCFe). ³¹P{¹H} NMR (C₆D₆, 81 MHz): δ 101.2

(s). ¹H NMR (C₆D₆, 200 MHz): δ 7.95 (m, 4H), 7.55−6.90 (m, 24H,

H_Ar/dppe), 2.75 (s, 1H, CCH), 2.65 (m, 2H, CH_2/dppe), 1.62 (m, 2H,

CH_2/dppe), 1.52 (s, 15H, H_Cp*). ¹³C{¹H} NMR (C₆D₆, 50 MHz): δ

Synthesis of Fe(κ²-dppe)(η⁵-C₅Me₅){[CC(1,4-C₆H₄)]₂CCSiMe₃}

(5b). A solution of Fe(κ²-dppe)(η⁵-C₅Me₅)Cl (4; 630 mg, 1.01 mmol),

((4-((4-ethynylphenyl)ethynyl)phenyl)ethynyl)trimethylsilane (302

mg, 1.01 mmol), and KPF₆(375 mg, 2.03 mmol) in THF (20 mL)

and MeOH (20 mL) was stirred for 16 h. The solvent was removed in

vacuo. The brown solid was extracted with CH₂Cl₂and the solvent

removed in vacuo, providing the crude vinylidene complex 5b-vin[PF₆]

as a brown precipitate. FT-IR (KBr, cm⁻¹): ν 2210 (w, CC), 2155

−

(m, CC), 1625 (s, FeCCH), 824 (vs, PF₆). ³¹P{¹H} NMR

−

1

(CDCl₃, 81 MHz): δ 86.5 (s, 2P, dppe), −143.0 (sept, 1P, PF₆). H

NMR (CDCl₃, 200 MHz): δ 7.95−7.00 (m, 26H, H_Ar/dppe), 6.29 (m,

2H, H_Ar) 5.07 (m, 1H, FeCCH), 3.14 (m, 2H, CH_2/dppe), 2.55

(m, 2H, CH_2/dppe), 1.64 (s, 15H, H_Cp*), 0.28 (s, 9H, H_SiMe3). This

crude vinylidene complex was dissolved in THF (20 mL), DBU (0.3

mL, 2.01 mmol) was added, and the mixture was stirred for 10 min.

The solvent was removed in vacuo, yielding the crude product as a red

solid. The product was puriﬁed on a short column of deactivated silica

under an inert atmosphere, eluting with toluene. The solvent was

removed and the red solid washed with n-pentane, aﬀording 5b as an

orange solid in 74% yield (628 mg, 0.75 mmol). HRMS: m/z 886.2968

[M]^+•, m/z calcd for [C₅₇H₅₆P₂Si⁵⁶Fe] 886.29705. FT-IR (KBr,

cm⁻¹): ν 2206 (w, CC), 2155 (w, CC), 2046 (vs, CCFe).

2

146.9 (t, J_P,C= 38 Hz, FeCC), 139.9−128.3 (C_Ar, CH_Ar), 125.7

(C_Ar), 125.1 (C_Ar), 121.7 (FeCC), 116.9 (C_Ar), 94.0, 89.8 (CC),

88.9 (C₅(CH₃)₅), 83.8 (CCH), 79.2 (CCH), 31.0 (m, CH₂), 10.5

(C₅(CH₃)₅).

1

³¹P{¹H} NMR (C₆D₆, 81 MHz): δ 101.2 (s). H NMR (C₆D₆, 200

Synthesis of Fe(κ²-dppe)(η⁵-C₅Me₅){[CC(1,4-C₆H₄)]₃CCSiMe₃}

MHz): δ 7.95 (m, 4H, H_dppe), 7.52 (m, 2H, H_Ar), 7.35−6.90 (m, 22H,

H_Ar/dppe), 2.65 (m, 2H, CH_2/dppe), 1.81 (m, 2H, CH_2/dppe), 1.52 (s,

(5c). A solution of Fe(κ²-dppe)(η⁵-C₅Me₅) (4; 255 mg, 0.41 mmol),

4377

dx.doi.org/10.1021/om400515g | Organometallics 2013, 32, 4366−4381

Organometallics

Article

3

9H, H_SiMe). ¹³C{¹H} NMR (C₆D₆, 101 MHz): δ 141.9 (t, J_C,P= 39

((4-((4-((4-ethynylphenyl)ethynyl)phenyl)ethynyl)phenyl)ethynyl)-

trimethylsilane (165 mg, 0.41 mmol), and KPF₆(136 mg, 0.74 mmol)

in THF (30 mL) and MeOH (30 mL) was stirred for 16 h. The

solvent was removed in vacuo, and the brown solid was extracted with

CH₂Cl₂. The solvent volume was reduced to 10 mL, n-pentane (50

mL) was added to the stirring solution, and the crude vinylidene

complex 5c-vin[PF₆] was collected as a brown precipitate. FT-IR

(KBr, cm⁻¹): ν 2205 (w, CC), 2154 (m, CC), 1618 (s, FeC

3

Hz, FeCC), 140.3 and 138.6 (m, C_Ar/dppe), 135.1 (m, CH_Ar/dppe),

134.8 (CH_Ar), 132.3 (C_Ar), 131.5 (CH_Ar), 129.9 and 129.7 (C_Ar/dppe),

129.0 (CH_Ar), 128.1 (m, C_Ar/dppe), 127.2 (CH_Ar), 124.1 (C_Ar), 120.3

(FeCC), 107.9 (CC), 92.6, (CC), 88.5 (C₅(CH₃)₅), 31.6 (m,

CH₂), 11.0 (C₅(CH₃)₅), 0.8 (Si(CH₃)₃). UV−vis (CH₂Cl₂, λ_max/nm

[ε/10³M⁻¹cm⁻¹]): 354 [10.3], 406 [sh, 6.2].

Synthesis of Fe(κ²-dppe)(η⁵-C₅Me₅)[CC(1,3-C₆H₄)CCH] (2). A

solution of 6 (420 mg, 0.53 mmol) and K₂CO₃(89 mg, 0.64 mmol) in

THF (20 mL) and MeOH (20 mL) was stirred for 16 h. The solvent

was removed in vacuo. The product was puriﬁed on a short column of

deactivated silica under an inert atmosphere, eluting with toluene. The

solvent was removed in vacuo and the orange solid was washed with n-

pentane, aﬀording 2 as an orange solid in 94% yield (360 mg, 0.50

mmol). Crystals of the complex were grown by slow evaporation of a

diethyl ether solution of the complex. HRMS: m/z 714.2260 [M]⁺, m/

z calcd for [C₄₆H₄₄P₂⁵⁶Fe] 714.2268. FT-IR (KBr, cm⁻¹): ν 3298 (m,

C−H), 2105 (w, CCH), 2044 (vs, FeCC). Raman (neat,

cm⁻¹): ν 2107 (w, CCH), 2044 (vs, FeCC). ³¹P{¹H} NMR

(C₆D₆, 81 MHz): δ 101.3 (s). ¹H NMR (C₆D₆, 200 MHz): δ 7.96 (m,

4H, H_dppe), 7.54 (m, 1H, H_Ar), 7.35−6.90 (m, 19H, H_Ar/dppe), 2.73 (s,

1H, CCH), 2.56 (m, 2H, CH_2/dppe), 1.79 (m, 2H, CH_2/dppe), 1.50 (s,

15H, H_Cp*). ¹³C{¹H} NMR (CD₂Cl₂, 50 MHz): δ 139.5−119.3

(FeCC, C_Ar, CH_ArH, FeCC), 88.0 (C₅(CH₃)₅), 84.6(CCH),

76.2 (CCH), 30.8 (m, CH₂), 10.2 (C₅(CH₃)₅). UV−vis (CH₂Cl₂,

λ_max/nm [ε/10³M⁻¹cm⁻¹]): 358 [8.6], 398 [sh, 6.1].

−

CH), 839 (vs, PF₆). ³¹P{¹H} NMR (CDCl₃, 81 MHz): δ 87.7 (s, 2P,

dppe), −143.0 (sept., 1P, PF₆). ¹H NMR (CDCl₃, 200 MHz): δ 8.10−

7.10 (m, 30H, H_Ar/dppe), 6.30 (m, 2H, H_Ar), 5.10 (m, 1H, FeC

CH), 3.15 (m, 2H, CH_2/dppe), 2.65 (m, 2H, CH_2/dppe), 1.62 (s, 15H,

C₅(CH₃)₅), 0.30 (s, 9H, Si(CH₃)₃). This vinylidene complex (367 mg,

0.324 mmol) was dissolved in THF (30 mL), DBU (0.15 mL, 1.0

mmol) was added, and the mixture was stirred for 30 min. The solvent

was removed in vacuo, yielding the crude product as a red solid. The

product was puriﬁed on a short column of deactivated silica under an

inert atmosphere, eluting with toluene. The solvent was removed and

the red solid washed twice with n-pentane, aﬀording 5c as an orange

solid in 76% yield (243 mg, 0.246 mmol). HRMS: m/z 986.3276

[M]⁺, m/z calcd for [C₆₅H₆₀SiP₂⁵⁶Fe] 986.32835. FT-IR (KBr, cm⁻¹):

ν 2205 (vw, CC), 2156 (w, CCSi), 2043 (s, FeCC). ³¹P{¹H}

1

NMR (C₆D₆, 81 MHz): δ 101.0 (s). H NMR (C₆D₆, 200 MHz): δ

7.95 (m, 4H, H_dppe), 7.58−6.95 (m, 28H, H_Ar/dppe), 2.56 (m, 2H,

CH_2/dppe), 1.82 (m, 2H, CH_2/dppe), 1.52 (s, 15H, H_Cp*), 0.25 (m, 9H,

H_SiMe). ¹³C{¹H} NMR (C₆D₆, 125 MHz): δ 139.2−127.4 (C_Ar, CH_Ar),

3

125.6 124.9, 123.6, 123.5, 122.4 (C_Ar); 121.7 (FeCC); 116.0 (C_Ar);

104.8 (CCSi); 96.8 (CCSi); 93.4, 91.5, 90.8, 89.3 (CC); 88.2

(C₅(CH₃)₅); 30.9 (m, CH₂); 10.2 (C₅(CH₃)₅); 0.1 (Si(CH₃)₃)). UV−

vis (CH₂Cl₂, λ_max/nm [ε/10³M⁻¹cm⁻¹]): 340 [51.8], 476 [8.1].

Synthesis of Fe(κ²-dppe)(η⁵-C₅Me₅){[CC(1,4-C₆H₄)]₃CCH}

(1d). A solution of 5c (211 mg, 0.21 mmol) and K₂CO₃(62 mg,

0.45 mmol) in THF (40 mL) and MeOH (30 mL) was stirred for 16

h, and the solvent was removed in vacuo. The product was extracted

with toluene, the solvent was removed, and the orange solid was

washed twice with n-pentane, aﬀording 1d as an orange solid in 72%

yield (138 mg, 0.15 mmol). HRMS: m/z 914.2879 [M]⁺, m/z calcd for

[C₆₂H₅₂P₂⁵⁶Fe] 914.2894. FT-IR (KBr, cm⁻¹): ν 3280 (w, C−H),

2203 (w, CC), 2107 (vw, CC), 2040 (vs, FeCC). Raman

(neat, cm⁻¹): ν 2207 (s, CC), 2175 (w sh, CC), 2104 (w, C

CH), 2037 (m, FeCC). ³¹P{¹H} NMR (C₆D₆, 81 MHz): δ 101.2

Synthesis of Fe(κ²-dppe)(η⁵-C₅Me₅)[CC(1,4-C₆H₄)CHCH₂] (3).

A solution of Ph₃PCH₃Br (500 mg, 1.40 mmol) in THF (25 mL) was

cooled to 0 °C, and n-BuLi in hexanes (2.0 M solution, 0.7 mL, 1.40

mmol) was added. After 30 min the temperature was lowered to −90

°C, and a solution of 7 (400 mg, 0.56 mmol) in THF (25 mL) was

added. After 30 min at −90 °C the solution was allowed to return to

room temperature and stirred there for 16 h. The solvent was removed

in vacuo. The product was puriﬁed on a short column of deactivated

silica under an inert atmosphere, eluting with toluene, and the solvent

was removed in vacuo. The product was washed with n-pentane,

aﬀording 3 as an orange solid in 88% yield (350 mg, 0.49 mmol).

Crystals of the complex could also be grown by slow diﬀusion of n-

pentane in a CH₂Cl₂solution of the complex. Anal. Calcd for

C₄₆H₄₆P₂Fe: C, 77.09; H, 6.47. Found: C, 77.09; H, 6.38. HRMS: m/z

716.2420 [M]⁺, m/z calcd for [C₄₆H₄₆P₂⁵⁶Fe] 716.2420. FT-IR (KBr,

cm⁻¹): ν 2054, 2036 (vs, FeCC), 1620 (CC). Raman (neat,

cm⁻¹): ν 2053, 2038 (vs, FeCC), 1622 (w, CC). ³¹P{¹H} NMR

(C₆D₆, 81 MHz): δ 101.4 (s). ¹H NMR (C₆D₆, 200 MHz): δ 8.01 (m,

4H, H_dppe), 7.36−7.04 (m, 20H, H_Ar/dppe), 6.66 (dd, 1H, CHCH₂),

5.59 (dd, 1H, CHCH₂), 5.03 (dd, 1H, CHCH₂), 2.56 (m, 2H,

CH_2/dppe), 1.79 (m, 2H, CH_2/dppe), 1.54 (s, 15H, H_Cp*). ¹³C{¹H} NMR

(CDCl₃, 50 MHz): δ 139.6−120.4 (FeCC, C_Ar, CH_Ar, Fe−CC,

−CHCH₂), 111.0 (−CHCH₂), 87.9 (C₅(CH₃)₅), 30.7 (m, CH₂),

10.2 (C₅(CH₃)₅.

1

(s). H NMR (CD₂Cl₂, 200 MHz): δ 7.95 (m, 4H, H_dppe), 7.55−6.90

(m, 19H, H_Ar/dppe), 3.11 (s, 1H, CCH) 2.56 (m, 2H, CH_2/dppe), 1.79

(m, 2H, CH_2/dppe), 1.49 (s, 15H, H_Cp*). ¹³C{¹H} NMR (CD₂Cl₂, 125

MHz): δ 138.8−127.4 (C_Ar, CH_Ar), 123.8 (C_Ar), 123.4 (C_Ar), 122.5

(C_Ar), 122.4 (FeCC), 115.6 (C_Ar), 91.2−91.1 (CC), 88.3

(C₅(CH₃)₅), 83.4 (CCH), 79.4 (CCH), 30.7 (m, CH₂), 10.2

(C₅(CH₃)₅), FeCC not detected.

Synthesis of Fe(κ²-dppe)(η⁵-C₅Me₅){CC(1,3-C₆H₄)CCSiMe₃}

(6). A solution of Fe(κ²-dppe)(η⁵-C₅Me₅)(Cl) (4; 500 mg, 0.80

mmol), ((3-ethynylphenyl)ethynyl)trimethylsilane (198 mg, 1.00

mmol), and KPF₆(184 mg, 1.00 mmol) in THF (20 mL) and

MeOH (20 mL) was stirred for 16 h. The solvent was removed in

vacuo. The brown solid was extracted with CH₂Cl₂and the solvent

removed in vacuo, providing the crude vinylidene complex 6-vin[PF₆]

as a brown-orange solid. This crude vinylidene complex was dissolved

in CH₂Cl₂(20 mL), DBU (0.24 mL, 1.61 mmol) was added, and the

mixture was stirred for 1 h. The solvent was removed in vacuo, yielding

the crude product as a red solid. The product was puriﬁed on a short

column of deactivated silica under an inert atmosphere, eluting with

toluene. The solvent was removed and the red solid washed twice with

n-pentane, aﬀording 6 as an orange solid in 78% yield (490 mg, 0.62

mmol). Crystals of the complex were grown by slow diﬀusion of n-

pentane vapors in a toluene solution of the complex. Anal. Calcd for

C₄₉H₅₂SiP₂Fe: C, 74.80; H, 6.66. Found: C, 75.11; H, 6.87. HRMS:

m/z 786.2653 [M]⁺, m/z calcd for [C₄₉H₅₂SiP₂⁵⁶Fe] 786.2663. FT-IR

(KBr, cm⁻¹): ν 2149 (s, CCSi), 2043 (vs, FeCC). ³¹P{¹H} NMR

(C₆D₆, 81 MHz): δ 101.3 (s). ¹H NMR (C₆D₆, 200 MHz): δ 7.95 (m,

4H, H_dppe), 7.54 (m, 1H, H_Ar), 7.35−6.90 (m, 19H, H_Ar/dppe), 2.56 (m,

2H, CH_2/dppe), 1.79 (m, 2H, CH_2/dppe), 1.49 (s, 15H, H_Cp*), 0.24 (m,

Synthesis of [Fe(κ²-dppe)(η⁵-C₅Me₅)(CC(1,4-C₆H₄)CCH)][PF₆]

(1b[PF₆]). [Fe(η⁵-C₅H₅)₂][PF₆] (57 mg, 0.172 mmol) and 1b (130

mg, 0.182 mmol) were dissolved in CH₂Cl₂(20 mL) and stirred for 1

h. The solvent volume was reduced to ∼5 mL in vacuo, and n-pentane

was added (40 mL) to give a dark precipitate. Decantation and

subsequent washings with toluene (2 × 5 mL), diethyl ether (2 × 5

mL), and n-pentane (2 × 5 mL) yielded [Fe(κ²-dppe)(η⁵-C₅Me₅)-

{CC(1,4-C₆H₄)CCH}][PF₆] (1b[PF₆]) (78 mg, 0.091 mmol,

53%). FT-IR (KBr, cm⁻¹): ν 3274 (m, C−H), 2102 (w, CCH),

1994 (s, FeCC). ¹H NMR (CD₂Cl₂, 500 MHz): δ 30.4 (s, H₂), 7.9

(s, H_dppe), 7.2 (s, H_dppe), 6.8 (s, H_dppe), 6.2 (s, H_dppe), 3.6 (s, H_dppe), 1.3

(s, H_dppe), −2.8 (s, CH_2/dppe), −10.3 (br s, C₅(CH₃)₅), −20.1 (s, C

CH), −39.6 (br s, H₃).

Synthesis of [Fe(κ²-dppe)(η⁵-C₅Me₅)({CC(1,4-C₆H₄)}₂CCH)]-

[PF₆] (1c[PF₆]). [Fe(η⁵-C₅H₅)₂][PF₆] (64 mg, 0.193 mmol) and 1c

(165 mg, 0.203 mmol) were dissolved in CH₂Cl₂(20 mL) and stirred

for 1 h. The solvent volume was reduced to ∼5 mL in vacuo, and n-

pentane was added (50 mL) to give a dark precipitate. Decantation

and subsequent washings with toluene (3 × 5 mL), diethyl ether (3 ×

5 mL), and n-pentane (2 × 5 mL) yielded [Fe(κ²-dppe)(η⁵-

4378

dx.doi.org/10.1021/om400515g | Organometallics 2013, 32, 4366−4381

Organometallics

Article

C₅Me₅){CC(1,4-C₆H₄)CC(1,4-C₆H₄)CCH}][PF₆] (1c[PF₆])

(123 mg, 0.128 mmol, 66%). FT-IR (KBr, cm⁻¹): ν 3269 (s, C−H),

2207 (w, CC), 2107 (vw, CC−H), 1991 (s, CC−Fe). ¹H

NMR (CD₂Cl₂, 500 MHz): δ 31.3 (s, H₂), 9.3 (s, H₄), 7.9 (s, H_dppe),

7.2 (s, H_dppe), 6.8 (s, H_dppe), 6.2 (s, H_dppe), 3.6 (s, H_Ar/dppe+ H₅), 1.7 (s,

H_dppe), 0.9 (s, CCH), −2.8 (s, CH_2/dppe), −10.3 (br s, C₅(CH₃)₅),

−41.8 (br s, H₃).

optimization was checked by calculations of the harmonic vibrational

frequencies using the revPBE functional. Computed EPR properties

were accomplished using the ESR procedure developed by van Lenthe

and co-workers.⁶³The g-tensor components were obtained using self-

consistent spin-unrestricted DFT calculations after incorporating the

relativistic spin−orbit coupling by ﬁrst-order perturbation theory from

a ZORA Hamiltonian, using the PW91 functional⁶⁰for nonlocal

corrections to the exchange and correlation energies.

Time-dependent density functional theory (TD-DFT) calcula-

tions⁶⁴were performed on the optimized structures of neutral and

open-shell systems using the hybrid PBE0 functional⁶⁵and taking into

account the solvation eﬀects using the conductor-like screening model

(COSMO)⁶⁶with a dielectric constant simulating dichloromethane

solvent. Molecular orbitals and spin densities were plotted with the

ADF-GUI package.⁵⁸

Crystallography. Crystals of 1b, 1c·C₇H₈, 2, 2[PF₆], 3, 5′a, and 6

were studied on an Oxford Diﬀraction Xcalibur Saphir 3 with graphite-

monochromatized Mo Kα radiation. The cell parameters were

obtained with Denzo and Scalepack with 10 frames (psi rotation: 1°

per frame).⁶⁷The data collection⁶⁸details (2θ_max, number of frames, Ω

rotation, scan rate, and HKL range) for 1b, 1c·C₇H₈, 2, 2[PF₆], 3, 5′a,

and 6 are given in the Supporting Information (Table S1). Subsequent

data reduction with Denzo and Scalepack⁶⁷gave the independent

reﬂections. The structures were solved with SIR-97, which revealed the

non-hydrogen atoms.⁶⁹After anisotropic reﬁnement, the remaining

atoms were found in Fourrier diﬀerence maps. The complete

structures were then reﬁned with SHELXL97⁷⁰by the full-matrix

least-squares methods against F². All non-hydrogen atoms were reﬁned

with anisotropic thermal displacement parameters. Hydrogen atoms

were included as a riding model based on the atom to which they are

bonded. Atomic scattering factors were taken from the literature.⁷¹

Thermal ellipsoid plots of 1b, 1c, 2, 2[PF₆], 3, 5′a, and 6 were realized

with ORTEP.

Synthesis of [Fe(κ²-dppe)(η⁵-C₅Me₅)({CC(1,4-C₆H₄)}₃CCH)]-

[PF₆] (1d[PF₆]). [Fe(η⁵-C₅H₅)₂][PF₆] (33 mg, 0.100 mmol) and 1d

(95 mg, 0.104 mmol) were dissolved in CH₂Cl₂(10 mL) and stirred

for 30 mn. The solvent volume was reduced to ∼5 mL in vacuo, and n-

pentane was added (30 mL) to precipitate a dark brown precipitate.

Decantation and subsequent washings with toluene (2 × 5 mL),

diethyl ether (2 × 5 mL), and n-pentane (2 × 5 mL) yielded [Fe(κ²-

dppe)(η⁵-C₅Me₅){CC(1,4-C₆H₄)CC(1,4-C₆H₄)CC(1,4-

C₆H₄)CCH}][PF₆] (1d[PF₆]; 79 mg, 0.070 mmol, 70%). FT-IR

(KBr, cm⁻¹): ν 3275 (s, C−H), 2208 (w, CC), 2105 (vw, C

1

CH), 1989 (s, CCFe). H NMR (CD₂Cl₂, 500 MHz): δ 31.6 (s,

H₂), 9.5 (s, H₄), 7.9 (s, H_dppe*), 7.7−6.8 (s, H_Ar/dppe), 3.6 (s, H_Ar/dppe

+

H₆*), 2.5 (s, CCH*), 1.7 (s, H_dppe*), −2.8 (s, CH_2/dppe), −10.3 (br

s, C₅(CH₃)₅), −42.4 (br s, H₃); “*” indicates a tentative attribution.

Synthesis of [Fe(κ²-dppe)(η⁵-C₅Me₅){CC(1,3-C₆H₄)CCH}][PF₆]

(2[PF₆]). [Fe(η⁵-C₅H₅)₂][PF₆] (40 mg, 0.121 mmol) and 2 (90 mg,

0.126 mmol) were dissolved in CH₂Cl₂(20 mL) and stirred for 2 h.

The solvent volume was reduced to ∼5 mL in vacuo, and n-pentane

was added (50 mL) to give a red-brown precipitate. Decantation and

subsequent washings with toluene (2 × 5 mL), diethyl ether (2 × 5

mL), and n-pentane (2 × 5 mL) yielded [Fe(κ²-dppe)(η⁵-C₅Me₅)-

{CC(1,3-C₆H₄)CCH}][PF₆] (2[PF₆]; 74 mg, 0.086 mmol,

71%). Crystals of the complex were grown by slow diﬀusion of n-

pentane in a CH₂Cl₂solution of the complex. FT-IR (KBr, cm⁻¹): ν

3287 (s, C−H), 2008 (w, CCFe). ¹H NMR (CD₂Cl₂, 500 MHz):

δ 27.8 (s, H₃), 8.9 (s, CCH), 7.9 (s, H_dppe), 6.9 (s, H_dppe), 6.4 (s,

H_dppe), 6.2 (s, H_dppe), 3.6 (s, H_dppe), 1.6 (s, H_dppe), −2.7 (s, CH_2/dppe),

−10.5 (br s, C₅(CH₃)₅), −36.7 (br s, H₂), −38.2 (br s, H₄+ H₄′).

Synthesis of [Fe(κ²-dppe)(η⁵-C₅Me₅){CC(1,4-C₆H₄)CHCH₂}]-

[PF₆] (3[PF₆]). [Fe(η⁵-C₅H₅)₂][PF₆] (55 mg, 0.166 mmol) and 3

(125 mg, 0.174 mmol) were dissolved in CH₂Cl₂(20 mL) and stirred

for 2 h. The solvent volume was reduced to ∼5 mL in vacuo, and n-

pentane was added (50 mL) to give a dark brown precipitate.

Decantation and subsequent washings with toluene (2 × 5 mL),

diethyl ether (2 × 5 mL), and n-pentane (2 × 5 mL) yielded [Fe(κ²-

dppe)(η⁵-C₅Me₅){CC(1,4-C₆H₄)CHCH₂})][PF₆] (3[PF₆]; 97

mg, 0.113 mmol, 68%). FT-IR (KBr, cm⁻¹): ν 1992 (w, CCFe),

1595 (w, CCH). ¹H NMR (CD₂Cl₂, 500 MHz): δ 32.5 (s, H₃), 28.1

(s, CHCH₂), 10.0 (s, H_dppe), 7.2 (s, H_dppe), 6.8 (s, H_dppe), 6.3 (s,

H_dppe), 3.7 (s, H_dppe), 2.0 (s, H_dppe), −2.9 (s, CH_2/dppe), −10.3 (br s,

C₅(CH₃)₅), −22.3 (s, CHCH₂), −24.6 (s, CHCH₂), −46.6 (br s,

H₅), −46.6 (br s, H_Ar).

ASSOCIATED CONTENT

* Supporting Information

■

S

Synthesis of selected organic precursors, selected spectroscopic

and crystallographic data for 1b−d and 1b−d[PF₆] complexes.

Crystallographic (CIF) ﬁle for 1b, 1c·C₇H₈, 2, 2[PF₆], 3, 5′a,

and 6. Cartesian coordinates for all calculated geometries. Final

atomic positional coordinates, with estimated standard

deviations, bond lengths, angles, and anisotropic thermal

parameters have been deposited at the Cambridge Crystallo-

graphic Data Centre and were allocated the deposition

numbers CCDC 299572, 942561, 676885, 918574, 668871,

298157, and 664966, respectively. This material is available free

of charge via the Internet at http://pubs.acs.org.

Solvent-Glass ESR Measurements. A 1−2 mg sample of an

Fe(II) complex was introduced in an ESR tube under an argon-ﬁlled

atmosphere along with an excess of [FcH][PF₆], and a 1:1 mixture of

degassed dichloromethane/1,2-dichloroethane was transferred to

dissolve the solid. For solvent glass measurements at 77 K, the

solvent mixture was frozen immediately in liquid nitrogen, and the

tubes were sealed and transferred in the ESR cavity.

AUTHOR INFORMATION

Corresponding Author

*E-mail: bruno.fabre@univ-rennes1.fr, jean-francois.halet@

univ-rennes1.fr, frederic.paul@univ-rennes1.fr.

■

Author Contributions

Computational Details. DFT calculations were carried out on the

1b−d^0/+, 2^0/+, and 3^0/+complexes using the Amsterdam Density

Functional (ADF) program.^57,58Electron correlation was treated

within the local density approximation (LDA) in the Vosko−Wilk−

Nusair parametrization.⁵⁹Nonlocal corrections were added to the

exchange and correlation energies using the Perdew−Wang 1991

(PW91)⁶⁰or revPBE⁶¹functionals. Calculations were performed using

the standard ADF triple-ζ quality basis set for the atom valence shells

augmented with a 2p polarization function for H, a 3d polarization

function for C and P, and a 4p for Fe. Orbitals up to 1s, 2p, and 3p

were kept frozen for C, P, and Fe, respectively. Full geometry

optimizations (assuming C₁symmetry) were carried out on each

complex, using the analytical gradient method implemented by

Versluis and Ziegler.⁶²The nature of the stationary points after

The manuscript was written through contributions of all

authors. All authors have given approval to the ﬁnal version of

the manuscript.

Notes

The authors declare no competing ﬁnancial interest.

ACKNOWLEDGMENTS

■

F.P. and N.G. thank Region Bretagne for ﬁnancial support. H.S.

acknowledges a Ph.D. fellowship from the PROFAS French-

Algerian Programme. The Agence Nationale de la Recherche

(ANR 09-BLAN-0109) and CNRS are fully acknowledged for

ﬁnancial support. T. Guizouarn and F. Justaud (Rennes) are

4379

dx.doi.org/10.1021/om400515g | Organometallics 2013, 32, 4366−4381

Organometallics

Article

T. E.; Fischer, D. A.; Wudl, F.; Kramer, E. J. Langmuir 2010, 26,

17000−17012.

acknowledged for experimental assistance in ESR and A. Barnes

for her editorial assistance.

(19) Le Narvor, N.; Toupet, L.; Lapinte, C. J. Am. Chem. Soc. 1995,

117, 7129−7138.

REFERENCES

(20) Green, K.; Gauthier, N.; Sahnoune, H.; Halet, J.-F.; Paul, F.;

Fabre, B. To be submitted.

(21) Dahlenburg, L.; Weiss, A.; Bock, M.; Zhal, A. J. Organomet.

■

(1) (a) Liu, Z.; Yasseri, A. A.; Lindsey, J. S.; Bocian, D. F. Science

2003, 302, 1543−1545. (b) Lindsey, J. S.; Bocian, D. F. Acc. Chem. Res.

2011, 44, 638−650.

Chem. 1997, 541, 465−471.

(22) Courmarcel, J.; Le Gland, G.; Toupet, L.; Paul, F.; Lapinte, C. J.

Organomet. Chem. 2003, 670, 108−122.

(2) Roth, K. M.; Yasseri, A. A.; Liu, Z.; Dabke, R. B.; Malinovskii, V.;

Schweikhart, K.-H.; Yu, L.; Tiznado, H.; Zaera, F.; Lindsey, J. S.; Kuhr,

W. G.; Bocian, D. F. J. Am. Chem. Soc. 2003, 125, 505−517.

(3) (a) Tour, J. M. Acc. Chem. Res. 2000, 33, 791−803. (b) Joachim,

C. New J. Chem. 1991, 15, 223−229.

(4) Li, Q.; Mathur, G.; Gowda, G.; Surthi, S.; Zhao, Q.; Yu, L.;

Lindsey, J. S.; Bocian, D. F.; Misra, V. Adv. Mater. 2004, 16, 133−137.

Li, C.; Ly, J.; Lei, B.; Fan, W.; Zhang, D.; Han, J.; Meyyapan, M.;

Thompson, M.; Zhou, C. J. Phys. Chem. B 2004, 108, 9646−9649.

(5) Qi, H.; Sharma, S.; Li, Z.; Snider, G. L.; Orlov, A. O.; Lent, C. S.;

Fehlner, T. P. J. Am. Chem. Soc. 2003, 125, 15250−15259.

(6) Qi, H.; Ghupta, A.; Noll, B. C.; Snider, G. L.; Lu, Y.; Lent, C. S.;

Fehlner, T. P. J. Am. Chem. Soc. 2005, 127, 15218−15227.

(7) (a) Aguirre-Etcheverry, P.; O’Hare, D. Chem. Rev. 2010, 4839−

4864. (b) Ren, T. Organometallics 2005, 24, 4854−4870. (c) Ceccon,

A.; Santi, S.; Orian, L.; Bisello, A. Coord. Chem. Rev. 2004, 248, 683−

724.

̂

(23) Roger, C.; Hamon, P.; Toupet, L.; Rabaa, H.; Saillard, J.-Y.;

Hamon, J.-R.; Lapinte, C. Organometallics 1991, 10, 1045−1054.

(24) Denis, R.; Toupet, L.; Paul, F.; Lapinte, C. Organometallics

2000, 19, 4240−4251.

(25) Rodríguez, J. G.; Tejedor, J. L.; La Parra, T.; Díaz, C.

Tetrahedron 2006, 62, 3355−3361.

(26) Taratula, O.; Rochford, J.; Piotrowiak, P.; Galoppini, E.; Carlisle,

R. A.; Meyer, G. J. J. Phys. Chem. B 2006, 110, 15734−15741.

(27) We also report the synthesis of the known complexes 5a and 5′a

by this route, since these were previously isolated by catalytic coupling

from 1a.²²

(28) Makowska-Janusik, M.; Kityk, I. V.; Gauthier, N.; Paul, F. J. Phys.

Chem. C 2007, 111, 12094−12099.

(29) Wittig, G.; Schollkopf, U. Chem. Ber. 1954, 87, 1318−1330.

̈

(30) Costuas, K.; Paul, F.; Toupet, L.; Halet, J.-F.; Lapinte, C.

Organometallics 2004, 23, 2053−2068.

(8) (a) Paul, F.; Lapinte, C. Coord. Chem. Rev. 1998, 178/180, 431−

509. (b) Halet, J.-F.; Lapinte, C. Coord. Chem. Rev. 2013, 257, 1584−

1613.

(31) This signal exhibits a diagnostic coupling of ca. 39 Hz to the

phosphorus atoms of dppe.

(32) Cifuentes, M. P.; Humphrey, M. G.; Morrall, J. P.; Samoc, M.;

Paul, F.; Roisnel, T.; Lapinte, C. Organometallics 2005, 24, 4280−

4288.

(9) (a) Higgins, S. J.; Nichols, R. J.; Martin, S.; Cea, P.; van der Zant,

H. S. J.; Richter, M. M.; Low, P. J. Organometallics 2011, 30, 7−12.

(b) Humphrey, M. G.; Cifuentes, M. P.; Samoc, M. Top. Organomet.

Chem. 2011, 28, 57−73. (c) Akita, M.; Koike, T. J. Chem. Soc., Dalton

Trans. 2008, 3523−3530.

(33) Paul, F.; Mevellec, J.-Y.; Lapinte, C. J. Chem. Soc., Dalton Trans.

2002, 1783−1790.

(34) Hansch, C.; Leo, A.; Taft, R. W. Chem. Rev. 1991, 91, 165−195.

(10) Wong, K. M.-C.; Lam, S. C.-F.; Ko, C.-C.; Zhu, N.; Yam, V. W.-

́

(35) Paul, F.; Toupet, L.; Thepot, J.-Y.; Costuas, K.; Halet, J.-F.;

́

W.; Roue, S.; Lapinte, C.; Fathallah, S.; Costuas, K.; Kahlal, S.; Halet,

J.-F. Inorg. Chem. 2003, 42, 7086−7097.

(11) (a) Buriak, J. M. Chem. Rev. 2002, 102, 1271−1308. (b) Wayner,

D. D. M.; Wolkow, R. A. J. Chem. Soc., Perkin Trans. 2 2002, 23−34.

(c) Ciampi, S.; Harper, J. B.; Gooding, J. J. Chem. Soc. Rev. 2010, 39,

2158−2183. (d) Boukherroub, R. Curr. Opin. Solid State Mater. Sci.

2005, 9, 66−72.

Lapinte, C. Organometallics 2005, 24, 5464−5478.

(36) Paul, F. Work in progress.

(37) Paul, F.; Lapinte, C. In Unusual Structures and Physical Properties

in Organometallic Chemistry; Gielen, M., Willem, R., Wrackmeyer, B.,

Eds.; Wiley: San Francisco, CA, 2002; pp 219−295.

(38) Paul, F.; da Costa, G.; Bondon, A.; Gauthier, N.; Sinbandhit, S.;

Toupet, L.; Costuas, K.; Halet, J.-F.; Lapinte, C. Organometallics 2007,

26, 874−896.

(12) (a) Scheres, L.; Giesbers, M.; Zuilhof, H. Langmuir 2010, 26,

4790−4795. Scheres, L.; Rijksen, B.; Giesbers, M.; Zuilhof, H.

Langmuir 2011, 27, 972−980. (b) Scheres, L.; Giesbers, M.; Zuilhof,

H. Langmuir 2010, 26, 10924−10929. (c) Ng, A.; Ciampi, S.; James,

M.; Harper, J. B.; Gooding, J. J. Langmuir 2009, 25, 13934−13941.

(13) Fabre, B. Acc. Chem. Res. 2010, 43, 1509−1518.

(39) Bertini, I.; Luchinat, C.; Parigi, G. Solution NMR of Paramagnetic

Molecules. Application to Metallobiomolecules and Models; Elsevier:

Amsterdam, 2001.

(40) In this connection, the isotropic shifts obtained by subtracting

the shifts of the corresponding protons in the diamagnetic Fe(II)

parents³⁸roughly correlate with the spin densities computed on the

attenant carbon atoms (see the Supporting Information).

(41) Paul, F.; Bondon, A.; da Costa, G.; Malvolti, F.; Sinbandhit, S.;

Cador, O.; Costuas, K.; Toupet, L.; Boillot, M.-L. Inorg. Chem. 2009,

48, 10608−10624.

(42) (a) Ibn Ghazala, S.; Paul, F.; Toupet, L.; Roisnel, T.; Hapiot, P.;

Lapinte, C. J. Am. Chem. Soc. 2006, 128, 2463−2476. (b) Ibn Ghazala,

S.; Gauthier, N.; Paul, F.; Toupet, L.; Lapinte, C. Organometallics

2007, 26, 2308−2317. (c) Justaud, F.; Argouarch, G.; Gazalah, S. I.;

Toupet, L.; Paul, F.; Lapinte, C. Organometallics 2008, 27, 4260−4264.

(d) Costuas, K.; Cador, O.; Justaud, F.; Le Stang, S.; Paul, F.; Monari,

A.; Evangelisti, S.; Toupet, L.; Lapinte, C.; Halet, J.-F. Inorg. Chem.

2012, 50, 12601−12622. (e) Drouet, S.; Merhi, A.; Grelaud, G.;

Cifuentes, M. P.; Humphrey, M. G.; Matczyszyn, K.; Samoc, M.;

Toupet, L.; Paul-Roth, C. O.; Paul, F. New J. Chem. 2012, 36, 2192−

2195. (f) Grelaud, G.; Cador, O.; Roisnel, T.; Argouarch, G.;

Cifuentes, M. P.; Humphrey, M. G.; Paul, F. Organometallics 2012,

31, 1635−1642. (g) Trujillo, A.; Veillard, R.; Argouarch, G.; Roisnel,

T.; Singh, A.; Ledoux, I.; Paul, F. Dalton Trans. 2012, 41, 7454−7456.

(43) In the latter complex (2[PF₆]), the bond length found (1.135

0.009 Å) for the terminal ethynyl bond is much shorter than expected

(14) Cummings, S. P.; Savchenko, J.; Ren, T. Coord. Chem. Rev.

2011, 255, 1587−1616.

(15) (a) Zigah, D.; Herrier, C.; Scheres, L.; Giesbers, M.; Fabre, B.;

Hapiot, P.; Zuilhof, H. Angew. Chem., Int. Ed. 2010, 49, 3157−3160.

(b) Hauquier, F.; Ghilane, J.; Fabre, B.; Hapiot, P. J. Am. Chem. Soc.

2008, 130, 2748−2749. (c) Fabre, B.; Hauquier, F. J. Phys. Chem. B

2006, 110, 6848−6855. (d) Decker, F.; Cattaruzza, F.; Coluzza, C.;

Flamini, A.; Marrani, A. G.; Zanoni, R.; Dalchiele, E. A. J. Phys. Chem.

B 2006, 110, 7374−7379. (e) Dalchiele, E. A.; Aurora, A.; Bernardini,

G.; Cattaruzza, F.; Flamini, A.; Pallavicini, P.; Zanoni, R.; Decker, F. J.

Electroanal. Chem. 2005, 579, 133−142. (f) Tajimi, N.; Sano, H.;

Murase, K.; Lee, K.-H.; Sugimura, H. Langmuir 2007, 23, 3193−3198.

(16) (a) Yasseri, A. A.; Syomin, D.; Loewe, R. S.; Lindsey, J. S.; Zaera,

F.; Bocian, D. F. J. Am. Chem. Soc. 2004, 126, 15603−15612.

(b) Huang, K.; Duclairoir, F.; Pro, T.; Buckley, J.; Marchand, G.;

Martinez, E.; Marchon, J.-C.; De Salvo, B.; Delapierre, G.; Vinet, F.

Chem. Phys. Chem. 2009, 10, 963−971.

(17) Gauthier, N.; Argouarch, G.; Paul, F.; Humphrey, M. G.;

Toupet, L.; Ababou-Girard, S.; Sabbah, H.; Hapiot, P.; Fabre, B. Adv.

Mater. 2008, 20, 1952−1957.

(18) (a) Cicero, R. L.; Linford, M. R.; Chidsey, C. E. D. Langmuir

2000, 16, 5688−5695 and references therein. (b) Kondo, M.; Mates,

4380

dx.doi.org/10.1021/om400515g | Organometallics 2013, 32, 4366−4381

Organometallics

Article

(1.181 Å),⁴⁴while the C43−C45 bond is slightly longer (1.465

0.012 Å vs 1.436 Å). This artifactual feature results possibly from the

thermal motions of the terminal carbon atoms.⁴⁵

page 9. The corrected version was reposted on August 12,

2013.

(44) Allen, F. H.; Kennard, O.; Watson, D. G.; Brammer, L.; Orpen,

A. G.; Taylor, R. J. Chem. Soc., Perkin Trans. 1987, 2, S1−S19.

(45) See for instance: Dunitz, J. D. J. Chem. Soc., Chem. Commun.

1999, 2547−2547.

(46) Szafert, S.; Gladysz, J. A. Chem. Rev. 2006, 106, PR1−PR33.

(47) Gauthier, N.; Argouarch, G.; Paul, F.; Toupet, L.; Ladjarafi, A.;

Costuas, K.; Halet, J.-F.; Samoc, M.; Cifuentes, M. P.; Corkery, T. C.;

Humphrey, M. G. Chem.Eur. J. 2011, 17, 5561−5577.

(48) A model of compound 1b has been theoretically studied

previously; see ref 10.

(49) Koopmans, T. Physica 1934, 1, 104−113.

(50) Gendron, F.; Burgun, A.; Skelton, B. W.; White, A. H.; Roisnel,

T.; Bruce, M. I.; Halet, J.-F.; Lapinte, C.; Costuas, K. Organometallics

2012, 31, 6796−6811.

(51) Schauer, P. A.; Low, P. J. Eur. J. Inorg. Chem. 2012, 390−411.

(52) Paul, F.; Toupet, L.; Roisnel, T.; Hamon, P.; Lapinte, C. C. R.

Chim. 2005, 8, 1174−1185.

(53) For related observations with isoelectronic Ru(II/III) analogues,

́

see: Khairul, W. M.; Fox, M. A.; Schauer, P. A.; Albesa-Jove, D.; Yufit,

D. S.; Howard, J. A. K.; Low, P. J. Inorg. Chim. Acta 2011, 374, 461−

471.

(54) Bruce, M. I.; Burgun, A.; Gendron, F.; Grelaud, G.; Halet, J.-F.;

Skelton, B. W. Organometallics 2011, 30, 2861−2868.

(55) Paul, F.; Malvolti, F.; da Costa, G.; Stang, S. L.; Justaud, F.;

Argouarch, G.; Bondon, A.; Sinbandhit, S.; Costuas, K.; Toupet, L.;

Lapinte, C. Organometallics 2010, 29, 2491−2502.

(56) Connelly, N. G.; Geiger, W. E. Chem. Rev. 1996, 96, 877−910.

(57) (a) te Velde, G.; Bickelhaupt, F. M.; Fonseca Guerra, C.; van

Gisbergen, S. J. A.; Baerends, E. J.; Snijders, J.; Ziegler, T. Theor. Chem.

Acc. 2001, 22, 931−967. (b) Fonseca Guerra, C.; Snijders, J.; te Velde,

G.; Baerends, E. J. Theor. Chem. Acc. 1998, 99, 391−403.

(58) ADF2010.02, SCM; Theoretical Chemistry, Vrije Universiteit:

Amsterdam, The Netherlands, http://www.scm.

(59) Vosko, S. D.; Wilk, L.; Nusair, M. Can. J. Phys. 1980, 58, 1200−

1211.

(60) Perdew, J. P.; Wang, Y. Phys. Rev. B 1992, 45, 13244−13249.

(61) Zhang, Y.; Yang, W. Phys. Rev. Lett. 1998, 80, 890.

(62) Versluis, L.; Ziegler, T. J. Chem. Phys. 1988, 88, 322−328.

(63) (a) van Lenthe, E.; van der Avoird, A.; Wormer, P. E. S. J. Chem.

Phys. 1998, 108, 4783−4796. (b) van Lenthe, E.; van der Avoird, A.;

Wormer, P. E. S. J. Chem. Phys. 1997, 107, 2488−2498. (c) Autschbach,

J.; Pritchard, B. Theor. Chem. Acc. 2011, 129, 453−466.

(64) van Gisbergen, S. J. A.; Snijders, J. G.; Baerends, E. J. Comput.

Phys. Commun. 1999, 118, 119−138.

(65) Adamo, C.; Barone, V. J. Chem. Phys. 1999, 110, 6158−6178.

(66) (a) Pye, C.; Ziegler, T. Theor. Chem. Acc. 1999, 396−408.

(b) Klamt, A. J. Phys. Chem. 1995, 99, 2224−2235.

(67) Otwinowski, Z.; Minor, W. In Methods in Enzymology; Carter, C.

W., Sweet, R. M., Eds.; Academic Press: London, 1997; Vol. 276, pp

307−326.

(68) Nonius, B. V. Kappa CCD Software; Delft: The Netherlands,

1999.

(69) Altomare, A.; Foadi, J.; Giacovazzo, C.; Guagliardi, A.; Moliterni,

A. G. G.; Burla, M. C.; Polidori, G. J. Appl. Crystallogr. 1998, 31, 74−

77.

(70) Sheldrick, G. M. SHELX97-2. Program for the Reﬁnement of

Crystal Structures; Univ. of Gottingen: Germany, 1997.

̈

(71) Reidel, D. International Tables for X-ray Crystallography; Kynoch

Press (present distrib. D. Reidel, Dordrecht): Birmingham, 1974; Vol.

IV.

NOTE ADDED AFTER ASAP PUBLICATION

■

This paper was published on the Web on July 21, 2013, with

errors in the ﬁrst paragraph on page 4 and the last paragraph on

4381

dx.doi.org/10.1021/om400515g | Organometallics 2013, 32, 4366−4381

Article Doi

Synthesis and characterization of redox-active mononuclear Fe(κ2-dppe)(η5-C5Me 5)-Terminated π-conjugated wires

DOI: 10.1021/om400515g

Source and publish data:

Authors:

Article abstract of DOI:10.1021/om400515g

Full text of DOI:10.1021/om400515g

Products guided by the article

R&D Labs maybe for 1448755-65-4

Relevant to this article

Hot Product