Mono(η5-cyclopentadienyl)metal(II) complexes with thienyl acetylide chromophores: Synthesis, electrochemical studies, and first hyperpolarizabilities

Article

pubs.acs.org/Organometallics

Mono(η⁵‑cyclopentadienyl)metal(II) Complexes with Thienyl

Acetylide Chromophores: Synthesis, Electrochemical Studies, and

First Hyperpolarizabilities

Tiago J. L. Silva,^†Paulo J. Mendes,*^,‡Ana M. Santos,^†M. Helena Garcia,^†M. Paula Robalo,^§,∥

J. P. Prates Ramalho,^‡A. J. Palace Carvalho,^‡Marina Buchert,^⊥Christian Wittenburg,^⊥and Jurgen Heck^⊥

̈

^†Centro de Cien

Lisboa, Portugal

̂

cias Moleculares e Materiais, Faculdade de Cien

̂

cias, Universidade de Lisboa, Ed. C8, Campo Grande, 1749-016

‡

́

Centro de Quımica de Evora, Departamento de Quımica, Escola de Cien

̂

cias e Tecnologia, Universidade de Evora, Rua Romao

̃

́

Ramalho 59, 7002-554 Evora, Portugal

§

́

́ ́

Centro de Quımica Estrutural, Instituto Superior Tecnico, Universidade Tecnica de Lisboa, Av. Rovisco Pais, 1049-001 Lisboa,

Portugal

∥

́

Area Departamental de Engenharia Quımica, Instituto Superior de Engenharia de Lisboa, Rua Conselheiro Emídio Navarro, 1,

1959-007 Lisboa, Portugal

^⊥Institut fur Anorganische und Angewandte Chemie, Universitat Hamburg, Martin-Luther-King-Platz 6, D-20146 Hamburg

̈

S

* Supporting Information

ABSTRACT: A series of mono(η⁵-cyclopentadienyl)metal-

(II) complexes with nitro-substituted thienyl acetylide ligands

of general formula [M(η⁵-C₅H₅)(L)(CC{C₄H₂S}_nNO₂)]

(M = Fe, L = κ²-DPPE, n = 1, 2; M = Ru, L= κ²-DPPE, 2 PPh₃,

n = 1, 2; M = Ni, L = PPh₃, n = 1, 2) has been synthesized and

fully characterized by NMR, FT-IR, and UV−Vis spectrosco-

py. The electrochemical behavior of the complexes was

explored by cyclic voltammetry. Quadratic hyperpolarizabilities

(β) of the complexes have been determined by hyper-Rayleigh

scattering (HRS) measurements at 1500 nm. The eﬀect of

donor abilities of diﬀerent organometallic fragments on the

quadratic hyperpolarizabilities was studied and correlated with

spectroscopic and electrochemical data. Density functional

theory (DFT) and time-dependent DFT (TDDFT) calculations were employed to get a better understanding of the second-

order nonlinear optical properties in these complexes. In this series, the complexity of the push−pull systems is revealed; even so,

several trends in the second-order hyperpolarizability can still be recognized. In particular, the overall data seem to indicate that

the existence of other electronic transitions in addition to the main MLCT clearly controls the eﬀectiveness of the organometallic

donor ability on the second-order NLO properties of these push−pull systems.

are needed. These systems can be obtained by combining a π-

conjugated chain with electron donor and/or acceptor groups

(D-π-A) in which metal centers can behave either as acceptor

or donor groups by simply varying the metal and/or its

oxidation state. During the last two decades, it has been found

that high β values could be found for molecules in which the

metal center is coplanar with the π-conjugated chain.

Concerning this feature, systematic studies were made by our

group and others on mono(η⁵-cyclopentadienyl)metal and

pseudo-octahedral complexes with benzene- or thiophene-

INTRODUCTION

■

Organometallic compounds have given rise to a great deal of

interest owing to their application in the ﬁeld of nonlinear

optics (NLO).¹⁻¹¹Organometallic complexes (and organic

molecules) have been shown as potential alternatives to the

traditional inorganics due to fast and large NLO response, high

optical damage thresholds, and structural diversity. In

comparison to organic chromophores, organometallics can

oﬀer additional ﬂexibility due to the presence of metal−ligand

charge transfer excitations, usually at low energy and of high

intensity, which are tunable due to the diversity of metal

centers, oxidation states, ligand environments, and coordination

geometries.

Special Issue: Organometallic Electrochemistry

Received: February 11, 2013

In order to obtain high second-order responses (molecular

quadratic hyperpolarizability, β), strongly asymmetric systems

A

dx.doi.org/10.1021/om4001204 | Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

Scheme 1. Synthesis and Atom Labeling of the Fe(II), Ru(II), and Ni(II) Complexes

based conjugated chains, coordinated to the metal centers

through nitrile or acetylide linkages.^7,12−23

conjugation length of the chromophores also showed contra-

dictory results. For benzene-based acetylides an increase in β_corr

value was found (β_corrtaking into account the dispersion eﬀects

using a two-level model, TLM²⁹) upon an increase in chain

length, in spite of some saturation of the β_corrsignal for long

chains.^7,13In the case of benzene-based nitrile metal complexes,

an attempt to increase the observed intrinsic quadratic

hyperpolarizability by replacing a phenyl with a biphenyl unit

was unsuccessful due to the signiﬁcant torsion angle in the

latter ligand, which leads to the conjugation breaking.^21,17The

higher quadratic hyperpolarizability obtained when the

conjugation length is increased without aﬀecting planarity by

insertion of a vinylene unit further supported this assumption.

In the case of metal complexes presenting oligo-thiophene

chromophores, our studies on mono(η⁵-cyclopentadienyl)iron-

(II) nitrile compounds showed some constancy of β_corrvalues

upon chain lengthening due to a compromise between the

conjugation length and the metal to ligand charge transfer.¹⁶In

addition, β values do not increase regularly upon chain

lengthening on (porphinato)zinc(II) thiophene acetylides.³⁰

The eﬀect of diﬀerent coligands on quadratic hyperpolariz-

abilities has been also studied. In fact, diﬀerent coligands can

modify donor strength of the organometallic moieties. For

example, replacing two CO ligands by DPPE results in a

signiﬁcant increase in the quadratic hyperpolarizability of

mono(η⁵-cyclopentadienyl)iron(II) benzene acetylides,⁷while

subtle variations in the hyperpolarizability are observed on

replacing DPPE by (+)-DIOP in related mono(η⁵-

cyclopentadienyl)iron(II) nitriles¹⁷or on replacing DPPE by

DPPM or two PPh₃ligands in pseudo-octahedral acetylide

complexes.⁷However, when TLM is taken into account, the

eﬀect of replacing DPPE (β_corr= 161 × 10⁻³⁰esu) by two PPh₃

In many cases, the structure−NLO activity trends were

revealed to be similar to those observed in organic compounds:

i.e., nonlinearities can be enhanced by either increasing the

conjugation length and ease of the electronic delocalization of

the hyperpolarizable chromophore or increasing the strength of

donor or acceptor groups. For example, very eﬃcient NLO

responses were found when strong electron donors such as

readily oxidizable 18-electron iron(II) and ruthenium(II)

organometallic moieties were coupled with strong electron

acceptors such as a nitro group.^5,7,21,23In addition, compounds

derived from metal half-sandwich moieties possessing benze-

noid acetylide ligands revealed better NLO properties than the

nitrile analogues, probably due to more favorable metal−ligand

coupling through enhanced metal to ligand π back-bond-

ing.^4,20,24,25

However, some aspects of the relationship between structural

features and NLO activity in these systems remain unclear.

Replacing phenyl rings with heterocycles improves the ease of

delocalization, thus yielding a good basis for high second-order

NLO responses, as already proven in organic chromo-

phores²⁶⁻²⁸and in a series of mono(η⁵-cyclopentadienyl)-

ruthenium(II) thiophene acetylide²²and mono(η⁵-

cyclopentadienyl)iron(II) thiophene nitrile¹⁶complexes. How-

ever, the eﬀect on the optical nonlinearities was found to be not

easily understandable. For instance, quadratic nonlinearities on

metal acetylides were found to be⁷(i) not consistently

increased on replacment of phenyl by a pyridyl ring, (ii)

decreased on incorporation of a furanyl ring, and (iii) almost

independent of the relative location of the thienyl and phenyl

rings in two-ring acetylide ligands. The eﬀect of increasing the

B

dx.doi.org/10.1021/om4001204 | Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

(β_corr= 96 × 10⁻³⁰esu) or PMe₃(β_corr= 38 × 10⁻³⁰esu) on the

quadratic hyperpolarizability of mono(η⁵-cyclopentadienyl)-

ruthenium(II) acetylides cannot be underestimated.⁷

which the deprotection of the trimethylsilyl alkynes 1 and 2

with KF in methanol in the presence of the organo-ruthenium

chloride gave the ruthenium alkynyls in a single step. Finally,

the nickel derivatives were obtained by transmetalation

reactions catalyzed by copper(I) iodide of the suitable terminal

alkyne 1′ or 2′ with [Ni(η⁵-C₅H₅)(PPh₃)Cl], in the presence of

triethylamine. The compounds are fairly stable toward

oxidation in air and to moisture both in the solid state and in

The eﬀect of metal variation on quadratic hyperpolariz-

abilities needs also some attention. For mono(η⁵-

cyclopentadienyl)metal complexes, for example, studies suggest

a trend on quadratic nonlinearity as β(Ni) < β(Fe) ≤ β(Ru) for

benzene acetylides having one aryl ring.^20,31−33When the

dispersion eﬀects are taken into account using the TLM, the

corresponding corrected β values for ruthenium complexes

were revealed clearly to be larger than those found for iron and

Ni complexes: β_corr(Ni) ≤ β_corr(Fe) < β_corr(Ru).⁷For the

corresponding nitrile derivatives, however, both β and β_corr

values for iron are clearly higher than those found for

ruthenium and nickel complexes, thus leading to Ni < Ru <

Fe.²¹For the compounds presenting two benzene rings in the

conjugated system, the limited data available for direct

comparison seem to indicate that both β and β_corrfollow the

same trend (Ni < Ru < Fe),^7,21but further studies will be

necessary to conﬁrm these results.

In our continuing eﬀort to establish detailed structure−NLO

activity correlations and to complement the previously reported

studies on mono(η⁵-cyclopentadienyl)metal(II) complexes

presenting benzene- or thiophene-based conjugated chromo-

phores, it seems that it is of importance to further study the

quadratic hyperpolarizability of these systems incorporating

acetylide thiophene chromophores. Our goal is to contribute to

the clariﬁcation of the aforementioned issues with particular

relevance for the role of diﬀerent organometallic donor

moieties, on the quadratic hyperpolarizabilities. Thus, we

report herein the syntheses of a series of mono(η⁵-

cyclopentadienyl)metal(II) complexes with nitro-substituted

thienyl acetylide ligands of general formula [M(η⁵-C₅H₅)(L)-

(CC{C₄H₂S}_nNO₂)] (M = Fe, L = κ²-DPPE, n= 1, 2; M =

Ru, L = κ²-DPPE, 2 PPh₃, n = 1, 2; M = Ni, L = PPh₃, n = 1, 2).

Spectroscopic and electrochemical data and theoretical

calculations on model complexes using DFT and TDDFT are

used to explain the experimental quadratic hyperpolarizabilities

of these compounds. Measurements of these properties were

carried out by HRS at the fundamental wavelength of 1500 nm

to avoid superposition of the UV−vis absorptions and the

second harmonic signal to ensure more reliable analysis of the

results on the basis of the well-known two-level model.

1

solution and were characterized by IR and H, ¹³C, and ³¹P

NMR spectroscopy and satisfactory microanalyses.

For all compounds, a characteristic ν(CC) band in the IR

spectrum was found in the range 2005−2074 cm⁻¹. A

signiﬁcant low-energy shift was observed in this band upon

coordination of the ethynyl ligands, in particular for iron and

ruthenium complexes (Table 1). The negative shifts have been

related to enhanced π back-donation from the metal d orbitals

to the π* orbital of the CC group, which leads to a decrease

in the CC bond order. It is well accepted that π back-

bonding has a synergic eﬀect with the dominant σ bond with

the reinforcement of the binding. This π-back-bonding eﬀect

was found to be relatively small in comparison to the M−C σ

bond by theoretical calculations on similar mono(η⁵-cyclo-

pentadienyl)-derived molecules³⁵and should be also the case in

the compounds studied in this work. In the present study, the

observed negative shifts show the expected trend, considering

the π-donor ability of the organometallic fragment ([Fe] >

[Ru] ≫ [Ni]). In addition, the thiophene chain lengthening

leads to a less eﬀective π back-donation interaction and no

signiﬁcant diﬀerences are observed for Ru complexes by the

presence of diﬀerent phosphane coligands. The magnitude of

the π back-donation interaction is higher for the compounds

studied in this work in comparison to related thiophene

nitrile^16,15and benzene acetylide^{4,7,20,25,33,36}complexes.

¹H and ¹³C NMR resonances for the cyclopentadienyl ring

are in the range usually observed for neutral iron(II),^20,32,37

ruthenium(II),^22,25,32and nickel(II)^33,38,39complexes and are

relatively insensitive to the chain lengthening of the acetylide

ligand. Considering the thiophene ligand protons, an overall

shielding eﬀect upon coordination was observed, especially for

the H4 protons (see Scheme 1 for numbering), indicating an

electronic ﬂow toward the aromatic ligand due to π back-

donation involving the metal center. The shifts in the

resonances are consistent with the possibility of some

contribution of an allenylidene form in solution. ¹³C NMR

data show an expected deshielding on the CC carbons (C1

and C2) upon coordination, whereas for the ring carbons more

signiﬁcant changes were observed for those in the ring adjacent

to the acetylide group, in agreement with the trends observed

RESULTS AND DISCUSSION

■

Synthesis and Spectroscopic Studies. Trimethyl((5-

nitrothiophen-2-yl)ethynyl)silane (1) and trimethyl((5′-nitro-

2,2′-bithiophen-5-yl)ethynyl)silane (2) ligands were obtained

in good yields by Sonogashira cross-coupling of ethynyltrime-

thylsilane with the corresponding nitro-thiophene bromides.

Subsequent removal of the trimethylsilyl protecting group with

base aﬀorded 2-ethynyl-5-nitrothiophene (1′) and 5-ethynyl-5′-

nitro-2,2′-bithiophene (2′).

1

in the H NMR spectra. ³¹P resonances for the phosphane

coligands are in the range usually observed for mono(η⁵-

cyclopentadienyl)iron(II), -ruthenium(II) and -nickel(II) com-

plexes and are relatively insensitive to the chain lengthening of

the acetylide ligand.

The optical absorption spectra of all complexes were

recorded using 1.0 × 10⁻⁵M solutions in chloroform. The

spectrum for 2aRu typiﬁes the behavior of the compounds

studied in this work (Figure 1), and the optical data are

summarized in Table 2. The main feature of these spectra is the

presence of an intense band in the range 484−613 nm and

bands at higher energy (with peak positions in the range 286−

410 nm), depending on the organometallic fragment and

thiophene chromophore. The band at lower energy has been

attributed, in parent benzene compounds, to a MLCT (metal to

The acetylide complexes were prepared in good yields (54−

80%) using the synthetic methodologies successfully utilized for

the preparation of iron(II),²⁰ruthenium(II),³⁴and nickel(II)³³

σ-arylacetylides (Scheme 1). Reaction of [Fe(η⁵-C₅H₅)-

(DPPE)I] with 1′ and 2′ in reﬂuxing methanol, in the presence

of NH₄PF₆, aﬀorded iron vinylidene complexes, which were

deprotonated in situ using sodium methoxide to give the σ-

acetylide products. The vinylidene route has been widely used

in the preparation of ruthenium alkynyls with NLO properties.⁷

However, we adopted the method of Xia and Selegue,³⁴in

C

dx.doi.org/10.1021/om4001204 | Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

Figure 1. UV−visible spectrum for 2aRu in chloroform.

Table 2. Optical Spectral Data for Fe, Ru, and Ni Complexes

in Chloroform Solution (ca. 1.0 × 10⁻⁵mol dm⁻³)

compd

λ/nm (ε/10⁴M⁻¹cm⁻¹

)

1Fe

578 (1.3), 363 (0.7)

613 (1.5), 400 (1.1)

517 (2.0), ∼320 (sh)

536 (2.2), ∼310 (sh)

566 (2.8), 373 (2.0)

580 (2.4), 374 (1.2)

2Fe

1aRu

1bRu

2aRu

2bRu

1Ni

a

484 (0.7), 404 (0.2), 286 (0.9)

506 (2.3), 410 (sh), 350 (sh)

2Ni

a

Obtained from a ﬁt with multiple Gaussian bands.

ligand charge transfer) transition, whereas the high-energy

bands are ascribed to internal transitions occurring at the

acetylide chromophores and or/organometallic moiety.^20,24,25,36

In fact, according to the results of our TDDFT calculations (see

below) the lower energy bands, which are the key for second-

order nonlinear optical properties, can be attributed to MLCT.

The energy and intensity of these bands depend mainly on the

metal and acetylide chromophore chain length and, to a lesser

extent, on the diﬀerent phosphane coligands in the Ru

complexes. The energy of these bands follows the expected

trend, considering the diﬀerences on the electronic properties

of the metal (Ni > Ru > Fe). These energies are lower than

those found in already reported benzene-based iron(II),²⁰

ruthenium(II)^24,25and nickel(II)³³acetylide analogues and

related iron(II)¹⁶and ruthenium(II)¹⁵nitrile compounds and

are within the range of those observed in a series of ruthenium

acetylides with an end-capping nitro group and thienyl and

benzene entities in the conjugation chain.²²The chain

lengthening of the chromophore led to a bathochromic eﬀect

on these bands for all complexes. This eﬀect is opposite of that

already reported for benzene-based iron(II),²⁰ruthenium-

(II)^24,25and nickel(II)³³acetylide analogues and also for

related iron(II)¹⁶and ruthenium(II)¹⁵nitrile compounds. This

can be explained by an improved planarity that can be expected

from: (i) the limited steric hindrance of dithienyl in comparison

to the biphenyl moiety and (ii) the higher level of electronic

coupling between the organometallic fragment and the

thiophene chromophores through an acetylide linkage in

comparison to a nitrile bond. Considering point ii, IR data

revealed that the magnitude of π back-donation interaction is

stronger for the compounds studied in this work, thus favoring

a more eﬀective conjugation and improved planarity in

comparison to the related thiophene nitrile and benzene-

based acetylide complexes (see above).

D

dx.doi.org/10.1021/om4001204 | Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

UV/vis absorption spectra were also recorded for all

complexes in other solvents with diﬀerent polarities (ethyl

ether and DMF), in the wavelength range 200−800 nm, to

examine the solvatochromic behavior of the lowest-energy band

(Table 3). The results showed a bathochromic shift for all

the Supporting Information (Figures S1−S5 and Tables S1−

S2). The computed data showed that the use of phenylated

phosphanes and solvated media does not result in a signiﬁcantly

better match between the calculated and experimental data. In

addition, no diﬀerences in the simulated electronic spectra for

the lowest energy bands, which are the important ones for

NLO, were achieved by the use of a larger number of excitation

states in TDDFT studies. Thus, the use of these approx-

imations gives a good description of the experimental results

with signiﬁcant gain in computational eﬀort.

Table 3. Solvatochromic Behavior of the Lowest-Energy

Bands of All Complexes

λ/nm

compd

Et₂O

CHCl₃

DMF

̃

Δλ (Δν )

/10³cm⁻¹

̅

Figure 2 shows the structure and atom labeling of the model

complexes used in calculations, and Table 4 gives selected

1Fe

524

582

484

498

534

554

468

494

578

613

517

536

566

580

484

506

601

631

538

577

585

501

510

77 (2.4)

49 (1.3)

54 (2.1)

40 (1.5)

43 (1.4)

31 (1.0)

33 (1.4)

16 (0.6)

2Fe

1aRu

1bRu

2aRu

2bRu

1Ni

2Ni

complexes upon increasing solvent polarity. This positive

solvatochromic behavior is characteristic of transitions with an

increase of the dipole moment upon photoexcitation. The

magnitude of this red shift depends on the organometallic

fragment ([Fe] > [Ru] > [Ni]; DPPE > 2 PPh₃) and chain

lengthening of the coordinated acetylide ligand ([M]-C

C{C₄H₂S}_nNO₂: n = 1 > n = 2). The absolute values of the

solvatochromic shift are almost comparable to those found for

benzene-based acetylide ruthenium(II)^24,25(35−39 nm)

analogues but are large in comparison to those for benzene-

based iron(II)²⁰(13−23 nm) acetylide analogues and also for

related iron(II)¹⁶(23−32 nm) and ruthenium(II)¹⁵(14−30

nm) nitrile compounds.

DFT Calculations. Recently, DFT calculations on a series of

mono(η⁵-cyclopentadienyl)iron(II) and -ruthenium(II) thio-

phene acetylide complexes with general formula M(η⁵-C₅H₅)-

(H₂PCH₂CH₂PH₂)(CC{C₄H₂S}Y) (Y = donor and accept-

or groups) have been carried out by our group in order to study

structure−static quadratic hyperpolarizability relationships.^23,40

The results have shown that the traditional qualitative

arguments for enhancing second-order nonlinear optical

responses were applicable to these complexes, thus leading to

the prediction of higher hyperpolarizabilities for M(η⁵-C₅H₅)-

(H₂PCH₂CH₂PH₂)(CC{C₄H₂S}NO₂) (M = Fe, Ru), in

which the strong electron donor organometallic fragment was

combined with the better acceptor group NO₂. In this work we

extend the DFT study to nickel complexes and to compounds

with two thienyl rings. In addition, the eﬀect of diﬀerent

phosphane coligands in the Ru complexes was evaluated. As a

compromise between accuracy and computational eﬀort, three

simpliﬁcations were used in the calculations: (i) only the ﬁrst

24 lower excitation states were computed in TDDFT studies,

(ii) dephenylated phosphanes were used, i.e., H₂PCH₂CH₂PH₂

instead of DPPE and PH₃instead of PPh₃, and (iii) isolated

molecules in the gas phase were used (i.e., no solvent

corrections were employed). The adequacy of the simpliﬁcation

(i) was checked by the use of a larger number of excited states

for simulating the electronic spectra of 1a′Ru, and the results

were compared to those obtained with 24 excited states. The

adequacy of the simpliﬁcations (ii) and (iii) was checked by a

comparison between the calculated optical data obtained for

the model complexes 1a′Ru and 2a′Ru and real 1aRu and

2aRu with experimental data. All these results are included in

Figure 2. Structure and atom labeling of the model complexes used in

DFT calculations.

structural data for the optimized structures in the gas phase.

The optimized structures for all complexes are shown in the

Supporting Information (Figures S6−S13).

The calculated angles and bond lengths are consistent with

experimental crystal data for related mono(η⁵-

cyclopentadienyl)metal σ-arylacetylides. For instance, the

calculated M−Cp, M−C1, and C1−C2 bond lengths and

M−C1−C2 and P−M−C1 bond angles are within or close to

the range of experimental reported values for iron (Fe−Cp,

2.060−2.122 Å; Fe−C1, 1.856−1.910 Å; C1−C2, 1.202−1.236

Å; Fe−C1−C2, 177.0−178.7°; P−Fe−C1, 85.1−85.5°),^20,37

ruthenium (Ru−Cp, 2.222−2.260 Å; Ru−C1, 1.974−2.017 Å;

C1−C2, 1.214−1.232 Å; Ru−C1−C2, 173.7−178.0°; P−Ru−

C1, 85.9−90.8°),^22,24,25,36and nickel (Ni−Cp, 2.063−2.146 Å;

Ni−C1, 1.840−1.856 Å; C1−C2, 1.191−1.209 Å; Ni−C1−C2,

173.9−177.7°)³⁶complexes. The calculations predict a decrease

of the C1−C2 bond order and a decrease of the C2−C3 bond

length of the acetylide ligands upon coordination, which is

consistent with a metal to ligand π back-donation interaction

that was observed experimentally (see above). It is interesting

to note that, in spite of some overestimation of ν(CC), DFT

calculations predict the experimental trend on the magnitude of

the metal to ligand π-back-donation interaction when they

consider the eﬀect of diﬀerent metal and chain lengthening of

the acetylide chromophores. The concept of bond length

E

dx.doi.org/10.1021/om4001204 | Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

Table 4. Selected Calculated Structural Data for the Fe, Ru, and Ni Model Complexes

a

b

compd

1′Fe

2′Fe

1a′Ru

1b′Ru

2a′Ru

2b′Ru

1′Ni

2.154

2′Ni

c

Bond Lengths (Å)

d

M−Cp

M−P

2.128

2.295

1.901

2.264

2.258

2.400

2.001

2.263

2.257

2.154

2.258

1.854

e

2.297

1.898

2.391

1.998

2.391

2.001

2.400

2.260

1.853

M−C1

C1−C2

C2−C3

BLA

2.004

1.237 (1.211) 1.237 (1.212)

1.393 (1.406) 1.394 (1.405)

0.022 (0.037) 0.031 (0.042)

1.236 (1.211) 1.235 (1.211) 1.236 (1.212)

1.392 (1.406) 1.393 (1.406) 1.392 (1.405)

0.019 (0.037) 0.021 (0.037) 0.029 (0.042)

1.235 (1.212)

1.393 (1.405)

0.030 (0.042)

1.230 (1.211) 1.230 (1.212)

1.398 (1.406) 1.398 (1.405)

0.027 (0.037) 0.035 (0.042)

c

Bond Angles (deg)

M−C1−C2

C1−C2−C3

P−M−C1

179.3

179.2

179.6

179.2

179.9

178.6

178.2

178.6

179.9 (179.4) 179.4 (179.5)

179.4 (179.4) 179.3 (179.4) 179.8 (179.5)

179.6 (179.5)

84.6

179.6 (179.4) 179.9 (179.5)

f

86.8

86.6

84.8

85.0

84.7

90.3

P1−M−P2

P1−M−Cp

86.3

83.5

92.7

83.5

92.7

g

129.1

124.6

129.2

131.2

125.2

127.7

125.0

131.2

125.1

127.9

137.1

132.3

137.3

C1−M−Cp

124.7

124.8

132.1

h

Dh

−179.7 (−179.9)

179.9 (−179.9) 179.7 (−179.9)

−179.5 (−179.9)

i

ν(CC) (cm⁻¹

2081 (−58)

)

2072 (−67)

2071 (−63)

2074 (−65)

2072 (−62)

2078 (−56)

2116 (−23)

2114 (−20)

a

b

c

Bond lengths and angles taken from ref 23. Bond lengths and angles taken from ref 40. Calculated bond lengths and angles of thiophene ligands

d

e

f

HCC{C₂H₄S}_n-NO₂are shown in parentheses. Average M−C(Cp). Average M−P1/P2 (M = Fe, Ru) bond lengths. Average P1/P2−M−C1

g

h

i

(M = Fe, Ru) bond angles. Average P1/P2−M−Cp (M = Fe, Ru) bond angles. Dihedral angle (C5−C6−C7−C8). Diﬀerences in calculated

ν(CC) (ν_complex− ν_ligand) upon coordination to the organometallic fragments are shown in parentheses.

Table 5. Optical Data for 1′ and 2′ and the Studied Model Fe(II), Ru(II), and Ni(II) Complexes Obtained by Using TDDFT

Calculations

a

b

λ_exp

λ_eg

c

d

e

compd

(nm)

f

attribution

character of the CT

1′

2′

346

410

281

578

363

341

429

316

468

452

321

559

378

470

451

310

473

446

308

304

568

380

552

378

372

0.356 H → L (97)

0.648 H → L (99)

CC (58), T1 (42) → NO₂(100)

CC (33), T1 (67) → T2 (18), NO₂(82)

0.115 H-1 → L (37), H → L+1 (53)

0.390 H → L (75), H-1 → L+3 (10)

CC (14), T1 (79), T2 (7) → NO₂(100)

f

1′Fe

Fe (39), CC (25), T1 (30), Cp (6) → NO₂(87), P (13)

0.080 H-1 → L+3 (40), H-1 → L+2 (18), H → L (20) Fe (57), Cp (3), CC (40) → NO₂(27), P (67), T1 (6)

0.106 H-3→L (92)

Fe (75), Cp (15), CC (10) → NO₂(83), T1 (17)

Fe (36), CC (37), Cp (5), T1 (20), P (2) → NO₂(61), T2 (39)

Fe (47), CC (47), Cp (6) → NO₂(58), T2 (26), P (16)

Ru (46), CC (54) → T1 (5), NO₂(77), P (18)

2′Fe

613

400

517

320

0.710 H→L (98)

0.242 H-3 → L (33), H → L+1 (46)

0.355 H → L (72), H → L+1 (24)

0.172 H → L (24), H → L+1 (67)

0.102 H-4 → L (21), H-3 → L (74)

0.091 H → L (21), H → L+1 (71)

0.472 H → L (77), H → L+1 (20)

0.050 H-4 → L (72), H-3 → L (24)

g

1a′Ru

T1 (40), CC (60) → Cp (27), NO₂(17), P (56)

Ru (56), P (2), T1 (23), Cp (19) → NO₂(100)

1b′Ru

536

310

T1 (43), CC (57) → Ru (9), Cp (33), NO₂(11), P (47)

Ru (41), CC (59) → T1 (5), NO₂(82), P (10), Cp (3)

Ru (16), T1 (80), Cp (4) → NO₂(86), CC (12), P (2)

0.060 H-4 → L (17), H-3 → L (19), H-3 → L+1 (17) Ru (43), CC (6), T1 (48), Cp (3) → NO₂(76), P (24)

2a′Ru

2b′Ru

566

373

580

374

0.760 H → L (99)

Ru (32), Cp (7), CC (38), T1 (20), P (3) → T2 (39), NO₂(61)

0.158 H-3 → L (38), H → L+2 (30), H → L+3 (10) Ru (54), Cp (8), CC (38) → T1 (3), T2 (26), NO₂(55), P (16)

0.773 H → L (99)

Ru (30), Cp (5), CC (39), T1 (24), P (2) → T2 (38), NO₂(62)

Ru (64), Cp (18), CC (13), P (5) → T1 (9), T2 (32), NO₂(59)

0.103 H-2 → L (75)

0.221 H-3 → L (49), H → L+2 (44)

Ru (53), Cp (12), CC (29), P (6) → T1 (14), T2 (31), NO₂

(55)

1′Ni

2′Ni

484

404

286

506

410

438

350

252

521

510

0.395 H → L (96)

Ni (34), Cp (41), CC (25) → T1 (26), NO₂(74)

Ni (37), P (2), Cp (50), CC (11) → T1 (25), NO₂(75)

NO₂(26), Cp (44), CC (30) → Ni (33), P (60), T1 (7)

T1 (77), CC (23) → P (36), Cp (9), NO₂(55)

0.183 H-1 → L (92)

0.160 H-12 → L+1 (10), H-7 → L+1 (42)

0.273 H → L (36), H-1 → L+1 (20)

0.364 H-4 → L+1 (16), H → L (57)

Ni (24), Cp (3), CC (24), T1 (49) → P (15), T2 (24), NO₂

(61)

350

410

351

0.141 H-1 → L (95)

Ni (38), P (1), Cp (57), CC (4) → T1 (8), T2 (39), NO₂(53)

0.172 H → L+2 (81)

Ni (26), Cp (28), CC (46) → T1 (16), T2 (24), NO₂(60)

a

b

c

d

Experimental data. Absorption wavelength of the main transitions. Oscillator strength. Abbreviations: H, HOMO; L, LUMO. The percentage of

the contribution is shown in parentheses. Based on the represented molecule fragments (overall percentage of the charge transfer given in

e

parentheses). Abbreviations: Cp, cyclopentadienyl; CC, acetylide group; T, thiophene rings (T1, close to CC; T2, close to NO₂); P, phosphane

f

g

coligands. Reference 23. Reference 40.

F

dx.doi.org/10.1021/om4001204 | Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

alternation (BLA), deﬁned as the diﬀerence between the

average carbon−carbon adjacent bond lengths along a

conjugated backbone, was found to be an important parameter

that can be correlated with the optical nonlinearities of organic

and organometallic/coordination complexes.⁴¹⁻⁴³The calcu-

lated BLA parameter for the model complexes indicates a

relative enhanced contribution of the allenylidene form in the

organometallic compounds in comparison to the ethynyl

ligands, in particular for Ru and Fe complexes:

−

[M]−C C−th−NO₂↔ [M]⁺CCthNO₂

The contribution of the allenylidene form in the ground state

has already been suggested by NMR spectroscopic data (see

above). The overall data suggest an enhanced ground-state

polarization due to an increasing of the donor−acceptor

strength, i.e., the better Ru(II) and Fe(II) electron-donor

organometallic moiety (compared to H in acetylide ligands)

can form a very eﬀective push−pull system in combination with

acceptor nitro substituent in the thiophene acetylide

chromophore.

In order to achieve a deep knowledge of the relevant

electronic transitions involved, which can be helpful in the

further discussion of the experimental ﬁrst hyperpolarizabilities,

we performed TDDFT calculations in the model complexes

(TDDFT calculations of thiophene ligands HCC-

{C₄H₂S}_nNO₂are also included for comparison). The energies,

oscillator strengths, and compositions of the main electronic

transitions, in terms of the contributions of groups of atoms

involved, are given in Table 5. The results show that, in general,

TDDFT calculations overestimate the energy of the ETs for the

studied complexes, in particular for 1′Fe, in comparison to

experimental data. However, a reasonable accuracy was

obtained for HCC{C₄H₂S}_nNO₂ligands and for complexes

with two thienyl rings. In spite of the diﬀerences in the ET

energies, TDDFT calculations reproduce the number of ET

bands in the experimental UV−vis spectra. As an example, the

calculated spectrum for 2a′Ru, which can be compared to the

experimental spectrum for 2aRu (Figure 1), is given in the

Supporting Information (Figure S14).

Figure 3. Representation of the main orbitals involved in electronic

transitions of the complexes 1′Fe, 1a′Ru, 1b′Ru, and 1′Ni.

The analysis of the orbitals mainly involved in the ETs of the

studied model complexes, depicted in Figures 3 and 4, together

with the TDDFT data given in Table 5, allows us to predict

reliable assignments of the bands observed in UV−vis

experimental spectra of the corresponding synthesized

complexes (Table 2).

Figure 4. Representation of the main orbitals involved in electronic

transitions of the complexes 2′Fe, 2a′Ru, 2b′Ru, and 2′Ni.

First, we will consider the attribution of the higher energy

bands for the complexes, in the range λ 286−410 nm. On the

basis of the DFT calculations on the model 1′Fe, 1a′Ru, 1b′Ru,

and 1′Ni complexes, the CT processes observed experimentally

in the corresponding synthesized complexes with one

thiophene ring, 1Fe, 1aRu, 1bRu, and 1Ni, seem to be

somewhat complex. For 1Fe this band seems to clearly be a

MLCT, for 1aRu some contribution of an intraligand CT is

predicted, in addition to the MLCT, for 1bRu an internal

electronic charge transfer occurring at the organometallic

moiety could also contribute to this band, in addition to MLCT

and ILCT character, and ﬁnally for 1Ni the experimental band

at λ 404 nm can be clearly assigned to MLCT and the band at λ

286 nm can be assigned mainly to internal transitions occurring

at the organometallic moiety together with some LMCT

contribution. Considering the complexes with two thiophene

rings, the experimental high-energy bands for 2bRu and 2Ni

can be assigned to purely MLCT (on the basis of the DFT

calculations on the model complexes 2b′Ru and 2′Ni,

respectively), whereas for 2Fe and 2aRu the theoretical

calculations on the corresponding models 2′Fe and 2a′Ru

predict some contribution of internal charge transfer occurring

at the organometallic moiety, in addition to the MLCT

character.

In view of the second-order NLO properties it is important

to consider the lower energy charge transfer excitations in some

detail. For the synthesized Fe and Ru complexes with one

thiophene ring, 1Fe, 1aRu, and 1bRu, the experimental broad

absorption band could be due to the existence of two CT

contributions closer in energy, as predicted by TDDFT

calculations (468 and 452 nm for 1′Fe, 470 and 451 nm for

1′Ru, and 473 and 446 nm for 1b′Ru). This band can be

viewed mainly as MLCT from the organometallic moiety to the

NO₂group, as expected considering the donor and acceptor

character of the organometallic fragments and the nitro group,

G

dx.doi.org/10.1021/om4001204 | Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

respectively. In addition, the thiophene ring linked to the

acetylide group (labeled T1 in Table 5) acts as a donor during

charge transfer, as in the HCC{C₄H₂S}NO₂ligand, and thus

some contribution of ILCT character is observed. For these

compounds, an electronic charge transfer within the organo-

metallic fragment could also contribute to this band. In fact,

TDDFT calculations revealed that the phosphane coligand also

increases its electronic density during the excitation. For Ru

complexes, in addition to the phosphane coligands, the Cp ring

also slightly participates in this charge transfer process. In either

case, the electronic transfer from the organometallic moiety to

the acceptor NO₂group during excitation is dominant for all

complexes (ca. 60−80%). In contrast to that found for Fe and

Ru complexes, in the synthesized nickel compound 1Ni only

one excitation is predicted, according to the calculations on the

model complex 1′Ni. This band can be assigned clearly to a

MLCT, since the TDDFT results show that both NO₂and the

thiophene ring increase their electronic density during the

transfer and the phosphane coligand seems not to participate in

the process. In addition, the Cp ring has an active involvement

as a donor in this electronic transition. Opposite to what was

found for the synthesized 1Fe, 1aRu, and 1bRu, in the case of

complexes with two thiophene rings, and according to the

calculations in the model complexes 2′Fe, 2a′Ru, 2b′Ru, only

one excitation is predicted and no contribution of the

phosphane coligands and/or Cp ring as acceptors during the

electronic transitions is observed. To a moderate extent,

phosphane coligands and the Cp ring in fact act as electron

donors during the CT process instead of electron acceptors.

Thus, the experimental band for these complexes can be viewed

to have higher MLCT character, in comparison to those for the

1Fe, 1aRu, and 1bRu counterparts, in addition to the

contribution of some internal electronic density transfer within

the acetylide chromophore. The MLCT character is the result

of a CT from the “CpM(PP)(CC)” fragment to the T2-NO₂

moiety and contributes ca. of 80% to the overall charge transfer

process. The remaining charge transfer has an intraligand

component and, as for the synthesized 1Fe, 1aRu, and 1bRu,

corresponds to the transfer of electronic density from the

thiophene ring T1 (see Table 5) to the T2-NO₂moiety. This

ILCT contribution resembles the character of the CT occurring

at the ligand HCC{C₄H₂S}₂NO₂, according to the TDDFT

calculations. In the case of the ligand, T1 strongly decreases its

electronic density during the excitation (Table 5). For the

complexes, however, the participation of T1 is reduced at the

expense of the organometallic moiety that plays an important

role as a donor on the overall charge transfer process. In the

case of the synthesized complex 2Ni, the experimental

absorption band (λ 506 nm) can consist of two CT

contributions closer in energy, as predicted by TDDFT

calculations on the model complex 2′Ni (521 and 510 nm).

The CT process is somewhat complex, but the overall behavior

resembles that found for 1Fe, 1aRu, and 1bRu: i.e., in addition

to the MLCT and ILCT character, an electronic charge transfer

to the phosphane coligand could also contribute to this band.

In either case, the electronic transfer to the T2-NO₂moiety

during excitation is dominant for this complex (ca. 75%).

However, in comparison to what was found for 1Fe, 1aRu, and

1bRu, a major contribution of a charge transfer occurring at the

acetylide chromophore (ca. 60%) is predicted for 2Ni,

according to the theoretical calculations on the model 2′Ni

complex. Finally, as expected, the overall CT observed on the

complexes with two thiophene rings led to an increase of the

electronic density mainly at the nitro group (ca. 60%).

The TDDFT data show that the chain lengthening of the

chromophore led to a bathochromic eﬀect on the calculated

low-energy ETs for all of the model complexes (see Table 5),

thus reproducing the experimental data (Table 2). An analysis

of the energies of the frontier orbitals of the model complexes

(Figure 5) shows that the introduction of a second thiophene

Figure 5. Energy levels of selected orbitals for the model complexes.

ring causes an increase of the energy of HOMOs and a decrease

in the energy of LUMOs, leading to the expected reduction of

the HOMO−LUMO gap and the observed bathochromic

eﬀect. The major contribution to the reduced HOMO−LUMO

gap for iron and ruthenium complexes is due to the stabilization

of LUMOs, whereas that for nickel is the destabilization of the

HOMO. The overall result is an improved reduction eﬀect on

the HOMO−LUMO gap for the iron and ruthenium

complexes in comparison to the nickel complexes.

Electrochemical Studies. In order to get an insight on the

electron richness of the organometallic fragment and the

coordinated chromophores, the electrochemical behavior of all

complexes was studied by cyclic voltammetry in dichloro-

methane and acetonitrile, between the limits imposed by the

solvents. As an example, the electrochemical response for 1bRu

in dichloromethane is shown in Figure 6, and the most relevant

data for redox changes exhibited by all the complexes in

Figure 6. Cyclic voltamogramm of 1bRu (−) showing the isolated

oxidative process (- - -) in dichloromethane (scan rate 200 mV s⁻¹).

H

dx.doi.org/10.1021/om4001204 | Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

dichloromethane and acetonitrile at a scan rate of 200 mV s⁻¹

are summarized in Tables 6 and 7, respectively.

process.⁴⁴⁻⁴⁹The data are consistent with an oxidation process

sensitive to the variation of the organometallic moiety and the

chain lengthening of the acetylide ligand. The relative oxidation

potentials for iron (E_1/2= 0.15−0.29 V), ruthenium (E_1/2

=

Table 6. Cyclic Voltammetric Data for Compounds 1Fe,

1aRu, 1bRu, 1Ni, 2Fe, 2aRu, 2bRu, and 2Ni in CH₂Cl₂

a

0.43−0.63 V), and nickel (E_pa= 0.78−0.86 V) complexes are in

agreement with the expected ease of oxidation of the

organometallic moiety ([Fe] > [Ru] > [Ni]), considering

their electronic properties. In the case of ruthenium complexes,

replacing 2 PPh₃by DPPE gives an expected decrease of the

oxidation potential (40 mV), in accordance with the relative

donor strength of the phosphane coligands. The chain

lengthening of the acetylide ligand leads to a signiﬁcant

decrease in oxidation potentials, with the most signiﬁcant

variation (140 mV) being for iron complexes. This trend is

analogous to that observed for the related benzene-based

acetylide iron(II),²⁰ruthenium(II),^24,25and nickel(II)³³com-

plexes and also for analogous iron(II)¹⁶and ruthenium(II)¹⁵

nitrile compounds. These results conﬁrm the evidence of the

spectroscopic data discussed above: namely, the magnitude of

the metal to ligand π back-donation, which was found to

decrease as the chain lengthening increases. The reversibility of

the oxidation process depends on the organometallic moiety: it

is reversible or quasi-reversible for iron and ruthenium

complexes and irreversible for nickel complexes. In the case

of ruthenium complexes, the i_pc/i_paratio (in the range 0.3−0.6)

suggests some instability of the 17e species [Ru (η⁵-C₅H₅)-

(PP)(CC{C₄H₂S}_nNO₂)]⁺at the electrode surface (Figure

6). Scan rate studies showed that the process became more

reversible when the scan direction was immediately reversed

after the oxidation potential for high scan rates (1000 mV s⁻¹).

In contrast, for the lower scan rate (50 mV s⁻¹) the process

became more irreversible (see the Supporting Information,

Figure S15). This behavior can be associated with the

formation of an unstable oxidized ruthenium species followed

by fast decomposition. At higher scan rates, the scan direction

was reversed before an appreciable amount of decomposition

occurred. Furthermore, experiments performed at −20 °C also

showed some improvement in the reversibility, the values of the

i_pc/i_paratio being augmented to 0.5−0.8 (see the Supporting

Information, Figure S16), which suggests that the lower

temperature diminishes the kinetics of the decomposition

process.

As mentioned above, all complexes show one redox process,

in the range of −1.01 to −1.20 V, which can be assigned mainly

to reduction at the nitro group of the coordinated acetylide. In

fact, according to the DFT calculations discussed above, the

LUMOs of the model complexes are mainly located at this

group (Figures 3 and 4). It is well-known that HOMO−LUMO

gaps from electrochemical data can be correlated with optical

properties.⁵⁰⁻⁵²In the present study, care should be taken in

the quantitative correlation between HOMO−LUMO gaps

from electrochemical data and optical properties due to the

quasi-reversible/irreversible nature of the redox processes.

However, a trend seems to be found between the HOMO−

LUMO gap and the lowest-energy band observed in UV−vis

spectra: the sequence of nickel, ruthenium, and iron complexes

and chain lengthening of the acetylide ligands leads to a

decrease in the HOMO−LUMO gap.

compd

E_pc(V)

E_pa(V)

E_1/2(V)

E_pa− E_pc(mV)

i_pc/i_pa

1Fe

1.59

0.33

0.25

0.29

80

1.0

−1.0

−1.10

1.60

2Fe

0.10

0.20

0.15

0.59

100

140

1.0

0.3

−1.08

−0.94

1.28

1aRu

0.55

−1.09

−1.33

0.64

100

−1.13

1.29

1bRu

0.58

−1.06

−1.24

0.67

0.63

90

140

100

0.5

−1.10

1.10

2aRu

2bRu

0.38

0.48

0.43

0.47

0.5

0.6

−1.07

1.12

0.52

0.42

100

−1.08

1Ni

2Ni

0.86

b

−1.17

−0.99

0.78

−1.07

a

b

All values vs SCE (v = 200 mV s⁻¹). Small waves.

Table 7. Cyclic Voltammetric Data for Compounds 1Fe, 1Ni,

2Fe, and 2Ni in NCMe

a

compd

E_pc(V)

E_pa(V)

E_1/2(V)

E_pa− E_pc(mV)

i_pc/i_pa

1Fe

0.20

−0.99

−1.19

0.07

0.28

0.24

80

1.0

b

−1.09

0.16

−1.14

0.12

100

90

0.9

2Fe

1Ni

2Ni

1.0

−0.77

−1.0

−0.88

0.73

140

120

−0.52

−1.08

b

−0.99

0.72

−1.03

0.6

−0.53

−1.09

b

−0.97

−1.03

0.8

a

b

All values vs SCE (v = 200 mV s⁻¹). i_pa/i_pc

The main electrochemical behavior in dichloromethane is

characterized by the presence of one oxidation process, whose

reversibility depends on the organometallic fragment, and one

irreversible reductive process. In general, the oxidation process

is usually viewed as essentially involving the M^II/M^IIIcouple,

but in the present case it is also associated with the acetylide

ligand. In fact, according to our DFT calculations the mixing of

the metal and acetylide frontier orbitals leads to HOMOs with

both metal-centered and acetylide ligand contributions (Figures

3 and 4). These results are in accordance with the behavior

found for several arylalkynyl complexes that have shown a

signiﬁcant participation of the acetylide ligand in the oxidation

In addition to the oxidation and reduction processes

discussed above, one irreversible oxidation, with E_pa= 1.59−

1.60 V (for iron complexes) and E_pa= 1.10−1.29 V (for

ruthenium complexes) is also observed, probably due to an

additional oxidation at the thiophene moiety. For compound

I

dx.doi.org/10.1021/om4001204 | Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

1bRu a small irreversible cathodic wave, with E_pc= ca. −0.5 V,

is also present and is attributed to decomposition products

originating at higher oxidation potentials. In fact, this process

vanishes when the direction of the cyclic voltammogram scan is

moved toward negative potentials before proceeding to the

oxidation of the complex (see the Supporting Information,

Figure S17).

Table 8. Quadratic Hyperpolarizabilities and Relevant

Spectroscopic Data

λ_exp(nm) (ε

(10⁴M⁻¹

cm⁻¹))

a

b

g

compd

β

β_corr

β₀

ref

h

1Fe

578 (1.3)

613 (1.5)

504 (0.8)

479 (0.5)

232

434

80

97

2Fe

120

325

c

The electrochemical experiments in acetonitrile showed the

same general behavior that was observed in dichloromethane.

The ruthenium complexes, however, decompose in this solvent

in the presence of the supporting electrolyte, and no results are

available for these complexes. The redox processes assigned to

reduction of the coordinated acetylide ligands, in the range

−0.95 to −1.14 V, became more reversible for iron and nickel

complexes. No signiﬁcant changes were observed in the

reversibility of the oxidation process. For the studied

complexes, the oxidation potential is lower in acetonitrile,

which reﬂects a relative destabilization of the HOMOs in

comparison to that observed in dichloromethane. On the other

hand, reduction potentials are shifted to lower negative values

in acetonitrile, which reveals the stabilization of the LUMOs.

The overall result is that HOMO−LUMO gaps are lower in

acetonitrile in comparison to those observed in dichloro-

methane. This trend seems to agree with the solvatochromic

behavior of the lower-energy bands in UV−vis spectra (see

above), where a bathochromic shift upon an increase in the

polarity of the solvent was found.

[Fe](CC-ph-NO₂)

1160

92

20

c

[Fe](CC-(ph)₂-

1150

174

NO₂)

d

[Fe](NC-th-

579 (sh)

535 (sh)

255

97

16

NO₂)⁺PF₆

−

d

[Fe](NC-(th)₂-

177

82

NO₂)⁺PF₆

−

i

1aRu

517 (2.0)

536 (2.2)

566 (2.8)

580 (2.4)

326

192

506

655

151

82

99

1bRu

96

356

321

2aRu

187

224

2bRu

e

ce

,

[Ru](CC-ph-NO₂)

460 (1.1)

468

560

96

25

e

ce

,

ce

,

[Ru](CC-(ph)₂-

448 (1.6)

134

NO₂)

f

df

,

df

,

[Ru](CC-ph-(E)-

533 (2.4)

294

333

210

419

138

163

109

195

22

CHCH-th-NO₂)

f

df

,

df

,

[Ru](CC-th-(E)-

CHCH-ph-NO₂)

522 (3.4)

f

df

,

df

,

[Ru](CC-ph- C

505 (2.0)

C-th-NO₂)

f

df

,

df

,

[Ru](CC-(th-(E)-

CHCH)₂-ph-

NO₂)

536 (4.5)

Quadratic Hyperpolarizabilities. Experimental data for

all complexes at 1500 nm obtained by hyper-Rayleigh scattering

and two-level-model (TLM) corrected values are collected in

Table 8, together with data for the previously reported

benzene-based iron(II), ruthenium(II), and nickel(II) acetylide

complexes and also for analogous iron(II) nitrile compounds

for comparison. Relevant experimental spectroscopic data and

static quadratic hyperpolarizabilities calculated by DFT on our

chosen model complexes were also included.

1Ni

484 (0.7),

404 (0.2)

331

363

173

175

81

2Ni

506 (2.3)

242

e

ce

,

ce

,

[Ni](CCC₆H₄NO₂) 439 (0.9)

221

193

59

65

33

e

ce

,

ce

,

[Ni](C

413 (1.6)

C{C₆H₄}₂NO₂)

All measurements were performed in CHCl₃solution except where

otherwise indicated. The β values are expressed in units of 10⁻³⁰esu

and were measured at 1500 nm (β₁₅₀₀) except where otherwise

indicated (experimental error 15%). Abbreviations: [Fe],

CpFeDPPE; [Ru], CpRu(PPh₃)₂; [Ni], CpNiPPh₃; ph, C₆H₄; th,

a

Earlier studies on related benzene-based iron(II),²⁰

ruthenium(II)^24,25and nickel(II)³³acetylide complexes re-

ported β values measured at 1064 nm that showed substantial

resonance enhancement, due to the existence of an optical

transition in the visible range, very close to the second

harmonic at 532 nm. It is well-known that for a reliable

comparison of β between diﬀerent molecules, and for a

comparison with theoretical values, it is important to correctly

account for these resonance eﬀects and extrapolate β to the

static ﬁrst hyperpolarizability. The simple two-level model

(TLM)²⁹is widely used in the literature for this purpose but

ignores any kind of line-broadening mechanisms and is not

applicable near resonance,^18,53which points out the importance

of an appropriate theoretical β dispersion model for a reliable

extrapolation. Recently, a practical model for ﬁrst hyper-

polarizability dispersion, accounting for both homogeneous and

inhomogeneous broadening eﬀects, was proposed on the basis

of the relation between absorption and HRS spectra from

tunable wavelength HRS measurements.⁵⁴

Considering the λ_maxfor the compounds studied in this work,

very high resonance enhancements could be expected for β

measurements at 1064 nm. Therefore, to avoid this situation,

our HRS measurements here presented were performed at the

higher fundamental wavelength of 1500 nm to avoid super-

position of the UV−vis absorptions and the second harmonic

signal (750 nm), thus leading to more reasonable results

according to the TLM analysis. The use of the TLM formalism

b

C₄H₂S. β corrected for resonance enhancement using the two-level

model with β_corr= β[1 − (2λ_max/λ_HRS)²][1 − (λ_max/λ_HRS)²] (damping

c

d

e

f

g

factors not included). β₁₀₆₄

. β₁₅₅₀. In THF. In CH₂Cl₂. Static

quadratic hyperpolarizability of the model complexes calculated by

DFT (the β tensor components are shown in the Supporting

Information). Reference 23. Reference 40.

h

i

for the compounds studied in this work should be questioned

at this point. This model assumes that for the contribution to β

only one excited state is coupled strongly enough to the ground

state by the applied electric ﬁeld, and only one tensor

component dominates the second-order NLO response (i.e.,

an unidirectional charge-transfer transition). The computed

values of static hyperpolarizabilities of our model complexes

show that β₀is clearly dominated by the tensor component

along the charge transfer axis, β_xxx(the β tensor components

are shown in the Supporting Information, Table S3). However,

as discussed above, the low-energy bands of 1Fe, 1aRu, 1bRu,

and 2Ni are quite broad, revealing the existence of two CT

contributions close in energy, as predicted by the TDDFT

calculations in the corresponding model complexes. In addition,

non-negligible contributions to the optical nonlinearity could

be present for all complexes originating in the higher energy

bands in the range 304−410 nm. In fact, our TDDFT

calculations showed that these bands can also be assigned

J

dx.doi.org/10.1021/om4001204 | Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

mainly to MLCT transitions, with the expected large change in

dipole moment upon excitation (see Table 5). Thus, the TLM

formalism should be used with caution for our studied

complexes, in particular for 1Fe, 1aRu, 1bRu, and 2Ni.

Probably, a description of β using the dispersion model

proposed by Campo et al.⁵⁴mentioned above could be an

improvement to consider in future work.

ruthenium acetylides with an end-capping nitro group and

thienyl and benzene entities in the conjugation chain²²and

benzene-based ruthenium²⁵and iron²⁰acetylide complexes

(Table 8). Our results also show that iron and ruthenium

complexes are more sensitive to the extension of the aromatic

system than the nickel analogues. This result agrees with the

trend found for the relative inﬂuence of the chain lengthening

of the acetylide ligand on the degree of π back-donation

interaction (IR data), the bathochromic eﬀect of the low-energy

transition bands (UV−vis data), and oxidation potentials

(electrochemical data). Thus, for the same organometallic

fragment, higher hyperpolarizabilities are associated with

complexes having higher electron density at the organometallic

moiety and low-energy transition bands. In addition, a

hyperchromic eﬀect is observed for these bands, which could

also contribute to the enhanced hyperpolarizabilities. Consid-

ering the relative magnitude of the quadratic hyperpolariz-

abilities for compounds with two thiophene rings, the data

show that β and β_corrfollow respectively the trends 2Ni < 2Fe <

2aRu < 2bRu and 2Fe < 2Ni ≈ 2aRu ≈ 2bRu. However,

considering that β_corrvalues of 2aRu, 2bRu, and 2Ni are almost

within the experimental error, a coherent analysis of the results

is not straightforward. Some evidence pointed out by

theoretical calculations could give an insight into the behavior

of quadratic hyperpolarizabilities presented by the complexes

with two thiophene rings. For 2Fe, 2aRu, and 2bRu the lowest-

energy band, observed in UV−vis spectra, can be attributed to a

single transition (HOMO → LUMO) with a pure MLCT

character (see Table 5 and Figure 4), according to our TDDFT

calculations for the respective model complexes 2′Fe, 2a′Ru,

and 2b′Ru. For this reason, the use of the TLM formalism

seems to be reasonable for these complexes. In spite of the

expected small change in the dipole moment upon excitation

(according to the solvatochromic results) and the higher energy

of the main bands in the UV−vis spectra of ruthenium

complexes (λ_exp566−580 nm) in comparison to the iron

compound (λ_exp616 nm), the quadratic hyperpolarizabilities for

ruthenium complexes are higher. Probably the relative intensity

of these bands also plays an important role. In fact, a larger

oscillator strength was found for 2aRu and 2bRu (ε = 2.8 × 10⁴

and 2.4 × 10⁴M⁻¹cm⁻¹, respectively) in comparison to 2Fe

(1.5 × 10⁴M⁻¹cm⁻¹). Accordingly, 2bRu has also a higher

quadratic hyperpolarizability than 2aRu. As discussed for the

complexes with one thiophene ring, we cannot exclude the

possibility of a contribution to the overall hyperpolarizability of

the higher energy band observed in experimental UV−vis

spectra for the complexes with two thiophene rings (λ 373−410

nm). It is interesting to note that, according to the TDDFT

calculations on the model complexes, this band for 2aRu could

have a contribution of charge transfer within the organometallic

fragment, since the phosphane coligand increases its electronic

density upon excitation (Table 5; λ_eg380 nm). This can lead to

some hampering eﬀect on the overall hyperpolarizability in

comparison to that found for 2bRu. According to theoretical

calculations on the model complex 2b′Ru, a charge transfer

within the organometallic fragment seems to be absent, which

could result in an overall eﬃciency for this complex in

comparison to that found for 2aRu. Remarkably, 2bRu presents

a value of the corrected hyperpolarizability (β_corr= 224 × 10⁻³⁰

esu) that places this complex in the range found for the best

values of β_corrreported in the literature for an organometallic

compound with quadratic hyperpolarizabilities measured at

1500 nm.^4−7,9,10In the case of 2Ni the use of TLM analysis is

Our ﬁrst hyperpolarizability results show that β_corrfollows the

trend 1Fe ≈ 1bRu < 1aRu ≈ 1Ni for the complexes with one

thiophene ring. This trend is somewhat surprising, in particular

for 1Ni. Indeed, studies on related benzene-based iron(II),²⁰

ruthenium(II)^24,25and nickel(II)³³acetylide complexes sug-

gested that the more easily oxidizable iron and ruthenium

complexes lead to higher quadratic hyperpolarizabilities. The

corrected hyperpolarizability for 1Ni is almost 3 times that of

[Ni(η⁵-C₅H₅)(PPh₃)(CCC₆H₄NO₂)], and the correspond-

ing values of 1Fe and 1bRu, whose relative magnitudes are

within the experimental error, are comparable to those found

for related benzene-based iron(II) and ruthenium(II) acetylide

complexes and an iron thiophene nitrile derivative (see Table

8). According to TLM we would be expect a greater second-

order NLO response for iron and ruthenium complexes, since

they have higher oscillator strength at the main MLCT bands

(1Fe, λ_exp578 nm; 1bRu, λ_exp536 nm; 1aRu, λ_exp517 nm; 1Ni,

λ_exp484 and 404 nm). Furthermore, according to the

solvatochromic results a larger change in the dipole moment

is also expected for the iron and ruthenium complexes (1Fe,

1bRu, and 1aRu). Therefore, this apparently contradictory

trend was analyzed in light of TDDFT calculations. For 1Fe,

1aRu, and 1bRu, the experimental broad absorption band

could be due to the existence of two CT contributions closer in

energy, as predicted by TDDFT calculations (Table 5). In

addition to the main HOMO−LUMO character of the overall

CT process, also some contribution of a HOMO-1 to LUMO

+3 transition (for 1′Fe) and HOMO to LUMO+1 transition

(for 1a′Ru and 1b′Ru) could be present in the broad band

observed in the experimental spectra. In the case of 1′Fe this

charge transfer is within the organometallic fragment, while for

1a′Ru and 1b′Ru the charge transfer has mainly LMCT

character, as can be seen in the representation of the involved

orbitals (Figure 3). The resulting CT process is, thus,

somewhat complex and these additional transitions would

weaken the charge transfer eﬃciency from the metal donor

fragment to the NO₂acceptor group (a better CT eﬃciency

would be expected for a purely MLCT process). For 1Ni a very

broad and complex experimental band was observed with λ

centered at 484 and 404 nm (Table 2). According to the

TDDFT calculations on the model complex 1′Ni, the lowest-

energy band (λ 484 nm) and second band (λ 404 nm) can be

attributed respectively to the single transitions HOMO →

LUMO and HOMO-1 → LUMO, both clearly with a MLCT

character (see Table 5 and Figure 3). Since they are relatively

closer in energy, probably these two bands contribute to the

observed hyperpolarizability for 1Ni and the purely MLCT

processes certainly lead to improved CT eﬃciency. Thus, the

improved CT eﬃciency expected for 1Ni, in comparison to that

found for ruthenium and iron complexes, could help to explain

the overall trend on the quadratic hyperpolarizabilities for the

complexes with one thiophene ring.

The chain lengthening of the acetylide chromophores, with

the introduction of one thiophene ring, leads to higher

quadratic hyperpolarizabilities for all the complexes. The β_corr

values are in the range of those observed in a series of

K

dx.doi.org/10.1021/om4001204 | Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

also deceiving, since our TDDFT calculations on the model

complex 2′Ni (Table 5) predicted a contribution of two

electronic transitions closer in energy to the experimental

lowest energy band. In addition to the HOMO → LUMO

transition, the contribution of the HOMO-1, HOMO-4, and

LUMO+1 orbitals to the charge transfer could lead to a

hampering eﬀect on the overall hyperpolarizability, since these

orbitals are mainly centered at the organometallic fragment

(Figure 4). In comparison to the behavior found for 1Ni, where

a high quadratic hyperpolarizability was found, this hampering

eﬀect could explain the small increase in β (and almost

unchanged β_corr) with the chain lengthening of the acetylide

chromophore for this organometallic moiety.

In our previous studies on oligo-thiophene nitrile iron

complexes the resonant quadratic hyperpolarizabilities were

found to increase with the chain lengthening of the

chromophore.¹⁶The corrected values, however, were found

to be practically independent of the number of thiophene units

in the conjugated ligand. This constancy upon chain length-

ening was explained in terms of a competition between the

growing conjugation length, which tends to raise β, and a

decrease of the CT eﬃciency, having a β lowering eﬀect. In the

case of the present thiophene acetylide complexes a better CT

eﬃciency seems to be obtained upon chain lengthening. In fact,

as was discussed, TDDFT calculations on the model complexes

have shown that relevant optical transitions have clearly MLCT

character and a large amount of charge is transferred from the

organometallic moiety to the chromophore, in particular for the

acceptor NO₂group, which largely increases its electronic

density upon excitation.

The DFT approach has been increasingly used to calculate

static quadratic hyperpolarizabilities of organometallic com-

plexes.⁵⁵⁻⁵⁹In recent studies we calculated the static quadratic

hyperpolarizabilities of 1′Fe²³and 1a′Ru⁴⁰using the DFT/

B3LYP level of theory, and now we have extended these

calculations to the remaining model complexes using the same

level of theory for comparison. We are not as interested in

comparing the absolute values of calculated hyperpolarizabil-

ities to the experimental values, only in the observed trends,

since several approximations were used (model phosphane

coligands, isolated molecules in the gas phase (cf. experimental

data obtained in solution), calculated static-frequency quadratic

hyperpolarizability, β₀, vs experimental β₁₅₀₀). In comparison to

experimental data, the accuracy of calculated hyperpolariz-

abilities using the same level of theory depends on the studied

molecules. For instance, DFT/B3LYP was found to reproduce

reasonably well experimental quadratic hyperpolarizabilities for

iron and ruthenium sesquifulvalene complexes⁵⁸but can

signiﬁcantly overestimate or underestimate quadratic hyper-

polarizabilities of ferrocenyl derivatives.⁵⁵Also, it is well-known

that calculated hyperpolarizabilities strongly depend on the

level of theory used.^59,60,55The results show that the calculated

β₀overestimates the experimental β_corrvalues in most cases

(only for 1aRu and 1Ni was an underestimation observed). It is

diﬃcult to compare the trend found experimentally to that

observed from computed values, due to the fact that several

compounds have experimental quadratic hyperpolarizabilities

falling within the experimental error. The computed hyper-

polarizabilities increase ca. 3.0−3.6 times upon chain length-

ening of the acetylide chromophore, whereas experimentally

this enhancing eﬀect is slightly smaller (up to ca. 2.7 times),

probably due to the overestimation of quadratic hyper-

polarizabilities from DFT calculations in complexes with

extended π-conjugated systems.⁶⁰

CONCLUSIONS

■

A series of mono(η⁵-cyclopentadienyl)iron(II), -ruthenium(II)

and -nickel(II) complexes with nitro-substituted thienyl

acetylide ligands has been synthesized and fully characterized.

The spectroscopic and cyclic voltammetric data suggest an

improved coupling between the organometallic fragment and

the conjugated thienyl acetylide ligands, in comparison to the

previously reported parent benzene-based acetylides and

thiophene nitrile complexes. TDDFT studies on model

complexes allowed us to achieve a deep knowledge of the

relevant electronic transitions involved, which was very helpful

in the discussion of the experimental quadratic hyper-

polarizabilities. These properties have been determined by

HRS measurements at 1500 nm. It is well recognized that, as a

general behavior, better electron donor/acceptors and chain

lengthening of the conjugated chromophore in push−pull

systems should lead to higher quadratic hyperpolarizabilities.

However a complexity in the push−pull systems based on

organometallic donor moieties was found in the present study.

The results seem to indicate that the eﬀectiveness of an

organometallic donor on the second-order NLO properties

depends on the chain lengthening of the conjugated

chromophore: the nickel organometallic fragment originates

an enhanced eﬀect on the magnitude of quadratic hyper-

polarizability for short-length acetylide chromophores, whereas

iron and ruthenium have comparatively higher NLO merit for

long-length chromophores. In other words, iron and ruthenium

complexes seem to be more sensitive to the extension of the

aromatic system than the nickel analogues. In particular, the

existence of other electronic transitions in addition to the main

MLCT can clearly control the eﬀectiveness of the organo-

metallic donors on the second-order NLO properties of these

push−pull systems. This leads to a somewhat surprisingly high

quadratic hyperpolarizability for 1Ni in comparison with 1Fe,

1aRu, and 1bRu in spite of the better donor properties of iron

and ruthenium organometallic moieties. [Ru(η⁵-C₅H₅)-

(PPh₃)₂(CC{C₄H₂S}₂NO₂)] presents a value of the

corrected hyperpolarizability (224 × 10⁻³⁰esu) which places

this complex in the range found for the best values of β_corr

reported in the literature.

EXPERIMENTAL SECTION

■

General Procedures. Syntheses were carried out under a nitrogen

atmosphere using standard Schlenk techniques, and the solvents used

were dried by standard methods.⁶¹Commercial reagents were used

without further puriﬁcation, except for copper iodide, which was dried

and kept under nitrogen. Starting materials were prepared by following

the methods described in the literature: [Ru(η⁵-C₅H₅)(PPh₃)₂Cl] and

[Ru(η⁵-C₅H₅)(DPPE)Cl],⁶²[Fe(η⁵-C₅H₅)(DPPE)I],¹⁶[Ni(η⁵-C₅H₅)-

(PPh₃)Cl].⁶³2-(5-Bromothiophen-2-yl)-5-nitrothiophene was pre-

pared by bromination of 2-nitro-5-(thiophen-2-yl)thiophene with

NBS according to the general method described in the literature.⁶⁴2-

Nitro-5-(thiophen-2-yl)thiophene was prepared by nitration of 2,2′-

bithiophene with Claycop according to general procedures.⁶⁵The

syntheses of trimethyl((5-nitrothiophen-2-yl)ethynyl)silane (1),³⁴2-

ethynyl-5-nitrothiophene (1′),³⁴trimethyl((5′-nitro-2,2′-bithiophen-5-

yl)ethynyl)silane (2),³⁰5-ethynyl-5′-nitro-2,2′-bithiophene (2′),³⁰and

Ru(η⁵-C₅H₅)(PPh₃)₂(CC{C₄H₂S}NO₂) (1bRu)³⁴were previously

reported. In this work we used some modiﬁcations for the preparation

of 1, 1′, 2, and 2′ (see details below): trimethyl((5-nitrothiophen-2-

yl)ethynyl)silane (1) was prepared using Pd(PPh₃)₂Cl₂as precatalyst

L

dx.doi.org/10.1021/om4001204 | Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

instead of Pd(OAc)₂/PPh₃;³⁴trimethyl((5′-nitro-2,2′-bithiophen-5-

yl)ethynyl)silane (2) was prepared from the bromo-substituted

precursor instead of the reported iodo derivative³⁰using standard

procedures; a diﬀerent deprotection procedure for the trimethylsilane

group was used for obtaining 2-ethynyl-5-nitrothiophene (1′)³⁴and 5-

ethynyl-5′-nitro-2,2′-bithiophene (2′).³⁰¹H, ¹³C, and ³¹P NMR

spectra were recorded on a Bruker Avance 400 spectrometer at

(C6), 126.65 (C5), 129.71 (C9), 133.71 (C4), 135.77 (C3), 144.08

(C7), 149.73 (C10). UV−vis (CHCl₃): λ_max/nm (ε/10⁴M⁻¹cm⁻¹)

416 (2.2).

Synthesis of HCC{C₄H₂S}₂NO₂(2′). Compound 2 (0.25 g, 0.8

mmol) was dissolved in methanol (70 mL). Ammonium ﬂuoride

(0.033 g, 0.88 mmol) was added. After the mixture was stirred at room

temperature for 2 h, an orange solid precipitated. After 3 h of

additional stirring, the solvent was removed under vacuum to give an

orange solid. The solid was washed with dichloromethane and the

solvent evaporated to give the desired product as a light orange solid in

77% yield (0.145 g). ν(KBr)/cm⁻¹: 1325s and 1511s (NO₂), 2096m

1

probe temperature using CDCl₃or (CD₃)₂CO as solvents. The H

and ¹³C chemical shifts (s, singlet; d, doublet; t, triplet; m, multiplet

1

for H) are reported in parts per million (ppm) downﬁeld from the

residual solvent peak, and the ³¹P NMR spectra are reported in ppm

downﬁeld from the external standard 85% H₃PO₄. Coupling constants

are reported in Hz. Spectral assignments follow the numbering scheme

shown in Scheme 1 and are attributed using HMBC, HMQC, and

COSY NMR techniques. FT-IR spectra were recorded with a Perkin-

Elmer Paragon 1000 PC spectrophotometer as dry KBr pellets (only

signiﬁcant bands are cited). UV−vis spectra were recorded with a

Termo Electron, Nicolet Evolution 300 instrument, in dried solvents

in the range of 200−1000 nm. Elemental analyses were obtained at our

1

(CC). H NMR (CDCl₃): δ/ppm 3.50 (s, 1H, CH), 7.09 (d, 1H,

³J_HH= 4.4 Hz, H8), 7.21 (d, 1H, ³J_HH= 4.0 Hz, H4), 7.23 (d, 1H, ³J_HH

3

= 4.0 Hz, H5), 7.85 (d, 1H, J_HH= 4.0 Hz, H9). ¹H NMR

3

((CD₃)₂CO): δ/ppm 4.27 (s, 1H, CH), 7.39 (d, 1H, J_HH= 4.0 Hz,

H4), 7.44 (d, 1H, ³J_HH= 4.0 Hz, H8), 7.56 (d, 1H, ³J_HH= 4.0 Hz, H5),

3

8.03 (d, 1H, J_HH= 4.0 Hz, H9). ¹³C NMR (CDCl₃): δ/ppm 75.98

(C2), 84.09 (C1), 123.01 (C8), 124.33 (C3), 126.09 (C4), 129.61

(C9), 134.20 (C5), 136.16 (C6), 143.74 (C7), not attributed (C10).

¹³C NMR ((CD₃)₂CO): δ/ppm 76.54 (C2), 86.30 (C1), 123.03 (C3),

124.10 (C4), 128.09 (C8), 131.30 (C9), 135.57 (C5), 136.92 (C6),

143.18 (C7), 150.81 (C10). UV−vis (CHCl₃): λ_max/nm (ε/10⁴M⁻¹

cm⁻¹) 410 (1.3).

́ ́ ́

laboratories (Laboratorio de Analises, Instituto Superior Tecnico),

using a Fisons Instruments EA1108 system. Data acquisition,

integration, and handling were performed using a PC with the

software package EAGER-200 (Carlo Erba Instruments).

Synthesis of Me₃SiCC{C₄H₂S}NO₂(1). 5-Bromo-2-nitrothio-

phene (1.0 g; 5 mmol) was dissolved in THF (10 mL). Pd(PPh₃)₂Cl₂

(0.07 g, 0.10 mmol), CuI (0.019 g, 0.10 mmol), and NEt₃(1.2 g, 7.9

mmol) were added. A solution of ethynyltrimethylsilane (0.55 g, 5.5

mmol) in THF (20 mL) was added dropwise, and the mixture was

stirred for 2 h. The resultant dark brown suspension was ﬁltered, and

the solid was washed with dichloromethane (10 mL). The combined

THF and dichloromethane solvents were evaporated under vacuum,

and the resultant brown residue was then chromatographed on silica

gel. Elution with petroleum ether 40−60/Et₂O (9/1) gave a yellow

band from which microcrystalline trimethyl((5-nitrothiophen-2-yl)-

ethynyl)silane was isolated in 76% yield (0.856 g). ν(KBr)/cm⁻¹:

1345s and 1523s (NO₂), 2141m (CC). ¹H NMR (CDCl₃): δ/ppm

0.28 (s, 9H, CH₃), 7.11 (d, 1H, ³J_HH= 4.0 Hz, H4), 7.78 (d, 1H, ³J_HH

= 4.4 Hz, H5). ¹³C NMR (CDCl₃): δ/ppm 0.21 (CH₃), 95.36 (C2),

105.42 (C1), 128.32 (C5), 130.61 (C3), 131.54 (C4), 150.76 (C6).

UV−vis (CHCl₃): λ_max/nm (ε/10⁴M⁻¹cm⁻¹) 356 (1.1).

Synthesis of Fe(η⁵-C₅H₅)(DPPE)(CC{C₄H₂S}NO₂) (1Fe). To a

suspension of Fe(η⁵-C₅H₅)(DPPE)I (0.40 g, 0.62 mmol) and 1′ (0.11

g, 0.65 mmol) in methanol (30 mL) was added NH₄PF₆(0.11 g, 0.68

mmol), and the mixture was reﬂuxed for 2 h and then cooled. A

solution of NaOMe in methanol (7.4 mL, 0.1 M) was added, and the

mixture was stirred for 1 h at room temperature. The solvent was

removed under vacuum, and the resultant solid residue was then

chromatographed on ﬂash silica gel. Elution with n-hexane/Et₂O (1/1)

gave a dark blue band from which the desired compound was isolated

as a dark blue solid in 76% yield (0.32 g). ν(KBr)/cm⁻¹: 1421s and

1

1292s (NO₂), 2005m (CC). H NMR ((CD₃)₂CO): δ/ppm 2.51

(m, 2H, CH₂-DPPE), 2.63 (m, 2H, CH₂-DPPE), 4.39 (s, 5H, C₅H₅),

3

5.92 (d, 1H, J_HH= 4.6 Hz, H4), 7.60 (d, 1H, J_HH= 4.6 Hz, H5),

7.34−7.44 and 7.90 (m, 20H, H-Ph). ¹³C NMR ((CD₃)₂CO): δ/ppm

30.41 (CH₂-DPPE), 81.24 (s, C₅H₅), 118.52 (C2), 125.37 (C4),

2

128.68 (t, J_CP= 9.0 Hz, C_ortho-Ph), 129.13 (t, J_CP= 8.1 Hz, C_ortho

-

Ph), 130.40 (s, C_para-Ph), 131.67 (C5), 131.69 (s, C_para-Ph), 132.51 (t,

Synthesis of HCC{C₄H₂S}NO₂(1′). Compound 1 (0.338 g, 1.5

mmol) was dissolved in methanol (10 mL). Ammonium ﬂuoride

(0.061 g, 1.65 mmol) was added. After the mixture was stirred at room

temperature for 2 h, the solvent was removed under vacuum to give a

brown residue. The solid was washed with petroleum ether 40−60 and

the solvent evaporated to give the desired product as a light yellow

microcrystalline powder in 72% yield (0.165 g). ν(KBr)/cm⁻¹: 1341s

³J_CP= 8.8 Hz, C_meta-Ph), 134.24 (t, ³J_CP= 8.8 Hz, C_meta-Ph), 136.45 (t,

1

¹J_CP= 13.6 Hz, C_ipso-Ph), 138.18 (t, J_CP= 14.9 Hz, C_ipso-Ph), 140.05

2

(C3), 141.90 (t, J_CP= 28.5 Hz, C1), 143.34 (C6). ³¹P NMR

((CD₃)₂CO): δ/ppm 103.93 (s, 2P-DPPE). UV−vis (CHCl₃): λ_max

/

nm (ε/10⁴M⁻¹cm⁻¹) 578 (1.3). Anal. Calcd for C₃₇H₃₁FeNO₂P₂S: C,

66.18; H, 4.65; N, 2.09; S, 4.78. Found: C, 65.80; H, 5.00; N, 1.99; S,

5.00.

1

and 1528s (NO₂), 2101m (CC). H NMR (CDCl₃): δ/ppm 3.57

Synthesis of Fe(η⁵-C₅H₅)(DPPE)(CC{C₄H₂S}₂NO₂) (2Fe). To a

suspension of Fe(η⁵-C₅H₅)(DPPE)I (0.40 g, 0.62 mmol) and 2′ (0.16

g, 0.68 mmol) in methanol (50 mL) was added NH₄PF₆(0.11 g, 0.68

mmol), and the mixture was reﬂuxed for 3 h and then cooled. A

solution of NaOMe in methanol (7.4 mL, 0.1 M) was added, and the

mixture was stirred for 1 h at room temperature. The solvent was

removed under vacuum, and the resultant solid residue was then ﬂash

chromatographed on silica gel. Elution with n-hexane/Et₂O (2/3) gave

a dark blue band from which the desired compound was isolated as a

blue-purple solid in 78% yield (0.36 g). ν(KBr)/cm⁻¹: 1422s and

(s, 1H, CCH), 7.17 (d, 1H, ³J_HH= 4.0 Hz, H4), 7.80 (d, 1H, ³J_HH

=

1

4.0 Hz, H5). H NMR ((CD₃)₂CO): δ/ppm 4.47 (s, 1H, CCH),

7.41 (d, 1H, ³J_HH= 4.0 Hz, H4), 8.00 (d, 1H, ³J_HH= 4.0 Hz, H5). ¹³

C

NMR (CDCl₃): δ/ppm 75.13 (C2), 86.04 (C1), 128.12 (C5), 129.15

(C3), 132.17 (C4), 151.29 (C6). ¹³C NMR ((CD₃)₂CO): δ/ppm

74.48 (C2), 87.45 (C1), 128.79 (C3), 129.08 (C5), 133.04 (C4),

147.75 (C6). UV−vis (CHCl₃): λ_max/nm (ε/10⁴M⁻¹cm⁻¹) 346 (0.8).

Synthesis of Me₃SiCC{C₄H₂S}₂NO₂(2). 2-(5-Bromothiophen-

2-yl)-5-nitrothiophene (0.14 g, 0.5 mmol) was dissolved in THF (5

mL). Pd(PPh₃)₂Cl₂(0.017 g, 0.025 mmol), CuI (0.008 g, 0.04 mmol),

and NEt₃(0.3 mL, 1.98 mmol) were added. A solution of

ethynyltrimethylsilane (0.1 g, 1.0 mmol) in THF (10 mL) was

added dropwise, and the mixture was stirred overnight. The solvent

was evaporated under vacuum, and the resultant brown residue was

then chromatographed on silica gel. Elution with petroleum ether 40−

60/Et₂O (9/1) gave an orange band from which trimethyl((5′-nitro-

2,2′-bithiophen-5-yl)ethynyl)silane was isolated as an orange powder

in 70% yield (0.108 g). ν(KBr)/cm⁻¹: 1327s and 1515s (NO₂), 2141m

1

1298s (NO₂), 2024m (CC). H NMR ((CD₃)₂CO): δ/ppm 2.45

(m, 2H, CH₂-DPPE), 2.62 (m, 2H, CH₂-DPPE), 4.31 (s, 5H, C₅H₅),

6.00 (d, 1H, ³J_HH= 3.9 Hz, H4), 6.99 (d, 1H, ³J_HH= 4.4 Hz, H8), 7.17

(d, 1H, ³J_HH= 3.9 Hz, H5), 7.88 (d, 1H, ³J_HH= 3.9 Hz, H9), 7.33−7.46

and 7.94 (m, 20H, H-Ph). ¹³C NMR ((CD₃)₂CO): δ/ppm 29.84

(CH₂-DPPE), 80.41 (s, C₅H₅), 115.24 (C2), 121.26 (C8), 126.73

3

(C4), 127.45 (C6), 128.42 (C5), 128.46 (t, J_CP= 9.0 Hz, C_meta-Ph),

2

128.89 (t, J_CP= 9.0 Hz, C_ortho-Ph), 129.84 (s, C_para-Ph), 130.07 (s,

C_para-Ph), 131.56 (C9), 132.47 (t, ²J_CP= 8.5 Hz, C_ortho-Ph), 134.29 (t,

1

(CC). H NMR (CDCl₃): δ/ppm 0.27 (s, 9H, CH₃), 7.08 (d, 1H,

²J_CP= 8.6 Hz, C_meta-Ph), 135.42 (C3), 138.34 (t, ¹J_CP= 49.5 Hz, C_ipso

-

³J_HH= 4.4 Hz, H8), 7.18 (d, 1H, ³J_HH= 4.0 Hz, H4), 7.20 (d, 1H, ³J_HH

= 3.6 Hz, H5), 7.85 (d, 1H, J_HH= 4.4 Hz, H9). ¹³C NMR (CDCl₃):

Ph), 142.28 (t, J_CP= 31.3 Hz, C1), 147.36 (C7), 152.41 (C10). ³¹P

NMR ((CD₃)₂CO): δ/ppm 104.5 (s, 2P-DPPE). UV−vis (CHCl₃):

3

2

δ/ppm 0.15 (CH₃), 96.36 (C2), 102.55 (C1), 122.84 (C8), 125.68

M

dx.doi.org/10.1021/om4001204 | Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

λ_max/nm (ε/10⁴M⁻¹cm⁻¹

C₄₁H₃₃FeNO₂P₂S₂: C, 65.34; H, 4.41; N, 1.86; S, 8.51. Found: C,

64.80; H, 4.40; N, 1.90; S, 9.00.

)

613 (1.5). Anal. Calcd for

(s, 2P-PPh₃). UV−vis (CHCl₃): λ_max/nm (ε/10⁴M⁻¹c m⁻¹) 536 (2.2).

Anal. Calcd for C₄₇H₃₇NO₂P₂RuS: C, 66.97; H, 4.42; N, 1.66; S, 3.80.

Found: C, 67.00; H, 4.50; N, 1.60; S, 4.00.

Synthesis of Ru(η⁵-C₅H₅)(DPPE)(CC{C₄H₂S}NO₂) (1aRu). To

a suspension of Ru(η⁵-C₅H₅)(DPPE)Cl (0.30 g, 0.5 mmol) and 1

(0.13 g, 0.58 mmol) in methanol (30 mL) was added KF (0.045 g,

0.75 mmol), and the mixture was reﬂuxed for 4 h. The solvent was

removed under vacuum, and the resultant solid residue was then ﬂash

chromatographed on silica gel. Elution with n-hexane/Et₂O (1/1) gave

a purple band, from which the desired compound was isolated as a

purple solid in 70% yield (0.25 g). ν(KBr)/cm⁻¹: 1297s and 1419s

Synthesis of Ru(η⁵-C₅H₅)(PPh₃)₂(CC{C₄H₂S}₂NO₂) (2bRu).

To a suspension of Ru(η⁵-C₅H₅)(PPh₃)₂Cl (0.27 g, 0.50 mmol) and

2 (0.17 g, 0.52 mmol) in methanol (30 mL) was added KF (0.14 g, 2.4

mmol), and the mixture was reﬂuxed for 4.5 h. The solvent was

removed under vacuum, and the resultant solid residue was then ﬂash

chromatographed on silica gel. Elution with n-hexane/Et₂O (1/1) gave

a dark blue band, from which the desired compound was isolated as a

dark blue solid in 54% yield (0.25 g). ν(KBr)/cm⁻¹: 1309s and 1431s

(NO₂), 2037m (CC). ¹H NMR (CDCl₃): δ/ppm 4.36 (s, 5H,

1

(NO₂), 2023m (CC). H NMR (CDCl₃): δ/ppm 2.33 (m, 2H,

3

CH₂-DPPE), 2.58 (m, 2H, CH₂-DPPE), 4.84 (s, 5H, C₅H₅), 5.81 (d,

C₅H₅), 6.53 (d, 1H, J_HH= 3.9 Hz, H4), 6.90 (d, 1H, J_HH= 4.3 Hz,

3

1H, J_HH= 4.4 Hz, H4), 7.53 (d, 1H, J_HH= 4.4 Hz, H5), 7.30−7.42

and 7.80 (m, 20H, H-Ph). ¹³C NMR (CDCl₃): δ/ppm 28.26 (t, ¹J_CP

47.2, CH₂-DPPE), 83.05 (s, C₅H₅), 108.76 (C2), 125.47 (C4), 127.96

H8), 7.13−7.43 (m, 30H, H-PPh₃), 7.19 (d, 1H, J_HH= 3.9 Hz, H5),

7.81 (d, 1H, ³J_HH= 4.3 Hz, H9). ¹³C NMR (CDCl₃): δ/ppm 85.67 (s,

C₅H₅), 108.43 (C2), 120.25 (C8), 126.70 (C4), 127.41 (t, ³J_CP= 17.5

Hz, C_meta-PPh₃), 127.92 (C5), 128.69 (s, C_para-PPh₃), 130.44 (C9),

=

3

(t, J_CP= 27.2 Hz, C_meta-Ph), 128.13 (t, J_CP= 25.0 Hz, C_meta-Ph),

129.33 (s, C_para-Ph), 129.80 (s, C_para-Ph), 130.76 (C5), 131.41 (t, ²J_CP

2

133.70 (t, J_CP= 19.7 Hz, C_ortho-PPh₃), 135.30 (C7), 137.94 (C6),

2

138.36 (t, ¹J_CP= 84.4 Hz, C_ipso-PPh₃), 140.21 (t, ²J_CP= 83.1 Hz, C1),

146.80 (C3), 147.37 (C10). ³¹P NMR (CDCl₃): δ/ppm 49.98 (s, 2P-

PPh₃). UV−vis (CHCl₃): λ_max/nm (ε/10⁴M⁻¹cm⁻¹) 580 (2.4). Anal.

Calcd for C₅₁H₃₉NO₂P₂RuS₂: C, 66.22; H, 4.25; N, 1.51; S, 6.93.

Found: C, 66.44; H, 4.24; N, 1.53; S, 6.91.

= 28.8 Hz, C_ortho-Ph), 133.38 (t, J_CP= 27.6 Hz, C_ortho-Ph), not

attributed (C_ipso-Ph), 140.51 (C3), 142.43 (C6), 143.36 (t, ²J_CP= 21.0

Hz, C1). ³¹P NMR (CDCl₃): δ/ppm 84.62 (s, 2P-DPPE). UV−vis

(CHCl₃): λ_max/nm (ε/10⁴M⁻¹cm⁻¹) 517 (2.0). Anal. Calcd for

C₃₇H₃₁NO₂P₂RuS: C, 62.00; H, 4.36; N, 1.95; S, 4.47. Found: C,

62.12; H, 4.60; N, 1.90; S, 4.30.

Synthesis of Ni(η⁵-C₅H₅)(PPh₃)(CC{C₄H₂S}NO₂) (1Ni). Ni(η⁵-

C₅H₅)(PPh₃)Cl (0.42 g, 1.0 mmol), 1′ (0.17 g, 1.10 mmol), and CuI

(0.015 g, 0.08 mmol) were suspended in NEt₃/THF 4/1 (25 mL).

After the mixture was stirred at room temperature for 16 h, the solvent

was removed under vacuum and the resultant solid residue was

extracted with dichloromethane and then ﬂash chromatographed on

silica gel. Elution with n-hexane/Et₂O (4/1) gave a red band, from

which the desired compound was isolated as a wine red solid in 80%

yield (0.47 g). ν(KBr)/cm⁻¹: 1339s and 1516s (NO₂), 2074m (C

Synthesis of Ru(η⁵-C₅H₅)(DPPE)(CC{C₄H₂S}₂NO₂) (2aRu).

Ru(η⁵-C₅H₅)(DPPE)Cl (0.15 g, 0.25 mmol) and 2 (0.08 g, 0.26

mmol) were suspended in methanol (40 mL), and a few drops of

dichloromethane were added to ensure complete solubilization. After

the mixture was stirred at room temperature for 2.5 h and reﬂuxed for

an additional 3.5 h, TlPF₆(0.09 g, 0.27 mmol) was added and,

immediately, the yellow vinylidene derivative [Ru(η⁵-C₅H₅)(DPPE)-

(CCH{C₄H₂S}₂NO₂)][PF₆] precipitated. The solid was ﬁltered,

washed with diethyl ether, and dried under vacuum. Subsequent

redissolution in methanol (30 mL) with 5% acetone and addition of

sodium methoxide (0.016 g, 0.3 mmol) aﬀorded a dark purple

solution. The mixture was stirred for 1 h at room temperature. The

solvents were removed under vacuum, and the resultant solid residue

was then ﬂash chromatographed on silica gel. Elution with n-hexane/

Et₂O (1/1) gave a purple band, from which the desired compound was

isolated as a purple solid in 72% yield (0.14 g). ν(KBr)/cm⁻¹: 1291s

1

3

C). H NMR (CDCl₃): δ/ppm 5.26 (s, 5H, C₅H₅), 6.17 (d, 1H, J_HH

= 4.2 Hz, H4), 7.53 (d, 1H, ³J_HH= 4.2 Hz, H5), 7.39−7.47 and 7.60−

7.69 (m, 15H, H-PPh₃). ¹³C NMR (CDCl₃): δ/ppm 92.98 (s, C₅H₅),

3

111.35 (C2), 126.92 (C4), 128.42 (t, J_CP= 10.5 Hz, C_meta-PPh₃),

2

130.56 (s, C_para-PPh₃), 129.36 (C5), 115.24 (d, J_CP= 45.2 Hz, C1),

133.67 (t, ²J_CP= 11.1 Hz, C_ortho-PPh₃), 133.22 (t, ¹J_CP= 49.3 Hz, C_ipso

-

PPh₃), 137.65 (C3), 145.30 (C6). ³¹P NMR (CDCl₃): δ/ppm 41.65

(s, P-PPh₃). UV−vis (CHCl₃): λ_max/nm (ε/10⁴M⁻¹cm⁻¹) 484 (0.7).

Anal. Calcd for C₂₉H₂₂NNiO₂PS·¹/₅CH₂Cl₂: C, 63.17; H, 4.07; N,

2.52; S, 5.78. Found: C, 63.00; H, 4.10; N, 2.60; S, 6.00.

1

and 1412s (NO₂), 2037m (CC). H NMR (CDCl₃): δ/ppm 2.26

(m, 2H, CH₂-DPPE), 2.68 (m, 2H, CH₂-DPPE), 4.83 (s, 5H, C₅H₅),

5.63 (d, 1H, ³J_HH= 3.9 Hz, H4), 6.81 (d, 1H, ³J_HH= 4.5 Hz, H8), 7.00

Synthesis of Ni(η⁵-C₅H₅)(PPh₃)(CC{C₄H₂S}₂NO₂) (2Ni). Ni(η⁵-

C₅H₅)(PPh₃)Cl (0.42 g, 1.0 mmol), 2′ (0.26 g, 1.10 mmol), and CuI

(0.015 g, 0.08 mmol) were suspended in NEt₃/THF 4/1 (30 mL).

After the mixture was stirred at room temperature for 16 h, the solvent

was removed under vacuum and the resultant solid residue was

extracted with dichloromethane and then ﬂash chromatographed on

silica gel. Elution with n-hexane/Et₂O (1/1) gave a red band, from

which the desired compound was isolated as a dark red solid in 78%

3

(d, 1H, J_HH= 3.9 Hz, H5), 7.76 (d, 1H, J_HH= 4.5 Hz, H9), 7.32−

7.34 and 7.56 (m, 20H, H-Ph). ¹³C NMR (CDCl₃): δ/ppm 27.69 (t,

¹J_CP= 50.0 Hz, CH₂-DPPE), 83.78 (s, C₅H₅), 119.17 (C8), 119.66

(C4), 125.82 (C6), 127.55 (C2), 127.79 (t, ³J_CP= 17.4 Hz, C_meta-Ph),

128.01 (t, ³J_CP= 18.5 Hz, C_meta-Ph), 129.07 (s, C_para-Ph), 126.90 (C5),

2

129.11 (s, C_para-Ph), 130.81 (C9), 132.13 (t, J_CP= 20.0 Hz, C_ortho

-

2

1

Ph), 132.26 (t, J_CP= 19.1 Hz, C_ortho-Ph), 138.41 (t, J_CP= 85.9 Hz,

C_ipso-Ph), 141.28 (C7), 141.76 (C3), 145.25 (t, ²J_CP= 105.6 Hz, C1),

149.57 (C10). ³¹P NMR (CDCl₃): δ/ppm 85.33 (s, 2P-DPPE). UV−

vis (CHCl₃): λ_max/nm (ε/10⁴M⁻¹cm⁻¹) 566 (2.8). Anal. Calcd for

C₄₁H₃₃NO₂P₂RuS₂: C, 61.64; H, 4.16; N, 1.75; S, 8.03. Found: C,

61.20; H, 4.30; N, 1.70; S, 8.00.

1

yield (0.48 g). ν(KBr)/cm⁻¹: 1314s and 1517s, 2074m (CC). H

3

NMR (CDCl₃): δ/ppm 5.26 (s, 5H, C₅H₅), 6.27 (d, 1H, J_HH= 4.2

Hz, H4), 6.80 (d, 1H, ³J_HH= 4.0 Hz, H8), 6.94 (d, 1H, ³J_HH= 4.2 Hz,

3

H5), 7.74 (d, 1H, J_HH= 4.0 Hz, H9), 7.39−7.49 and 7.62−7.72 (m,

15H, H-PPh₃). ¹³C NMR (CDCl₃): δ/ppm 92.85 (s, C₅H₅), 110.79

3

Synthesis of Ru(η⁵-C₅H₅)(PPh₃)₂(CC{C₄H₂S}NO₂) (1bRu). To

a solution of Ru(η⁵-C₅H₅)(PPh₃)₂Cl (0.11 g, 0.15 mmol) and 1 (0.036

g, 0.16 mmol) in methanol (30 mL) was added KF (0.022 g, 0.30

mmol), and the mixture was reﬂuxed for 3 h. The solvent was removed

under vacuum, and the resultant solid residue was then ﬂash

chromatographed on silica gel. Elution with n-hexane/Et₂O (1/1)

gave a purple band, from which the desired compound was isolated as

a purple microcrystalline solid in 77% yield (0.10 g). ν(KBr)/cm⁻¹:

1294s and 1419s (NO₂), 2019m (CC). ¹H NMR (CDCl₃): δ/ppm

4.39 (s, 5H, C₅H₅), 6.39 (d, 1H, ³J_HH= 4.2 Hz, H4), 7.78 (d, 1H, ³J_HH

= 4.2 Hz, H5), 7.12−7.28 (m, 30H, H-PPh₃). ¹³C NMR (CDCl₃): δ/

ppm 86.10 (s, C₅H₅), 111.83 (C2), 125.38 (C4), 127.54 (t, ³J_CP= 20.0

Hz, C_meta-PPh₃), 128.92 (s, C_para-PPh₃), 131.04 (C5), 133.10 (C1),

(C2), 120.98 (C8), 126.26 (C5), 128.33 (t, J_CP= 10.5 Hz, C_meta

-

PPh₃), 128.98 (C4), 130.00 (C9), 130.36 (C6), 130.37 (s, C_para-PPh₃),

1

2

132.11 (C3), 133.29 (t, J_CP= 49.0 Hz, C_ipso-PPh₃), 133.75 (t, J_CP

=

2

11.0 Hz, C_ortho-PPh₃), 134.46 (t, J_CP= 21.8 Hz, C1), 146.21 (C7),

147.78 (C10). ³¹P NMR (CDCl₃): δ/ppm 41.14 (s, P-PPh₃). UV−vis

(CHCl₃): λ_max/nm (ε/10⁴M⁻¹cm⁻¹) 506 (2.3). Anal. Calcd for

C₃₃H₂₄NNiO₂PS₂·¹/₄CH₂Cl₂: C, 62.25; H, 3.85; N, 2.18; S, 10.00.

Found: C, 62.10; H, 3.90; N, 2.20; S, 10.00.

Electrochemical Studies. Electrochemical measurements were

performed on an EG&G Princeton Applied Research potentiostat/

galvanostat Model 273A equipped with Electrochemical PowerSuite

v2.51 software for electrochemical analysis, in anhydrous dichloro-

methane or acetonitrile with tetrabutylammonium hexaﬂuorophos-

phate (0.1 M) as supporting electrolyte. The electrochemical cell was a

homemade three-electrode conﬁguration cell with a platinum-disk

133.60 (t, ²J_CP= 20.0 Hz, C_ortho-PPh₃), 137.94 (t, ¹J_CP= 88.0 Hz, C_ipso

PPh₃), 142.39 (C6), 143.00 (C3). ³¹P NMR (CDCl₃): δ/ppm 49.94

-

N

dx.doi.org/10.1021/om4001204 | Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

working electrode (1.0 mm) probed by a Luggin capillary connected

to a silver-wire pseudoreference electrode and a platinum-wire

auxiliary electrode. All of the experiments were performed under a

nitrogen atmosphere at room temperature and in some special cases

also at −20 °C. All of the potentials reported were measured against

the ferrocene/ferrocenium redox couple as internal standard and

normally quoted relative to SCE (using the ferrocenium/ferrocene

redox couple E_1/2= 0.46 or 0.40 V versus SCE for dichloromethane or

acetonitrile, respectively⁶⁶). The electrochemical grade electrolyte was

purchased from Aldrich Chemical Co. and dried under vacuum at 110

°C for 24 h. Reagent grade solvents were dried, puriﬁed by standard

procedure,s⁶¹and distilled under a nitrogen atmosphere before use.

HRS Measurements of the First Hyperpolarizabilities. β

measurements were carried out using the harmonic light scattering

technique (also named hyper-Rayleigh) in chloroform solutions. The

10⁻³−10⁻⁵M solutions of the complexes were placed into a 4 cm long

ﬂuorimetric cell, after being carefully ﬁltered through a 0.2 μm ﬁlter in

order to eliminate the white light noise resulting from microburning of

any the remaining dust particles by the incoming laser beam. The

measurements were performed at a fundamental wavelength of 1500

nm as described in ref 67, using a Q-switched Nd:YAG laser operating

in the 10 Hz repetition range. Fluorescence checks were made by

replacement of the interference ﬁlter at the entrance of the

photomultiplier tube using ﬁlters with the transmittances at 650,

700, and 800 nm. Disperse Red 1 (DR1) was used as an external

standard dissolved in chloroform. The reference hyperpolarizability β

of DR1 in CHCl₃was measured by comparison of the slopes of I_2wvs

concentration plot of the standard in CH₂Cl₂and CHCl₃. Using the

hyperpolarizability of DR1 in dichloromethane (β_DR1(CH₂Cl₂) = 70 ×

10⁻³⁰esu⁶⁸) the hyperpolarizability of DR1 in CHCl₃was estimated to

be 80 × 10⁻³⁰esu, which is very close to the published value of 74 ×

10⁻³⁰esu.⁶⁷Assuming that the scattering contribution from the solvent

is negligibly small, this external reference method is used to calculate

the β values of complexes according to eq 1, where s is the slope of the

appropriate I_2wvs concentration plot and β_refis the orientational

average of the ﬁrst hyperpolarizability of the reference sample.

hyperpolarizabilities. The ﬁrst 24 lower excitation energies were

computed, and the simulated absorption bands were obtained by

convolution of Gaussian functions centered at the calculated excitation

energies using the GaussSum⁷⁹(version 2.2.4) software. The

Chemcraft⁸⁰program (version 1.6) was used for the visualization of

the computed results, including the representation of the geometries

and the orbitals.

ASSOCIATED CONTENT

■

S

* Supporting Information

Text, ﬁgures, and tables, giving the justiﬁcation for the use of

the simpliﬁcations in DFT calculations, TDDFT calculated

absorption spectrum for 2a′Ru, optimized geometries and β

tensors of all computed structures, cyclic voltammetry plots,

and selected NMR spectra. This material is available free of

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

AUTHOR INFORMATION

■

Corresponding Author

*P.J.M.: e-mail, pjgm@uevora.pt; tel, +351 266745318; fax,

+351 26645303.

Notes

The authors declare no competing ﬁnancial interest.

ACKNOWLEDGMENTS

■

The authors thank the FCT for funding of the project

FCOMP-01-0124-FEDER-007433. T.J.L.S. is also grateful for

his Ph.D. grant.

REFERENCES

■

(1) Nalwa, H. S. Appl. Organomet. Chem. 1991, 5, 349−377.

(2) Long, N. J. Angew. Chem., Int. Ed. Engl. 1995, 34, 21−38.

(3) Verbiest, T.; Houbrechts, S.; Kauranen, M.; Clays, K.; Persoons,

A. J. Mater. Chem. 1997, 7, 2175−2189.

s_sample

β =

β

reference

s_reference

(1)

(4) Whittall, I. R.; McDonagh, A. M.; Humphrey, M. G.; Samoc, M.

In Organometallic Complexes in Nonlinear Optics I: Second-Order

Nonlinearities; Stone, F. G. A., Robert, W., Eds.; Academic Press: New

York, 1998; Advances in Organometallic Chemistry 42, pp 291−362.

(5) Di Bella, S. Chem. Soc. Rev. 2001, 30, 355−366.

(6) Goovaerts, E.; Wenseleers, W. E.; Garcia, M. H.; Cross, G. H.

Design and characterization of organic and organometallic molecules

for second order nonlinear optics. In Handbook of Advanced Electronic

and Photonic Materials and Devices; Nalwa, H. S., Ed.; Academic Press:

Burlington, 2001; Chapter 3, pp 127−191.

(7) Powell, C. E.; Humphrey, M. G. Coord. Chem. Rev. 2004, 248,

725−756.

(8) Cariati, E.; Pizzotti, M.; Roberto, D.; Tessore, F.; Ugo, R. Coord.

Chem. Rev. 2006, 250, 1210−1233.

(9) Bozec, H.; Guerchais, V. Molecular Organometallic Materials for

Optics; Springer-Verlag: Berlin and Heidelberg, Germany, 2010; p 230.

(10) Morrall, J. P.; Dalton, G. T.; Humphrey, M. G.; Samoc, M. In

Organotransition Metal Complexes for Nonlinear Optics; West, R., Hill,

A., Fink, M. J., Eds.; Academic Press: New York, 2007; Advances in

Organometallic Chemistry 55, pp 61−136.

(11) Heck, J.; Dede, M. Ferrocene-Based Electro-Optical Materials.

In Ferrocenes; Wiley: Chichester, U.K., 2008; pp 319−392.

(12) Green, K. A.; Cifuentes, M. P.; Samoc, M.; Humphrey, M. G.

Coord. Chem. Rev. 2011, 255, 2025−2038.

DFT Calculations. All calculations were performed at the DFT

level using the Gaussian 09 package.⁶⁹The hybrid functional

B3LYP^70,71(Becke’s three-parameter functional with Lee−Yang−

Parr exchange correlations) were used for the calculations. As a

compromise between accuracy and computational eﬀort we have

adopted the 6-31G* basis set (for geometry optimizations) and the 6-

31+G* basis set (for the calculation of hyperpolarizabilities) for C, H,

N, O, and H and the LANL2DZ eﬀective core potential basis set for S,

P, Fe, Ru, and Ni.^72,73In the case of the hyperpolarizability

calculations the LANL2DZ basis set was also augmented with a

polarization function (exponents of 0.496 and 0.364) and a diﬀuse

function (exponents of 0.0347 and 0.0298) for elements S and P,

respectively.⁷⁴⁻⁷⁶Geometry optimizations were performed without

any symmetry constraints. In all cases, the Hessian was computed to

conﬁrm the stationary points of the potential energy surfaces (PES) as

true minima. The static ﬁrst hyperpolarizability, β_tot, for all compounds

was calculated by means of the analytic gradient methodology adopted

in the Gaussian 09 program package by using the equation

β

=

β ²+ β ²+ β ²

x y z

tot

(2)

upon calculating the individual static components

(13) Babgi, B.; Rigamonti, L.; Cifuentes, M. P.; Corkery, T. C.;

Randles, M. D.; Schwich, T.; Petrie, S.; Stranger, R.; Teshome, A.;

Asselberghs, I.; Clays, K.; Samoc, M.; Humphrey, M. G. J. Am. Chem.

Soc. 2009, 131, 10293−10307.

(14) Garcia, M. H.; Florindo, P.; Piedade, M. F. M.; Duarte, M. T.;

Robalo, M. P.; Goovaerts, E.; Wenseleers, W. J. Organomet. Chem.

2009, 694, 433−445.

1

3

β = β +

(β + β + β )

ijj jij jji

∑

i

iii

i≠j

(3)

Time-dependent density functional theory (TDDFT)^77,78was used

to compute the electronic spectra of the studied molecules, applying

the same theory level and basis sets used for the calculation of the

O

dx.doi.org/10.1021/om4001204 | Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

(15) Garcia, M. H.; Mendes, P. J.; Robalo, M. P.; Duarte, M. T.;

Lopes, N. J. Organomet. Chem. 2009, 694, 2888−2897.

(16) Garcia, M. H.; Mendes, P. J.; Robalo, M. P.; Dias, A. R.; Campo,

J.; Wenseleers, W.; Goovaerts, E. J. Organomet. Chem. 2007, 692,

3027−3041.

(44) Olivier, C.; Kim, B.; Touchard, D.; Rigaut, S. Organometallics

2008, 27, 509−518.

(45) Olivier, C.; Costuas, K.; Choua, S.; Maurel, V.; Turek, P.;

Saillard, J.-Y.; Touchard, D.; Rigaut, S. J. Am. Chem. Soc. 2010, 132,

5638−5651.

(46) Armitt, D. J.; Bruce, M. I.; Gaudio, M.; Zaitseva, N. N.; Skelton,

B. W.; White, A. H.; Le Guennic, B.; Halet, J.-F.; Fox, M. A.; Roberts,

R. L.; Hartl, F.; Low, P. J. Dalton Trans. 2008, 6763−6775.

(47) Powell, C. E.; Cifuentes, M. P.; Morrall, J. P.; Stranger, R.;

Humphrey, M. G.; Samoc, M.; Luther-Davies, B.; Heath, G. A. J. Am.

Chem. Soc. 2002, 125, 602−610.

(48) Wuttke, E.; Pevny, F.; Hervault, Y.-M.; Norel, L.; Drescher, M.;

Winter, R. F.; Rigaut, S. Inorg. Chem. 2012, 51, 1902−1915.

(49) Costuas, K.; Rigaut, S. Dalton Trans. 2011, 40, 5643−5658.

(50) Heck, J.; Dabek, S.; Meyer-Friedrichsen, T.; Wong, H. Coord.

Chem. Rev. 1999, 190−192, 1217−1254.

(51) Loukova, G. V. Chem. Phys. Lett. 2002, 353, 244−252.

(52) Loukova, G. V.; Strelets, V. V. J. Organomet. Chem. 2000, 606,

203−206.

(53) Campo, J.; Wenseleers, W.; Goovaerts, E.; Szablewski, M.;

Cross, G. H. J. Phys. Chem. C 2007, 112, 287−296.

(54) Campo, J.; Wenseleers, W.; Hales, J. M.; Makarov, N. S.; Perry,

J. W. J. Phys. Chem. Lett. 2012, 3, 2248−2252.

(55) Mang, C.; Wu, K.; Zhang, M.; Hong, T.; Wei, Y. J. Mol. Struct.

(THEOCHEM) 2004, 674, 77−82.

(17) Garcia, M. H.; Robalo, M. P.; Dias, A. R.; Piedade, M. F. M.;

Galvao, A.; Wenseleers, W.; Goovaerts, E. J. Organomet. Chem. 2001,

̃

619, 252−264.

(18) Robalo, M. P.; Teixeira, A. P. S.; Garcia, M. H.; Minas da

Piedade, M. F.; Duarte, M. T.; Dias, A. R.; Campo, J.; Wenseleers, W.;

Goovaerts, E. Eur. J. Inorg. Chem. 2006, 2006, 2175−2185.

(19) Wenseleers, W.; Goovaerts, E.; Hepp, P.; Garcia, M. H.; Paula

Robalo, M.; Dias, A. R.; Piedade, M. F. M.; Duarte, M. T. Chem. Phys.

Lett. 2003, 367, 390−397.

(20) Garcia, M. H.; Robalo, M. P.; Dias, A. R.; Duarte, M. T.;

Wenseleers, W.; Aerts, G.; Goovaerts, E.; Cifuentes, M. P.; Hurst, S.;

Humphrey, M. G.; Samoc, M.; Luther-Davies, B. Organometallics 2002,

21, 2107−2118.

(21) Wenseleers, W.; W. Gerbrandij, A.; Goovaerts, E.; Helena

Garcia, M.; Paula Robalo, M.; J. Mendes, P.; C. Rodrigues, J.; R. Dias,

A. J. Mater. Chem. 1998, 8, 925−930.

(22) Wu, I.-Y.; Lin, J. T.; Luo, J.; Li, C.-S.; Tsai, C.; Wen, Y. S.; Hsu,

C.-C.; Yeh, F.-F.; Liou, S. Organometallics 1998, 17, 2188−2198.

(23) Mendes, P. J.; Carvalho, A. J. P.; Ramalho, J. P. P. J. Mol. Struct.

(THEOCHEM) 2009, 900, 110−117.

(24) Whittall, I. R.; Humphrey, M. G.; Hockless, D. C. R.; Skelton, B.

W.; White, A. H. Organometallics 1995, 14, 3970−3979.

(25) Whittall, I. R.; Cifuentes, M. P.; Humphrey, M. G.; Luther-

Davies, B.; Samoc, M.; Houbrechts, S.; Persoons, A.; Heath, G. A.;

Hockless, D. C. R. J. Organomet. Chem. 1997, 549, 127−137.

(26) Branger, C.; Lequan, M.; Lequan, R. M.; Barzoukas, M.; Fort, A.

J. Mater. Chem. 1996, 6, 555−558.

(27) Chou, S.-S. P.; Sun, D.-J.; Lin, H.-C.; Yang, P.-K. J. Chem. Soc.,

Chem. Commun. 1996, 0, 1045−1046.

(28) Rao, V. P.; Jen, A. K. Y.; Wong, K. Y.; Drost, K. J. J. Chem. Soc.,

Chem. Commun. 1993, 0, 1118−1120.

(29) Oudar, J. L.; Chemla, D. S. J. Chem. Phys. 1977, 66, 2664−2668.

(30) Zhang, T.-G.; Zhao, Y.; Asselberghs, I.; Persoons, A.; Clays, K.;

Therien, M. J. J. Am. Chem. Soc. 2005, 127, 9710−9720.

(31) Whittall, I. R.; Humphrey, M. G.; Persoons, A.; Houbrechts, S.

Organometallics 1996, 15, 1935−1941.

(32) Powell, C. E.; Cifuentes, M. P.; McDonagh, A. M.; Hurst, S. K.;

Lucas, N. T.; Delfs, C. D.; Stranger, R.; Humphrey, M. G.;

Houbrechts, S.; Asselberghs, I.; Persoons, A.; Hockless, D. C. R.

Inorg. Chim. Acta 2003, 352, 9−18.

(33) Whittall, I. R.; Cifuentes, M. P.; Humphrey, M. G.; Luther-

Davies, B.; Samoc, M.; Houbrechts, S.; Persoons, A.; Heath, G. A.;

(56) Curreli, S.; Deplano, P.; Faulmann, C.; Ienco, A.; Mealli, C.;

Mercuri, M. L.; Pilia, L.; Pintus, G.; Serpe, A.; Trogu, E. F. Inorg. Chem.

2004, 43, 5069−5079.

(57) Li, Q.; Sa, R.; Liu, C.; Wu, K. J. Phys. Chem. A 2007, 111, 7925−

7932.

(58) Mang, C.; Wu, K. Int. J. Quantum Chem. 2006, 106, 2529−2535.

(59) Cardoso, C.; Abreu, P. E.; Nogueira, F. J. Chem. Theory Comput.

2009, 5, 850−858.

(60) Bruschi, M.; Fantucci, P.; Pizzotti, M. J. Phys. Chem. A 2005,

109, 9637−9645.

(61) Armarego, W. L. F.; Chai, C. L. L. Puriﬁcation of Laboratory

Chemicals, 5th ed.; Elsevier: Amsterdam, 2003.

(62) Ashby, G. S.; Bruce, M. I.; Tomkins, I. B.; Wallis, R. C. Aust. J.

Chem. 1979, 32, 1003−1016.

(63) Barnett, K. W. J. Chem. Educ. 1974, 51, 422.

(64) Bauerle, P.; Wurthner, F.; Gotz, G.; Effenberger, F. Synthesis

̈

1993, 1993 (1099), 1103.

(65) Laszlo, P.; Cornelis, A. Aldrichim. Acta 1988, 21, 6.

(66) Connelly, N. G.; Geiger, W. E. Chem. Rev. 1996, 96, 877−910.

(67) Stadler, S.; Dietrich, R.; Bourhill, G.; Brauchle, C.; Pawlik, A.;

̈

Grahn, W. Chem. Phys. Lett. 1995, 247, 271−276.

(68) Ipaktschi, J.; Mohseni-Ala, J.; Dulmer, A.; Steffens, S.;

̈

́

Bogsanyi, D. Organometallics 1997, 16, 2631−2637.

Wittenburg, C.; Heck, J. Organometallics 2004, 23, 4902−4909.

(69) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;

Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci,

B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H.

P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.;

Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima,

T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A.;

Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin,

K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.;

Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega,

N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.;

Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.;

Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.;

Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.;

Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas; Foresman, J.

B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian 09, Revision A.01;

Gaussian, Inc., Wallingford, CT, 2009.

(34) Xia, A.; Selegue, J. P. Inorg. Chim. Acta 2002, 334, 219−224.

(35) McGrady, J. E.; Lovell, T.; Stranger, R.; Humphrey, M. G.

Organometallics 1997, 16, 4004−4011.

(36) Whittall, I. R.; Humphrey, M. G.; Hockless, D. C. R. Aust. J.

Chem. 1998, 51, 219−228.

(37) Pilar Gamasa, M.; Gimeno, J.; Lastra, E.; Lanfranchi, M.;

Tiripicchio, A. J. Organomet. Chem. 1991, 405, 333−345.

(38) H. Naulty, R.; P. Cifuentes, M.; G. Humphrey, M.; Houbrechts,

S.; Boutton, C.; Persoons, A.; A. Heath, G.; C. R. Hockless, D.; Luther-

Davies, B.; Samoc, M. J. Chem. Soc., Dalton Trans. 1997, 4167−4174.

(39) Whittall, I. R.; Humphrey, M. G.; Houbrechts, S.; Maes, J.;

Persoons, A.; Schmid, S.; Hockless, D. C. R. J. Organomet. Chem. 1997,

544, 277−283.

(40) Mendes, P. J.; Silva, T. J. L.; Carvalho, A. J. P.; Ramalho, J. P. P.

J. Mol. Struct. (THEOCHEM) 2010, 946, 33−42.

(41) Marder, S. R.; Gorman, C. B.; Tiemann, B. G.; Perry, J. W.;

Bourhill, G.; Mansour, K. Science 1993, 261, 186−189.

(42) Janowska, I.; Zakrzewski, J.; Nakatani, K.; Palusiak, M.; Walak,

M.; Scholl, H. J. Organomet. Chem. 2006, 691, 323−330.

(43) Liu, C.-G.; Qiu, Y.-Q.; Sun, S.-L.; Li, N.; Yang, G.-C.; Su, Z.-M.

Chem. Phys. Lett. 2007, 443, 163−168.

(70) Becke, A. D. J. Chem. Phys. 1993, 98, 5648−5652.

(71) Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. B 1988, 37, 785−789.

(72) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 270−283.

(73) Hay, P. J.; Wadt, W. R. J. Chem. Phys. 1985, 82, 299−310.

P

dx.doi.org/10.1021/om4001204 | Organometallics XXXX, XXX, XXX−XXX

Organometallics

Article

(74) Check, C. E.; Faust, T. O.; Bailey, J. M.; Wright, B. J.; Gilbert, T.

M.; Sunderlin, L. S. J. Phys. Chem. A 2001, 105, 8111−8116.

(75) Feller, D. J. Comput. Chem. 1996, 17, 1571−1586.

(76) Schuchardt, K. L.; Didier, B. T.; Elsethagen, T.; Sun, L.;

Gurumoorthi, V.; Chase, J.; Li, J.; Windus, T. L. J. Chem. Inf. Model.

2007, 47, 1045−1052.

(77) Nalewajski, R. Density Functional Theory. In Perspectives in

Electronic Structure Theory; Springer: Berlin and Heidelberg, Germany,

2011; pp 255−368.

(78) Casida, M. E.; Jamorski, C.; Casida, K. C.; Salahub, D. R. J.

Chem. Phys. 1998, 108, 4439−4449.

(79) O’Boyle, N. M.; Tenderholt, A. L.; Langner, K. M. J. Comput.

Chem. 2008, 29, 839−845.

(80) Zhurko, G. A. Z. D. A. ChemCraft: Tool for treatment of chemical

data, build 342; http://www.chemcraftprog.com.

Q

dx.doi.org/10.1021/om4001204 | Organometallics XXXX, XXX, XXX−XXX

Article Doi

Mono(η5-cyclopentadienyl)metal(II) complexes with thienyl acetylide chromophores: Synthesis, electrochemical studies, and first hyperpolarizabilities

DOI: 10.1021/om4001204

Source and publish data:

Authors:

Article abstract of DOI:10.1021/om4001204

Full text of DOI:10.1021/om4001204

Products guided by the article

R&D Labs maybe for 189190-61-2

Relevant to this article

Hot Product