Electron-Deficient Imidazolium Substituted Cp Ligands and their Ru Complexes

Full Paper

doi.org/10.1002/chem.202002801

Chemistry—A European Journal

&

Organometallic Chemistry |Hot Paper|

Electron-Deficient Imidazolium Substituted Cp Ligands and their

Ru Complexes

Fabio Mazzotta,^[a]Georg Zitzer,^[b]Bernd Speiser,^[b]and Doris Kunz*^[a]

Abstract: The synthesis of electron-poor mono-, di- and tri-

(imidazolium)-substituted Cp-ylides is presented and their

electronic properties are discussed based on NMR spectros-

copy, X-ray structure analyses, electrochemical investigations

and DFT calculations as well as by their reactivity toward

[Ru(CH₃CN)₃Cp*](PF₆). With mono- and di(imidazolium)-sub-

stituted cyclopentadienides the respective monocationic and

dicationic ruthenocences are formed (X-ray), whereas tri(imi-

dazolium) cyclopentadienides are too electron-poor to form

the ruthenocenes. Cyclic voltammetric analysis of the ruthe-

nocenes shows reversible oxidation at a potential that in-

creases with every additional electron-withdrawing imidazo-

lium substituent at the Cp ligand by 0.53–0.55 V in an elec-

trolyte based on a weakly coordinating anion. A reversible

oxidation can be observed for the free 1,3-disubstituted

ligand as well.

Introduction

Cp ligands are among the most common ligands in organome-

tallic chemistry and tuning of the steric and electronic proper-

ties by introducing different substituents is well known.^[1–3]

Each substituent at the Cp ring will increase its steric demand

and in most cases also the electron density in the Cp ring so

that the MꢀCp bond strength increases. Substituents that

weaken the MꢀCp bond (e.g. Ar,^[4]F,^[5]Cl,^[5]CN, COOR, NO₂,

CF₃,^[6]phosphonium,^[7]pyridinium or ammonium groups) are

much less studied.^[8,9]

One interesting ligand class in this regard is that of cyclo-

pentadienylides. These zwitterionic compounds are thermody-

namically stable Cp compounds containing at least one a-cat-

+

ionic substituent at the Cp ring, e.g. -SR₂⁺, -NR₃⁺, -PR₃or car-

Figure 1. Selection of literature known cyclopentadienylides A-E and the

new di- and tri(imidazolium) Cps of this work.

benium.^[7,9]The latter can also be described with the resonance

structure of ylenes, for example, as fulvenes or fulvalenes

(Figure 1). In case of an enhanced stabilisation of the separated

charges by mesomeric effects or aromaticity, the ylidic charac-

ter can become dominant. In early works, Mꢀller-Westerhoff in-

troduced the diamino fulvene A as a cationic Cp ligand in fer-

rocene by simple addition to iron(II) chloride at 608C.^[10]This

indicates a sufficient polarity of the exocyclic double bond and

a Cp like reactivity of these compounds. With the idea of fur-

ther enhancing the ylidic character by aromatic stabilization,

we reported in 2008 on the first imidazolium cyclopentadieny-

lide B, which can also be regarded as a diazafulvalene, and the

formation of its ferrocene.^[11]This compound exhibits a highly

ylidic character and can be best described as an imidazolium

substituted Cp without significant ylene character. Six years

later, we introduced the dipyrido-anellated ligand C, which

provides an even stronger charge separation.^[12]An analogous

class of compounds are phosphonium cyclopentadienylides,^[7]

with the Ramirez ylide D being its most prominent representa-

tive.^[13]The first examples of di(phosphonium) substituted Cp

rings E have been reported recently.^[14,15]While the coordina-

tion chemistry of the monosubstituted Cp ligands A–D has

[a] F. Mazzotta, Prof. Dr. D. Kunz

Institut fꢀr Anorganische Chemie

Eberhard Karls Universitꢁt Tꢀbingen

Auf der Morgenstelle 18, 72076 Tꢀbingen (Germany)

E-mail: Doris.Kunz@uni-tuebingen.de

[b] G. Zitzer, Prof. Dr. B. Speiser

Institut fꢀr Organische Chemie

Eberhard Karls Universitꢁt Tꢀbingen

Auf der Morgenstelle 18, 72076 Tꢀbingen (Germany)

Supporting information and the ORCID identification number(s) for the au-

thor(s) of this article can be found under:

https://doi.org/10.1002/chem.202002801.

ꢂ 2020 The Authors. Published by Wiley-VCH GmbH. This is an open access

article under the terms of the Creative Commons Attribution License, which

permits use, distribution and reproduction in any medium, provided the

original work is properly cited.

Chem. Eur. J. 2020, 26, 1 – 16

1

ꢁ 2020 The Authors. Published by Wiley-VCH GmbH

&

These are not the final page numbers! ÞÞ

Full Paper

doi.org/10.1002/chem.202002801

Chemistry—A European Journal

been already investigated, that of the cationic disubstituted Cp

ligand E remains unexplored so far.

tivity and decomposition of 1 in polar solvents,^[21]which we

found not to be the case in acetonitrile.

The imidazolium cyclopentadienylides B and C were used as

ligands for half-sandwich and sandwich complexes^[11,16,17]by

simple addition to a free coordination site of a suitable metal

precursor or by ligand substitution. As the cyclopentadieny-

lides contain a positive charge at the a-position of the sub-

stituents, their electron density is reduced compared to regular

Cps. At this point, no detailed investigation of the donor prop-

erties of cyclopentadienylide B had been made. Therefore, we

will present this together with an improved synthesis of B as

well as the synthesis and properties of the first di(imidazolium)

cyclopentadienylides and tri(imidazolium) cyclopentadieny-

lides. To elucidate the electronic properties of these Cp ligands,

we carried out DFT calculations and, if possible, prepared the

respective ruthenocenes by reaction with [Ru(CH₃CN)₃Cp*](PF₆)

and analyzed the oxidation potential of the ligands as well as

of the ruthenocenes.

With the chloride 1 in hand, we synthesized cyclopentadie-

nylide 2 achieving an improved yield and purity. As our previ-

ous work showed a fast H/D exchange of the Cp protons of 2

with D₂O but no decomposition,^[11]we could remove residual

1

lithium salts with degassed water. The H NMR chemical shifts

of the Cp signals of 2 are strongly depending on the polarity

of the solvent (Figure 2). Except for tetrahydrofuran, the signals

are shifted downfield with lower polarity of the solvent. Inter-

estingly, the b-signals (3/4-H) are more strongly affected than

the a-signals (2/5-H). The signal of LiCp shows qualitatively the

same trend.

In a previous work, we reported on the facile coordination

of cyclopentadienylide C to LiNTf₂under formation of Li(h⁵-Cp)

complexes that are in a fast equilibrium between a half-sand-

wich and a lithiocene complex in toluene already at ꢀ808C.^[17]

Therefore, we tested ylide 2 in an analogous NMR experiment,

and used Li[B(C₆F₅)₄]·2.5Et₂O as well as LiNTf₂as lithium salt

and deuterated dichloromethane as a polar, weakly coordinat-

ing solvent (Scheme 2). In the case of LiNTf₂we observe one

Results and Discussion

7

broad signal in the Li NMR at ꢀ7.0 ppm (LiNTf₂) that is diag-

Improved synthesis for imidazolium cyclopentadienylide 2

nostic for a half-sandwich lithium complex. In the case of

Li[B(C₆F₅)₄]·2.5Et₂O we observe at room temperature two broad

signals. The first at 0.8 ppm can be assigned to a lithium

cation and the second at ꢀ7.3 ppm (Li[B(C₆F₅)₄]) again to the

half-sandwich complex. At ꢀ808C (Li[B(C₆F₅)₄]) we observe

three lithium signals that can be assigned to a Li⁺cation at

ꢀ0.80 ppm, and in a 1:5 ratio to the half-sandwich lithium

complex 3 at ꢀ6.0 ppm and to the lithiocene 4 at ꢀ11.7 ppm.

The broad signals indicate a fast equilibrium between these

Until now, the synthesis of the monosubstituted compounds B

or C required the synthesis of the corresponding uronium or

guanidinium salts^[11,12]that were reacted with two equivalents

of either LiCp or NaCp (Scheme 1). One equivalent was needed

7

species that is confirmed by a Li EXSY NMR experiment. To

compare it directly with C, we performed the experiment with

Scheme 1. Previous (left) and improved (right) synthesis of the imidazolium

cyclopentadienylide 2.

as nucleophile to obtain the protonated Cp intermediates and

the second to deprotonate them under generation of the de-

sired imidazolium cyclopentadienylides. However, in the case

of the uronium salts side reactions at the alkoxy group can

lead to the corresponding imidazolone.^[18]To avoid also the for-

mation of LiBF₄as a tediously to remove byproduct, we

changed the leaving group from ethoxide to chloride. Chloro-

imidazolium salts are in general obtained from chlorination of

the free carbene if the backbone is alkylated.^[19]In our case

Kuhn’s 2-chloro-tetramethylimidazolium chloride (1)^[20–22]

proved to be a suitable precursor for introducing the imidazoli-

um substituent to Cps.

Figure 2. Solvent-dependent chemical shift of the CpꢀH signals (black and

red) of 2 in comparison with those of LiCp (blue).

However, we found it advantageous to change the reaction

conditions of the synthesis of 1 to firstly cooling the solution

of a slight excess of hexachloroethane in diethyl ether to

ꢀ308C and then adding dropwise a solution of the free car-

bene under vigorous stirring. We were also able to elucidate

the molecular structure of 1 by X-ray crystal structure analysis,

which exhibits intermolecular halogen-halogen interactions

(see Supporting Information). Kuhn mentioned the high reac-

Scheme 2. Coordination of ylide 2 to Li[B(C₆F₅)₄]·2.5Et₂O.

&

Chem. Eur. J. 2020, 26, 1 – 16

www.chemeurj.org

2

ꢁ 2020 The Authors. Published by Wiley-VCH GmbH

ÝÝ These are not the final page numbers!

Full Paper

doi.org/10.1002/chem.202002801

Chemistry—A European Journal

LiNTf₂in toluene, but even at ꢀ808C only the half-sandwich

complex can be observed.

Synthesis of 1,3-di(imidazolium)cyclopentadienylide (5)

To synthesize the imidazolium analog of the di(phosphonium)

cyclopentadienylide E, we used the monosubstituted ylide 2 as

a nucleophile to react with the chloroimidazolium salt 1. The

reaction is analogous to the synthesis of the ylide 2, but with

the difference that LiHMDS is used as a bulky and weakly nu-

cleophilic auxiliary base (Scheme 3). The reaction is much

Figure 3. ORTEP plot of the molecular structure of 5 (anisotropic atomic dis-

placement parameters at 50% probability). Hydrogen atoms and the PF₆

ꢀ

counterion are omitted for clarity. Selected bond lengths [ꢂ] and angles [8]:

C1ꢀC6 1.440(3), C3ꢀC13 1.439(2), C1ꢀC2 1.406(2), C2ꢀC3 1.411(2), C3ꢀC4

1.431(2), C4ꢀC5 1.378(3), C5ꢀC1 1.427(2); N1-C6-N2 106.2(2), N3-C13-N4

106.1(1), C5-C1-C2 108.1(2), C1-C2-C3 107.7(2), C2-C3-C4 107.6(1), C3-C4-C5

108.4(2), C4-C5-C1 108.2(2), N1-C6-C1-C2 43.4, N3-C13-C3-C2 35.6.

Characterization of side product 6

During the synthesis of 5, we could observe from the crude

product a peak in the ESI–MS⁺at m/z=286.1, which fits to a

combination of two imidazolium and one (deprotonated) ace-

tonitrile moiety. Although, we were not able to optimize the

synthesis and isolate this interesting air-stable compound with

the desired purity, we obtained single crystals of 6 suitable for

X-ray structure analysis, which confirms the anticipated struc-

ture (see Figure 4). The bond lengths N5ꢀC16 1.153(2) ꢂ and

C15ꢀC16 1.410(2) ꢂ are comparable with the lengths of aceto-

nitrile (CꢀN 1.150(4) ꢂ and CꢀC 1.442(4) ꢂ)^[23]and the calculat-

ed structure (CꢀN 1.176 ꢂ and CꢀC 1.405 ꢂ). The most plausi-

ble resonance structure is that of a CꢁN triple bond and the

negative charge mainly localized on the adjacent carbon atom.

Analogous di(phosphonium)ylides are known from Erker and

co-workers.^[24]

Scheme 3. Synthesis of the di(imidazolium) cyclopentadienylide 5 and H/D

exchange of the Cp protons.

slower, due to the weaker nucleophilicity of 2 compared to

LiCp. The reduced nucleophilicity caused by each additional

imidazolium substituent is the reason why a second substitu-

tion during the synthesis of 2 or a third substitution during

the synthesis of 5 are not observed. For the same reason, the

work up of 5 can be performed under air. Notably, only the

1,3-substituted product is formed. A 1,2-substitution pattern

might be impeded by steric effects and Coulomb repulsion.

Changing the solvent from acetonitrile to tetrahydrofuran re-

sulted in longer reaction times, possibly due to the poor solu-

bility of both starting materials, but in the same purity of 5

(NMR).

We could reproduce this peak and therefore the formation

of 6 by mixing the free carbene with the chloroimidazolium

salt 1 in acetonitrile (Scheme 4). After short workup, we ob-

tained a mixture of two species in acetonitrile. The ESI–MS⁺re-

vealed again the peak at m/z=286.1 for the product 6 and

two peaks at m/z=124.1 for a dicationic species and m/z=

393.1 for a mono cationic species. Furthermore, we could

remove the second so far unknown species 7 by extraction of

the raw product 6 with dichloromethane. As it is known, di-

(imidazolium) salts can be formed under these conditions.^[25,26]

Attempts to deprotonate compound 5 to form an NHC ana-

logue failed so far due to the very low acidity of the CꢀH pro-

tons. Moreover, the coordination to lithium salts under the

same conditions as with 2 did not take place. However, we ob-

served a slow H/D exchange of the Cp protons in D₂O already

at room temperature. To accelerate the exchange, we heated

the solution to 608C. Most probably the exchange proceeds

by a protonation–deprotonation mechanism. We also tried to

freeze the rotation of the imidazolium substituents at ꢀ808C

and quantify the rotation barrier, but even at ꢀ808C no

second signal set is observed. This observation matches with a

high single bond character in the exocyclic CꢀC bonds.

The molecular structure could be obtained from single crys-

tals by X-ray structure analysis (Figure 3). The bond lengths of

the ImiꢀCp bonds (C1ꢀC6 and C3ꢀC13, 1.440(3) and

1.439(3) ꢂ) are longer than for 2 (C1ꢀC3 1.430(3) ꢂ).^[11]This fact

can either be attributed to a higher single bond character

compared to 2 per se and/or to a lower Coulomb attraction

due to the additional cationic moiety. The bond lengths of the

Cp ring lie between 1.431(2) and 1.378(3) ꢂ, and are compara-

ble with those of 1. The angles of the Cp ring are very close to

1088, the value of a regular pentagon.

Figure 4. ORTEP plot of the molecular structure of 6 (anisotropic atomic dis-

ꢀ

placement parameters at 50% probability). Hydrogen atoms and the PF₆

counterion are omitted for clarity. Selected bond lengths [ꢂ] and angles [8]:

N5ꢀC16 1.153(2), C15ꢀC16 1.410(2), C1ꢀC15 1.432(2), C8ꢀC15 1.432(2); N1-

C1-N3 106.9(1), N3-C8-N4 107.0(1), C1-C15-C8 123.0(1).

Chem. Eur. J. 2020, 26, 1 – 16

www.chemeurj.org

3

ꢁ 2020 The Authors. Published by Wiley-VCH GmbH

&

These are not the final page numbers! ÞÞ

Full Paper

doi.org/10.1002/chem.202002801

Chemistry—A European Journal

Scheme 4. Independent synthesis of 6, which also leads to formation of 7.

We confirmed formation of 7 by independent synthesis and

characterization (Supporting Information).

Figure 5. ORTEP plot of the molecular structure of 9 (anisotropic atomic dis-

placement parameters at 50% probability). Hydrogen atoms, one molecule

of dichloromethane, and the PF₆^ꢀcounterion are omitted for clarity. Selected

bond lengths [ꢂ] and angles [8]: C13ꢀC14 1.414(3), C18ꢀC19 1.418(3), C14ꢀ

C15 1.402(3), C15ꢀC16 1.399(3), C16ꢀC17 1.403(3), C17ꢀC18 1.400(3), C14ꢀ

C18 1.457(3); N1-C19-N2 103.5(2), N3-C13-N4 103.7(2), C18-C14-C15 107.6(2),

C14-C15-C16 107.5(2), C15-C16-C17 110.1(2), C16-C17-C18 107.7(2), C17-C18-

C14 107.3(2), N1-C19-C18-C17 3.0, N4-C13-C14-C15 0.8.

Synthesis of 1,2-di(imidazolium)cyclopentadienylide 9

To obtain also a 1,2-substitution pattern of a di(imidazolium)

Cp, we reacted the di(chloroimidazolium) salt 8^[12,27]with two

equivalents of LiCp or NaCp (see Scheme 5). In both cases, full

conversion was obtained within one hour and the desired

product was detected, but some brown impurities impeded its

isolation. Therefore, we used the less nucleophilic thallium(I)

cyclopentadienide and after two days full conversion at no ap-

parent side reactions was achieved (NMR). Just like with com-

pound 5, the coordination of the cationic ylide 9 to Li⁺is not

observed under the same conditions as applied for 2. We ob-

served a H/D exchange with D₂O however, not of the Cp but

of the imidazo hydrogen atoms (see Scheme 5). As these posi-

tions are known to be acidic from the synthesis of abnormal

NHCs, they seem to be the most acidic hydrogen atoms of

9.^[28]

Interestingly, no monoylide 10 was observed, even when

only one equivalent of LiCp was used (see Scheme 5). This

shows that the intramolecular S_Nreaction with the second

chloroimidazolium moiety proceeds much faster than the inter-

molecular S_Nreaction to form 10, or an analogue of the 1,3-

disubstituted Cp compound 5. Moreover, no diylide 11 can be

observed with a 40-fold excess of LiCp, even if the reaction is

performed in tetrahydrofuran, in which 8 is poorly soluble

compared the LiCp. The reaction of dilithioferrocene with the

di(chloroimidazolium) salt 8 to form directly the corresponding

dicationic ansa-ferrocene led to no defined product.

Figure 5). The bond lengths of the ImiꢀCp bonds (C13ꢀC14

1.414(3) ꢂ and C18ꢀC19 1.418(3) ꢂ) are shorter compared to

compound 5. Tentatively, this is caused by electronic effects of

the conjugated, fixed planar p-system formed. In contrast, the

annellated Cp bond C14ꢀC18 1.457(3) ꢂ is much longer than

any Cp bond in compound 5. All other four Cp bonds have

equal bond lengths of 1.40 ꢂ within the standard deviation.

The angles of the Cp ring are close to 1088 as in 5.

Synthesis of tri(imidazolium) cyclopentadienylides 12 and

13

As the nucleophilicity of the disubstituted compounds is re-

duced, an additional nucleophilic substitution by the chloro-

imidazolium salt 1 was not observed during the synthesis of 5.

However, as the intramolecular reaction is much faster, we

took advantage of it and reacted the di(chloroimidazolium)

compound 8 with 2 which is a weaker nucleophile than LiCp.

In addition, we used LiHMDS as a sterically hindered base to

form the tri-substituted Cp-ylides 12 and 13 (see Scheme 6).

The molecular structure of 9 could be elucidated with single

crystal X-ray structure analysis. As expected, 9 is planar (except

for the n-propyl groups) due to the conjugated p-system (see

Scheme 5. Synthesis of the 1,2-substituted Cp-ylide 9 and H/D exchange of

the Cp protons. The potential di(Cp-ylides) 11 and 11-Fe were not obtained.

Scheme 6. Synthesis of the tri(imidazolium) Cp-ylides 12 and 13 and H/D ex-

change at the imidazo moieties.

&

Chem. Eur. J. 2020, 26, 1 – 16

www.chemeurj.org

4

ꢁ 2020 The Authors. Published by Wiley-VCH GmbH

ÝÝ These are not the final page numbers!

Full Paper

doi.org/10.1002/chem.202002801

Chemistry—A European Journal

Monitoring the reaction by NMR spectroscopy showed full

conversion of the starting material along with some side prod-

uct. The ratio of 4.5:1 for 12:13 is equal for the reaction at

room temperature and at ꢀ308C. Separation of these two salts

of similar polarity failed until now.

from the additional substituent is longer compared to the re-

spective bonds of 2 or 5. In compound 13 the mean value of

the ImiꢀCp bonds C13ꢀC14 and C18ꢀC19 is comparable to

that of 9 as well. The bond C15ꢀC20 to the third imidazolium

substituent is longer compared to 2 or 5, which can be ex-

plained with the reduced Coulomb attraction of the Cp-ring

due to the additional imidazolium substituent. In both mole-

cules 12 and 13 the angles in the Cp ring are still close to

1088.

The trisubstituted compounds were also tested in a H/D-ex-

change experiment. As it was already revealed for 9, the hy-

drogen atoms at the imidazo moieties of 12 and 13 were ex-

changed.

1

We were able to obtain the molecular structures of 12 and

13 by X-ray structure analysis (see Figure 6 and 7). In both

cases a planar anellated ring system with a third, almost per-

pendicular imidazolium substituent is revealed. The ImiꢀCp

bond lengths C13ꢀC14 and C18ꢀC19 of 12 are almost identical

to those of compound 9. Only the ImiꢀCp bond C16ꢀC20

When comparing the H NMR chemical shifts of the Cp sig-

nals of LiCp and the Cpꢀylides 2, 5, 9, 12 and 13 as well as the

signal of benzene in deuterated acetonitrile, we noticed a

downfield shift with every additional imidazolium substituent

but also a downfield shift from the 1,3-disubstituted system 5

to the planar 1,2-disubstituted system 9 (see Figure 8).

Figure 8. ¹H NMR chemical shifts of CpꢀH signals of LiCp, the ylides 2, 5, 9,

12, 13 and benzene in deuterated acetonitrile.

Figure 6. ORTEP plot of the molecular structure of 12 (anisotropic atomic

displacement parameters at 50% probability). Hydrogen atoms, one acetoni-

trile molecule and two PF₆^ꢀcounterions are omitted for clarity. Selected

bond lengths [ꢂ] and angles [8]: C13ꢀC14 1.413(5), C18ꢀC19 1.422(5), C16ꢀ

C20 1.452(5), C14ꢀC15 1.390(5), C15ꢀC16 1.403(5), C16ꢀC17 1.414(5), C17ꢀ

C18 1.385(5), C14ꢀC18 1.456(4); N1-C19-N2 104.0(3), N3-C13-N4 103.6(3), N5-

C20-N6 106.7(3), C18-C14-C15 107.5(3), C14-C15-C16 107.6(3), C15-C16-C17

109.7(3), C16-C17-C18 107.1(3), C17-C18-C14 108.1(3), N1-C19-C18-C17 2.0,

N4-C13-C14-C15 2.5, N5-C20-C16-C17 52.9.

DFT calculations

The Cp-like reactivity of these compounds should be highly

depending on the electron density in the Cp ring. Therefore,

we carried out an NBO population analysis of the geometry

optimized ligands (DFT, BP86/def2-TZVP, COSMO: e=37.5),

which describes very well delocalized systems. For a reasona-

ble comparison of the charges at the Cp ring, we summed up

the natural charges of the atoms from the Cp fragment (q_Cp),

which can be correlated with the electron density. Compared

to the non-substituted Cp, the negative charge at the carbon

atoms is reduced with every additional imidazolium substitu-

ent, except for the third substituent in 12 compared to 9 (see

Figure 9). Besides the electron density, it should be considered

that the Coulomb interactions also contribute significantly to

the stability of the metal-Cp bond.

Figure 7. ORTEP plot of the molecular structure of 13 (anisotropic atomic

In addition, we carried out the NBO population analysis of

the hypothetical 1,2-disubstituted non-annellated com-

pound 14 to compare the influence of the substitution pattern.

A 1,2-substitution leads to a different dipole moment of the

molecule, but interestingly to the same charge of the Cp frag-

ment. Therefore, we can assign the observed lower electron

density of the Cp ring in the annellated system 9 to the fixed

displacement parameters at 50% probability). Hydrogen atoms and two

PF₆counterions are omitted for clarity. Selected bond lengths [ꢂ] and

ꢀ

angles [8]: C13ꢀC14 1.424(2), C18ꢀC19 1.414(2), C15ꢀC20 1.457(2), C14ꢀC15

1.414(2), C15ꢀC16 1.411(2), C16ꢀC17 1.390(2), C17ꢀC18 1.400(2), C14ꢀC18

1.453(2); N1-C19-N2 104.2(1), N3-C13-N4 103.1(1), N5-C20-N6 107.1(1), C18-

C14-C15 106.0(1), C14-C15-C16 107.8(1), C15-C16-C17 110.1(1), C16-C17-C18

107.2(1), C17-C18-C14 108.9(1), N1-C19-C18-C17 9.3, N4-C13-C14-C15 7.8,

N5-C20-C16-C17 76.6.

Chem. Eur. J. 2020, 26, 1 – 16

www.chemeurj.org

5

ꢁ 2020 The Authors. Published by Wiley-VCH GmbH

&

These are not the final page numbers! ÞÞ

Full Paper

doi.org/10.1002/chem.202002801

Chemistry—A European Journal

Synthesis of the ruthenocenes

To quantify the influence of our new ligands on the metal

center, we decided to synthesize their respective mixed Cp*Ru-

complexes and compare their redox potential with the mixed

ruthenocene [RuCp*Cp] (15). The synthesis of the literature

known complex starts from [Ru(CH₃CN)₃Cp](PF₆) using LiCp*^[29]

and involves a low reaction temperature (ꢀ788C) and 14 h re-

action time in total. However, we obtained a higher yield and

comparable purity, when we started from [Ru(CH₃CN)₃Cp*](PF₆)

and LiCp (see Scheme 7) in acetonitrile as solvent and ran the

reaction at room temperature for 10 min. The work up was

kept identical to the literature procedure.

Figure 9. Selected natural charges and the sum of the natural charges of the

Cp-moiety q_Cpfrom the natural bond orbital (NBO) population analysis of

the Cp anion, the ylides 2, 5, 14, 9, 12, and 13 as well as of the benzene

ring. Only the s bonds between the C_sp₂atoms are shown for clarity.

planar geometry of the imidazolium substituents and not to

the substitution pattern.

Scheme 7. Synthesis of the ruthenocenes starting from

[Ru(CH₃CN)₃Cp*](PF₆).

The NBO population analysis of 2 reveals a slightly more

negatively charged 3-position of the Cp ring which explains

the preferred 1,3-substitution pattern in the nucleophilic sub-

stitution reaction.

For the synthesis of the corresponding imidazolium substi-

tuted ruthenocenes, we also used [RuCp*(CH₃CN)₃](PF₆) (a salt

metathesis route is not possible) and reacted it with the li-

gands 2 and 5 in acetonitrile at room temperature. Both com-

plexes 16 and 17 are air-stable already during the workup of

the reaction mixture.

With every additional imidazolium substituent, the energy of

the highest-occupied molecular orbital (HOMO) is decreasing

(see Figure 10) as it is to be expected with electron withdraw-

The h⁵-coordination of both ligands 2 and 5 was confirmed

by their molecular structure obtained from X-ray structure

analysis of suitable single crystals (see Figure 11 and 12) and

the unhindered rotation of the imidazolium substituents in

complex 17 is observed even down at ꢀ808C, as it is also ob-

served for the free ligand 5.

Compared to the free ligands the exocyclic ImiꢀCp bonds

are longer in the complexes. In complex 16 it measures

1.454(2) ꢂ compared to the exocyclic bond of the free ligand 2

of 1.430(3) ꢂ.^[11]In complex 17 the exocyclic bonds C1ꢀC6 and

C3ꢀC13 of 1.453(2) and 1.457(3) ꢂ are also longer than in the

free ligand 5 of 1.440(3) and 1.439(3) ꢂ. This might be caused

by a higher single bond character but, what is more plausible,

by the weaker Coulomb attraction of the charges due to coor-

dination of the Cp-ring to the metal.

Figure 10. Calculated HOMO energies of the Cp anion, the ylides 2, 5, 14, 9,

12, 13 and benzene.

ing substituents. For all Cp ylides the HOMO is located at the

Cp ring and has p-character, but also with some contribution

of the imidazolium substituents (Supporting Information). In

comparison, the HOMO of benzene lies at considerably lower

energies. The 1,2-substitution of the virtual compound 14,

leads to a HOMO with pronouncedly lower energy compared

to 5. The LUMOs of the compounds are in some cases not Cp-

like and are therefore not discussed.

We tried to synthesize complex 18, bearing the planar 1,2-

disubstituted ligand 9, in the same way as complexes 16 and

17. However, in NMR experiments we recognized only partial

conversion to the metal complex 18, so that an equilibrium

mixture containing also the free ligand 9, acetonitrile and

[Ru(CD₃CN)₃Cp*](PF₆) is formed. To shift this equilibrium to the

product side, we changed the solvent to dichloromethane so

that complex 18 precipitated as a yellow crystalline solid

during the reaction. To confirm the existence of such an equi-

&

Chem. Eur. J. 2020, 26, 1 – 16

www.chemeurj.org

6

ꢁ 2020 The Authors. Published by Wiley-VCH GmbH

ÝÝ These are not the final page numbers!

Full Paper

doi.org/10.1002/chem.202002801

Chemistry—A European Journal

Figure 13. ORTEP plot of the molecular structure of 18 (anisotropic atomic

displacement parameters at 50% probability). Hydrogen atoms, one acetoni-

trile molecule and two PF₆^ꢀanions are omitted for clarity. Selected bond

lengths [ꢂ] and angles [8]: C13ꢀC14 1.432(3), C18ꢀC19 1.436(3), C14ꢀC15

1.427(3), C15ꢀC16 1.415(3), C16ꢀC17 1.424(3), C17ꢀC18 1.423(3), C14ꢀC18

1.472(3), Ru1ꢀC14 2.165(2), Ru1ꢀC15 2.189(2), Ru1ꢀC16 2.200(2), Ru1ꢀC17

2.216(2), Ru1ꢀC18 2.198(2); N1-C19-N2 104.2(2), N3-C13-N4 104.5(2), N1-C19-

C18-C17 5.1, N4-C13-C14-C15 1.9.

Figure 11. ORTEP plot of the molecular structure of 16 (anisotropic atomic

displacement parameters at 50% probability). Hydrogen atoms and the PF₆

anion are omitted for clarity. Selected bond lengths [ꢂ] and angles [8]: C1ꢀ

C6 1.454(2), C1ꢀC2 1.443(2), C2ꢀC3 1.426(2), C3ꢀC4 1.424(2), C4ꢀC5

1.423(2), C5ꢀC1 1.443(2), Ru1ꢀC1 2.195(1), Ru1ꢀC2 2.181(1), Ru1ꢀC3

ꢀ

2.193(1), Ru1ꢀC4 2.212(1), Ru1ꢀC5 2.207(1); N1-C6-N2 106.9(1), N1-C6-C1-C2

36.5.

The molecular structure of complex 18, obtained from X-ray

structure analysis, confirms the h⁵-coordination mode of 9 (see

Figure 13). The ImiꢀCp bonds of the metal complex 18

(1.432(3) and 1.436(3) ꢂ) are longer compared to those of the

free ligand 9 (1.414(3) and 1.418(3) ꢂ). This can be explained

again with the reduced Coulomb attraction.

In all solid state structures, the carbenium ion is not bent to

the metal center, as it would be the case if further stabilization

was required,^[30,31]but rather points away from it for steric rea-

sons. This behavior was already observed in the ferrocene

complex bearing a derivative of 2.^[11]The effect is stronger for

16 and 17, in which the imidazolium substituents are freely ro-

tating than for complex 18 bearing the rigid ligand 9. But in all

cases, it confirms again the good stabilization of the carbeni-

um ions.

Figure 12. ORTEP plot of the molecular structure of 17 (anisotropic atomic

displacement parameters at 50% probability). Hydrogen atoms, one acetoni-

trile molecule and two PF₆^ꢀanions are omitted for clarity. Selected bond

lengths [ꢂ] and angles [8]: C1ꢀC6 1.457(3), C3ꢀC13 1.453(2), C1ꢀC2 1.429(2),

C2ꢀC3 1.437(2), C3ꢀC4 1.442(2), C4ꢀC5 1.415(3), C5ꢀC1 1.437(2), Ru1ꢀC1

2.214(2), Ru1ꢀC2 2.219(2), Ru1ꢀC3 2.203(2), Ru1ꢀC4 2.173(2), Ru1ꢀC5

2.177(2); N1-C6-N2 107.1(2), N3-C13-N4 106.9(2), N1-C6-C1-C2 49.1, N3-C13-

C3-C2 36.6.

Attempts to form the corresponding ruthenocene with the

tri(imidazolium) substituted Cp ligands 12 and 13 in dichloro-

methane failed. Due to the double positive charge of these li-

gands as well as the weak electron density in the Cp ring, the

acetonitrile ligands cannot be replaced by these Cp com-

pounds.

librium, we isolated the crystals and redissolved them in

[D₃]acetonitrile. After 30 min the formation of 4% the free

metal precursor [Ru(CD₃CN)₃Cp*](PF₆) was observed in the

¹H NMR spectrum. After 25 h the spectrum shows the final

equilibrium ratio with 22% of [Ru(CD₃CN)₃Cp*](PF₆), which was

confirmed after 9 days. Obviously, the monocationic ligand 9

binds much weaker to the metal fragment than the monocat-

ionic ligand 5. In contrast to complex 17, complex 18 is air sen-

sitive and decomposes over time, most likely due to hydrolysis.

All ruthenocenes 16–18 were observed in the ESI–MS spec-

trum using a solution of complexes 16 and 17 in acetonitrile

and a solution of 18 in dichloromethane. Only in the case of

complex 18 the free ligand 9 is observed in high intensity. This

also confirms the weak ligand-metal bond and the poor donor

property of ligand 9.

Recycling of the air stable ligand 9

As 18 is not air-stable, it is possible to recover the free ligand

by dissolving complex 18 in acetone under air and stir the mix-

ture for several days. Once the complex is decomposed, the

formed precipitate can be filtered off and the ligand can be

isolated from the filtrate in crystalline form using the same

workup procedure as described in the synthesis of ligand 9.

Primarily this is important to reduce the use of thallium salts

significantly.

Side product 20 from the synthesis of [(C₆H₆)RuCp*](PF₆) (19)

For the following electrochemical investigations we synthe-

sized the literature known [(C₆H₆)RuCp*](PF₆) (19) starting from

[(C₆H₆)RuCl₂]₂. Theoretically, this synthesis should require only a

Chem. Eur. J. 2020, 26, 1 – 16

www.chemeurj.org

7

ꢁ 2020 The Authors. Published by Wiley-VCH GmbH

&

These are not the final page numbers! ÞÞ

Full Paper

doi.org/10.1002/chem.202002801

Chemistry—A European Journal

salt metathesis with LiCp* however, there is no report on the

use of LiCp* in the literature for this example. Instead,

TlCp*^[32,33]and (nBu)₃SnCp*^[34]were used as Cp* source, most

likely to avoid the nucleophilic attack at the coordinated ben-

zene.^[33,35–37]Inspired by the literature known reaction with

(nBu)₃SnCp*,^[34]we changed the solvent to tetrahydrofuran and

added 2 equivalents of LiCp* relative to [(C₆H₆)RuCl₂]₂and

stirred the mixture at room temperature for 2 h (see

Scheme 8). For the workup, we substituted excess NH₄PF₆by

2 equivalents of KPF₆. Although the formation and isolation of

the desired benzene complex 19 in high purity proved suc-

cessful, the yield of 12% was significantly lower than reported

for the other methods (31%). Examination of the mother

liquor after precipitation of the Ru complex by NMR revealed a

second well-defined compound with a similar solubility. Com-

plex 20 could not be fully separated from 19, but unambigu-

ously characterized by NMR spectroscopy, X-ray structure anal-

ysis (see Supporting Information) and ESI-MS (m/z=449.2 in

acetonitrile).

Figure 14. Cyclic voltammograms of 2 (magenta, 0.26 mm), 5 (red, 0.30 mm),

9 (green, 0.28 mm), 15 (black, 0.30 mm), 16 (blue, 0.26 mm), 17 (purple,

0.34 mm) and 19 (orange, 0.29 mm); DCM/0.1m NBu₄PF₆, v=0.1 Vs^ꢀ1

.

ruthenocene(III) center.^[38]A similar mechanism might operate

here. In both electrolytes, oxidation of ligand 2 is also chemi-

cally irreversible. In contrast, for ligands 5 and 9 the shapes of

the voltammograms indicate ECE-type mechanisms. Here, the

primary oxidation products undergo a chemical reaction and

the resulting species can be reduced. The identity of these in-

termediates was not further investigated. Only 15 shows a re-

verse peak directly connected to the primary oxidation in di-

chloromethane, as indicated by the peak potentials. With a

ox

peak current ratio I_p^red/I_pof 0.6 and a peak potential differ-

ence DE_p>100 mV at a scan rate of 0.1 Vs^ꢀ1(78 mV at

0.05 Vs^ꢀ1), we assume a quasi-reversible electron transfer step

followed by a chemical reaction. Voltammograms of 15 in the

acetonitrile electrolyte are comparable to 5 and 9, thus again

indicating an ECE-mechanism.

Scheme 8. Unexpected formation of the ruthenium benzene complex 20.

Presumably, complex 20 is formed by nucleophilic attack of

LiCp* at the benzene ligand of either the starting material or

complex 19. In both cases a cyclohexadienyl anion should be

formed. These complexes are already known from litera-

ture.^[33–37]In our case however, a hydride abstraction is fol-

lowed so that the benzene ligand is rearomatized, whereas the

cyclopentadiene fragment cannot gain back its aromaticity.

ꢀ

The interaction between the oxidation product and the PF₆

anion might be avoided if Li[B(C₆F₅)₄] with its weakly coordinat-

ꢀ

ing anion were used. The PF₆counter anion of compounds 5

and 16–19 was removed by in situ precipitation of solid LiPF₆

in dichloromethane. The replacement of the anion leads

indeed to the occurrence of reverse peaks for complexes 15,

16 and 17 (Figure 15). The voltammograms of ligands 2 and 9

keep their irreversible shape. The peak potentials of ligands 2,

5, and 9, as well of the complexes 15, 16 and 17 depend on

Electrochemical investigations

The electrochemical behavior of pentamethylruthenocene (15),

the imidazolium-substituted ruthenocene derivates 16–18 and

the cationic pentamethylcyclopentadienylbenzeneruthenium

complex (19) as well as the free ligands 2, 5 and 9 has been in-

vestigated by cyclic voltammetry (CV) in dichloromethane and

acetonitrile. Three different electrolytes were used, based on

tetrabutylammonium hexafluorophosphate (NBu₄PF₆; both sol-

vents) or lithium tetrakis(pentafluorophenyl)borate (Li[B(C₆F₅)₄]

(containing 2.5 equiv. Et₂O; dichloromethane only).

Complexes 15–17 and 19 exhibit a chemically irreversible

anodic oxidation at scan rates of 0.05 to 2 Vs^ꢀ1with the sup-

porting electrolyte NBu₄PF₆in dichloromethane (Figure 14) or

acetonitrile. Traditional anions, such as PF₆^ꢀ, that may interact

coordinatively with redox active species have been shown to

be responsible for promoting a two-electron ruthenocene oxi-

dation process by anion attack at the strongly electrophilic

Figure 15. Cyclic voltammograms of 2 (magenta, 0.14 mm), 5 (red, 0.14 mm),

9 (green, 0.13 mm), 15 (black, 0.15 mm), 16 (blue, 0.2 mm) and 17 (purple,

0.2 mm); DCM/0.008m Li[B(C₆F₅)₄], v=0.1 Vs^ꢀ1

.

&

Chem. Eur. J. 2020, 26, 1 – 16

www.chemeurj.org

8

ꢁ 2020 The Authors. Published by Wiley-VCH GmbH

ÝÝ These are not the final page numbers!

Full Paper

doi.org/10.1002/chem.202002801

Chemistry—A European Journal

ox

Figure 16. E_pof 2, 5, 9, 15, 16, 17, and 19; CH₃CN/0.1m NBu₄PF₆(black),

DCM/0.1m NBu₄PF₆(red), DCM/0.008m Li[B(C₆F₅)₄] (green).

Figure 17. Cyclic voltammograms of 16 (0.26 mm) in DCM/ 0.008m

Li[B(C₆F₅)₄] before (black) and after (red) adding of NBu₄PF₆to the solution.

the electrolyte composition, with most positive potentials for

ꢀ

the B(C₆F₅)₄electrolyte (Figure 16). At the same time, there is

increases moderately, indicating the influence of electrochemi-

cal quasireversibility (electron-transfer kinetics) at these faster

time scales. Concentration normalized currents I_p/c⁰are almost

independent of c⁰and increase linearly with the square root of

the scan rate. Thus, the electrode reaction is diffusion con-

a clear trend to more positive potentials for more electron de-

ficient species, that is, those bearing a larger number of imida-

zolium substituents, in the two series of compounds. Note,

that peak potentials of even partially irreversible peaks have a

thermodynamic and a kinetic component. The thermodynami-

cally more meaningful formal potentials of compounds 15, 16

and 17 will be discussed below. Cyclic voltammograms of 18

show an intense, symmetric oxidation peak above 1.57 V, with

a peak current that is independent of the starting concentra-

tion c⁰. A reduction peak is not found. During multiple poten-

ox

trolled. The peak current ratio I_p^red/I_pis close to unity for v be-

tween 0.05 and 2 Vs^ꢀ1, indicating chemical reversibility, that is,

the absence of follow-up reactions of the primary oxidation

products. Under conditions where the processes are chemically

and electrochemically reversible (or mildly quasi-reversible),

formal potentials E⁰of 15, 16, and 17 were determined from

the cyclic voltammograms as the „mid-point potentials“ and

confirmed from differential pulse voltammetry (DPV) data

(Table 1).

ox

tial cycles, I_pdecreases and E_pshifts. We assume adsorption

of the compound and progressive deactivation of the elec-

trode surface. Further analysis of the data was not attempted.

Complex 19 does not show any electrochemical activity in the

ꢀ

B(C₆F₅)₄electrolyte between ꢀ2.3 and 1.7 V. From the oxida-

Table 1. E⁰of 15, 16 and 17 vs. Fc/Fc⁺; DCM/0.008m Li[B(C₆F₅)₄].

tion peak potentials in Figure 16 we estimate a value for 19

under these conditions more positive than the accessible po-

tential range of the electrolyte. Thus, 5 (in complex 17) with

two imidazolium substituents seems still to be a better donor

ligand than benzene (in 19).

E⁰/V

15

16

17

CV

DPV

0.219ꢃ0.003

0.219ꢃ0.001

0.749ꢃ0.002

0.750ꢃ0.002

1.299ꢃ0.002

1.301ꢃ0.003

The oxidation current of precursor ligands 2 and 5 decreases

during subsequent potential cycles until the electrochemical

activity disappears, possibly due to slow formation of the re-

spective half-sandwich complexes of lithium and lithiocenes

(see Scheme 2 for compound 2). This could be promoted by

the fact that the Li salt is used in excess for the measurement.

Another decomposition pathway could be the protonation of

the corresponding Cp ring by HCl generated from decompos-

ing dichloromethane.

We conclude that every additional imidazolium substituent

increases the oxidation potential of ruthenocene complexes

from 15 to 16 to 17 by more than 500 mV owing to its elec-

tron-withdrawing effect and insertion of an additional positive

charge.

While none of the compounds could be reduced in the ac-

cessible potential range of DCM, 5, 9, 16, 17 and 19 showed

reduction peaks at very negative potentials in MeCN/ 0.1m

NBu₄PF₆. The reduction of the complexes under these condi-

tions is irreversible, possibly owing to the high reactivity of a

Addition of NBu₄PF₆in a concentration equimolar to the

ꢀ

B(C₆F₅)₄supporting electrolyte in dichloromethane changes

the electrochemical behaviour back to an irreversible two-elec-

tron process, (for example 16, Figure 17). The negative poten-

tial shift could be a result of a coupled follow-up reaction or

increased ion pairing of the oxidation product with PF₆^ꢀ, as it

was proposed by Geiger et al.^[39]

red

primarily formed 19-electron ruthenocene. The values for E_p

(5: ꢀ3.145 V, 9: ꢀ2.244 V, 16: ꢀ2.890 V, 17: ꢀ2.775 V, 19:

ꢀ2.599 V; see Supporting Information) are consistent with the

substituent effect and the decrease of the electron density in

this order.

At 0.05ꢂvꢂ0.2 Vs^ꢀ1, the peak potentials of 15, 16, and 17

ꢀ

in the B(C₆F₅)₄electrolyte are independent of v and the con-

centration c⁰, with DE_paround 70 mV. At higher scan rates DE_p

Chem. Eur. J. 2020, 26, 1 – 16

www.chemeurj.org

9

ꢁ 2020 The Authors. Published by Wiley-VCH GmbH

&

These are not the final page numbers! ÞÞ

Full Paper

doi.org/10.1002/chem.202002801

Chemistry—A European Journal

nium

hexafluorophosphate

(8),^[27,45]

B(C₆F₅)₃

and

[44,46–48]

Conclusions

Li[B(C₆F₅)₄]·2.5Et₂O^[49]were synthesized according to the published

procedure. nBuLi, dicyclopentadiene, hexachloroethane, lithium

bis(trimethylsilyl)amide, potassium hexafluorophosphate and deu-

terated water were purchased and used without further purifica-

tion.

In this work we improved the synthesis of imidazolium cyclo-

pentadienylides like 2 and introduced di(imidazolium)- (5 and

9) and even tri(imidazolium) cyclopentadienylides (12 and 13).

The electron-withdrawing effect of the imidazolium substituent

could already be noticed in the synthesis of the compounds

and comparing their reactivity against air and moisture. The

different reactivity manifests itself also in the synthesis of the

ruthenocenes of the mono substituted ylide 16 and of the dis-

ubstituted ylides 17 and 18. Complexes of the trisubstituted

ylides could not be prepared. The electrochemical data sup-

port the electron withdrawing substituent effect for the li-

gands and the complexes. The experimental data and DFT cal-

culations show a pronounced difference between free rotating

and planar fixed imidazolium moieties. We also could observe

a rare equilibrium between a coordinated and a non-coordi-

nated Cp ligand, as the metal Cp bond is weakened due to the

lower electron density of the ligand 9. The trisubstituted com-

pounds 12 and 13 provide even less electron density in the Cp

ring which prevents coordination to the ruthenium centre.

Currently we are investigating the CO stretching frequencies

of the corresponding half-sandwich tricarbonyl complexes of

chromium(0), molybdenum(0) and tungsten(0) to quantify the

donor properties of our novel ligands.

Electrochemical investigations: Dichloromethane and acetonitrile

for electrochemical experiments were purchased from Alfa Aesar.

Dichloromethane was distilled over P₂O₅and then over K₂CO₃. Ace-

tonitrile was distilled successively over P₂O₅, CaH₂and again over

P₂O₅. Silver perchlorate (AgClO₄), tetra-n-butylammonium hexa-

fluorophosphate (NBu₄PF₆) and ferrocene were purchased from

Alfa Aesar (reagent grade). The supporting electrolytes NBu₄PF₆

were used in 0.1m (NBu₄PF₆) and in 0.008m (LiB(C₆F₅)₄) concentra-

tion. The electrolyte solutions were degassed by freeze-pump-thaw

cycles (CH₂Cl₂) or by bubbling of argon (acetonitrile). For the elec-

trochemical experiments, an Eco Chemie BV Autolab PG-STAT100

(Metrohm, Filderstadt, Germany) was used with control software

GPES (v. 4.9). Cyclic staircase voltammograms were recorded with

a=0.5 defining the sampling point at 178C with a glassy carbon

electrode. All experiments were carried out under argon with a

gas-tight full-glass cell in a three-electrode arrangement.^[50]iR drop

was compensated by positive feedback through the GPES soft-

ware. Cyclic voltammetric scan rates ranged from 0.05 to 2 Vs^ꢀ1

.

All CV experiments were carried out for several concentrations.

Background curves were recorded for various scan rates in the first

part of an experimental session for later use. Then the substrate

was added in the form of several aliquots from a stock solution. i-

E-curves were recorded at all scan rates after each addition. Finally,

background currents were subtracted from these data. Differential

pulse voltammograms were recorded with a pulse height of 5 mV.

Formal potentials were determined as mean values of peak poten-

tials („mid-point potentials“) from cyclic voltammograms of reversi-

ble shape, or from the peak potentials in differential pulse voltam-

mograms after correction by half the pulse height. All potentials

were determined vs. a Ag/Ag⁺(0.01m in CH₃CN/0.1m NBu₄PF₆)

electrode with a Haber-Luggin dual reference electrode system.^[51]

The values were rescaled to the Fc/Fc⁺standard for each electro-

Experimental Section

General remarks: All experiments, if not other otherwise noted,

were performed under inert argon atmosphere using standard

Schlenk technique. Solvents were dried and degassed using an

MB-SPS Solvent Purification System from MBraun. Deuterated sol-

vents were dried with standard purification methods and de-

gassed·^[40]The NMR spectra were recorded at 268C with a Bruker

DRX-250, Bruker AVII+400 spectrometer, a Bruker AVII+500, a

lyte (E⁰

(Fc/Fc⁺)=212ꢃ1 mV, E⁰

(Fc/Fc⁺)=82ꢃ

DCM=NBu₄PF₆

MeCN=NBu₄PF₆

1

1 mV, E⁰

(Fc/Fc⁺)=302ꢃ3 mV, all vs. Ag/Ag⁺).

Bruker AVII HDX 600 and with a Bruker Avance III HDX 700. H and

DCM=LiBðC₆F₅Þ₄

¹³C{¹H} NMR spectra were referenced to the respective solvent

signal as the internal standard. For ¹⁹F{¹H} NMR pure CFCl₃and for

³¹P{¹H} NMR 85% aqueous H₃PO₄was used as external standard.

Modified synthesis of lithium cyclopentadienide:^[52]To a solution of

nBuLi (1.6m, 50 mL, 80 mmol, 1 equiv) in n-hexane at 08C a solu-

tion of freshly distilled CpH (9.9 mL, 0.12 mol, 1.5 equiv) in 100 mL

of pentane was added dropwise under stirring within 1 h. After the

addition was completed, the suspension was stirred for 30 min at

room temperature. Then the residue was filtered off and washed

three times with 20 mL of pentane. After drying in vacuo, the

product (5.5 g, 77 mmol, 96%) was obtained as a white powder.

¹H NMR (400 MHz, [D₆]DMSO): d [ppm]=5.32 (s, Cp-H). ¹H NMR

(400 MHz, CD₃CN): d [ppm]=5.51 (s, Cp-H). ¹H NMR (400 MHz,

[D₈]THF): d [ppm]=5.68 (s, Cp-H). ¹H NMR (400 MHz, [D₈]toluene:

[D₈]THF 11:1): d [ppm]=6.32 (s, Cp-H). ¹H NMR (400 MHz, C₆D₆:

[D₈]THF 8:1): d [ppm]=6.46 (s, Cp-H).

7

The Li{¹H} NMR chemical shifts are calibrated to a 1m LiCl solution

in water as the external standard. Signals were assigned using 2D

NMR (COSY, NOESY, HSQC, HMBC) experiments and reported in

ppm. The numbering scheme of the signals is depicted in the Sup-

porting Information. A slight deviation of the integrals of the Cp

protons compared to the methyl groups can be observed in the

¹H NMR spectrum of the ligands. This is not due to H/D exchange

with deuterated solvents, but to the different relaxation times of

the protons. Changing the delay time from 1 s to 60 s leads to

matching of observed and theoretical integral values. The IR spec-

tra were collected with a Bruker Vertex 70. The elemental analyses

were determined using a varioMICRO cube by the elemental analy-

sis section of the Institut fꢀr Anorganische Chemie at the Universi-

ty of Tꢀbingen and mass spectra were recorded using a Bruker Es-

quire 3000 Plus by the mass spectra section of the Institut fꢀr Or-

ganische Chemie at the University of Tꢀbingen. Melting points

were measured using a Bꢀchi Melting Point M-560. UV/VIS spectra

were recorded with a Jasco V-770 UV/Visible/NIR spectrophotome-

Modified synthesis of 2-chloro-1,3,4,5-tetramethylimidazolium

chloride (1):^[21,22]To

a solution of hexachloroethane (1.19 g,

5.03 mmol, 1.25 equiv) in 20 mL of diethyl ether at ꢀ308C a solu-

tion of 1,3,4,5-tetramethylimidazolin-2-ylidene (500 mg, 4.03 mmol,

1 equiv) in 40 mL diethyl ether was added dropwise over a period

of 30 min. The suspension was stirred for additional 20 min at

room temperature. After filtration the residue was washed with

40 mL of diethyl ether and dried in vacuo. The product was ob-

tained as colorless microcrystals (0.713 g, 3.65 mmol, 91%). Single

crystals suitable for X-ray analysis were obtained by diffusion of di-

ter.

1,3,4,5-tetramethylimidazolin-2-ylidene,^[20]

Cp*H,^[41]

[RuCp*Cl₂]_n,^[42][RuCp*(CH₃CN)₃](PF₆),^[43]thallium cyclopentadie-

nide,^[44]N,N’-di(n-propyl)-1,8-dichlorodiimidazo[1,5-b:5’,1’-f]pyridazi-

&

Chem. Eur. J. 2020, 26, 1 – 16

www.chemeurj.org

10

ꢁ 2020 The Authors. Published by Wiley-VCH GmbH

ÝÝ These are not the final page numbers!

Full Paper

doi.org/10.1002/chem.202002801

Chemistry—A European Journal

ethyl ether into a solution of acetonitrile at ꢀ308C. ¹H NMR

(400 MHz, CD₃CN): d [ppm]=3.67 (s, 6H, NCH₃), 2.26 (s, 6H, CCH₃).

¹³C{¹H} NMR (100 MHz, CD₃CN): d [ppm]=130.1 (CCl, signal from

HMBC), 128.8 (CCH₃), 33.9 (NCH₃), 9.3 (CCH₃). CHN: calcd C 43.10, H

6.20, N 14.36 found C 43.25, H 5.93, N 14.47. MS (ESI⁺, CH₃CN):

m/z=159.0 [M]⁺.

trile and the product precipitated with 15 mL water. After filtration,

washing with water and drying in vacuo, the product was obtained

as a pale heather, air stable crystalline solid (369 mg, 0.808 mmol,

63%). Single crystals suitable for X-ray analysis were obtained by

slow evaporation of acetonitrile from a solution of 5. ¹H NMR

3

(400 MHz, CD₃CN): d [ppm]=6.44 (t, J_HH=2.3 Hz, 1H, 2-H), 6.32 (d,

³J_HH=2.3 Hz, 2H, 4/5-H), 3.68 (s, 12H, NCH₃), 2.22 (s, 12H, CCH₃).

¹³C{¹H} NMR (100 MHz, CD₃CN): d [ppm]=147.2 (C-6/7), 124.9

Modified synthesis of 1,3,4,5-tetramethylimidazolium cyclopenta-

dienide (2):^[11]To

a suspension of LiCp (488 mg, 6.77 mmol,

(CCH₃), 114.7 (C-2), 113.4 (C-4/5), 103.4 (C-1/3), 33.8 (NCH₃), 9.1

2.10 equiv) in 15 mL of acetonitrile at room temperature a solution

of 1 (629 mg, 3.23 mmol, 1 equiv) in 5 mL of acetonitrile was

added under stirring. The suspension was stirred for 1 h, where-

upon the color turned brownish and a white precipitate formed,

which was filtered off. The filtrate was dried in vacuo and extracted

three times with each 7 mL of dichloromethane. The combined or-

ganic layers were washed three times with 5 mL of degassed water

and then concentrated to dryness. The residue was washed with

15 mL of pentane and dried in vacuo to obtain the product as a

1

(CCH₃). ¹⁹F{¹H} NMR (376 MHz, CD₃CN): d [ppm]=ꢀ72.9 (d, J_PF

=

707 Hz, PF₆). ³¹P{¹H} NMR (162 MHz, CD₃CN): d [ppm]=ꢀ144.6 (spt,

¹J_PF=707 Hz, PF₆). ¹H NMR (400 MHz, CD₂Cl₂): d [ppm]=6.41 (t,

³J_HH=2.3 Hz, 1H, 2-H), 6.34 (d, ³J_HH=2.3 Hz, 2H, 4/5-H), 3.72 (s,

12H, NCH₃), 2.25 (s, 12H, CCH₃). CHN: calcd C 50.00, H 5.96, N

12.28 found C 50.11, H 5.70, N 12.26. MS (ESI⁺, CH₃CN): m/z=311.2

[M]⁺. UV/VIS (CH₂Cl₂): l₁=284 nm (e=3.0ꢃ10³Lmol^ꢀ1·cm^ꢀ1), l₂=

339 nm

10²Lmol^ꢀ1·cm^ꢀ1).

(e=7.2ꢃ10³Lmol^ꢀ1·cm^ꢀ1),

UV/VIS (CH₃CN):

l₃=524 nm

l₁=275 nm

(e=1ꢃ

(e=1.2ꢃ

1

lilac solid (517 mg, 2.74 mmol, 85%). H NMR (400 MHz, [D₆]DMSO):

10⁴Lmol^ꢀ1·cm^ꢀ1), l₂=336 nm (e=3.5ꢃ10⁴Lmol^ꢀ1·cm^ꢀ1), l₃=

509 nm (e=2.5ꢃ10⁴Lmol^ꢀ1·cm^ꢀ1). IR (KBr): v˜ [cm^ꢀ1]=2961 (w),

2929 (w), 1653 (m), 1548 (s), 1504 (m), 1440 (m), 1348 (m), 1245

(w), 1213 (w), 1126 (w), 1047 (w), 910 (m), 840 (vs., PF₆^ꢀ), 710 (w),

647 (w), 558 (s, PF₆^ꢀ). Mp.: 2668C.

d [ppm]=6.04–6.02 (m, 2H, 2/5-H), 5.84–5.82 (m, 2H, 3/4-H), 3.63

(s, 6H, NCH₃), 2.18 (s, 6H, CCH₃). ¹H NMR (400 MHz, CD₃CN): d

[ppm]=6.14–6.12 (m, 2H, 2/5-H), 5.94–5.93 (m, 2H, 3/4-H), 3.66 (s,

1

6H, NCH₃), 2.17 (s, 6H, CCH₃). H NMR (400 MHz, CD₂Cl₂): d [ppm]=

6.21–6.20 (m, 2H, 2/5-H), 6.05–6.04 (m, 2H, 3/4-H), 3.70 (s, 6H,

NCH₃), 2.17 (s, 6H, CCH₃). ¹H NMR (400 MHz, [D₈]THF): d [ppm]=

6.08–6.07 (m, 2H, 2/5-H), 5.93–5.91 (m, 2H, 3/4-H), 3.66 (s, 6H,

NCH₃), 2.17 (s, 6H, CCH₃). ¹H NMR (400 MHz, CDCl₃): d [ppm]=

6.35–6.34 (m, 2H, 2/5-H), 6.31–6.29 (m, 2H, 3/4-H), 3.76 (s, 6H,

NCH₃), 2.20 (s, 6H, CCH₃). ¹H NMR (400 MHz, [D₈]toluene): d

[ppm]=6.92–6.91 (m, 2H, 3/4-H), 6.54–6.52 (m, 2H, 2/5-H), 3.00 (s,

H/D exchange of 5: To a solution of 5 (23.0 mg, 50.4 mmol, 1 equiv)

in 0.4 mL of CD₃CN 91 mL D₂O (101 mg, 5.04 mmol, 100 equiv)

were added. The mixture was then heated for 5 d at 608C. The sol-

vent was removed in vacuo and the residue dissolved in 0.4 mL of

CD₃CN to confirm the formation of 5-d₃. ¹H NMR (400 MHz,

CD₃CN): d [ppm]=6.45 (s, residual proton signal, 2-H), 6.32 (s, re-

sidual proton signal, 4/5-H), 3.69 (s, 12H, NCH₃), 2.23 (s, 12H,

CCH₃). ¹³C{¹H} NMR (100 MHz, CD₃CN): d [ppm]=147.2 (C-6/7),

1

6H, NCH₃), 1.14 (s, 6H, CCH₃). H NMR (400 MHz, C₆D₆): d [ppm]=

7.14–7.12 (m, 2H, 3/4-H), 6.69–6.67 (m, 2H, 2/5-H), 2.98 (s, 6H,

NCH₃), 1.25 (s, 6H, CCH₃). CHN: calcd C 76.55, H 8.57, N 14.88

found C 76.63, H 8.30, N 14.56. MS (ESI⁺, CH₃CN): m/z=189.1

[M+H]⁺. UV/VIS (CH₃CN): l₁=332 nm (e=2.5ꢃ10⁴Lmol^ꢀ1·cm^ꢀ1),

l₂=491 nm (e=1ꢃ10¹Lmol^ꢀ1·cm^ꢀ1). Contrary to earlier observa-

tions the compound was stable over months under argon at rt.

1

124.9 (CCH₃), 114.7 (C-2), 114.4 (t, J_CD=24.3 Hz, C-2), 113.2 (C-4/5),

1

113.0 (t, J_CD=24.3 Hz, C-4/5), 103.2 (C-1/3), 33.8 (NCH₃), 9.1 (CCH₃).

MS (ESI⁺, CH₃CN): m/z=314.2 [M]⁺. MS (ESI^ꢀ, CH₃CN): m/z=144.8

[PF₆]^ꢀ. The deuteration degree for the Cp of 91% was determined

from the ESI–MS spectrum.

Formation of the lithium adducts 3 and 4: Li[B(C₆F₅)₄]·2.5Et₂O

(18 mg, 21 mmol, 1 equiv) and 2 (5.0 mg, 27 mmol, 1.3 equiv) were

dissolved in 0.4 mL of [D₂]dichloromethane. ¹H NMR (500 MHz,

CD₂Cl₂, RT): d [ppm]=6.13–6.12 (m, 2H, 2/5-H), 6.01–5.99 (m, 2H,

3/4-H), 3.68 (s, 6H, NCH₃), 2.23 (s, 6H, CCH₃). ⁷Li NMR (194 MHz,

CD₂Cl₂, RT): d [ppm]=0.8 (br s, Li), ꢀ7.3 (br s, LiCp). ⁷Li NMR

(194 MHz, CD₂Cl₂, ꢀ808C): d [ppm]=ꢀ0.8 (s, Li), ꢀ6.0 (br s, CpLi

(3)), ꢀ11.7 (s, LiCp₂(4)).

Formation of the lithium adducts 3-NTf₂: LiNTf₂(7.6 mg, 27 mmol,

1 equiv) and 2 (5.0 mg, 27 mmol, 1.0 equiv) were dissolved in

0.4 mL of [D₂]dichloromethane. ¹H NMR (250 MHz, CD₂Cl₂, RT): d

[ppm]=6.24–6.22 (m, 2H, 2/5-H), 6.12–6.9 (m, 2H, 3/4-H), 3.70 (s,

6H, NCH₃), 2.21 (s, 6H, CCH₃). ⁷Li NMR (97 MHz, CD₂Cl₂, RT): d

[ppm]=ꢀ7.0 (br s, LiCp).

Synthesis of 1,3-bis(1,3,4,5-tetramethylimidazolium)cyclopentadie-

nide hexafluorophosphate (5): A solution of 2 (241 mg, 1.28 mmol,

1 equiv) and 1 (250 mg, 1.28 mmol, 1 equiv) in 8 mL of acetonitrile

was stirred at room temperature for 20 h. Then LiHMDS (155 mg,

0.926 mmol, 0.72 equiv) was added to the formed suspension.

After 48 h a second portion of LiHMDS (81.0 mg, 0.484 mmol,

0.38 equiv) was added. After stirring the mixture overnight, the sol-

vent was removed in vacuo and the residue suspended in 5 mL of

water. The solid was removed by filtration and a concentrated so-

lution of KPF₆(259 mg, 1.41 mmol, 1.10 equiv) in water was added

under stirring to the filtrate. After stirring for 20 min at rt the pre-

cipitate was filtered off and washed with each 10 mL of water and

diethyl ether. The residue was then dissolved in 3 mL of acetoni-

Independent formation of 6 and 7: To a suspension of 1 (31.4 mg,

161 mmol, 1 equiv) in 0.5 mL of CD₃CN tetramethylimidazolin-2-yli-

dene (20.0 mg, 161 mmol, 1.00 equiv) was added. After 4 h 2 mL of

diethyl ether were added and the solid was filtered off, dissolved

in water and combined with a saturated aqueous solution of KPF₆

(70.0 mg, 380 mmol, 2.36 equiv). The formed off-white precipitate

was filtered off, washed with water and diethyl ether and dried in

vacuo. The air stable product is a mixture of 6 and 7 in the ratio

3:2. Single crystals suitable for X-ray analysis of 6 (colorless plates)

and 7 (colorless needles) were obtained by slow evaporation of

1

acetonitrile from a solution of the mixture. Compound 6: H NMR

(700 MHz, CD₃CN): d [ppm]=3.26 (s, 6H, NCH₃), 2.15 (s, 6H, CCH₃).

¹³C{¹H} NMR (176 MHz, CD₃CN):

d [ppm]=145.2 (NCN), 125.2

(CCH₃), 121.6 (NCC), 32.9 (NCH₃), 30.1 (NCC), 9.1 (CCH₃). ¹H NMR

(700 MHz, CD₂Cl₂): d [ppm]=3.33 (s, 6H, NCH₃), 2.20 (s, 6H, CCH₃).

¹³C{¹H} NMR (176 MHz, CD₃CN): d [ppm]=144.5 (NCN), 124.9

(CCH₃), 120.9 (NCC), 32.7 (NCH₃), 30.0 (NCC), 9.3 (CCH₃). MS (ESI⁺,

CH₃CN): m/z=286.2 [M]⁺. Compound 7: ¹H NMR (700 MHz,

CD₃CN): d [ppm]=3.58 (s, 6H, NCH₃), 2.36 (s, 6H, CCH₃). ¹³C{¹H}

NMR (176 MHz, CD₃CN): d [ppm]=134.0 (CCH₃), 124.1 (NCN), 34.8

(NCH₃), 9.4 (CCH₃). MS (ESI⁺, CH₃CN): m/z=124.1 [M]²⁺, 393.2 [M+

PF₆]²⁺

.

Synthesis of 7: To a suspension of 2 (100 mg, 513 mmol, 1 equiv) in

2.5 mL tetrahydrofuran at ꢀ308C was added dropwise a solution

of tetramethylimidazol-2-ylidene (64.0 mg, 515 mmol, 1.00 equiv) in

2.5 mL tetrahydrofuran. The mixture was stored at ꢀ308C for 16 h.

The solvent was removed in vacuo and the residue dissolved in

Chem. Eur. J. 2020, 26, 1 – 16

www.chemeurj.org

11

ꢁ 2020 The Authors. Published by Wiley-VCH GmbH

&

These are not the final page numbers! ÞÞ

Full Paper

doi.org/10.1002/chem.202002801

Chemistry—A European Journal

water. A saturated aqueous solution of KPF₆(234 mg, 1.27 mmol,

2.48 equiv) was added under stirring. The formed precipitate was

filtered off, washed with 5 mL water, twice with 5 mL diethyl ether

each and dried in vacuo to obtain 7 as an air stable off-white solid

(204 mg, 388 mmol, 74%). ¹H NMR (400 MHz, CD₃CN): d [ppm]=

3.59 (s, 6H, NCH₃), 2.36 (s, 6H, CCH₃). ¹³C{¹H} NMR (100 MHz,

CD₃CN): d [ppm]=134.0 (CCH₃), 124.2 (NCN, signal from HMBC),

34.8 (NCH₃), 9.4 (CCH₃). MS (ESI⁺, CH₃CN): m/z=124.1 [M]²⁺, 393.2

which the color turned orange. After stirring for 1 h LiHMDS

(65.2 mg, 390 mmol, 2 equiv) was added as solid together with

2 mL acetonitrile. After stirring for 1 h at room temperature, the

dark orange solution was concentrated in vacuo to 2 mL and

10 mL degassed water were added. After stirring for 30 min, the

formed brown precipitate was filtered off and dissolved in 2 mL of

acetonitrile. 30 mL of diethyl ether were added to precipitate the

product, which was filtered off, washed with 5 mL diethyl ether

and dried in vacuo. The brown air stable product (84.2 mg,

117 mmol, 60%) is a mixture of both isomers 12 and 13 in 4.5:1

ratio. Crystals suitable for X-ray analysis of 12 (orange plates) and

13 (oranges blocks) were obtained by diffusion of diethyl ether

[M+PF₆]²⁺

.

Synthesis of 9: To a suspension of TlCp (237 mg, 1.01 mmol,

3.05 equiv) in 4 mL acetonitrile a solution of N,N’-di(n-propyl)-1,8-

dichlorodiimidazo[1,5-b:5’,1’-f]pyridazinium

di(hexafluorophos-

1

into a solution of the product mixture in acetonitrile. 12: H NMR

phate) (8) (200 mg, 0.332 mmol, 1 equiv) in 6 mL of acetonitrile

was added. After stirring for 72 h in the dark, the reaction mixture

turned yellow and a grey precipitate had formed. The solid was fil-

tered off and the solvent removed in vacuo. The residue was dis-

solved in 10 mL acetonitrile and 20 mL of degassed water was

added to precipitate the yellow raw material, which was filtered

off, dissolved in 1.5 mL acetonitrile and 30 mL of ethyl ether were

added. The obtained yellow solid was filtered off again, dissolved

in 5 mL acetonitrile and 40 mL degassed water were added. The

yellow precipitate was filtered off, washed with 3 mL degassed

water and dried in vacuo. The product 9 was obtained as a yellow,

air-stable solid (108 mg, 0.240 mmol, 72%). Crystals suitable for X-

ray analysis were obtained by overlaying a solution of 9 in di-

(600 MHz, CD₃CN): d [ppm]=7.35 (s, 2H, H-2/5), 7.09 (s, 2H, H-7/9),

6.87 (s, 2H, H-3/4), 4.37 (t, ³J_HH=7.0 Hz, 4H, NCH₂), 3.65 (s, 6H,

NCH₃), 2.30 (s, 6H, CCH₃), 1.96 (tq, 4H, ³J_HH=7.4 Hz, ³J_HH=7.0 Hz,

3

CH₂), 1.00 (t, 6H, J_HH=7.4 Hz, CH₃). ¹³C{¹H} NMR (151 MHz, CD₃CN):

d [ppm]=143.2 (C-10), 132.9 (C-6a/9b), 127.4 (CCH₃), 121.3 (C-2a/

4a), 116.3 (C-3/4), 116.1 (C-2/5), 116.0 (C-6b/9a), 113.1 (s, C-7/9),

106.2 (C-8), 51.6 (NCH₂), 34.0 (NCH₃), 23.7 (CH₂), 10.9 (CH₃), 9.1

1

(CCH₃). 13: H NMR (600 MHz, CD₃CN): d [ppm]=7.34 (s, 1H, H-2),

3

7.24 (s, 1H, H-5), 7.12 (d, 1H, J_HH=4.2 Hz, H-9), 6.84 (d, 1H, J_HH

=

4.2 Hz, H-8), 6.83 (s, 1H, H-3 or H-4), 6.80 (s, 1H, H-3 or H-4), 4.40–

b

a

4.37 (m, 2H, NCH₂), 3.50 (s, 6H, NCH₃), 3.18–3.15 (m, 2H, NCH₂),

b

a

2.33 (s, 6H, CCH₃), 2.02–1.96 (m, 2H, CH₂), 1.59–1.53 (m, 2H, CH₂),

1.04 (t, 3H, ³J_HH=7.3 Hz, CH₃), 0.71 (t, 3H, ³J_HH=7.3 Hz, CH₃).

¹³C{¹H} NMR (151 MHz, CD₃CN): d [ppm]=142.6 (C-10), 128.6

(CCH₃), 127.9 (C-8), 116.5 (C-3 or C-4), 116.3 (C-3 or C-4), 116.2 (C-2),

b

a

chloromethane with pentane. ¹H NMR (400 MHz, CD₃CN):

d

3

[ppm]=7.16 (s, 2H, H-2/5), 6.91 (d, 2H, J_HH=3.8 Hz, H-7/9), 6.77 (t,

3

1H, J_HH=3.8 Hz, H-8), 6.73 (s, 2H, H-3/4), 4.29 (t, 4H, J_HH=7.2 Hz,

NCH₂), 1.96 (tq, 4H, ³J_HH=7.4 Hz, ³J_HH=7.2 Hz, CH₂), 1.00 (t, 6H,

³J_HH=7.4 Hz, CH₃). ¹³C{¹H} NMR (100 MHz, CD₃CN): d [ppm]=133.2

(C-6a/9b), 124.9 (C-8), 120.8 (C-2a/4a), 115.9 (C-3/4), 115.0 (C-2/5),

111.9 (C-7/9), 104.7 (C-6b/9a), 51.2 (NCH₂), 23.7 (CH₂), 11.0 (CH₃).

¹⁹F{¹H} NMR (376 MHz, CD₃CN): d [ppm]=ꢀ72.9 (d, ¹J_PF=707 Hz,

b

a

116.1 (C-5), 112.7 (C-9), 51.6 (NCH₂), 51.6 (NCH₂), 33.7 (NCH₃), 23.7

b

a

b

a

(CH₂), 23.4 (CH₂), 10.9 (CH₃), 10.8 (CH₃), 9.1 (CCH₃). The signals

for C-2a, C-4a, C-6a, C-6b, C-7, C-9a and C-9b could not be as-

signed unambiguously. Both: ¹⁹F{¹H} NMR (376 MHz, CD₃CN): d

1

[ppm]=ꢀ72.9 (d, J_PF=707 Hz, PF₆). ³¹P{¹H} NMR (162 MHz, CD₃CN):

1

d [ppm]=ꢀ144.6 (spt, J_PF=707 Hz, PF₆). MS (ESI⁺, CH₃CN): m/z=

PF₆). ³¹P{¹H} NMR (162 MHz, CD₃CN): d [ppm]=ꢀ144.6 (spt, J_PF

=

1

214.0 [M]²⁺. MS (ESI^ꢀ, CH₃CN): m/z=144.8 [PF₆]^ꢀ.

707 Hz, PF₆). H NMR (400 MHz, CD₂Cl₂): d [ppm]=7.08 (s, 2H, H-2/

5), 6.93–6.89 (m, 3H, H-7/9 and H-8), 6.73 (s, 2H, H-3/4), 4.29 (t, 4H,

³J_HH=7.3 Hz, NCH₂), 2.01 (tq, 4H, ³J_HH=7.4 Hz, ³J_HH=7.3 Hz, CH₂),

Synthesis of 12-d₂and 13-d₂: To a solution of the above obtained

mixture of 12 and 13 (20.0 mg, 27.8 mmol, 1 equiv) in 0.4 mL of

CD₃CN were added 100 mL D₂O (111 mg, 5.54 mmol, 199 equiv).

The mixture is then heated for 8 d at 608C. The solvent was re-

moved in vacuo and the residue dissolved in 0.4 mL of CD₃CN and

3

1.08 (t, 6H, J_HH=7.4 Hz, CH₃). CHN: calcd C 50.67, H 4.70, N 12.44

found C 50.64, H 4.95, N 12.18. MS (ESI⁺, CH₃CN): m/z=305.2 [M]⁺.

MS (ESI^ꢀ, CH₃CN): m/z=144.8 [PF₆]^ꢀ. UV/VIS (CH₃CN): l₁=284 nm

(e=2.2ꢃ10⁴Lmol^ꢀ1·cm^ꢀ1), l₂=339 nm (e=1.1ꢃ10⁴Lmol^ꢀ1·cm^ꢀ1).

IR (KBr): v˜ [cm^ꢀ1]=2971 (w), 2920 (w), 1630 (m), 1587 (m), 1459

(w), 1386 (w), 1308 (w), 1343 (w), 1308 (w), 1160 (w), 1047 (w),

1024 (w), 841 (vs., PF₆^ꢀ), 792 (w), 733 (w), 722 (w), 558 (s, PF₆^ꢀ).

Mp.: 2028C (dec.).

1

filtered. 12: H NMR (400 MHz, CD₃CN): d [ppm]=7.36 (s, residual

proton signal, H-2/5), 7.10 (s, 2H, H-7/9), 6.87 (s, 2H, H-3/4), 4.38 (t,

³J_HH=7.0 Hz, 4H, NCH₂), 3.66 (s, 6H, NCH₃), 2.30 (s, 6H, CCH₃), 1.98

(tq, 4H, ³J_HH=7.4 Hz, ³J_HH=7.0 Hz, CH₂), 1.01 (t, 6H, ³J_HH=7.4 Hz,

CH₃). ¹³C{¹H} NMR (100 MHz, CD₃CN): d [ppm]=143.2 (C-10), 132.8

(C-6a/9b), 127.3 (CCH₃), 121.2 (C-2a/4a), 116.2 (C-3/4), 116.0 (C-6b/

9a), 113.1 (C-7/9), 106.2 (C-2), 51.6 (NCH₂), 34.0 (NCH₃), 23.6 (CH₂),

10.9 (CH₃), 9.1 (CCH₃). The signal for C-2/5 could not be detected.

Synthesis of 9-d₂: To a solution of 9 (23.0 mg, 51.1 mmol, 1 equiv)

in 0.4 mL of CD₃CN 92 mL D₂O (102 mg, 5.11 mmol, 100 equiv) were

added. The mixture was then heated for 24 h at 608C. The solvent

was removed in vacuo and the residue dissolved in 0.4 mL of

13: ¹H NMR (400 MHz, CD₃CN): d [ppm]=7.34 (s, residual proton

3

CD₃CN and the solution filtered. ¹H NMR (400 MHz, CD₃CN): d

signal, H-2), 7.24 (s, residual proton signal, H-5), 7.11 (d, 1H, J_HH

=

3

4.3 Hz, H-9), 6.84 (d, 1H, J_HH=4.3 Hz, H-8), 6.82 (s, 1H, H-3 or H-4),

6.80 (s, 1H, H-3 or H-4), 4.41–4.37 (m, 2H, NCH₂), 3.50 (s, 6H,

NCH₃), 3.18–3.14 (m, 2H, NCH₂), 2.33 (s, 6H, CCH₃), 2.02–1.96 (m,

[ppm]=7.16 (s, residual proton signal, H-2/5), 6.89 (d, 2H, J_HH

=

b

3.8 Hz, H-7/9), 6.76 (t, 1H, ³J_HH=3.8 Hz, H-8), 6.72 (s, 2H, H-3/4),

a

3

4.28 (t, 4H, ³J_HH=7.2 Hz, NCH₂), 1.95 (tq, 4H, ³J_HH=7.4 Hz, J_HH

=

b

a

3

b

7.2 Hz, CH₂), 1.00 (t, 6H, J_HH=7.4 Hz, CH₃). ¹³C{¹H} NMR (100 MHz,

3

2H, CH₂), 1.61–1.51 (m, 2H, CH₂), 1.04 (t, 3H, J_HH=7.3 Hz, CH₃),

0.71 (t, 3H, ³J_HH=7.3 Hz, CH₃). The ¹³C signals could not be as-

a

CD₃CN): d [ppm]=133.1 (C-6a/9b), 124.9 (C-8), 120.7 (C-2a/4a),

signed unambiguously due to the low intensity. Both: MS (ESI⁺,

1

115.8 (C-3/4), 115.0 (C-2/5), 114.8 (t, J_CD=31.2 Hz, C-2/5), 111.9 (C-

CH₃CN): m/z=215.1 [M]²⁺

,

575.2 [M+PF₆]⁺. The deuteration

7/9), 104.7 (C-6b/9a), 51.2 (NCH₂), 23.7 (CH₂), 11.0 (CH₃). MS (ESI⁺,

CH₃CN): m/z=307.2 [M]⁺. The deuteration degree of 93% was de-

termined from the ESI-spectrum.

degree of 97% was determined from the ESI-spectrum.

Modified synthesis of 1,2,3,4,5-pentamethylruthenocene (15):^[29,53]

To a suspension of LiCp (15.0 mg, 208 mmol, 1.05 equiv) in 2 mL of

acetonitrile a solution of [Ru(CH₃CN)₃Cp*](PF₆) (100 mg, 198 mmol,

1 equiv) in 1 mL of acetonitrile was added under stirring. After

10 min, the solvent of the formed solution was removed in vacuo,

Synthesis of 12 and 13: To a solution of N,N’-di(n-propyl)-1,8-di-

chlorodiimidazo[1,5-b:5’,1’-f]pyridazinium di(hexafluorophosphate)

(7) (118 mg, 195 mmol, 1 equiv) in 2 mL acetonitrile, a solution of 2

(36.7 mg, 195 mmol, 1 equiv) in 2 mL acetonitrile is added upon

&

Chem. Eur. J. 2020, 26, 1 – 16

www.chemeurj.org

12

ꢁ 2020 The Authors. Published by Wiley-VCH GmbH

ÝÝ These are not the final page numbers!

Full Paper

doi.org/10.1002/chem.202002801

Chemistry—A European Journal

and the residue extracted three times with 3 mL pentane. The

combined extracts were filtered over neutral aluminum oxide and

concentrated to dryness in vacuo to obtain the product 15 as a

colorless, air-stable, crystalline solid (56.4 mg, 187 mmol, 95%).

¹H NMR (400 MHz, [D₆]acetone): d [ppm]=4.14 (s, 5H, Cp), 1.93 (s,

15H, Cp*). ¹H NMR (400 MHz, CD₃CN): d [ppm]=4.15 (s, 5H, Cp),

1.93 (s, 15H, Cp*). CHN: calcd C 59.78, H 6.69 found C 59.82, H

6.51. NMR data match with those from literature.^[29]

Synthesis of 18: To a solution of [Ru(CH₃CN)₃Cp*](PF₆) (30.0 mg,

59.5 mmol, 1 equiv) in 0.5 mL dichloromethane a suspension of 9

(26.8 mg, 59.5 mmol, 1 equiv) in 0.5 mL dichloromethane was

added under stirring. The solid dissolved and a yellow crystalline

solid formed. After 30 min the solvent was removed in vacuo and

the solid was suspended in 0.5 mL dichloromethane. After 30 min

the solvent is again removed in vacuo and the solid was washed

two times with 0.35 mL of dichloromethane. The residue was dried

in vacuo to obtain the product 18 as a yellow, crystalline solid

(40.5 mg, 48.7 mmol, 82%). Crystals suitable for X-ray structure

analysis were obtained by overlaying a solution of 18 in acetoni-

Synthesis of 1-(1,3,4,5-tetramethylimidazolium)-1’,2’,3’,4’,5’-pen-

tamethylruthenocene hexafluorophosphate (16): To a suspension

of 2 (29.9 mg, 159 mmol, 1 equiv) in 0.75 mL of acetonitrile a solu-

tion of [Ru(CH₃CN)₃Cp*](PF₆) (80.0 mg, 159 mmol, 1 equiv) in

0.75 mL of acetonitrile was added under stirring. The suspension

turned yellow and the solid dissolved after 30 min. The solvent

was removed in vacuo and the residue extracted 3 times with

0.5 mL dichloromethane. The product was precipitated from the

combined extract by addition of 5 mL of pentane, filtered off and

dried in vacuo to obtain 16 as an off-white, air-stable crystalline

solid (86.2 mg, 151 mmol, 95%). Crystals suitable for X-ray analysis

were obtained by diffusion of pentane into a solution of 16 in di-

1

trile with diethyl ether. H NMR (400 MHz, CD₃CN): d [ppm]=7.71

3

(s, 2H, H-2/5), 7.18 (s, 2H, H-3/4), 5.78 (d, 2H, J_HH=2.9 Hz, H-7/9),

3

5.38 (t, 1H, J_HH=2.9 Hz, H-8), 4.44–4.38 (m, 2H, NCH₂), 4.30–4.24

(m, 2H, NCH₂), 2.01 (ps sxt, 4H, ³J_HH=7.3 Hz, CH₂), 1.74 (s, 15H,

Cp*), 1.05 (ps t, 6H, ³J_HH=7.4 Hz, CH₃). ¹³C{¹H} NMR (126 MHz,

CD₃CN): d [ppm]=135.7 (C-6a/9b, signal from HMBC), 122.8 (C-2a/

4a), 118.9 (C-2/5), 116.8 (C-3/4), 92.4 (Cp*-CCH₃), 84.3 (C-8), 73.8 (C-

7/9), 68.0 (C-6b/9a), 52.6 (NCH₂), 23.9 (CH₂), 11.5 (Cp*-CCH₃), 10.8

1

(CH₃). ¹⁹F{¹H} NMR (376 MHz, CD₃CN): d [ppm]=ꢀ72.9 (d, J_PF

=

707 Hz, PF₆). ³¹P{¹H} NMR (162 MHz, CD₃CN): d [ppm]=ꢀ144.6 (spt,

¹J_PF=707 Hz, PF₆). CHN: calcd C 41.88, H 4.36, N 6.74 found C

41.73, H 4.12, N 6.81. MS (ESI⁺, CH₂Cl₂): m/z=270.8 [M]²⁺, 305.1

[M-RuCp*]⁺, 687.1 [M+PF₆]⁺. Mp.: 2308C (dec.).

1

chloromethane. H NMR (400 MHz, CD₃CN): d [ppm]=4.77–4.76 (m,

2H, 2/5-H), 4.55–4.54 (m, 2H, 3/4-H), 3.74 (s, 6H, NCH₃), 2.22 (s, 6H,

CCH₃), 1.86 (s, 15H, Cp*). ¹³C{¹H} NMR (100 MHz, CD₃CN): d [ppm]=

143.9 (C-6), 127.3 (CCH₃), 88.2 (Cp*-CCH₃), 77.5 (C-3/4), 74.8 (C-2/5),

71.6 (C-1), 34.4 (NCH₃), 11.8 (Cp*-CCH₃), 9.1 (CCH₃). ¹⁹F{¹H} NMR

(376 MHz, CD₃CN): d [ppm]=ꢀ72.9 (d, ¹J_PF=707 Hz, PF₆). ³¹P{¹H}

Formation of 19 and 20: To a suspension of LiCp* (900 mg,

6.33 mmol, 2.00 equiv) in 30 mL tetrahydrofuran [(C₆H₆)RuCl₂]₂

(1.58 g, 3.16 mmol, 1 equiv) was added as solid with 30 mL tetrahy-

drofuran and the mixture was stirred for 2 h. The solvent was re-

moved in vacuo and 200 mL water were added to the residue. The

solution was filtered and a concentrated solution of KPF₆(1.17 g,

6.35 mmol, 2.01 equiv) in water was added under stirring. The

formed off-white precipitate was filtered off and washed with

10 mL diethyl ether. The residue was dissolved in acetone, filtered

over neutral aluminum oxide and the filtrate concentrated to dry-

ness. After dissolving the residue in dichloromethane and addition

of diethyl ether, the formed precipitate was collected, washed with

diethyl ether and dried in vacuo to obtain 19 (360 mg, 784 mmol,

12%) as an off-white air stable solid. ¹H NMR (400 MHz,

[D₆]acetone): d [ppm]=6.03 (s, 6H, C₆H₆), 2.07 (s, 15H, Cp*).

¹H NMR (400 MHz, CD₃CN): d [ppm]=5.76 (s, 6H, C₆H₆), 1.97 (s,

1

NMR (162 MHz, CD₃CN): d [ppm]=ꢀ144.6 (spt, J_PF=707 Hz, PF₆).

¹H NMR (400 MHz, CD₂Cl₂): d [ppm]=4.65–4.64 (m, 2H, 2/5-H),

4.54–4.53 (m, 2H, 3/4-H), 3.78 (s, 6H, NCH₃), 2.26 (s, 6H, CCH₃), 1.87

(s, 15H, Cp*). CHN: calcd C 46.40, H 5.49, N 4.92 found C 46.41, H

5.37, N 5.14. MS (ESI⁺, CH₃CN): m/z=425.2 [M]⁺. UV/VIS (CH₃CN):

l₁=280 nm (e=3.5ꢃ10⁴Lmol^ꢀ1·cm^ꢀ1), l₂=332 nm (e=1.7ꢃ

10⁴Lmol^ꢀ1·cm^ꢀ1). IR (KBr): v˜ [cm^ꢀ1]=3137 (w), 2966 (w), 2901 (w),

1646 (m), 1552 (m), 1511 (m), 1452 (m), 1384 (m), 1262 (w), 1046

(w), 1028 (w), 850 (vs., PF₆^ꢀ), 827 (vs., PF₆^ꢀ), 710 (w), 558 (s, PF₆^ꢀ),

444 (w). Mp.: 1938C.

Synthesis of 1,3-bis(1,3,4,5-tetramethylimidazolium)-1’,2’,3’,4’,5’-

pentamethylruthenocene hexafluorophosphate (17): To a solution

of [Ru(CH₃CN)₃Cp*](PF₆) (50.0 mg, 99.1 mmol, 1 equiv) in 1.5 mL of

acetonitrile a solution of 5 (45.2 mg, 99.1 mmol, 1 equiv) in 1.5 mL

of acetonitrile was added under stirring. The solution turned

brownish and after 15 min the solvent was removed in vacuo. The

residue was extracted three times with 0.4 mL of dichloromethane

and the solvent removed in vacuo to obtain the ruthenocene 17

as an off-white, air-stable crystalline solid (75.5 mg, 90.1 mmol,

91%). Crystals suitable for X-ray analysis were obtained by slow

1

15H, Cp*). H NMR (400 MHz, CDCl₃): d [ppm]=5.80 (s, 6H, C₆H₆),

2.03 (s, 15H, Cp*). CHN: calcd C 41.83, H 4.61 found C 41.75, H

4.45. NMR data in acetone matches with that in the literature.^[34]

The mother liquor of the precipitation from dichloromethane with

diethyl ether was concentrated to dryness and the residue recrys-

tallized from dichloromethane/diethyl ether and subsequently

from acetone/diethyl ether to obtain 20 (262 mg) as an off-white

air stable solid which contains residual 19 (about 3%). Crystals suit-

able for X-ray structure analysis were obtained by crystallization

from acetone/diethyl ether. ¹H NMR (400 MHz, CDCl₃): d [ppm]=

5.69–5.67 (m, 2H, 3-H), 5.61–5.59 (m, 3H, 2-H and 4-H), 1.98 (s,

15H, Cp*), 1.76 (s, 6H, 7-H), 1.72 (s, 6H, 6-H), 1.22 (s, 3H, 5-H).

¹³C NMR (100 MHz, CDCl₃): d [ppm]=139.8 (CCH₃-6), 136.8 (CCH₃-7)

113.7 (C-1), 95.9 (Cp*-CCH₃), 86.6 (C-4), 86.5 (C-3), 83.5 (C-2), 58.0

(CCH₃-5), 18.1 (CH₃-5), 11.2 (CH₃-7), 10.9 (Cp*-CCH₃), 10.7 (CCH₃-6).

evaporation of acetonitrile from

a

solution of 17. ¹H NMR

3

(400 MHz, CD₃CN): d [ppm]=5.12 (d, J_HH=1.3 Hz, 2H, 4/5-H), 5.03

3

(t, J_HH=1.3 Hz, 1H, 2-H), 3.80 (s, 12H, NCH₃), 2.26 (s, 12H, CCH₃),

1.76 (s, 15H, Cp*). ¹³C{¹H} NMR (100 MHz, CD₃CN): d [ppm]=140.1

(C-6/7), 128.5 (CCH₃), 90.5 (Cp*-CCH₃), 77.6 (C-4/5), 76.9 (C-2), 75.6

(C-1/3), 34.7 (NCH₃), 11.1 (Cp*-CCH₃), 9.2 (CCH₃). ¹⁹F{¹H} NMR

(376 MHz, CD₃CN): d [ppm]=ꢀ72.9 (d, ¹J_PF=707 Hz, PF₆). ³¹P{¹H}

1

NMR (162 MHz, CD₃CN): d [ppm]=ꢀ144.6 (spt, J_PF=707 Hz, PF₆).

3

¹H NMR (400 MHz, CD₂Cl₂): d [ppm]=5.36 (t, J_HH=1.3 Hz, 1H, 2-H),

3

¹⁹F{¹H} NMR (376 MHz, CDCl₃): d [ppm]=ꢀ72.9 (d, ¹J_PF=713 Hz,

5.12 (d, J_HH=1.3 Hz, 2H, 4/5-H), 3.89 (s, 12H, NCH₃), 2.30 (s, 12H,

1

PF₆). ³¹P{¹H} NMR (162 MHz, CDCl₃): d [ppm]=ꢀ144.6 (spt, J_PF

=

CCH₃), 1.79 (s, 15H, Cp*). CHN: calcd C 41.58, H 5.05, N 6.69 found

C 41.37, H 4.93, N 6.66. MS (ESI⁺, CH₃CN): m/z=274.1 [M]²⁺. MS

(ESI^ꢀ, CH₃CN): m/z=144.9 [PF₆]^ꢀ. UV/VIS (CH₃CN): l=280 nm (e=

2.2ꢃ10⁴Lmol^ꢀ1·cm^ꢀ1). IR (KBr): v˜ [cm^ꢀ1]=2964 (w), 2918 (w), 1650

(m), 1520 (m), 1458 (m), 1387 (w), 1299 (w), 1287 (w), 1238 (w),

1128 (w), 1070 (w), 840 (vs., PF₆^ꢀ), 709 (w), 557 (s, PF₆^ꢀ), 501 (w),

451 (w). Mp.: 2648C (dec.).

713 Hz, PF₆). ¹H NMR (400 MHz, [D₆]acetone): d [ppm]=5.90–5.85

(m, 5H, 2-H, 3-H, 4-H), 2.06 (s, 15H, Cp*), 1.82 (s, 6H, 7-H), 1.79 (s,

6H, 6-H), 1.32 (s, 3H, 5-H). MS (ESI⁺, CH₃CN): m/z=449.2 [M]⁺.

Crystal structure analyses: X-ray diffraction data were collected on

a Bruker APEX II DUO instrument equipped with an ImS microfocus

sealed tube and QUAZAR optics for MoK_a(l=0.71073 ꢂ) radiation.

Chem. Eur. J. 2020, 26, 1 – 16

www.chemeurj.org

13

ꢁ 2020 The Authors. Published by Wiley-VCH GmbH

&

These are not the final page numbers! ÞÞ

Full Paper

doi.org/10.1002/chem.202002801

Chemistry—A European Journal

[11] D. Kunz, E. Ø. Johnsen, B. Monsler, F. Rominger, Chem. Eur. J. 2008 , 14 ,

1090 9 –1 0914.

[12] D. Schmid, A. Seyboldt, D. Kunz, Z. Naturforsch. B 2014, 69, 580–588.

[13] F. Ramirez, S. Levy, J. Org. Chem. 1956 , 21 , 48 8 –4 89.

[14] R. Lin, H. Zhang, S. Li, J. Wang, H. Xia, Chem. Eur. J. 2011 , 17 , 422 3 –

4231.

[15] C. Xu, Z.-Q. Wang, Z. Li, W.-Z. Wang, X.-Q. Hao, W.-J. Fu, J.-F. Gong, B.-M.

Ji, M.-P. Song, Organometallics 2012 , 31 , 79 8 –8 01.

[16] D. Schmid, A. Seyboldt, D. Kunz, Z. anorg. allg. Chem. 2015 , 641 , 222 8 –

2232.

[17] D. Schmid, A. Seyboldt, K. Eichele, D. Kunz, Dalton Trans. 2016 , 45 ,

12028.

Corrections for absorption effects were applied using SADABS or

TWINABS. All structures were solved by direct methods using the

ShelXle^[54–56]software package for structure solution and refine-

ꢀ

ment. In some cases, especially for the PF₆ions, the Disordered

Structure Refinement (DSR) tool was used for disordered fragment-

s.^[57,58]In the case of structures 12 and 18 the SQEEZE routine was

applied.^[59]

CCDC 2001145 (1 ), 2001139 (5 ), 2001146 (6 ), 2001141 (7 ), 2001148

(9 ), 2001147 (12 ), 2001143 (13 ), 2001140 (16 ), 2001142 (17 ),

2001149 (18 ) and 2001144 (20 ) contain the supplementary crystal-

lographic data of this paper. These data are provided free of

charge by the joint Cambridge Crystallographic Data Centre and

Fachinformationszentrum Karlsruhe Access Structures service

www.ccdc.cam.ac.uk/structures.

[18] D. Kunz, C. Deißler, V. Gierz, F. Rominger, T. Oeser, Z. Naturforsch. B

2010, 65, 861–872.

[19] A. J. Arduengo, F. Davidson, H. V. R. Dias, J. R. Goerlich, D. Khasnis, W. J.

Marshall, T. K. Prakasha, J. Am. Chem. Soc. 1997 , 119 , 1274 2 –1 2749.

[20] N. Kuhn, T. Kratz, Synthesis 1993, 1993, 561–562.

[21] N. Kuhn, J. Fahl, R. Fawzi, C. Maichle-Mçßmer, M. Steimann, Z. Natur-

forsch. B 1998, 53, 720–726.

[22] N. Kuhn, A. Abu-Rayyan, M. Gçhner, M. Steimann, Z. anorg. allg. Chem.

2002 , 628 , 172 1 –1 723.

[23] A. Olejniczak, A. Katrusiak, J. Phys. Chem. B 2008 , 112 , 718 3 –7 190.

[24] L. Braun, G. Kehr, T. Blçmker, R. Frçhlich, G. Erker, Eur. J. Inorg. Chem.

2007, 140, 3083–3090.

[25] N. Kuhn, G. Weyers, D. Blꢄser, R. Boese, Z. Naturforsch. B 2001, 56,

1235–1237.

[26] V. Gierz, J. Melomedov, C. Fçrster, C. Deissler, F. Rominger, D. Kunz, K.

Heinze, Chem. Eur. J. 2012 , 18 , 1067 7 –1 0688.

[27] B. Raible, D. Kunz, Z. Naturforsch. B 2016, 71, 659–666.

[28] S. C. Sau, P. K. Hota, S. K. Mandal, M. Soleilhavoup, G. Bertrand, Chem.

Soc. Rev. 2020 , 49 , 123 3 –1 252.

DFT calculations: Performed based on density functional theory at

the BP86/def2-TZVP^[60–65]level implemented in Turbomole.^[66–76]The

RI-approximation^[77–82]was used all over. Geometry optimization

and natural bond analysis were performed using the Conductor-

like Screening Model (COSMO)^[83,84]with the permittivity of acetoni-

trile (e=37.5). Minimum structures were verified at the BP86/def2-

TZVP level by calculating the Hessian matrix and ensuring that it

has no imaginary frequency. Despite numerous attempts, for exam-

ple, including the Grimme dispersion correction,^[85,86]to optimize

the geometry of 12 and 13, a negative frequency of ꢀ9.40 cm^ꢀ1

(12) and ꢀ17.33 cm^ꢀ1(13) remained. The negative vibrational

mode involves the rotation around the N-R bonds. The cartesian

coordinates of the geometry optimized structures are available as

xyz-file in the Supporting Information.

[29] D. V. Muratov, A. S. Romanov, A. R. Kudinov, Russ. Chem. Bull. 2014 , 63 ,

248 5 –2 492.

[30] R. Gleiter, R. Seeger, Helv. Chim. Acta 1971 , 54 , 121 7 –1 220.

[31] A. Z. Kreindlin, K. Y. Suponitsky, F. M. Dolgushin, Z. A. Starikova, M. I. Ry-

binskaya, Russ. Chem. Bull. 2001 , 50 , 161 6 –1 619.

Acknowledgements

[32] J. W. Hachgenei, R. J. Angelici, J. Organomet. Chem. 1988 , 355 , 35 9 –3 78.

[33] R. A. Zelonka, M. C. Baird, Can. J. Chem. 1972 , 50 , 306 3 –3 072.

[34] J. L. Schrenk, A. M. McNair, F. B. McCormick, K. R. Mann, Inorg. Chem.

1986 , 25 , 350 1 –3 504.

[35] I. U. Khand, P. L. Pauson, W. E. Watts, J. Chem. Soc. C 1969 , 202 4 –2 030.

[36] J. F. Helling, D. M. Braitsch, J. Am. Chem. Soc. 1970 , 92 , 720 7 –7 209.

[37] C. White, P. M. Maitlis, J. Chem. Soc. A 1971 , 332 2 –3 326.

[38] M. G. Hill, W. M. Lamanna, K. R. Mann, Inorg. Chem. 1991 , 30 , 468 7 –

4690.

We thank Kim S. Flaig, Ronja Jordan and Theo Maulbetsch for

crystal mounting and Cꢄcilia Maichle-Mçssmer for help with

the X-ray structure analysis. We also thank Felix Gauder, Taulant

Dema, Victoria Graf and Nicolꢅ Abis for contributing to the

synthesis of some of the compounds. Open access funding en-

abled and organized by Projekt DEAL.

[39] J. C. Swarts, A. Nafady, J. H. Roudebush, S. Trupia, W. E. Geiger, Inorg.

Chem. 2009 , 48 , 215 6 –2 165.

[40] W. L. F. Armarego, C. L. L. Chai, Purification of laboratory chemicals, Elsev-

ier/BH, Oxford, 2009.

[41] J. M. Manriquez, P. J. Fagan, L. D. Schertz, T. J. Marks, J. Bercaw, N.

Mcgrady, Reagents for transition metal complex and organometallic syn-

theses, Inorganic Syntheses, Vol. 28, Ed. R. J. Angelici, Wiley, New York,

1990, pp. 317–320.

[42] N. Oshima, H. Suzuki, Y. Moro-Oka, Chem. Lett. 1984 , 13 , 116 1 –1 164.

[43] B. Steinmetz, W. A. Schenk, Organometallics 1999 , 18 , 94 3 –9 46.

[44] A. J. Nielson, C. E. F. Rickard, J. M. Smith, B. N. Diel, Inorganic Syntheses,

Vol. 24, Ed. J. M. Shreeve, John Wiley & Sons, Inc, Hoboken, NJ, USA,

1986, pp. 97–99.

Conflict of interest

The authors declare no conflict of interest.

Keywords: cyclopentadienides

imidazoliums · ruthenocenes · ylides

·

electron-poor ligands

·

[1] P. Jutzi, J. Organomet. Chem. 1990 , 400 , 1 –1 7.

[2] P. Jutzi, N. Burford, Chem. Rev. 1999 , 99 , 96 9 –9 90.

[3] R. H. Crabtree, The Organometallic Chemistry of the Transition Metals,

John Wiley & Sons, Inc, Hoboken, NJ, USA, 2014.

[4] M. P. Thornberry, C. Slebodnick, P. A. Deck, F. R. Fronczek, Organometal-

lics 2001 , 20 , 92 0 –9 26.

[5] D. L. Lichtenberger, Y. Elkadi, N. E. Gruhn, R. P. Hughes, O. J. Curnow, X.

Zheng, Organometallics 1997 , 16 , 520 9 –5 217.

[45] V. Gierz, C. Maichle-Mçssmer, D. Kunz, Organometallics 2012 , 31 , 73 9 –

747.

[46] H. C. Strauch, PhD thesis, Universitꢄt Mꢀnster, 1999.

[47] A. G. Massey, A. J. Park, J. Organomet. Chem. 1964 , 2 , 24 5 –2 50.

[48] J. L. Pohlmann, F. E. Brinckmann, Z. Naturforsch. B 1965, 20, 5–11.

[49] M. Kuprat, M. Lehmann, A. Schulz, A. Villinger, Organometallics 2010 ,

29 , 142 1 –1 427.

[6] P. G. Gassman, C. H. Winter, J. Am. Chem. Soc. 1986 , 108 , 422 8 –4 229.

[7] J. Brownie, M. Baird, Coord. Chem. Rev. 2008 , 252 , 173 4 –1 754.

[50] S. Dꢀmmling, E. Eichhorn, S. Schneider, B. Speiser, M. Wꢀrde, Curr. Sep.

1996, 15, 53–56.

ˇ

˚ˇ ˇ

ˇ

[8] I. Honzꢆckovꢇ, J. Vinklꢇrek, C. C. Rom¼o, Z. Ruzickovꢇ, J. Honzꢆcek, New

J. Chem. 2016, 40, 245–256.

[51] B. Gollas, B. Krauß, B. Speiser, H. Stahl, Curr. Sep. 1994, 13, 42–44.

[52] J. B. G. Gluyas, N. J. Brown, J. D. Farmer, P. J. Low, Aust. J. Chem. 2017 ,

70 , 113.

[9] D. Kunz, F. Mazzotta, Advances in Organometallic Chemistry, Vol. 70, Ed.

P. J. Perez, Elsevier Academic Press, Cambridge, MA, 2018, pp. 181–231.

[10] U. Mueller-Westerhoff, Tetrahedron Lett. 1972 , 13 , 463 9 –4 642.

&

Chem. Eur. J. 2020, 26, 1 – 16

www.chemeurj.org

14

ꢁ 2020 The Authors. Published by Wiley-VCH GmbH

ÝÝ These are not the final page numbers!

Full Paper

doi.org/10.1002/chem.202002801

Chemistry—A European Journal

[53] A. R. Kudinov, M. I. Rybinskaya, Y. T. Struchkov, A. I. Yanovskii, P. V. Pet-

rovskii, J. Organomet. Chem. 1987 , 336 , 18 7 –1 97.

[74] M. K. Armbruster, F. Weigend, C. van Wꢀllen, W. Klopper, Phys. Chem.

Chem. Phys. 2008 , 10 , 174 8 –1 756.

[54] C. B. Hꢀbschle, G. M. Sheldrick, B. Dittrich, J. Appl. Crystallogr. 2011 , 44 ,

128 1 –1 284.

[75] D. Peng, N. Middendorf, F. Weigend, M. Reiher, J. Chem. Phys. 2013 , 138 ,

184105.

[55] G. M. Sheldrick, Acta Crystallogr. Sect. A 2013 , 69 , s74.

[56] G. M. Sheldrick, Acta Crystallogr. Sect. C 2015 , 71 , 3 –8.

[57] D. Kratzert, J. J. Holstein, I. Krossing, J. Appl. Crystallogr. 2015 , 48 , 93 3 –

938.

[76] A. E. Reed, R. B. Weinstock, F. Weinhold, J. Chem. Phys. 1985 , 83 , 73 5 –

746.

[77] K. Eichkorn, O. Treutler, H. ꢈhm, M. Hꢄser, R. Ahlrichs, Chem. Phys. Lett.

1995 , 240 , 28 3 –2 90.

[58] D. Kratzert, I. Krossing, J. Appl. Crystallogr. 2018 , 51 , 92 8 –9 34.

[59] A. L. Spek, Acta Crystallogr. Sect. C 2015 , 71 , 9 –1 8.

[60] A. D. Becke, Phys. Rev. A 1988 , 38 , 309 8 –3 100.

[78] K. Eichkorn, O. Treutler, H. ꢈhm, M. Hꢄser, R. Ahlrichs, Chem. Phys. Lett.

1995 , 242 , 65 2 –6 60.

[79] P. Deglmann, K. May, F. Furche, R. Ahlrichs, Chem. Phys. Lett. 2004 , 384 ,

10 3 –1 07.

[61] J. P. Perdew, Phys. Rev. B 1986 , 33 , 882 2 –8 824.

[62] A. Schꢄfer, H. Horn, R. Ahlrichs, J. Chem. Phys. 1992 , 97 , 257 1 –2 577.

[63] A. Schꢄfer, C. Huber, R. Ahlrichs, J. Chem. Phys. 1994 , 100 , 582 9 –5 835.

[64] F. Weigend, R. Ahlrichs, Phys. Chem. Chem. Phys. 2005 , 7 , 329 7 –3 305.

[65] F. Weigend, Phys. Chem. Chem. Phys. 2006 , 8 , 105 7 –1 065.

[66] C. Steffen, K. Thomas, U. Huniar, A. Hellweg, O. Rubner, A. Schroer, J.

Comput. Chem. 2010, 31, 2967–2970.

[67] TURBOMOLE V6.3.1 2011: a development of University of Karlsruhe and

Forschungszentrum Karlsruhe GmbH, 1989–2007, TURBOMOLE GmbH,

since 2007; available from http://www.turbomole.com.

[68] O. Treutler, R. Ahlrichs, J. Chem. Phys. 1995 , 102 , 34 6 –3 54.

[69] M. von Arnim, R. Ahlrichs, J. Comput. Chem. 1998 , 19 , 174 6 –1 757.

[70] C. van Wꢀllen, J. Comput. Chem. 2011 , 32 , 119 5 –1 201.

[71] P. Deglmann, F. Furche, R. Ahlrichs, Chem. Phys. Lett. 2002 , 362 , 51 1 –

518.

[80] F. Weigend, Phys. Chem. Chem. Phys. 2002 , 4 , 428 5 –4 291.

[81] M. Sierka, A. Hogekamp, R. Ahlrichs, J. Chem. Phys. 2003 , 118 , 913 6 –

9148.

[82] K. Eichkorn, F. Weigend, O. Treutler, R. Ahlrichs, Theor. Chem. Acc. 1997 ,

97 , 11 9 –1 24.

[83] A. Klamt, G. Schꢀꢀrmann, J. Chem. Soc. Perkin Trans. 2 1993 , 79 9 –8 05.

[84] A. Klamt, V. Jonas, J. Chem. Phys. 1996 , 105 , 997 2 –9 981.

[85] S. Grimme, J. Antony, S. Ehrlich, H. Krieg, J. Chem. Phys. 2010 , 132 ,

154104.

[86] S. Grimme, S. Ehrlich, L. Goerigk, J. Comput. Chem. 2011 , 32 , 145 6 –

1465.

[72] P. Deglmann, F. Furche, J. Chem. Phys. 2002 , 117 , 953 5 –9 538.

[73] R. Ahlrichs, M. Bꢄr, M. Hꢄser, H. Horn, C. Kçlmel, Chem. Phys. Lett. 1989 ,

162 , 16 5 –1 69.

Manuscript received: June 10, 2020

Version of record online: && &&, 0000

Chem. Eur. J. 2020, 26, 1 – 16

www.chemeurj.org

15

ꢁ 2020 The Authors. Published by Wiley-VCH GmbH

&

These are not the final page numbers! ÞÞ

Full Paper

doi.org/10.1002/chem.202002801

Chemistry—A European Journal

FULL PAPER

&

Organometallic Chemistry

F. Mazzotta, G. Zitzer, B. Speiser, D. Kunz*

&& – &&

Electron-Deficient Imidazolium

Substituted Cp Ligands and their Ru

Complexes

Ligand design and application: The

synthesis of electron-poor imidazolium

substituted Cp-ylides is presented. Their

electronic properties are elucidated

based on NMR spectroscopy, DFT and

reactivity studies. The first h⁵-Cp coordi-

nation of electron-poor cationic ylides is

confirmed by X-ray structure analyses

and an equilibrium for the coordination

of a Cp ligand is observed. Electrochem-

ical investigations of the ligands and of

the ruthenocenes quantify the decreas-

ing donor properties with increasing

number of imidazolium substituents.

&

Chem. Eur. J. 2020, 26, 1 – 16

www.chemeurj.org

16

ꢁ 2020 The Authors. Published by Wiley-VCH GmbH

ÝÝ These are not the final page numbers!

Article Doi

Electron-Deficient Imidazolium Substituted Cp Ligands and their Ru Complexes

DOI: 10.1002/chem.202002801

Source and publish data:

Authors:

Article abstract of DOI:10.1002/chem.202002801

Full text of DOI:10.1002/chem.202002801

Products guided by the article

R&D Labs maybe for 16733-97-4

Relevant to this article

Hot Product