C-C bond formation by cyanide addition to dinuclear vinyliminium complexes

Journal of Organometallic Chemistry 691 (2006) 4234–4243

www.elsevier.com/locate/jorganchem

C–C bond formation by cyanide addition to dinuclear

vinyliminium complexes

b

a

b

Vincenzo G. Albano , Luigi Busetto , Fabio Marchetti , Magda Monari ,

a

a,

*

Stefano Zacchini , Valerio Zanotti

a

`

Dipartimento di Chimica Fisica e Inorganica, Universita di Bologna, Viale Risorgimento 4, 40136 Bologna, Italy

b

`

Dipartimento di Chimica ‘‘G. Ciamician’’, Universita di Bologna, Via Selmi 2, 40126 Bologna, Italy

Received 23 May 2006; received in revised form 12 June 2006; accepted 12 June 2006

Available online 4 July 2006

Abstract

The diiron vinyliminium complexes [Fe₂{l-g¹:g³-C(R⁰)@C(H)C@N(Me)(R)}(l-CO)(CO)(Cp)₂][SO₃CF₃] (R=Me, R⁰= SiMe₃(1a);

R = Me, R⁰= CH₂OH (1b); R = CH₂Ph, R⁰= Tol (1c), Tol = 4-MeC₆H₄; R = CH₂Ph, R⁰= COOMe (1d); R = CH₂Ph, R⁰= SiMe₃

(1e)) undergo regio- and stereo-selective addition by cyanide ion (from NBuⁿ₄CN), aﬀording the corresponding bridging cyano-function-

alized allylidene compounds [Fe₂{l-g¹:g³-C(R⁰)C(H)C(CN)N(Me)(R)}(l-CO)(CO)(Cp)₂] (3a–e), in good yields. Similarly, the diiron

vinyliminium complexes [Fe₂{l-g¹:g³-C(R⁰)@C(R⁰)C@N(Me)(R)}(l-CO)(CO)(Cp)₂][SO₃CF₃] (R = R⁰= Me (2a); R = Me, R⁰= Ph

(2b); R = CH₂Ph, R⁰= Me (2c); R = CH₂Ph, R⁰= COOMe (2d)) react with cyanide and yield [Fe₂{l-g¹:g³-C(R⁰)C(R⁰)C(CN)N-

(Me)(R)}(l-CO)(CO)(Cp)₂] (9a–d). The reactions of the vinyliminium complex [Fe₂{l-g¹:g³-C(Tol)@CHC@N(Me)(4-C₆H₄CF₃)}-

(l-CO)(CO)(Cp)₂][SO₃CF₃] (4) with NaBH₄and NBu₄ⁿCN aﬀord the allylidene [Fe₂{l-C(Tol)C(H)@C(H)N(Me)(C₆H₄CF₃)}-

(l-CO)(CO)(Cp)₂] (5) and the cyanoallylidene [Fe₂{l-C(Tol)C(H)@C(CN)N(Me)(C₆H₄CF₃)}(l-CO)(CO)(Cp)₂] (6), respectively.

Analogously, the diruthenium vinyliminium complex [Ru₂{l-g¹:g³-C(SiMe₃)@CHC@N(Me)(CH₂Ph)}(l-CO)(CO)(Cp)₂][SO₃CF₃] (7)

reacts with NBuⁿCN to give [Ru₂{l-g¹:g³-C(SiMe₃)CHC(CN)N(Me)(CH₂Ph)}(l-CO)(CO)(Cp)₂] (8).

4

Finally, cyanide addition to [Fe₂{l-g¹:g³-C(COOMe)@C(COOMe)C@N(Me)(Xyl)}(l-CO)(CO)(Cp)₂][SO₃CF₃] (2e) (Xyl = 2,6-

Me₂C₆H₃), yields the cyano-functionalized bis-alkylidene complex [Fe₂{l-g¹:g²-C(COOMe)C(COOMe)(CN)CN(Me)(Xyl)}(l-CO)-

(CO)(Cp)₂] (10). The molecular structures of 3a and 9a have been elucidated by X-ray diﬀraction.

Keywords: Vinyliminium; Diiron complexes; Allylidene; Vinylalkylidene; Cyanide addition

1. Introduction

carbyne complexes [Fe₂{l-C@N(Me)(R)}(l-CO)(CO)-

(NCMe)(Cp)₂] [SO₃CF₃] [2].

Di- and polynuclear metal complexes are subject of

interest since the presence of bridging ligands and cooper-

ation among close metal centres is a potential source of

unique reactivity [1]. Our interest in the area has concerned

the chemistry of the diiron bridging vinyliminium com-

plexes 1 and 2 (Chart 1). These latter have been obtained

by insertion of mono and di-substituted alkynes, respec-

tively, in the metal–carbyne bond of the parent l-amino-

The bridging vinyliminium ligands display a remarkable

electrophilic character and have been shown to react with

NaBH₄undergoing hydride addition [3]. The addition

occurs selectively at the iminium carbon (C_a), except for

the complexes (1f–h) in which bulky Xyl group

(Xyl = 2,6-Me₂C₆H₃) was found to inhibit hydride attack

at C_aand direct the addition to the less hindered C_b[3a].

Attempts to extend our studies to the reactions with carbon

nucleophiles (i.e. lithium organyls) have evidenced a diﬀer-

ent reactivity consisting in the deprotonation of the C_b–H of

the vinyliminium ligand. The deprotonated intermediates

*

Corresponding author.

E-mail address: valerio.zanotti@unibo.it (V. Zanotti).

doi:10.1016/j.jorganchem.2006.06.033

V.G. Albano et al. / Journal of Organometallic Chemistry 691 (2006) 4234–4243

4235

complexes 3a–e has been determined by elemental analyses,

spectroscopy and X-ray diﬀraction for 3a. Two crystallo-

graphically independent but chemically equivalent mole-

cules have been found in the crystal (see Section 4). The

ORTEP molecular diagram of one of them is shown in

Fig. 1. Relevant bond lengths and angles are reported in

Table 1. It is worth noting that all the corresponding bond

distances in the two molecules are equal within experimen-

tal errors. Minor diﬀerences are observable in the confor-

mation of the ancillary groups. The molecular

stereogeometry of 3a is consistent with that of the parent

cation 1a, in which the C_a–C_b–C_cunit of the vinyliminium

ligand is bonded to the Fe₂unit in the non-conventional

mode shown in Scheme 1. The information conveyed by

the X-ray structure is primarily that CN^ꢀattacks the more

electrophilic C_acarbon [C(5)]. The generation of a C–C

bond turns the vinyliminium ligand into an allylidene

derivative, as one can infer from the double bond character

and the substantial equivalence of the C_a–C_b–C_cdistances

+

R'

H

R

Me

R

R'

C_β

N

C_γ

C_α

Me

CO

Fe

C

O

C

O

R

Me

R’

SiMe₃

CH₂OH 1b

1c

CH₂Ph COOMe 1d

CH₂Ph SiMe₃1e

COOMe 1f

R

Me

R’

Me

Ph

1a

2a

2b

2c

CH₂Ph Tol

CH₂Ph Me

CH₂Ph COOMe 2d

Xyl

COOMe 2e

Me

Et

Ph

Xyl

2f

2g

2h

Tol

SiMe₃

1g

1h

Chart 1.

˚

have been shown to rearrange aﬀording new mono and

polynuclear complexes [4]. On the other hand, type 2 com-

plexes, which do not contain acidic C_b–H hydrogens, have

been found to react with LiR undergoing nucleophilic addi-

tion at the C₅H₅ligand, rather than at the bridging vinyli-

minium. Initial attack is followed by an intramolecular

rearrangement, which involves hydride migration from the

cyclopentadiene ring (C₅H₅R) to the vinyliminium ligand

aﬀording l-vinylcarbene complexes [5].

Herein, we report on the reactions of 1–2, including a

similar diruthenium complex [6] with cyanide, in order to

determine to which extent the reactivity of the bridging

ligand is inﬂuenced by steric and electronic features of

the substituents (R and R⁰). The cyanide is a ‘stabilized

carbanion’ and, because of its lower basicity compared to

other carbanions, is the ideal candidate to give nucleophilic

addition at the bridging vinyliminium ligand, in spite of the

acidic character of C_b–H.

in both molecules [C(5)–C(4) 1.447 and 1.449(3) A; C(4)–

˚

C(3) 1.414 and 1.419(3) A]. Also the coordination interac-

tions Fe–(C_a–C_b–C_c) exhibit minor diﬀerences [Fe(1)–

˚

C(5) 2.068 and 2.077(2) A; Fe(1)–C(4) 2.012 and

˚

2.007(2) A; Fe(1)–C(3) 2.045 and 2.037(2) A].

H

N

R

H

+

R'

C_β

C

N

C_β

C_γ

C_α

C_γ

Me

R

CO

N

NBu₄CN

Fe

Me

C

O

C

O

R

Me

R’

SiMe₃

1a

3a

1b

1c

1d

1e

Me

CH₂OH

Tol

COOMe

SiMe₃

3b

3c

3d

3e

CH₂Ph

A further reason to investigate the functionalization of

the bridging ligand by cyanide is the high interest of the

cyanocarbon transition metal chemistry, due to various

potential applications in the ﬁeld of new materials and

organic synthesis [7].

Scheme 1.

2. Results and discussion

2.1. Cyanide addition to vinyliminium complexes containing

acidic C_b–H hydrogen

Thevinyliminiumcomplexes[Fe₂{l-g¹:g³-C(R⁰)@CHC@

N(Me)(R)}(l-CO)(CO)(Cp)₂][SO₃CF₃] (1a–e), containing

the acidic C_b–H hydrogen, react with (NBuⁿ₄ÞðCNÞ, in

CH₂Cl₂solution, aﬀording the complexes 3a–e, in about

80–90% yields (Scheme 1).

It is worth noting that cyanide exclusively attacks the

iminium carbon, rather than following other reaction

modes, including C_b–H proton removal, CO displacement

or addition at other sites of the l-ligand. The nature of

Fig. 1. ORTEP drawing (ellipsoids at 30% probability) of one of the two

conformers of 3a. Only the allylic C_bhydrogen is shown.

4236

V.G. Albano et al. / Journal of Organometallic Chemistry 691 (2006) 4234–4243

Table 1

tion of the methyl and benzyl groups around the C_a–N

bond, as just discussed for the NMR non-equivalence of

the NMe₂methyl groups in 3a–b. NOE experiments on

the major isomer of 3e suggest that the most favourable

geometry is that with the methyl and benzyl groups point-

ing toward the bridging and terminal CO, respectively

(Chart 2). Indeed, irradiation of the NMe resonance of

the major isomer (1.63 ppm) resulted in signiﬁcant

enhancement of one Cp resonance (5.03 ppm), whereas

irradiation of the NMe resonance of the secondary isomer

(2.43 ppm) gave no eﬀect on the Cp resonances. Moreover,

regarding the minor isomer, NOE eﬀect was detected on

the CH₂benzyl protons by irradiating the Cp resonance

at 4.97 ppm.

The reactions described in Scheme 1 parallel the corre-

sponding addition of NaBH₄[3a]. Indeed, complexes

1a–e undergo hydride attack at the C_aaﬀording the corre-

sponding l-allylidene complexes [Fe₂{l-g¹:g³–C(R⁰)CH-

C(H)N(Me)(R)}(l-CO)(CO)(Cp)₂] (R = Me, CH₂Ph),

similar to 3a–e. In both cases nucleophilic attack occurs

exclusively at the iminium carbon and is stereospeciﬁc.

However, the orientation of the entering groups is diﬀerent:

hydride is found anti with respect to the C_b–H, whereas

cyanide is syn. The conformational freedom of the NMe(R)

substituent is consequently diﬀerent in the two cases, with

hindered rotation observed only in the complexes obtained

by CN^ꢀaddition.

˚

Selected bond lengths (A) and angles (ꢁ) for 3a and 9a

3a^a

9a

Fe(1)–C(1)

Fe(1)–C(3)

Fe(1)–C(4)

Fe(1)–C(5)

Fe(2)–C(1)

Fe(2)–C(2)

Fe(2)–C(3)

Fe(1)–Fe(2)

C(1)–O(1)

C(2)–O(2)

C(5)–N(1)

C(9)–N(1)

C(10)–N(1)

C(3)–C(4)

C(4)–C(5)

1.910(2) [1.909(2)]

2.045(2) [2.037(2)

2.012(2) [2.007(2)]

2.068(2) [2.077(3)

1.935(2) [1.932(3)]

1.742(2) [1.753(3)]

1.969(2) [1.973(2)]

2.5408(5) [2.5488(5)]

1.168(3) [1.176(3)]

1.154(3) [1.150(3)]

1.445(3) [1.431(3)]

1.470(3) [1.461(3)]

1.459(3) [1.463(3)]

1.414(3) [1.419(3)]

1.447(3) [1.449(3)]

1.465(4) [1.470(4)]

1.137(3) [1.133(4)]

1.883(2) [1.886(2)]

1.137(3) [1.133(4)]

2.082 [2.106]

1.914(2)

2.027(1)

2.037(1)

2.059(1)

1.927(1)

1.765(2)

1.975(2)

2.5327(3)

1.173(2)

1.143(2)

1.437(2)

1.469(2)

1.462(2)

1.418(2)

1.469(2)

1.467(2)

1.143(2)

C(5)–C(11)/C(5)-C(8)

C(11)–N(2)/C(8)-N(2)

C(3)–Si

–

N(2)–C(11)/N(2)-C(8)

Fe(1)–C(Cp)(av)^b

Fe(2)–C(Cp)(av)^b

Fe(1)–C(1)–O(1)

Fe(2)–C(1)–O(1)

C(3)–C(4)–C(5)

N(1)–C(5)–C(4)

C(5)–C(8)–N(2)

C(5)–C(11)–N(2)

a

1.143(2)

2.102

2.114

140.9(1)

136.1(1)

119.2(1)

120.0(1)

175.2(2)

–

2.138 [2.130]

141.3(2) [140.0(2)]

135.4(2) [136.2(2)]

125.5(2) [125.5(2)]

122.1(2) [122.5(2)]

–

177.2(3) [178.0(4)]

Values in parentheses refer to the second conformer in the asymmetric

unit.

It is worth noting that the reaction described above

points out the similarities between H^ꢀ(from NaBH₄)

and CN^ꢀ(from NBu₄CN) as far as nucleophilic additions

at bridging ligands in cationic diiron complexes [8] are

concerned.

b

Main image of the Cp ligand.

The spectroscopic data of 3a–e are in good agreement

with the solid state structure. Their infrared spectra, in

CH₂Cl₂solution, exhibit absorptions for the terminal and

bridging carbonyls (e.g. at 1969 and 1787 cm^ꢀ1for 3a)

and a m(CN) absorption at about 2190 cm^ꢀ1. The NMR

spectra of 3a–b, in CDCl₃solution, show the presence of

a single isomer. Since the CN^ꢀaddition at C_acould occur

either on the same side of C_b–H or in the opposite one, gen-

erating syn and anti isomers, respectively, we conclude that

the reaction is stereoselective and assume that the geometry

adopted in solution is that observed in the solid state with

By contrast with the reactions shown in Scheme 1, the

complexes

[Fe₂{l-g¹:g³-C(R⁰)@CHC@N(Me)(Xyl)}(l-

CO)(CO)(Cp)₂][SO₃CF₃] (R⁰= CO₂Me (1f); R⁰= Tol

(1g); R⁰= SiMe₃(1h)), characterized by the presence of

the bulkier Xyl (2,6-Me₂C₆H₃) group, react with cyanide

yielding alkynyl complexes or fragmentation of the dinu-

clear precursor (Scheme 2). The reaction products are those

previously observed by deprotonation of 1f–h by NaH [4].

Therefore, the presence of the Xyl group in 1f–h inhibits

the attack at the iminium carbon and favours the C_b–H

proton removal.

1

the CN group syn to the C_b–H. The H NMR spectra of

3a–b show two distinct resonances for the NMe₂protons

(e.g. at d 2.31 and 1.71 ppm for 3a). Their inequivalence

suggests that free rotation around the C_a–N bond is inhib-

ited for steric reasons, because the C_a–N interaction does

not display double bond character [C(5)–N(1) 1.445 and

1.431(3) A]. In fact, the geometry observed for 3a evidences

that the CH₃groups in NMe₂would bump into the termi-

nal CO ligand [C(2)–O(2)] upon rotation around the C(5)–

The result is unexpected and markedly diﬀerent from

what observed in the corresponding reactions with NaBH₄:

˚

H

Me₃Si

C

CN

C

˚

Fe

N(1) axis [N(1)ꢁ ꢁ ꢁC(2) contact 2.761(3) and 2.695(4) A].

Me

Bz

N

Cp

N

Cp

CO

Other NMR features include the typical resonances for

C_a, C_band C_cof the allylidene skeleton (e.g. for 3a at

67.6 ppm, 86.5 ppm, 194.5 ppm, respectively) and the reso-

nance of the cyanide carbon, at about 121 ppm.

CO

Fe

Bz

Me

C

O

minor isomer

major isomer

Chart 2. isomers due to orientations of the N-substituents for 3e.

Complexes 3c–e exist as a mixture of two isomers in

about 1:5 ratio. These are due to the lack of free orienta-

V.G. Albano et al. / Journal of Organometallic Chemistry 691 (2006) 4234–4243

4237

1f–h were found to undergo hydride addition at C_baﬀord-

C₆H₄CF₃

N

ing bis–alkylidene complexes of the type [Fe₂{l-g¹:g²-

C(R⁰)CH₂CN(Me)(Xyl)}(l-CO)(CO)(Cp)₂]. The reason

why CN^ꢀreacts removing the C_b–H hydrogen rather than

giving addition at the C_bcarbon remains unclear. The ste-

ric inﬂuence exerted by the Xyl group is quite evident, but

electronic eﬀects due to its electron-withdrawing character

should also be taken into account. For a better discrimina-

tion between electronic and steric eﬀects, the reactivity

of [Fe₂{l-g¹:g³-C(Tol)@C(H)C@N(Me)(4-C₆H₄–CF₃)}(l-

CO)(CO)(Cp)₂][SO₃CF₃] (4), in which a para-triﬂuoro

methyl benzene (R = 4-C₆H₄–CF₃) replaces the Xyl group,

has been investigated. The synthesis of 4, which is a novel

compound, is detailed in Section 4. The p-C₆H₄–CF₃group

is less sterically demanding than Xyl, but a better electron-

withdrawer. The reactions of 4 with NaBH₄and NBu^t₄CN

have shown that, in both cases, the nucleophilic attack

occurs selectively at C_a(Scheme 3), suggesting that the ste-

ric bulk of the N substituents is dominant. Compounds

[Fe₂{l-C(Tol)C(H)@C(H)N(Me)(C₆H₄CF₃)}(l-CO)(CO)-

(Cp)₂] (5) and [Fe₂{l-C(Tol)C(H)@C(CN)N(Me)(C₆H₄-

CF₃)}(l-CO)(CO)(Cp)₂] (6) have been characterized by

H

Tol

Fe

C_β

Me

C_γ

C_α

O

C

H

Fe

C

O

C₆H₄CF₃

H

5

+

Tol

Fe

C_β

N

C_γ

C_α

Me

NaBH₄

CO

Fe

C

O

NBu₄CN

4

H

NC

C_α

Tol

Fe

C_β

C_γ

Me

O

C

N

Fe

CF₃C₆H₄

C

O

6

Scheme 3.

1

IR and H, ¹³C and ¹⁹F NMR spectroscopy. Complex 5

CH₂Ph

H

N

H

is analogous to previously reported compounds [3a],

whereas complex 6 resembles 3a–e.

+

SiMe₃

CO

C_β

N

SiMe₃

C

C_β

C_γ

C_α

Me

C_γ

PhCH₂

C_α

NBu₄CN

We have recently shown that diruthenium vinyliminium

complexes analogous to the diiron species 1 and 2 can also

be obtained [6]. Their chemistry parallels that of their

diiron counterparts, however some minor diﬀerences,

regarding the geometry of the complexes and the reactivity

of the terminally coordinated ligands have been evidenced.

Thus, in order to evidence possible eﬀects due to the pres-

ence of ruthenium in the place of iron, the reaction of the

diruthenium complex 7 with NBuⁿ₄CN has been investi-

gated. The result is the quantitative formation of the

cyano-functionalized allylidene complex 8, analogous to

3e (Scheme 4).

CO

N

Ru

C

O

Me

C

O

7

8

Scheme 4.

trum, an intense resonance at 121.9 ppm (C_a–CN), and a

doublet at d 67.9 ppm (¹J_CC= 63.3 Hz, C_a), indicating that

addition occurs selectively at the iminium carbon (C_a).

2.2. Cyanide addition to disubstituted vinyliminium

complexes in the absence of C_b–H

Therefore, ruthenium does not produce any noticeable

diﬀerence in the reactivity of the bridging ligand. The spec-

troscopic data of 8 are similar to those of the diiron coun-

terpart 3e, although only one isomer was observed.

Moreover, the reaction has also been performed with

¹³C-enriched cyanide (by using a 10:1 mixture of NBu₄CN

and K¹³CN, 99% ¹³C, from Aldrich). The corresponding

reaction product 8 has evidenced, in the ¹³C NMR spec-

Compounds 2a–d rapidly react with a slight excess of

NBuⁿ₄CN, in CH₂Cl₂solution, aﬀording the bridging

cyano-functionalized allylidene complexes [Fe₂{l-g¹:g³-

C(R⁰)C(R⁰)C(CN)N(Me)(R)}(l-CO)(CO)(Cp)₂] (9a–d), in

about 80–90% yields (Scheme 5).

Xyl

+

H

Me

Xyl

Me

Fe

H

N

R'

N

C_β

R'

N

C

C_β

C

C_α

Me

C_α

Fe

C_γ

Fe

O

CO

CN^-

C

CN^-

C

C_γ

Fe

R'

Fe

C

O

R' = COOMe

1f-h

Scheme 2.

R' = SiMe₃, Tol

4238

R

V.G. Albano et al. / Journal of Organometallic Chemistry 691 (2006) 4234–4243

absence of acidic C_b–H. An exception among these unreac-

tive vinyliminium complexes is represented by [Fe₂{l-

g¹:g³-C(COOMe)@C(COOMe)C@N(Me)(Xyl)}(l-CO)-

(CO)(Cp)₂][SO₃CF₃] (2e). Indeed, the latter compound

reacts with NBuⁿ₄CN, in CH₂Cl₂solution, yielding the

cyano-functionalized bis-alkylidene complex 10 (Scheme 6).

Compound 10, fully characterized by spectroscopy and

elemental analysis, is the result of selective cyanide addition

at the C_bposition and consequent adjustment of the coor-

dination mode of the bridging ligand, which remains

connected to the Fe atoms through the C_aand C_ccarbons.

Both carbons exhibit carbene character: the former is an

amino carbene ligand, terminally coordinated, whereas

the latter is a bridging alkylidene. This bis-alkylidene coor-

dination mode appears very stable and has been previously

found upon hydride addition at C_b[3] or rearrangements

promoted by C_b–H proton removal [4] in l-vinyliminium

complexes.

R'

N

R'

+

R'

C_β

R'

N

C

C_β

C_γ

C_α

Me

C_γ

R

NBu₄CN

CO

N

Fe

Me

C

O

C

O

R

Me

CH₂Ph

R’

Me

Ph

2a

2b

2c

2d

9a

9b

9c

9d

Me

COOMe

Scheme 5.

The structure of 9a has been determined by X-ray dif-

fraction: the ORTEP molecular diagram is shown in

Fig. 2 and relevant bond lengths and angles are reported

in Table 1. The molecular geometry is similar to that of

3a discussed above. The presence of a methyl substituent

at C_b[C(4)–C(7)] has not inﬂuenced the conﬁguration

around C_a[C(5)] and the CN group is oriented syn with

respect to C_b-Me. The bond parameters of interest are

strictly equivalent to the corresponding ones in 3a. It is

worth noting that also the NMe₂group is oriented as in 3a.

The spectroscopic data for 9a–b are very similar to those

of 3a–e. Again, for 9a–b the NMR data indicate the pres-

ence of a single isomer and the non-equivalence of the

two N-bonded methyl groups, whereas two isomers are

observed in solution, for the complexes 9c–d.

The addition of cyanide at the iminium carbon of 2a–d is

in accord with the observation that, in the absence of steric

protection oﬀered by the Xyl group, nucleophilic attack

occurs selectively at the iminium carbon (C_a). Conversely,

the complexes [Fe₂{l-g¹:g³-C(R⁰)@C(R⁰)C@N(Me)(Xyl)}-

(l-CO)(CO)(Cp)₂][SO₃CF₃] (R⁰= Me (2f); R⁰= Et (2g);

R⁰= Ph (2h)), containing the Xyl substituent, do not react

with CN^ꢀ. In this case, the iminium carbon is not accessi-

ble, neither deprotonation can occur, because of the

An explanation of the higher reactivity of 2e, compared

to 2f–h, should be related to the electron-withdrawing sub-

stituents (COOMe) at both C_band C_c.

The infrared spectrum of 10 (in CH₂Cl₂solution) exhibits

the usual pattern consisting of two strong absorptions for

terminal and bridging carbonyls (1947, 1777 cm^ꢀ1), and, in

addition, absorptions attributable to C„N (at 2175 cm^ꢀ1

)

and to the carboxylate (at 1733 and 1683 cm^ꢀ1). The

NMR spectra show only one isomer, indicating that the

addition is stereo- and regio-selective; the ¹³C NMR spec-

trum exhibits the resonances relative to C_a, C_band C_cin

the typical regions for an aminocarbene (d 252.7 ppm), a

quaternary sp³carbon (81.5 ppm) and a bridging alkylidene

bound to a carboxylate (129.4 ppm), respectively.

Labelling experiments with ¹³C-enriched CN^ꢀconﬁrms

that cyanide addition occurs selectively at C_b: the ¹³C

NMR spectrum of the ¹³C labelled product 10 displays dou-

blets at d 129.4 ppm (²J_CC= 5.0 Hz, C_c) and 81.5 ppm

(¹J_CC= 67.2 Hz, C_b), and an intense singlet at 116.5 ppm

(C_b–CN).

Since the C_aN(Me)(Xyl) moiety in 10 shows aminocar-

bene character, free rotation around the C_a–N bond is not

allowed, due to its partial double bond character. Two iso-

mers could consequently arise, according to the orienta-

tions of the Me and Xyl substituents. Since only one

isomer is observed, we assume that this is the one with

the NMe group pointing toward the C_bsubstituent. This

conformation is suggested by NOE experiments, which

CO₂Me

N

Me

MeO₂C

+

C

CO₂Me

C_β

N

Me

C_β

Fe

CO₂Me

C_γ

C_α

O

C

Xyl

N

C_α

C_γ

CO

Fe

Xyl

Fe

C

O

NBu₄CN

C

O

2e

10

Fig. 2. ORTEP drawing (ellipsoids at 30% probability) of 9a. Hydrogen

atoms have been omitted for clarity.

Scheme 6.

V.G. Albano et al. / Journal of Organometallic Chemistry 691 (2006) 4234–4243

4239

evidence enhancement only of one methyl group of the Xyl

4.2. Synthesis of [Fe₂{l-g¹:g³-C_c(R⁰)C_b(H)C_a

(CN)N(Me)(R)}(l-CO)(CO)(Cp)₂] (R = Me,

R⁰= SiMe₃(3a); R = Me, R⁰= CH₂OH (3b);

R = CH₂Ph, R⁰= Tol (3c); R = CH₂Ph, R⁰= COOMe

(3d); R = CH₂Ph, R⁰= SiMe₃(3e))

substituent (at 2.12 ppm) upon irradiation of the Cp

resonance at 4.19 ppm. A similar geometry has been previ-

ously observed in the X-ray structure of the related

bis-alkylidene complex [Fe₂{l-g¹:g²-C(Et)C(Et)(H)CN-

(Me)(Xyl)}(l-CO)(CO)(Cp)₂] [3b].

Although no structural investigation has been carried

out on complex 10, it is likely that the addition of CN^ꢀ,

which generates a stereogenic centre at C_b, is completely

stereoselective and occurs in the side opposite to the Cp

ligands, as previously observed for the addition of hydride

to similar complexes [3].

A solution of 1a (95 mg, 0.158 mmol), in CH₂Cl₂

(10 mL), was treated with NBuⁿ₄CN (64 mg, 0.239 mmol).

After stirring at room temperature for 20 min, the mixture

was ﬁltered on alumina. Solvent removal, under reduced

pressure, aﬀorded 3a as a brown microcrystalline powder.

Yield: 64 mg (91%). Crystals suitable for X-ray analysis

were collected by layering a CH₂Cl₂solution of 3a with

petroleum ether, at ꢀ20 ꢁC. Anal. Calc. for C₂₁H₂₆Fe₂-

N₂O₂Si: C, 52.74; H, 5.48; N, 5.86. Found: C, 52.80; H,

5.44; N, 5.98%. IR (CH₂Cl₂) m(C„N) 2188 (m), m(CO)

3. Conclusions

Diiron vinyliminium complexes undergo regio- and ste-

reo-speciﬁc additions by cyanide ion, under mild condi-

tions. The addition occurs normally at the iminium

carbon (C_a), aﬀording novel bridging cyano-functionalized

allylidene complexes, and is not inﬂuenced by the nature of

the substituent at C_b, neither by the presence of ruthenium

in place of iron. Conversely, in the presence of the xylyl

substituent at the iminium nitrogen, the CN^ꢀaddition to

C_ais inhibited by steric reasons. Diﬀerent outcomes are

consequently observed: in the presence of C_b–H hydrogen,

the cyanide ion promotes deprotonation; otherwise, the

vinyliminium is simply unreactive, unless activated by the

presence of electron-withdrawing substituents. In the latter

case, CN^ꢀaddition takes place selectively at C_b, aﬀording a

cyano-functionalized bis-alkylidene complex.

1969 (vs), 1787 (s) cm^ꢀ^{1 1}H NMR (CDCl₃) 4.94, 4.49

.

(s, 10H, Cp); 4.74 (s, 1H, C_bH); 2.32, 1.71 (s, 6H,

NMe); 0.63 (s, 9H, SiMe₃). ¹³C{¹H} NMR (CDCl₃) d

264.9 (l-CO); 212.2 (CO); 194.5 (C_c); 121.1 (CN); 87.4,

84.6 (Cp); 86.5 (C_b); 67.6 (C_a); 49.8, 42.2 (NMe); 3.6

(SiMe₃).

Compounds 3b–e were obtained by the same procedure

described for 3a, by reacting NBu^t₄CN with 1b–e,

respectively.

3b: Yield (81%). Anal. Calc. for C₁₉H₂₀Fe₂N₂O₃: C,

52.33; H, 4.62; N, 6.42. Found: C, 52.27; H, 4.71; N,

6.33%. IR (CH₂Cl₂) m(C„N) 2190 (m), m(CO) 1967 (vs),

1

1790 (s) cm^ꢀ1. H NMR (CDCl₃) 5.94 (m, 2H, CH₂OH);

5.05 (s, 1H, C_bH); 4.89, 4.49 (s, 10H, Cp); 2.29, 1.73 (s,

6H, NMe). ¹³C{¹H} NMR (CDCl₃) d 264.2 (l-CO);

212.8 (CO); 200.3 (C_c); 121.0 (CN); 87.4, 85.6 (Cp); 49.3,

44.1 (NMe).

4. Experimental details

4.1. General

3c: Yield (78%). Anal. Calcd. for C₃₁H₂₈Fe₂N₂O₂: C,

65.06; H, 4.93; N, 4.90. Found: C, 65.05; H, 4.97; N,

4.81%. IR (CH₂Cl₂)m(C„N) 2189 (m), m(CO) 1972 (vs),

All reactions were carried out routinely under nitrogen,

using standard Schlenk techniques. Solvents were distilled

immediately before use under nitrogen from appropriate

drying agents. Chromatography separations were carried

out on columns of deactivated alumina (4% w/w water).

Glassware was oven-dried before use. Infrared spectra were

recorded on a Perkin–Elmer Spectrum 2000 FT-IR spec-

trophotometer and elemental analyses were performed on

a ThermoQuest Flash 1112 Series EA Instrument. All

NMR measurements were performed on Varian Gemini

300 and Mercury Plus 400 instruments. The chemical shifts

for ¹H and ¹³C were referenced to internal TMS. The spec-

1784 (s) cm^ꢀ^{1 1}H NMR (CDCl₃) 7.55–7.18 (m, 9H,

.

C₆H₄Me and Ph); 4.73, 4.71, 4.65, 4.62 (s, 10H, Cp);

2

4.71, 3.29 (d, 2H, J_HH= 13 Hz, CH₂Ph); 4.47 (s, 1H,

C_bH); 2.47, 2.43 (s, 3H, C₆H₄Me); 2.20, 1.70 (s, 3H,

NMe); isomer ratio 7:1. ¹³C{¹H} NMR (CDCl₃) d 264.6

(l-CO); 213.3, 212.0 (CO); 200.9 (C_c); 156.1 (ipso-

C₆H₄Me); 135.4–126.8 (C₆H₄Me and Ph); 121.1, 121.0

(CN); 89.4, 86.3, 86.1 (Cp); 81.6 (C_b); 65.7 (C_a); 64.8,

59.8 (CH₂Ph); 46.4, 39.8 (NMe); 21.1 (C₆H₄Me).

3d: Yield (82%). Anal. Calc. for C₂₆H₂₄Fe₂N₂O₄: C,

57.81; H, 4.48; N, 5.19. Found: C, 57.93; H, 4.50; N,

5.23%. IR (CH₂Cl₂) m(C„N) 2191 (w), m(CO) 1982 (vs),

tra were fully assigned via DEPT experiments and H,¹³C

1

correlation through gs-HSQC and gs-HMBC experiments

[9]. All NMR spectra were recorded at 298 K; NMR sig-

nals due to a second isomeric form (where it has been pos-

sible to detect and/or resolve them) are italicized. NOE

measurements were recorded using the DPFGSE-NOE

sequence [10]. All the reagents were commercial products

(Aldrich Co.) of the highest purity available and used as

received, except for complexes 1a–h, 2a–e, and 7, which

were prepared by the published methods [2,6].

1795 (s), 1704 (m) cm^ꢀ1. H NMR (CDCl₃) 7.29–7.09 (m,

5H, Ph); 4.90, 4.84, 4.71, 4.61 (s, 10H, Cp); 4.86 (s, 1H,

C_bH); 4.51, 3.24, 3.07, 2.95 (d, 2H, ²J_HH= 12 Hz, CH₂Ph);

4.09, 4.05 (s, 3H, CO₂Me); 2.31, 1.60 (s, 3H, NMe); isomer

ratio 5:1. ¹³C{¹H} NMR (CDCl₃) d 262.4, 260.7 (l-CO);

211.2, 210.9 (CO); 180.8, 180.7 (C_cand CO₂Me); 137.2,

129.9, 127.8, 126.9 (Ph); 120.8 (CN); 88.3, 88.3, 87.1, 86.8

(Cp); 80.6 (C_b); 64.7 (C_a); 59.7 (CH₂Ph); 52.4, 52.2

(CO₂Me); 46.1, 38.6 (NMe).

4240

V.G. Albano et al. / Journal of Organometallic Chemistry 691 (2006) 4234–4243

3e: Yield (93%). Anal. Calc. for C₂₇H₃₀Fe₂N₂O₂Si: C,

58.50; H, 5.46; N, 5.05. Found: C, 58.60; H, 5.55; N,

4.5. Synthesis of [Fe₂{l-g¹:g³-C_c(Tol)@C_b

HC_a= N(Me)(p-C₆H₄CF₃)}(l-CO)(CO)(Cp)₂]

[SO₃CF₃] (4)

4.87%. IR (CH₂Cl₂) m(C„N) 2188 (m), m(CO) 1972 (vs),

1

1785 (s) cm^ꢀ1. H NMR (CDCl₃) 7.32–7.08 (m, 5H, Ph);

5.03, 4.97, 4.57, 4.50 (s, 10H, Cp); 4.78 (s, 1H, C_bH);

4.66, 3.26 (d, 2H, J_HH= 13 Hz, CH₂Ph); 2.43, 1.63 (s,

Compound [Fe₂{l-CN(Me)(p-C₆H₄CF₃)}(l-CO)(CO)-

(NCMe)(Cp)₂][SO₃CF₃] (200 mg, 0.297 mmol) was dis-

solved in CH₂Cl₂(25 mL), treated with HC„CTol

(0.33 mmol) and heated at boiling temperature for 4 h. Sol-

vent removal and chromatography of the residue on an alu-

mina column, with MeOH as eluent, gave a brown band

corresponding to 4. Yield: 156 mg (70%). Anal. Calc. for

C₃₁H₂₅F₆Fe₂NO₅S: C, 49.69; H, 3.36; N, 1.87. Found: C,

49.78; H, 3.45; N, 1.79%. IR (CH₂Cl₂) m(CO) 1992 (vs),

2

3H, NMe); 0.68, 0.58 (s, 9H, SiMe₃); isomer ratio 5:1.

¹³C{¹H} NMR (CDCl₃) d 264.6 (l-CO); 213.1, 211.9

(CO); 195.4, 194.3 (C_c); 138.0, 130.6, 129.3, 127.8, 126.6

(Ph); 121.5 (CN); 87.6, 87.5, 86.2, 84.8 (Cp); 85.0 (C_b);

68.1 (C_a); 64.9, 59.9 (CH₂Ph); 46.5, 40.1 (NMe); 3.6, 3.5

(SiMe₃).

1

4.3. Synthesis of [Fe₂{l-CN(Me)(p-C₆H₄CF₃)}

(l-CO)(CO)₂(Cp)₂][SO₃CF₃]

1813 (s), m(C_aN) 1607 (m). H NMR (CDCl₃) 7.84–7.26

(m, 8H, C₆H₄CF₃and C₆H₄Me); 5.26, 5.02, 4.94, 4.85 (s,

10H, Cp); 4.38 (s, 1H, C_bH); 4.30, 3.69 (s, 3H, NMe); 2.37

(s, 3H, C₆H₄Me). Z/E ratio 4:3. ¹³C{¹H} NMR (CDCl₃) d

255.5, 254.2 (l-CO); 231.6, 229.9 (C_a); 209.5, 209.4 (CO);

207.5, 205.8 (C_c); 153.3, 153.0 (ipso-C₆H₄Me); 148.1–122.5

(C₆H₄CF₃+ C₆H₄Me); 91.9, 91.5, 89.0, 88.5 (Cp); 56.1,

55.8 (C_b); 55.0, 52.3 (NMe); 20.9 (C₆H₄Me).

CN(p-C₆H₄CF₃) (1.00 g, 5.85 mmol) was added to a solu-

tion of [Fe₂(Cp)₂(CO)₄] (3.11 g, 8.79 mmol) in MeCN

(100 mL), and the mixture was heated at reﬂux temperature

for 24 h. Then, the solvent was removed under reduced pres-

sure and the residue was dissolved in CH₂Cl₂(50 mL).

Hence, CH₃SO₃CF₃(0.7 mL, 6.2 mmol) was added, and

the mixture was stirred for 5 h. The resulting solution

reduced in volume and treated with and diethyl ether

(200 mL). A dark red precipitate, corresponding to [Fe₂{l-

CN(Me)(p-C₆H₄CF₃)}(l-CO)(CO)₂(Cp)₂][SO₃CF₃], was

isolated. Yield: 1.74 g (45%). Anal. Calc. for C₂₃H₁₇F₆Fe_2-

NO₆S: C, 41.78; H, 2.59; N, 2.12. Found: C, 41.69; H, 2.44;

N, 2.16%. IR (CH₂Cl₂) m(CO) 2024 (vs), 1994 (s), 1840 (s),

4.6. Synthesis of [Fe₂{l-g¹:g³-C_c(Tol)C_b(H)@C_a

(H)N(Me)(p-C₆H₄CF₃)}(l-CO)(CO)(Cp)₂] (5)

Complex 4 (100 mg, 0.134 mmol) was dissolved in THF

(10 mL) and treated with NaBH₄(27 mg, 0.711 mmol). The

mixture was stirred for 15 min. Then, the solvent was

removed, the residue was dissolved in CH₂Cl₂(10 mL),

and ﬁltered on alumina. Complex 5 was obtained as orange

powder upon removal of the solvent. Yield: 64 mg (80%).

Anal. Calc. for C₃₀H₂₆F₃Fe₂NO₂: C, 59.93; H, 4.36; N,

2.33. Found: C, 60.03; H, 4.41; N, 2.27%. IR (CH₂Cl₂)

m(CO) 1958 (vs), 1777 (s). ¹H NMR (CDCl₃) 7.76–7.32

(m, 8H, C₆H₄CF₃and C₆H₄Me); 4.67 (d, 1H,

³J_HH= 8.42 Hz, C_bH); 4.51, 4.49 (s, 10H, Cp); 2.80 (s,

3H, NMe); 2.49 (s, 3H, C₆H₄Me); 1.54 (d, 1H,

³J_HH= 8.42 Hz, C_aH). ¹³C{¹H} NMR (CDCl₃) d 263.0

(l-CO); 214.9 (CO); 202.1 (C_c); 157.7 (ipso-C₆H₄Me);

153.0 (ipso-C₆H₄CF₃); 134.6-124.2 (C₆H₄Me); 88.9, 85.0

(Cp); 84.5 (C_a); 74.5 (C_b); 49.5 (NMe); 21.1 (C₆H₄Me).

¹⁹F{¹H} NMR (CDCl₃) ꢀ141.4 (s, 3 F, p-C₆H₄CF₃).

1

m(l-CN) 1539 (m). H NMR (CDCl₃) d 8.02–7.67 (m, 4H,

C₆H₄); 5.45, 4.74 (s, 10H, Cp); 4.55 (s, 3H, NMe). ¹³C{¹H}

NMR (CDCl₃) d 326.4 (l-C); 252.8 (l-CO); 208.5, 207.2

(CO); 152.5 (ipso-C₆H₄); 127.7, 126.1 (C₆H₄); 90.4, 90.0

(Cp); 57.2 (NMe). ¹⁹F{¹H} NMR (CDCl₃) d ꢀ63.0 (s, 3 F,

p-C₆H₄CF₃); ꢀ78.5 (s, 3 F, CF₃SO₃).

4.4. Synthesis of [Fe₂{l-CN(Me)(p-C₆H₄CF₃)}

(l-CO)(CO)(NCMe)(Cp)₂][SO₃CF₃]

A

solution of [Fe₂{l-CN(Me)(p-C₆H₄CF₃)}(l-CO)-

(CO)₂(Cp)₂][SO₃CF₃] (300 mg, 0.453 mmol), in MeCN

(20 mL), was treated with Me₃NO (44 mg, 0.587 mmol),

and the resulting mixture was stirred for 30 min. Then,

the solvent was removed under reduced pressure, the resi-

due was dissolved in CH₂Cl₂(10 mL) and ﬁltered through

a celite pad. Removal of the solvent gave a brown powder.

Yield: 275 mg (90%). Anal. Calc. for C₂₄H₂₀F₆Fe₂N₂O₅S:

C, 42.76; H, 2.99; N, 4.16. Found: C, 42.81; H, 3.04; N,

4.05%. IR (CH₂Cl₂)m(CO) 1985 (vs), 1819 (s), m(l-CN)

1524 (m). ¹H NMR (CDCl₃) 7.93–7.66 (m, 4H, C₆H₄);

5.07, 4.75, 4.39, 4.22 (s, 10H, Cp); 4.87, 4.57 (s, 3H,

NMe); 1.98 (s, 3H, NCMe). a/b ratio 3:2. ¹³C{¹H} NMR

(CDCl₃) d 337.9, 337.6 (l-C); 264.6, 264.4 (l-CO); 211.5,

210.5 (CO); 153.0, 152.5 (ipso-C₆H₄); 130.5, 130.1 (NCMe);

127.5, 127.3, 126.4 (C₆H₄); 88.8, 88.3, 87.3, 86.7 (Cp); 56.3,

55.6 (NMe); 4.00 (NCMe). ¹⁹F{¹H} NMR (CDCl₃)ꢀ62.7,

ꢀ62.8 (s, 3 F, p-C₆H₄CF₃); ꢀ78.5 (s, 3 F, CF₃SO₃).

4.7. Synthesis of [Fe₂{l-g¹:g³-C_c(Tol)C_b(H)C_a(CN)N-

(Me)(p-C₆H₄CF₃)}(l-CO)(CO)(Cp)₂] (6)

A solution of 4 (90 mg, 0.120 mmol), in CH₂Cl₂

(10 mL), was treated with NBuⁿ₄CN (42 mg, 0.157 mmol),

and the resulting mixture was stirred for 20 min. Then,

the solution was ﬁltered through an alumina pad. Complex

6 was obtained as an orange powder upon removal of the

solvent. Yield: 275 mg (90%). Anal. Calc. for

C₃₁H₂₅F₃Fe₂N₂O₂: C, 59.46; H, 4.02; N, 4.47. Found: C,

59.56; H, 4.08; N, 4.39%. IR (CH₂Cl₂)m(C„N) 2195 (w),

m(CO) 1972 (vs), 1788 (s). ¹H NMR (CDCl₃) 7.79–6.93

(m, 8H, C₆H₄CF₃and C₆H₄Me); 4.80, 4.79, 4.62, 4.61 (s,

10H, Cp); 3.38 (s, 1H, C_bH); 3.19, 2.60 (s, 3H, NMe);

V.G. Albano et al. / Journal of Organometallic Chemistry 691 (2006) 4234–4243

4241

2.49, 2.48 (s, 3H, C₆H₄Me). Isomer ratio 1:1. ¹³C{¹H}

NMR (CDCl₃) d 260.1 (l-CO); 212.8, 211.0 (CO); 202.1,

200.3 (C_c); 156.2 (ipso-C₆H₄Me); 153.6 (ipso-C₆H₄CF₃);

149.9–124.3 (C₆H₄Me and Ph); 123.0, 122.8 (CN); 117.9,

115.8 (C_b); 90.0, 86.5, 86.4 (Cp); 86.6, 85.8 (C_a); 48.1,

42.5 (NMe); 21.1 (C₆H₄Me). ¹⁹F{¹H} NMR (CDCl₃)

ꢀ61.6, ꢀ61.8 (s, 3 F, p-C₆H₄CF₃).

4.8. Synthesis of [Ru₂{l-g¹:g³-C_c-

(SiMe₃)C_b(H)C_a(CN)N(Me)(CH₂Ph)}

(l-CO)(CO)(Cp)₂] (8)

Complexes 9b–9d were prepared by the same procedure

described for 9a, by reacting NBuⁿ₄CN with 2b–d,

respectively.

9b: Yield (85%). Anal. Calc. for C₃₀H₂₆Fe₂N₂O₂: C,

64.55; H, 4.69; N, 5.02. Found: C, 64.63; H, 4.74; N,

5.10%. IR (CH₂Cl₂) m(C„N) 2187 (m), m(CO) 1968 (vs),

1788 (s) cm^ꢀ^{1 1}H NMR (CD₂Cl₂) 7.96–6.63 (m, 10H,

.

Ph); 4.76, 4.53 (s, 10H, Cp); 2.38, 2.03 (s, 6H, NMe).

¹³C{¹H} NMR (CD₂Cl₂) d 265.3 (l-CO); 214.1 (CO);

197.9 (C_c); 158.3, 140.7, 133.2, 129.0, 128.3, 128.1, 127.6,

127.2, 124.7 (Ph); 120.1 (CN); 96.5 (C_b); 91.3, 87.9 (Cp);

63.3 (C_a); 50.3, 43.1 (NMe).

NBuⁿ₄CN (40.0 mg, 0.149 mmol) was added to a solution

of 7 (64 mg, 0.083 mmol), in CH₂Cl₂(15 mL), and the solu-

tion stirred for 20 min. Solvent removal and chromatogra-

phy on an alumina column, with CH₂Cl₂as eluent, gave 8

as an orange fraction. Yield: 47 mg (88%). Anal. Calc. for

C₂₇H₃₀N₂O₂Ru₂Si: C, 50.30; H, 4.69; N, 4.34. Found: C,

50.59; H, 4.87; N, 4.21%. IR (CH₂Cl₂)m(C„N) 2189 (m),

9c: Yield (77%). Anal. Calc. for C₂₆H₂₆Fe₂N₂O₂: C,

61.21; H, 5.14; N, 5.49. Found: C, 61.15; H, 5.16; N,

5.50%. IR (CH₂Cl₂) m(C„N) 2183 (m), m(CO) 1965 (vs),

1

1779 (s) cm^ꢀ1. H NMR (CDCl₃) 7.27–7.04 (m, 5H, Ph);

4.93, 4.89, 4.44, 4.35 (s, 10H, Cp); 3.87, 3.76 (s, 3H, C_cMe);

2

4.43, 3.10, 3.04, 2.94 (d, 2H, J_HH= 12 Hz, CH₂Ph); 2.26,

2.21 (s, 3H, C_bMe); 2.08, 1.51 (s, 3H, NMe); Isomer ratio

6:5. ¹³C{¹H} NMR (CDCl₃) d 269.1, 267.4 (l-CO); 214.2,

212.7 (CO); 196.5, 195.5 (C_c); 137.7–126.7 (Ph); 120.3

(CN); 89.2, 89.0, 87.2, 87.0 (Cp); 90.4, 88.4 (C_b); 65.2,

64.6 (C_a); 59.5 (CH₂Ph); 45.1, 39.4 (NMe); 38.7 (C_cMe);

23.3, 22.5 (C_bMe).

m(CO) 1971 (vs), 1786 (s) cm^ꢀ^{1 1}H NMR (CDCl₃) d

.

7.01–7.30 (m, 5H, Ph); 5.34, 4.90 (s, 10H, Cp); 4.83 (s,

2

1H, C_bH); 4.39, 3.21 (d, 2H, J_HH= 12.6 Hz, CH₂Ph);

1.88 (s, 3H, NMe); 0.30 (s, 9H, SiMe₃). ¹³C{¹H} NMR

(CDCl₃) d 237.3 (l-CO); 200.5 (CO); 176.6 (C_c); 137.9

(ipso-Ph); 129.9, 127.8, 126.8 (Ph); 122.0 (CN); 89.2, 86.8

(Cp); 84.4 (C_b); 67.9 (C_a); 60.5 (CH₂Ph); 45.9 (NMe); 3.1

(SiMe₃). A ¹³C-enriched sample was obtained by using a

solution of K¹³CN (5.0 mg, 0.075 mmol) and NBuⁿ₄CN

(200 mg, 0.75 mmol) in CH₃CN/MeOH (3 + 3 mL). IR

(CH₂Cl₂) m(¹³CN) 2139 (m), m(CO) 1972 (vs), 1786 (s)

9d: Yield: (80%). Anal. Calc. for C₂₈H₂₆Fe₂N₂O₆: C,

56.22; H, 4.38; N, 4.68. Found: C, 56.30; H, 4.27; N,

4.72%. IR (CH₂Cl₂) m(C„N) 2191 (w), m(CO) 1987 (vs),

1

1801 (s), 1720 (m) cm^ꢀ1. H NMR (CDCl₃) 7.36–7.19 (m,

5H, Ph); 4.93, 4.89, 4.79, 4.70 (s, 10H, Cp); 4.35, 3.74,

2

3.64, 3.21 (d, 2H, J_HH= 12 Hz, CH₂Ph); 4.04, 4.01,

cm^ꢀ1

.

¹³C{¹H} NMR (CDCl₃) d 237.29 (l-CO); 200.5

3.91, 3.87 (s, 6H, CO₂Me); 2.07, 1.67 (s, 3H, NMe); Isomer

ratio 3:1. ¹³C{¹H} NMR (CDCl₃) d 260.2 (l-CO); 211.8,

210.2 (CO); 186.0, 184.7, 179.1, 170.8 (C_cand CO₂Me);

136.9–127.0 (Ph); 119.4, 119.2 (CN); 89.5, 89.2, 89.0, 88.6

(Cp); 88.1 (C_b); 65.3, 59.2 (CH₂Ph); 62.9 (C_a); 53.2, 52.9,

52.4, 52.2 (CO₂Me); 45.4, 36.9 (NMe).

3

(CO); 176.6 (d, J_CC= 5.1 Hz, C_c); 137.9 (ipso-Ph);

129.9, 127.8, 126.8 (Ph); 121.9 (CN); 89.2, 86.8 (Cp); 84.4

2

1

(d, J_CC= 5.1 Hz, C_b); 67.9 (d, J_CC= 63.3 Hz, C_a); 60.4

(CH₂Ph); 45.9 (NMe); 3.0 (SiMe₃).

4.9. Synthesis of [Fe₂{l-g¹:g³-C_c(R⁰)C_b(R⁰)C_a

(CN)N(Me)(R)}(l-CO)(CO)(Cp)₂] (R = Me, R⁰= Me

(9a); R = Me, R⁰= Ph (9b); R = CH₂Ph, R⁰= Me (9c);

R = CH₂Ph, R⁰= COOMe (9d))

4.10. Synthesis of [Fe₂{l-g¹:g²-C_c(CO₂Me)C_b(CO₂

Me)(CN)C_aN(Me)(Xyl)}(l-CO)(CO)(Cp)₂] (10)

A solution of 2e (98 mg, 0.133 mmol), in CH₂Cl₂

(10 mL), was treated at room temperature with NBuⁿ₄CN

(43 mg, 0.160 mmol). After 30 min of stirring, the solvent

was removed under reduced pressure. Chromatography

on alumina, with CH₂Cl₂as eluent, aﬀorded 10 as a brown

band. Yield: 59 mg (71%). Anal. Calc. for C₂₉H₂₈Fe₂N₂O₆:

C, 56.89; H, 4.61; N, 4.58. Found: C, 56.77; H, 4.62; N,

4.53%. IR (CH₂Cl₂) m(C„N) 2175 (w), m(CO) 1947 (vs),

A solution of 2a (90 mg, 0.162 mmol), in CH₂Cl₂

(10 mL), was treated at room temperature with NBuⁿ₄CN

(68 mg, 0.254 mmol). The mixture was stirred for 20 min,

then it was ﬁltered on alumina. Solvent removal, under

reduced pressure, gave 2a as microcrystalline powder.

Yield: 64 mg (91%). Crystals suitable for X-ray analysis

were collected by layering a diethyl ether solution of 9a

with petroleum ether (b.p. 40–60 ꢁC), at ꢀ20 ꢁC. Anal.

Calc. for C₂₀H₂₂Fe₂N₂O₂: C, 55.34; H, 5.11; N, 6.45.

Found: C, 55.39; H, 5.08; N, 6.49%. IR (CH₂Cl₂)m(C„N)

1777 (s), 1733 (m), 1683 (m) cm^ꢀ^{1 1}H NMR (CDCl₃)

.

7.33–7.14 (m, 3H, Me₂C₆H₃); 4.69, 4.19 (s, 10H, Cp);

3.95, 3.91 (s, 6H, CO₂Me); 3.36 (s, 3H, NMe); 2.15, 2.12

(s, 6H, Me₂C₆H₃). ¹³C{¹H} NMR (CDCl₃) d 272.3 (l-

CO); 252.7 (C_a); 214.6 (CO); 182.3 (C_c-CO₂Me); 168.5

(C_b-CO₂Me); 145.2 (ipso-Me₂C₆H₃); 134.1, 132.8, 129.7,

128.6, 128.3 (Me₂C₆H₃); 129.4 (C_c); 116.5 (C_bCN); 88.3,

87.6 (Cp); 81.5 (C_b); 53.5, 50.9 (CO₂Me); 44.9 (NMe);

18.1, 17.5 (Me₂C₆H₃). A ¹³C-enriched sample on the CN

2184 (m), m(CO) 1956 (vs), 1781 (s) cm^ꢀ^{1 1}H NMR

.

(CDCl₃) 4.85, 4.38 (s, 10H, Cp); 3.80 (s, 3H, C_cMe); 2.22

(s, 3H, C_bMe); 2.12, 1.60 (s, 6H, NMe). ¹³C{¹H} NMR

(CDCl₃) d 268.3 (l-CO); 213.3 (CO); 195.1 (C_c); 120.0

(CN); 88.9, 86.7 (Cp); 89.1 (C_b); 64.0 (C_a); 48.1, 41.2

(NMe); 39.3 (C_cMe); 22.3 (C_bMe).

4242

V.G. Albano et al. / Journal of Organometallic Chemistry 691 (2006) 4234–4243

position was obtained by using a solution of K¹³CN

(5.0 mg, 0.075 mmol) and NBuⁿ₄CN (200 mg, 0.75 mmol)

in CH₃CN (6 mL). IR (CH₂Cl₂) m(¹³C„N) 2090 (m),

thermal parameters to all the non-hydrogen atoms. In com-

plex 3a, the Cp ligand bound to Fe(1) was disordered over

two positions and the site occupation factors were reﬁned

yielding the values 0.60 and 0.40, respectively whereas in

complex 9a both Cp ligands were disordered over two posi-

tions and their site occupation factors were reﬁned yielding

the values 0.73 and 0.27 for the Cp bound to Fe(1) [0.64

and 0.36, respectively, for the Cp bound to Fe(2)]. The

methyl and aromatic hydrogen atoms were placed in calcu-

lated positions and reﬁned with idealized geometry,

whereas the H atom bound to C(4) in 3a was located in

the Fourier map and reﬁned isotropically. In order to be

sure about the mode of bonding of 3a and 9a, structure

models of alternative assignments of the carbon and nitro-

gen positions were reﬁned (cyanide, isocyanide). The isocy-

anide models for the independent molecules in 3a and 9a

showed worsening of the agreement indices and physically

unreasonable thermal motion ellipsoids (see Table 2).

m(CO) 1947 (vs), 1777 (s), 1733 (m), 1683 (m) cm^ꢀ1

.

¹³C{¹H} NMR (CDCl₃) d 272.7 (l-CO); 253.0 (C_a); 214.8

(CO); 182.5 (C_c-CO₂Me); 168.7 (C_b-CO₂Me); 145.3 (ipso-

Me₂C₆H₃); 134.2, 133.0, 129.8, 128.7, 128.4 (Me₂C₆H₃);

2

129.4 (d, J_CC= 5.0 Hz, C_c); 116.6 (C_bCN); 88.3, 87.6

(Cp); 81.5 (d, J_CC= 67.2 Hz, C_b); 53.5, 50.9 (CO₂Me);

1

44.8 (NMe); 18.0, 17.4 (Me₂C₆H₃).

4.11. X-ray crystallography for 3a and 9a

The diﬀraction experiments were carried out at room

temperature on a Bruker AXS SMART 2000 CCD based

diﬀractometer using graphite monochromated Mo Ka

˚

radiation (k = 0.71073 A). Intensity data were measured

over full diﬀraction sphere using 0.3ꢁ wide x scans. The

software ^SMART[11] was used for collecting frames of data,

indexing reﬂections and determination of lattice parame-

ters. The collected frames were then processed for integra-

tion by software ^SAINT[11] and an empirical absorption

correction was applied with ^SADABS[12]. The structures

were solved by direct methods (^SIR97) [13] and subsequent

Fourier syntheses, and reﬁned by full-matrix least-squares

calculations on F²(SHELXTL) [14] attributing anisotropic

Acknowledgements

`

We thank the Ministero dell’Universita e della Ricerca

Scientiﬁca e Tecnologica (MIUR) (Project: ‘New strategies

for the control of reactions: interactions of molecular frag-

ments with metallic sites in unconventional species’) and

the University of Bologna (’Funds for Selected Research

Topics’) for ﬁnancial support.

Table 2

Appendix A. Supplementary material

Crystal data and experimental details for 3a and 9a

Compound

3a

9a

Crystallographic data for the structures reported in this

paper have been deposited with the Cambridge Crystallo-

graphic Data Centre as supplementary publication nos.

CCDC 608048 and CCDC 608049 for 3a and 9a, respec-

tively. Copies of the data can be obtained free of charge

on application to CCDC, 12 Union Road, Cambridge

CB21EZ, UK, fax: +44 1223 336 033, e-mail: deposit@

ccdc.cam.ac.uk.

Formula

Fw

C

₂₁H₂₆Fe₂N₂O₂Si

C₂₀H₂₂Fe₂N₂O₂

434.10

293(2)

0.71073

monoclinic

P2₁/n

478.23

293(2)

0.71073

monoclinic

P2₁/n

T (K)

˚

k (A)

Crystal symmetry

Space group

Unit cell dimensions

˚

a (A)

10.5739(3)

28.4979(8)

15.2823(4)

90

109.319(1)

90

4345.8(2)

8

1.462

7.9451(3)

16.3339(7)

14.3416(6)

90

99.369(1)

90

1836.4(1)

4

1.570

˚

b (A)

˚

c (A)

References

a (ꢁ)

b (ꢁ)

c (ꢁ)

[1] (a) R.A. Adams, F.A. Cotton (Eds.), Catalysis by Di-and Polynuclear

Metal Complexes, Wiley–VCH, New York, 1998;

3

˚

Cell volume (A )

Z

`

(b) P. Braunstein, J. Rose, in: P. Braunstein, L.A. Oro, P.R. Raithby

D_calc(Mg m^ꢀ3

)

(Eds.), Metal Cluster in Chemistry, Wiley–VCH, Weinheim, 1999, p.

616;

l(Mo K_a) (mm^ꢀ1

F(000)

)

1.409

1984

1.596

896

´

(c) M.V. Jimenez, E. Sola, F.J. Lahoz, L.A. Oro, Organometallics 24

Crystal size (mm)

h Limits (ꢁ)

Reﬂections collected

0.35 · 0.30 · 0.28

1.43–30.02

56753( h, k, l)

0.30 · 0.28 · 0.25

1.90–30.03

23289( h, k, l)

(2005) 2722;

(d) R.D. Adams, B. Captain, J. Organomet. Chem. 689 (2004) 4521;

(e) B. Bosnich, Inorg. Chem. 38 (1999) 2554;

Unique observed reﬂections

(f) B.D. Rowsell, R. McDonald, M. Cowie, Organometallics 23

(2004) 3873.

[2] (a) V.G. Albano, L. Busetto, F. Marchetti, M. Monari, S. Zacchini,

V. Zanotti, Organometallics 22 (2003) 1326;

(b) V.G. Albano, L. Busetto, F. Marchetti, M. Monari, S. Zacchini,

V. Zanotti, J. Organomet. Chem. 689 (2004) 528.

[3] (a) V.G. Albano, L. Busetto, F. Marchetti, M. Monari, S. Zacchini,

V. Zanotti, Organometallics 23 (2004) 3348;

[F_o> 4r(F_o)] (R_int

)

12692 (0.0652)

0.934

0.0419, 0.0964

0.504/ꢀ0.471

5364 (0.0264)

1.034

0.0286, 0.0772

0.270/ꢀ0.485

Goodness-of-ﬁt-on F²

R₁(F)^a, wR₂(F²)^b

Largest diﬀerence in

peak and hole (e A

ꢀ3

˚

)

P

a

R₁= jjF_oj ꢀ jF_cj/ jF_oj.

wR₂¼

h

i

P

2

b

wðF ²_oꢀ F _c²Þ = wðF _o²Þ

, where w ¼ 1=½r²ðF _o²Þþ ðaPÞ þ

(b) V.G. Albano, L. Busetto, F. Marchetti, M. Monari, S. Zacchini,

V. Zanotti, J. Organomet. Chem. 690 (2005) 837.

1=2

bPꢂ, where P ¼ ðF ²_oþ 2F _c²Þ=3.

V.G. Albano et al. / Journal of Organometallic Chemistry 691 (2006) 4234–4243

4243

[4] L. Busetto, F. Marchetti, S. Zacchini, V. Zanotti, Organometallics 24

(2005) 2297.

[10] K. Stott, J. Stonehouse, J. Keeler, T.L. Hwang, A.J. Shaka, J. Am.

Chem. Soc. 117 (1995) 4199.

[5] V.G. Albano, L. Busetto, F. Marchetti, M. Monari, S. Zacchini, V.

Zanotti, J. Organomet. Chem. 690 (2005) 4666.

[6] L. Busetto, F. Marchetti, S. Zacchini, V. Zanotti, J. Organomet.

Chem. 691 (2006) 2424.

[7] R.B. King, Coordin. Chem. Rev. 248 (2004) 1533.

[8] (a) V. Zanotti, S. Bordoni, L. Busetto, L. Carlucci, A. Palazzi, R.

Serra, V.G. Albano, M. Monari, F. Prestopino, F. Laschi, P. Zanello,

Organometallics 14 (1995) 5232;

[11] ^SMART& SAINT Software Reference Manuals, Version 5.051 (Windows

NT Version), Bruker Analytical X-ray Instruments Inc., Madison,

WI, 1998.

[12] G.M. Sheldrick, ^SADABS, Program for Empirical Absorption Correc-

tion, University of Go¨ttingen, Germany, 1996.

[13] A. Altomare, M.C. Burla, M. Cavalli, G.L. Cascarano, C. Giaco-

vazzo, A. Guagliardi, A.G.G. Moliterni, G. Polidori, R. Spagna, J.

Appl. Cryst. 32 (1999) 115.

(b) L. Busetto, V. Zanotti, J. Organomet. Chem. 690 (2005) 5430.

[9] W. Wilker, D. Leibfritz, R. Kerssebaum, W. Beimel, Magn. Reson.

Chem. 31 (1993) 287.

[14] G.M. Sheldrick, ^SHELXTLplus Version 5.1 (Windows NT Version)

Structure Determination Package, Bruker Analytical X-ray Instru-

ments Inc., Madison, WI, 1998.

Article Doi

C-C bond formation by cyanide addition to dinuclear vinyliminium complexes

DOI: 10.1016/j.jorganchem.2006.06.033

Source and publish data:

Authors:

Article abstract of DOI:10.1016/j.jorganchem.2006.06.033

Full text of DOI:10.1016/j.jorganchem.2006.06.033

Products guided by the article

R&D Labs maybe for 852632-13-4

Relevant to this article

Hot Product