The metal-carbon multiple bond in iron(I)- and iron(II)-dibenzotetramethyltetra[14]azaannulene: Carbene, carbonyl, and isocyanide derivatives

www.elsevier.nl/locate/jorganchem

Journal of Organometallic Chemistry 591 (1999) 45–62

The metal–carbon multiple bond in iron(I)– and

iron(II)–dibenzotetramethyltetra[14]azaannulene: carbene, carbonyl,

and isocyanide derivatives

Alain Klose ^a, Joe¨lle Hesschenbrouck ^a, Euro Solari ^a, Mario Latronico ^a,1,

Carlo Floriani ^a,*, Nazzareno Re ^b, Angiola Chiesi-Villa ^c, Corrado Rizzoli ^c

^aInstitut de Chimie Mine´rale et Analytique, BCH, Uni6ersite´ de Lausanne, CH-1015 Lausanne, Switzerland

^bFacolta` di Farmacia, Uni6ersita` degli Studi ‘G. D’Annunzio’, I-66100 Chieti, Italy

^cDipartimento di Chimica, Uni6ersita` di Parma, I-43100 Parma, Italy

Received 25 May 1999; accepted 13 June 1999

Abstract

The two parent compounds used for studying the iron–carbon multiple bond interactions are [Fe(tmtaa)] (1), and [Fe(tm-

taa)Na(THF)₃] (2) [tmtaaꢀdibenzotetramethyltetraaza[14]annulene dianion], the latter being obtained by reduction of 1. The

reaction of 1 with CO led to the corresponding monocarbonyl derivative [Fe(tmtaa)(CO)(L)] [L=THF, 3; L=Py, 4], while the

reaction with RNC allowed us to isolate mono-isocyanide [Fe(tmtaa)(o-Me₃Si–C₆H₄NC)(THF)] (5), [Fe(tmtaa)(ⁿBuNC)(THF)]

(6), and bis-isocyanide [Fe(tmtaa)(^tBuNC)₂] (7) derivatives. Reduction of 6 with sodium metal or the reaction of 1 with NaCN led

to a monocyano derivative bridged into a dimeric form by sodium cations in [{Fe(tmtaa)(CN)}₂(m-NaL_n)₂] [L=THF, n=3, 8a;

n

L=DME, n=2, 8b], while the reaction of 1 with Bu₄N⁺CN⁻led to the monomeric form [Fe(tmtaa)(CN)]⁻(ⁿBu₄N)⁺(9). A

detailed magnetic analysis of 1–10, the last one being the bis-pyridine derivative [Fe(tmtaa)(Py)₂] (10) showed a variety of low and

intermediate spin states, and spin crossovers (with a minor role played by high spin states) as a function of the axial ligands. A

remarkable difference was observed with the analogous porphyrin derivatives. The d⁷iron(I) derivative 2 occurs in tight ion-pair

form, both iron and sodium being bonded to the tmtaa ligand. The reaction of 2 with carbon monoxide led to a monocarbonyl

derivative bridged in a dimeric form by sodium cations bonded to the oxygen atoms in [{Fe(tmtaa)}₂{m-CONa(THF)₂}₂] (11).

Both 2 and 11 showed a spin conversion between S=1/2 and S=3/2, with a small antiferromagnetic coupling in the latter case,

due to the dimeric form. The reaction of 1 with diazoalkane RR%CN₂led to the corresponding low-spin diamagnetic carbene

derivatives [Fe(tmtaa)(CRR%)] [R=R%=Ph, 12; R=Ph, R%=H, 13], the ﬁrst one being by far more thermally stable, while the

second one decomposes at room temperature to 1 and a mixture of cis and trans-stilbene. Both react with O₂giving Ph₂CO and

PhCHO and the m-oxo dimer [{Fe(tmtaa)}(m-O)] (14). The proposed structures are supported by the X-ray analyses on complexes

Keywords: Metalla-macrocycle; Iron carbene; Iron–carbon monoxide; Iron macrocycle redox

transition metals. In the case of middle transition

metals, however, the metal–carbon functionalities have

1. Introduction

been essentially considered for their relationship with

Macrocycles are increasingly considered as ancillary

biologically occurring systems, namely vitamin B₁₂in

ligands in organometallic chemistry [1–7]. This trend

the case of cobalt [8] and the cytochrome P450 in the

has been developed particularly in recent years for early

case of iron [9]. The macrocycle par excellence used in

such a context has been the porphyrin skeleton, with

* Corresponding author. Fax: +41-21-6923905.

important contributions in Ru [10a] and Rh [10b]

E-mail address: carlo.ﬂoriani@icma.unil.ch (C. Floriani)

chemistry. A number of other model ligands, however,

have been considered and among them particularly the

tetradentate Schiff bases [6] and the dibenzo-

¹Present address: Dipartimento di Ingegneria e Fisica dell’Ambi-

ente, Universita` della Basilicata, via della Tecnica 3, I-85100 Potenza,

Italy.

PII: S0022-328X(99)00354-X

46

A. Klose et al. / Journal of Organometallic Chemistry 591 (1999) 45–62

2. Results and discussion

2.1. Synthesis of iron(II)– and iron(I)–tmtaa

deri6ati6es

Two different approaches for synthesizing complex

[Fe(tmtaa)] 1 are known [11a,h]. Our method, which is

reproducible on

a 20 g scale, involved reacting

[FeCl₂(THF)_1.5] with the deprotonated tmtaa ligand

(Scheme 1) [2d].

Complex 1 formed as a poorly soluble crystalline

solid in THF. This feature made its isolation easy since

the related inorganic salt is very soluble in the solvent.

An NMR spectrum could only be obtained in pyridine-

d₅and showed broad signals due to the paramagnetic

behavior of the molecule (see below). Elemental analy-

sis suggested the inclusion of a THF molecule. The

X-ray analysis has already been reported by Goedken

et al. [11h] and shows that the ligand has the familiar

saddle shape conformation, with iron displaced slightly

Scheme 1.

,

from the N₄plane (0.114 A). The axial positions are

not occupied suggesting their potential for coordinating

other substrates.

tetramethyltetraaza[14]annulene [2]. In the latter case,

some fundamental work has been developed by the

groups of Goedken [11] and Cotton [12]. The present

report deals with the synthesis and the properties

of iron(II) and iron(I)–dibenzotetramethyltetraaza-

[14]annulene [tmtaa] complexes, containing a multiple

bond to carbon ligands, namely CO, RNC, carbene,

and cyanides. The nature of some starting iron(II)–tm-

taa complexes has been elucidated by Goedken and

co-workers [11], though large-scale and improved syn-

theses are reported here. Particularly remarkable are: (i)

the stabilization of the iron–carbene multiple bond

functionalities; and (ii) the magnetic properties associ-

ated both to the iron(I) and iron(II) derivatives. Some

results have been published in preliminary communica-

tions [13,14].

When 1 was reduced with exactly one equivalent of

sodium in THF, complex 2 formed in high yield

(Scheme 1). The magnetic analysis of this very air-sensi-

tive complex is reported in the following section. The

X-ray analysis is given in Fig. 1 and shows the ion pair

structure of 2, the molecule B between the two indepen-

dent ones being only considered. The ion pairing phe-

nomenon has already been observed in the reduced

form of metal Schiff base complexes [15] and in por-

phyrin derivatives [16].

Crystallographic data and details associated with

data collection are given in Table 1. Selected bond

distances and angles for all complexes are quoted in

Table 2. Relevant conformational parameters within

the [Fe(tmtaa)] moieties are given in Table 3. The

labeling scheme adopted for the tmtaa ligand is de-

picted below.

Both iron and sodium cations are bonded to the

macrocycle. The distance of the iron to the N₄coordi-

,

nation plane of 0.092(1) A is indicative of a change in

Fig. 1. SCHAKAL view of molecule A in complex 2.

the oxidation state and supports a d⁷conﬁguration.

A. Klose et al. / Journal of Organometallic Chemistry 591 (1999) 45–62

47

Table 1

Experimental data for the X-ray diffraction studies on crystalline complexes 2, 8b, 9, 11b and 12

Complex

Formula

2

8b

9

11b

12

C₃₄H₄₆FeN₄NaO₃

C

₆₂H₈₄Fe₂N₁₀Na₂O₈·

C₂₃H₂₂FeN₅·

C₁₆H₃₆N·2C₄H₈O

12.256(2)

22.060(4)

17.302(2)

90

107.87(2)

90

4452.2(13)

4

C₆₂H₈₄Fe₂N₈Na₂O₁₀

5C₆H₆

·

C₃₅H₃₂FeN₄·C₇H₈

C₄H₁₀O₂

11.484(2)

15.037(3)

11.360(2)

106.43(2)

101.96(1)

106.38(1)

1715.2(6)

1

,

a (A)

10.441(2)

17.648(4)

34.799(7)

90

92.51(3)

90

13.879(2)

15.549(2)

21.326(3)

99.07(2)

98.80(2)

98.02(2)

4427.6(12)

2

11.670(3)

13.593(4)

11.234(3)

106.72(2)

96.21(2)

77.62(2)

1665.0(8)

2

,

b (A)

,

c (A)

h (°)

i (°)

k (°)

3

,

V (A )

6406(2)

8

Z

Formula weight

Space group

T (°C)

637.6

P2₁/c (no. 14)

22

0.71069

1.322

5.21

1345.2

811.0

P2₁/n (no. 14)

−140

0.71069

1.210

3.79

1649.6

656.7

(

P1 (no. 2)

22

P1 (no. 2)

−140

0.71069

1.302

4.94

−140

0.71069

1.310

4.86

,

u (A)

0.71069

1.237

3.94

z_calc.(g cm⁻³

)

v (cm⁻¹

)

Transmission coefﬁcients

0.758–1.000

0.082

0.166

1.208

5897

11168

5897

776

0.674–1.000

0.073

0.203

1.066

2793

3295

3082

392

0.647–1.000

0.055

0.164

0.988

4809

9915

8286

505

0.921–1.000

0.086

0.249

1.118

8549

17976

10999

979

0.562–1.000

0.059

0.163

1.021

4428

5831

5293

424

a

R

wR₂

b

Goodness-of-ﬁt

N-observed ^c

N-independent ^d

N-reﬁnement ^e

Variables

^aCalculated on the unique observed data having I=2|(I) for 8b, 9, 11b, 12 and I=3|(I) for 2.

^bCalculated on the unique data with I=0 for 8b, 9, 11b, 12 and with I=3|(I) for 2.

^cN-observed is the total number of the independent reﬂections having I=2|(I) for 8b, 9, 11b, 12 and I=3|(I) for 2.

^dN-independent is the number of independent reﬂections.

^eN-reﬁnement is the number of reﬂection used in the reﬁnement having I=0 for 8b, 9, 11b, 12 and I=3|(I) for 2.

Table 2

Selected bond lengths (A) and angles (°) for complexes 2, 8b, 9, 11b and 12

,

2

8b

9

11b

12

Mol A

Mol B

Mol A

Mol B

Bond lengths

FeꢁN(1)

FeꢁN(2)

FeꢁN(3)

FeꢁN(4)

1.898(5)

1.895(5)

1.914(6)

1.912(5)

1.886(6)

1.894(5)

1.904(5)

1.920(6)

1.928(5)

1.924(5)

1.920(6)

1.912(6)

1.844(8)

1.349(8)

1.401(9)

1.345(9)

1.404(10)

1.413(9)

1.355(7)

1.367(7)

1.389(9)

1.167(10)

2.061(2)

2.056(3)

2.050(2)

2.057(4)

2.084(4)

1.327(4)

1.421(5)

1.338(4)

1.410(4)

1.429(5)

1.329(5)

1.324(4)

1.421(4)

1.160(6)

1.986(4)

2.004(4)

1.994(4)

1.998(4)

1.752(5)

1.322(7)

1.410(7)

1.326(8)

1.414(6)

1.401(8)

1.325(6)

1.337(6)

1.417(6)

1.202(7)

2.053(4)

2.044(4)

2.045(5)

2.043(4)

1.758(5)

1.328(8)

1.405(7)

1.322(8)

1.408(7)

1.414(6)

1.330(6)

1.318(8)

1.402(8)

1.196(7)

1.921(3)

1.926(2)

1.911(3)

1.920(2)

1.794(3)

1.333(4)

1.418(4)

1.321(5)

1.418(5)

1.422(4)

1.327(5)

1.333(5)

1.409(5)

FeꢁC(23)

N(1)ꢁC(2)

N(1)ꢁC(22)

N(2)ꢁC(4)

N(2)ꢁC(6)

N(3)ꢁC(11)

N(3)ꢁC(13)

N(4)ꢁC(15)

N(4)ꢁC(17)

XꢁC(23) ^a

C(23)ꢁC(31)

C(23)ꢁC(41)

1.375(9)

1.396(9)

1.356(9)

1.419(9)

1.435(9)

1.341(9)

1.343(9)

1.416(9)

1.360(10)

1.409(9)

1.342(9)

1.431(9)

1.407(9)

1.364(9)

1.325(10)

1.425(10)

1.487(7)

1.472(4)

Bond angles

FeꢁC(23)ꢁX ^a

FeꢁC(23)ꢁC(41)

FeꢁC(23)ꢁC(31)

C(31)ꢁC(23)ꢁC(41)

179.6(6)

179.3(4)

177.8(4)

178.2(4)

122.6(2)

121.7(2)

115.7(3)

^aX should be read N(5) for 8b, 9 and O(1) for 11b.

48

A. Klose et al. / Journal of Organometallic Chemistry 591 (1999) 45–62

Table 3

Comparison of relevant conformational parameters within Fe(tmtaa) moiety for complexes 2, 8b, 9, 11b and 12

2

8b

9

11b

12

Mol A

Mol B

Mol A

Mol B

,

(a) Distances (A) of atoms from the N₄mean plane

N(1)

N(2)

N(3)

N(4)

Fe(1)

0.003(7)

−0.003(7)

0.003(7)

−0.003(7)

0.092(1)

0.018(7)

−0.018(7)

0.018(7)

−0.018(7)

0.093(1)

−0.004(5)

0.005(5)

−0.008(6)

0.005(5)

0.014(2)

−0.014(2)

0.014(2)

−0.014(2)

0.610(1)

0.032(5)

−0.031(5)

0.031(5)

−0.025(4)

0.553(2)

0.003(4)

−0.003(4)

0.003(4)

−0.003(4)

0.655(2)

−0.057(2)

0.057(2)

−0.057(2)

0.057(2)

0.266(1)

0.335(1)

(b) Dihedral angles (°) between signiﬁcant planes

N₄N(1)C₃N(2)

N₄N(3)C₃N(4)

N₄N(1)C₆N(4)

N₄N(2)C₆N(3)

N(1)C₃N(2)N(3)C₃N(4)

N(1)C₆N(4)N(2)C₆N(3)

C(6)···C(11)C(17)···C(22)

158.7(3)

158.1(2)

161.6(2)

159.7(2)

136.8(3)

141.2(1)

138.2(2)

159.2(3)

158.6(2)

160.1(2)

160.4(2)

137.7(3)

140.6(1)

136.8(2)

154.7(2)

160.3(1)

160.4(2)

129.4(2)

140.7(1)

136.1(2)

146.2(1)

144.6(1)

167.0(1)

166.5(1)

110.8(1)

153.5(1)

149.5(1)

149.9(1)

148.0(2)

163.5(1)

161.9(1)

117.9(2)

145.4(1)

140.9(2)

146.5(2)

145.3(2)

162.3(1)

162.2(1)

111.8(2)

144.6(1)

139.1(2)

153.1(1)

154.7(1)

155.2(1)

154.2(1)

127.8(1)

130.1(1)

123.1(1)

The same value was obtained for the isoelectronic

complex [Co(tmtaa)] [11a]. One other factor to consider

is the average Fe–N bond length. This distance de-

creases with respect to the metal center–N₄plane dis-

tance or as its dⁿconﬁguration increases, varying from

accessible, and were all isolated in crystalline form and

in good yield. A general overview of the reactions is

presented in Scheme 2.

The reaction of 1 with CO readily formed monocar-

bonyl complexes either in the presence or the absence of

other axial ligand like pyridine or THF. Complex 3 was

obtained from the reaction of 1 with CO in THF. The

strong binding of CO is indicated by the characteristic

IR band displaced from 2155 cm⁻¹for free CO to 1913

cm⁻¹for 3. This indicates a relevant back-donation

process. The reactivity of CO was also observed by its

ability to displace one coordinated pyridine in the

complex [Fe(tmtaa)(Py)₂] (10) formed in situ. The re-

,

2.054 A for [Fe(tmtaa)₂O]·MeCN [11o] to 2.002 A for

,

[Fe(tmtaa)Cl]·MeCN [11a], to 1.918 A in 1 [11h] and

,

1.904(5) A in 2. It is worth noting that, at variance with

the iron porphyrin anion [Fe(TPP)]⁻[16a], no signiﬁ-

cant change at the bond distance pattern in the tetraaza

macrocycle ligand is observed. This supports the pro-

posed [Fe^I(tmtaa)]⁻iron(I) formulation over the alter-

II

−

native [Fe (tmtaa )] iron(II)-p radical anion, which is

instead suggested in the porphyrin analogue [16a],

though a later study was more in favor of the iron(I)-d⁷

formulation for [Fe(TPP)]⁻[16c]. A more signiﬁcant

difference is observed between iron(II) and iron(I) com-

plexes. The tmtaa ligand shows the usual saddle shape

conformation (Table 3). The sodium cation binds to the

tmtaa ligand interacting with the N(1) and N(2) nitro-

gen atoms at rather long distances (Na(1)–N(1),

,

2.784(7) [2.747(8)] A; Na(1)–N(2), 2.810(8) [2.845(7)]

,

A). Complex 2 is bifunctional and contains, in close

proximity, two different reactive metal sites, namely

iron and sodium. The nucleophilic iron shows very

interesting reactivity with small molecules and organic

functionalities (see below).

2.2. Reacti6ity of [Fe(tmtaa)] (1), with CO, isocyanides,

and cyanides

Coordination to 1 of strong ﬁeld ligands, such as CO,

cyanides, and isonitriles, gave complexes that could be

characterized with NMR spectroscopy, although their

magnetic measurement do not suggest a diamagnetic

behavior. These complexes (see Scheme 1) were easily

Scheme 2.

A. Klose et al. / Journal of Organometallic Chemistry 591 (1999) 45–62

49

observed at 2076 cm⁻¹, 5, and 1989 cm⁻¹, 6. The com-

parison with the free isonitriles (2123 cm⁻¹, o-

t

Me₃SiC₆H₄NC; 2134 cm⁻¹, BuNC) suggests a strong

back-donation. The ¹H-NMR spectra of 5 does not

show any unusual behavior, in contrast to 6, which gave

additional peaks of low integration similar to the car-

bonyl compound 3. These signals do not belong to the

bis-isonitrile complex 7 or to 1. The explanation of a

paramagnetic contribution can similarly be suggested,

although in this case the compound is essentially dia-

magnetic with a little residual paramagnetism, probably

due to paramagnetic impurities or temperature-indepen-

dent paramagnetism (TIP). A quite common method

was used to form complex 6 and involves the displace-

t

ment of the carbonyl ligand in 3 by BuNC.

A crystalline product of better solubility was obtained

t

when 1 was reacted with two equivalents of BuNC. As

a result of a second coordination, [Fe(tmtaa)(^tBuNC)₂]

(7) showed an IR band shifted to 2089 cm⁻¹. The

t

¹H-NMR spectrum conﬁrms the presence of two Bu

groups. This spectrum does not show any unusual be-

havior, except for the broadness of the signals. This may

be a consequence of the paramagnetic behavior of the

molecule (v_eff.=2.55v_Bat 298 K). The prediction of

whether the conﬁguration is cis or trans can be answered

by the iron out-of-plane distance, the electron rich

metals being almost in the N₄-plane and they favor a

trans arrangement [2d]. The IR spectrum with a single

CꢀN band conﬁrms such an assignment. The reduction

of 6 with Na metal in THF is a slow reaction leading to

8a. Recrystallization of 8a in DME gave crystals suit-

able for an X-ray analysis, which conﬁrmed the forma-

tion of the cyanide complex 8b (Fig. 2). In light of this

structure, a straightforward synthesis was tested by re-

acting 1 with sodium cyanide. Because of the poor solu-

bility of NaCN in the organic solvent, a slow reaction

was observed under reﬂux conditions in THF. The ana-

lytical and spectroscopic data were in agreement with

the cyanide complex 8a.

Fig. 2. ^SCHAKALview of complex 8b. Disorder affecting the DME

molecules has been omitted for clarity. A prime denotes a transfor-

mation of −x, −1−y, −z.

sulting compound 4 shows a slight variation in its IR

spectrum with the CO band centered at 1927 cm⁻¹. ¹H-

NMR spectra were obtained for both complexes despite

the paramagnetic behavior of 3 (v_eff.=3.10v_Bat 298 K)

(see below). These two spectra were recorded in

pyridine-d₅and are the same except that 3 includes the

THF peaks. Each set of tmtaa signals also shows similar

weak resonances, which do not correspond to 10. This

can be explained by the formation of a very small

amount of [Fe(tmtaa)(CO)₂], although rarely observed

because of the high lability of the second coordinated

CO [17], or by the possibility of a spin state equilibrium,

which will give rise to a paramagnetic contribution.

Complexes [Fe(tmtaa)(CO)], [Fe(tmtaa)(CO)(Py)] (4)

and [Fe(tmtaa)(CO)(N₂H₄)] have been reported by

Goedken et al. [11h]. However, they have not been ob-

tained from the carbonylation of 1, except during the in

situ synthesis of 1 under a carbon monoxide atmo-

sphere. In addition, no detail is given concerning the

magnetic properties of such compounds.

The molecular structure of 8b is given in Fig. 2 and

consists of centrosymmetric dimers. Each monomeric

unit is bridged to the other through sodium atoms,

which are each surrounded by two DME molecules in

an octahedral arrangement. The Fe(1)ꢁC(23) distance

,

(1.853(8) A) is slightly shorter than for other iron(II)

,

complexes (see [Na₄Fe(CN)₆]: 1.909 A (average value)

,

[18]; [H₄Fe(CN)₆]: 1.890 A [19]; [Na₂Fe(CN)₅NO]: 1.913

A) [20]. In contrast the C(23)ꢁN(5) distance of 1.167(10)

A in 8b is longer (see [Na₄Fe(CN)₆]: 1.163 A; [H₄Fe-

(CN)₆]: 1.149 A; [Na₂Fe(CN)₅NO]: 1.15 A). This is a re-

sult of the higher electronic density of 8 and its stronger

back-donation to the cyanide ligand. The out-of-plane

,

In contrast to CO, the stronger s-donor isonitriles

can selectively give mono- or bis-coordinated com-

plexes. When 1 was reacted with one equivalent of RNC

in THF, the resulting microcrystalline products of poor

solubility 5 (R=o-Me₃SiC₆H₄NC) and 6 (R=^tBuNC)

were isolated. Characteristic isonitrile IR bands were

,

distance of the iron from the N₄-plane is 0.266(1) A and

is in agreement with a d⁶conﬁguration. The tmtaa lig-

and, which shows the usual saddle shape conformation,

has structural parameters quite close to those reported

for the other iron complexes (Tables 2 and 3).

50

A. Klose et al. / Journal of Organometallic Chemistry 591 (1999) 45–62

saddle-shape conformation of the tmtaa ligand differs

from those observed in the previous complexes mainly

in the mutual orientation of the opposite NC₃N sys-

tems, which are pushed toward the cavity to form a

dihedral angle of 110.8(1)° versus 136.8(3) [137.7(3)]°

and 129.4(2)° in 2 and 8b, respectively (Table 3). The

,

FeꢁC(23) distance (2.084(4) A) is also remarkably

longer than those observed in 8b. The remarkable dif-

ferences observed between the structural parameters of

8 and 9, along with the moving of the CꢀN stretching

vibration from 2089 cm⁻¹in 9 to 1987 cm⁻¹in 8a, are

due to the different spin states occurring for the

monomeric and dimeric species (see below).

Fig. 3. ^SCHAKALview of the anion in complex 9.

The peculiar magnetic properties of 8a prompted us

2.3. Reacti6ity of [Fe(tmtaa)Na(THF)₃] (2), with

to prove if they were essentially associated to the

dimeric nature of the complex. In this light, we synthe-

sized the mononuclear cyanide complex 9, according to

the reaction in Scheme 1. Complex 9 displays a CꢀN

stretching band at 2089 cm⁻¹and magnetic properties

different from those of the dimer 8a (see below).

The structure of 9 consists of discrete [Fe(tm-

taa)(CN)]⁻anions (Fig. 3), [NⁿBu₄]⁺cations and THF

solvent molecules of crystallization in the stoichiometric

molar ratio of 1:1:2. Iron exhibits an almost regular

tetragonal pyramidal coordination, the base being

deﬁned by the four nitrogen atoms from the N₄core

and the apex by the C(23) carbon atom from the

cyanide ligand. The N₄core is nearly planar, the metal

carbon monoxide and carbon dioxide

2.3.1. Reaction with CO

There has been considerable interest in the coordina-

tion of carbon monoxide to iron and cobalt surrounded

by a macrocyclic ligand [11h,21]. In contrast, its reac-

tivity toward transition metal organic anions has re-

ceived very limited attention, probably due to the fact

that their isolation is difﬁcult. Few examples have been

reported and these generally involved normal coordina-

tion of CO [22]. An exception is seen with

[Co(salen)Na], which in the ﬁrst step of the reaction

formed [Co(salen)(CO)Na], and then decomposed to

other products [23a,b]. Another remarkable example is

the carbonylation of [Fe^I(TPP)]⁻carried out in situ and

analyzed with spectroscopic and electrochemical tools,

though isolation of any CO derivative has not been

mentioned [23c]. A rapid reaction was observed when 2

was reacted with CO in chilled (−30°C) THF. The

resulting product was very air sensitive. The CO IR

band is shifted from 1913 cm⁻¹in 3 to 1736 cm⁻¹in

11, thus suggesting that coordination had occurred on a

Fe(I) center. The remarkable back-donation from Fe(I)

(Scheme 3) is coupled with the formation of an iron–

carbon multiple bond. Such a decrease in the CO

stretching frequency is a further proof of the iron

oxidation state (I) in the [Fe(tmtaa)]⁻anion, and it

should be compared with that at much higher fre-

quency in [Fe(TPP)(CO)]⁻(1931 cm⁻¹) [23c]. The high

back-donation from iron(I) to carbon monoxide makes

the oxygen electron rich, thus suitable for binding the

Lewis acid Na⁺. This is particularly relevant in the

context of the so-called bifunctional activation of car-

bon monoxide by a simultaneous interaction with an

electron-rich and an electron-poor center. In addition,

as recently singled out in the reactivity and structural

study of Li₃[Co(CN)₅] [24], the alkali cation interacting

with the CN⁻ligand can be used for tuning the elec-

tron density at the transition metal center.

,

being displaced by 0.610(1) A (Table 3). The FeꢁC(23)

vector is nearly perpendicular to the N₄core, the dihe-

dral angle it forms with the normal to the N₄mean

plane being 2.7(1)°. The FeꢁN bond distances are not

signiﬁcantly different from each other (mean value

,

2.056(3) A) and are remarkably longer than those ob-

served in 8b, in agreement with the increased out-of-

plane distance of the metal from the N₄. The

The structure of 11 is displayed in Fig. 4. The dimers

crystallize with benzene solvent molecules in a stoichio-

Scheme 3.

A. Klose et al. / Journal of Organometallic Chemistry 591 (1999) 45–62

51

values usually observed for FeꢁCO (terminal CO) frag-

ments. Coordination around each sodium cation is

completed to distorted octahedral through the oxygen

atoms from two DME molecules. The magnetic proper-

ties of 11 have been analyzed as reported in the follow-

ing section.

2.3.2. Reaction with CO₂

The general strategies for ﬁxing and subsequently

activating CO₂involved the use of highly basic metals

and the utilization of bifunctional systems which con-

tain in their structure the metal in a low oxidation state

(basic center) along with an acidic center [25]. A num-

ber of metal carbonate complexes were formed on

reaction with CO₂. This can be explained by a reductive

disproportionation mechanism [25]. A second transfor-

mation was also observed and is the deoxygenation

process, which liberates CO and forms metal-oxo spe-

cies [25]. The reaction of 2 with CO₂is solvent depen-

dent (Scheme 3). When 2 was reacted with CO₂in

THF, two products were detected. The IR spectrum

showed clearly 3 having a sharp CO band at 1913

1

cm⁻¹. The H-NMR, however, showed the 1:1 mixture

of 1 and 3 and the precipitation of Na₂CO₃. In con-

trast, when 2 was reacted with CO₂in toluene, only 1

was isolated and sodium oxalate detected. The reaction

of 2 with CO₂is the ﬁrst example where the pathway

reduction of CO₂is a function of the solvent [25].

Fig. 4. ^SCHAKALview of complex 11b.

metric molar ratio of 1:5. The two independent tmtaa

ligands, for which the labeling shown in Chart 1 has

been applied, are referred to as molecule A and B,

respectively. Values referring to molecule B will be

hereafter given in square brackets. The coordination

environment of iron is a distorted tetragonal pyramid,

2.4. Carbene functionalities bonded to [Fe(tmtaa)]

The synthesis of the carbene derivative 12 was per-

formed by reacting 1 with the corresponding dia-

zoalkane. The reaction was carried out at −30°C in

THF and then at room temperature, and after workup

gave a good yield of 12 (Scheme 4).

,

with the metal displaced by 0.553(2) A [0.655(2) A]

from the N₄mean plane (Table 3). The FeꢁC(23) vector

forms a dihedral angle of 1.9(2) [1.6(2)]° with the

normal to the N₄mean plane. The FeꢁC(23) bond

Complex 12 is diamagnetic and easily characterized

1

by H- and ¹³C-NMR spectroscopies. An X-ray analy-

sis of this compound conﬁrmed the spectroscopic data

(Fig. 5). The coordination environment of iron is a

distorted tetragonal pyramid, unlike the Feꢁporphyrin

,

distance (1.752(5) [1.758(5)] A) falls in the lower end of

Fig. 5. SCHAKAL view of complex 12.

Scheme 4.

52

A. Klose et al. / Journal of Organometallic Chemistry 591 (1999) 45–62

The question as to whether the oxidation state of

iron in 12 is (II) or (IV) can be reasonably answered, in

the absence of Mo¨ssbauer measurements, by consider-

ing the Fe distance from the N₄plane. A correlation in

tmtaa complexes has been established between the out

of plane of the metal and its dⁿconﬁguration [2d]. All

the M(tmtaa) units presented show the saddle shape

conformation with differences depending principally on

the metal ion radius and on its oxidation state. The

observation is that for electron-rich metals (d⁶–d⁸), the

saddle shape conformation almost disappears and the

metals lie in the plane. Metals with d⁰and d¹conﬁgura-

tions, on the other hand, have the most pronounced

out-of-plane and saddle shape conformation. The value

,

found in the present case for 12 [0.335(1) A] is expected

6

,

for a low-spin d ﬁve-coordinate iron(II), i.e. 0.29 A in

[Fe(tmtaa)CO] [11h]. Much longer distances would be

found for a d⁴conﬁguration. The extended Hu¨ckel

calculations (see below) further support the +II oxida-

tion state for iron in 12 and the use of carbene rather

than alkylidene terminology for the present complex.

The carbene complex 12 showed a very high thermal

stability and even a high resistance to hydrolysis. In

order to test the metathesis reaction and polymeriza-

tion, we reacted 12 with benzophenone, dipheny-

lacetylene and norbornene. In all cases no reaction

occurred even under reﬂux conditions. Oxidation and

reduction reactions were then carried out, in order to

change the kinetically inert d⁶conﬁguration. When

carbene 12 was reacted with dioxygen in THF, the

poorly soluble m-oxo dimer 14 [11o] formed. A GC

analysis of the mother liquor shows the presence of

benzophenone. In contrast, the reduction of 12 with

sodium did not give a well-deﬁned product. The stabil-

ity and the nature of the ﬁnal compound derived from

the reaction of 1 with diazoalkanes is strongly depen-

dent on the substituents at the carbene carbon. In the

case of PhCHN₂, the carbene derivative 2 forms only at

low temperature. It decomposes at room temperature

leading to the starting material, trans-stilbene (81%)

and cis-stilbene (19%) as shown by GC analysis. In the

presence of dioxygen at low temperature, benzaldehyde

and the m-oxo dimer 14 are formed (Scheme 4), which

proves the formation of the intermediate carbene spe-

cies at low temperature.

Fig. 6. Magnetic susceptibilities (ꢀ) and magnetic moments ( ) as a

function of the temperature: (a) for complex 1; (b) for complex 7.

[9a] and Osꢁporphyrin [26] examples, where the metal is

always hexa-coordinate. The metal is displaced by

,

0.335(1) A from the N₄average plane. The FeꢁC(23)

vector is perpendicular to the N₄core, the dihedral

angle with the normal to the N₄plane being 1.0(1)°.

The carbene plane C(23),C(31),C(41) is almost parallel

to the N(2)···N(4) vector, the torsion angles

C(31)ꢁC(23)ꢁFeꢁN(4) and C(41)ꢁC(23)ꢁFeꢁN(2) being

−15.5(3) and −15.2(3)°, respectively. The Fe–C(23)

2.5. Magnetic properties

The magnetic susceptibilities of complexes 1–11 were

collected in the temperature range 1.9–300 K and those

of 1, 2, 7–9, and 11 are shown in Figs. 6–8.

,

bond distance [1.794(3) A] is particularly short com-

pared with the only available iron–porphyrin [FeꢂCCl₂,

,

1.83(3) A] [9a] or the organometallic derivatives of iron,

,

where it ranges from 1.978(3) to 1.85(3) A [27]. The

2.5.1. Iron(II) complexes

The d⁶iron(II) ion can exhibit three spin states, i.e.

the S=0 low-spin state, the S=1 intermediate-spin

state and the S=2 high-spin state. Although the inter-

tmtaa ligand, which shows the usual saddle shape con-

formation, has structural parameters close to those

reported for the other iron complexes (Tables 2 and 3).

A. Klose et al. / Journal of Organometallic Chemistry 591 (1999) 45–62

53

mediate spin cannot be observed in an ideal octahedral

environment, it has been observed for some tetragonal

or severely distorted pyramidal coordinations, see dis-

cussion below.

The temperature dependence of the magnetic mo-

ments of 1 is illustrated in Fig. 6(a). The magnetic

moment is almost constant between 100 and 300 K with

a room temperature value of 3.45v_Bat 298 K and

shows a sudden decrease below 50 K reaching 0.95v_Bat

2 K. The room temperature value of the effective

magnetic moment is consistent with an S=1 intermedi-

ate-spin state, while the decrease at low temperature

can be attributed to a large zero-ﬁeld splitting with a

non-magnetic level lying lowest. This behavior is quite

Fig. 8. Magnetic susceptibilities (ꢀ) and magnetic moments ( ) as a

function of the temperature: (a) for complex 2; (b) for complex 11a.

similar to that of other tetracoordinated Fe^IIcomplexes

with a square-planar coordination constituted by a N₄

macrocyclic ligand [28,29]. The room-temperature value

of 1 is very close to that reported for Fe(Pc) (Pc=ph-

thalocyanine), v_eff=3.9v_B[28], and slightly lower than

that of Fe(TPP) (TPP=tetraphenylporphyrin), v_eff

=

4.2v_B[29]. The magnetic data have been analyzed in

terms of an isolated S=1 state with an isotropic g-fac-

tor and an axial zero-ﬁeld splitting described by the

following spin hamiltonian [30]: H=igH·S+D[S²_z−

S(S+1)/3], where S=1 and D is the zero-ﬁeld splitting

constant. The spin degeneracy of the S=1 state is

partly lift by spin-orbit coupling into the M_s=0 and

M_s= 91 components, separated by the zero-ﬁeld split-

ting parameter D. The magnetic susceptibility can be

derived by using the Van-Vleck equation and is given

by [30]

Fig. 7. (a) Inverse magnetic susceptibility (ꢀ) and magnetic moment

(

) as a function of the temperature for complex 8a; (b) magnetic

susceptibility (ꢀ) and magnetic moment ( ) as a function of the

temperature for complex 9.

54

A. Klose et al. / Journal of Organometallic Chemistry 591 (1999) 45–62

Table 4

Summary of the magnetic properties of [Fe(tmtaa)(L)(L%)] complexes

L

L%

v_eff(2 K)

v_eff(298 K)

Spin state

Features

1

3

4

None

CO

None

THF

Py

0.95

0.70

–

3.45

3.10

–

1

Intermediate-spin

Spin-crossover

Diamagnetic

0l1

0

5

6

7

8b

9

CNR ^a

CN^tBu

CN⁻

Py

THF

CN^tBu

None

Py

–

0.55

0.60

3.20

0.50

–

2.55

3.95

5.80

2.40

0

Diamagnetic

Spin-crossover

0l1

0l1l2

1l2

0l1

10

^aR=o-Me₃SiC₆H₄.

Ng²v²_B2 exp(−x)+(2/x) exp(−x)

For the porphyrin complexes the spin state of the

iron(II) center is determined by the nature and the num-

ber of axial ligands: the coordination of strong-ﬁeld lig-

ands (Py, CO) leads to low-spin six-coordinate

complexes, while the coordination of weaker ﬁeld lig-

ands (THF, 2-MeIm) leads to high-spin ﬁve- or six-coor-

dinate complexes; the absence on any axial ligands leads

to a intermediate spin complex. Therefore, the decrease

of the axial ligand ﬁeld give sequentially low-spin, high-

spin and intermediate-spin states. These observations

have been interpreted in terms of ligand ﬁeld effects on

the d-orbital splitting [32], which for an iron porphyrin is

given by three nearly degenerate low-lying d_xy, d_xz, d_yz,

dim

=

kT

1−2 exp(−x)

where x=D/kT. A temperature-independent paramag-

netism (TIP) contribution has been added to account for

the slight decrease between 300 and 100 K. A good ﬁt

was found for g=2.22, D=42.5 cm⁻¹, and Nh=9×

10⁻⁴(see solid line in Fig. 6(a)).

The coordination of one or two axial ligands to 1

changes drastically its magnetic behavior. A summary of

the magnetic properties of polycrystalline samples of the

characterized [Fe(tmtaa)(L)(L%)] complexes is given in

Table 4. We see that most of these compounds are either

diamagnetic (4–6), although with a little residual para-

magnetism due to TIP contribution or to paramagnetic

impurities, or show magnetic moments which are com-

patible with only low or intermediate spin states (3, 7,

10). Only the cyanide complexes 8 and 9 show magnetic

moments compatible with the presence of some high-

spin character (see below).

Compounds 3, 7 and 10 show an uncommon magnetic

behavior which is exempliﬁed by the temperature depen-

dence of the magnetic moment for 7 illustrated in Fig.

6(b). The moment of 7 has a value of 2.55v_Bat 298 K,

which is slightly lower than that expected for an interme-

diate spin state, and shows a continuous decrease with

temperature reaching a value of 0.55 at 2 K. This behav-

ior is compatible with: (i) an intermediate-spin triplet

ground state with a large zero-ﬁeld splitting; (ii) the pres-

ence of a spin crossover involving a low-spin S=0 and

an intermediate spin S=1 states. Since the zero-ﬁeld in-

teraction necessary to cause a decrease of the magnetic

moment already at room temperature would be anoma-

lously too high, and it was not possible to ﬁt the mag-

netic data to either an axial or rhombic spin hamiltonian,

we favor the latter explanation, i.e. an S=0–1 spin-

crossover.

the high-lying d_x

₂whose lobes point towards the por-

2

phyrinato nitroge⁻n^yatoms, and the d_zwhose energy is

2

affected strongly by the axial ligand ﬁeld. In particular,

the high energy of the d_x

₂-orbital stabilizes the

2

0

(d )²(d )²(d , d )²(d

)

⁻i^yntermediate-spin state,

2

x

2

xy

xz

yz

z

while the strong depende⁻nc^ye of the d -orbital on the ax-

2

z

ial ligand ﬁeld may determine a low-spin or high-spin

ground state.

The main difference between the magnetic behavior of

the iron(II) porphyrins and our iron(II) tetraaza com-

plexes is the higher occurrence of the intermediate-spin

state, while the high-spin state is nearly absent, being ob-

served only in the cyanide compounds 8a and 9, and

never as ground state. Such an effect can be easily

rationalized on the basis of the different sizes of the

porphyrin and tetraaza macrocycle coordination cavity

[2m]. Indeed, the tetraaza system has only a 14-mem-

bered inner ring as compared to the larger 16-membered

,

porphyrin ring and this gives rise to a 1.90–1.93 A ring

,

size for tetraaza, which is ca. 0.1 A less than that for the

porphyrins. The small ring rises the d_x

₂-orbital en-

2

ergy well above the other four, thus fav⁻or^ying the low-

spin and intermediate-spin states over the high-spin

state.

It is interesting to compare the magnetic behavior of

these four-, ﬁve-, and six-coordinate iron(II) tetraaza

complexes with the related iron(II) porphyrin complexes,

whose structural and electronic features have been

probed in considerable depth in the past years [31,32].

The magnetic behavior of the Na⁺-bridged cyanide

dimer 8a is quite peculiar as illustrated by the tempera-

ture dependence of the magnetic moment in Fig. 7(a),

showing the occurrence of an almost complete double

A. Klose et al. / Journal of Organometallic Chemistry 591 (1999) 45–62

55

2.5.2. Iron(I) complexes

step spin transition (note the presence of a second

−1

Magnetic susceptibilities data for complex 2 are dis-

played in Fig. 8(a). At room temperature the magnetic

moment of 2 (4.15v_Bat 298 K) is almost consistent with

a S=3/2 high spin state, whereas the moment de-

creases at lower temperatures reaching a value of

2.04v_Bat 1.9 K, which is slightly greater than the spin

only value for a S=1/2 low spin state. This behavior is

quite different from that observed for the iron por-

phyrin anion [Fe(TPP)]⁻, which shows an isolated dou-

blet ground state [16a].

maximum in the

versus T plot). Such a magnetic

behavior is quite rare, and was ﬁrst reported in 1969 for

the iron(III) compound H[Fe(5-Br–thsa)₂] (thsa=2-

bromo-salicylaldheydethiosemicarbazone) [33] and only

in 1982 for the iron(II) complex [Fe(2-pic)₃Cl₂]×EtOH

(2-pic=2-picolylamine) [34]. Since then few rare exam-

ples of two-step spin conversions for iron(II) complexes

have been described [35–37] and only one involves a

dimeric system [36]. Two possible explanation have

been proposed for this behavior: (i) the occurrence of

the following two transitions LS–LSlLS–HSlHS–

HS where the two LS (low-spin) or HS (high-spin)

states refers to two non-equivalent sites of the lattice

[38] or to the two metal centers within the dimer [36a];

and (ii) the occurrence of a two-step transition involv-

ing the S=0, S=1 and S=2 spin states of the single

metal centers [37]. The former explanation have been

proposed in all two-steps spin conversions described to

date, except for the recently characterized [Fe(4,4%-

dpb)₂(NCS)₂] complex (4,4%-dpb=4,4%-diphenyl-2,2%-

bipyridine), whose magnetic behavior was interpreted

assuming the involvement of the S=1 state [37]. Al-

though the dimeric nature of 8a would suggest the

occurrence of a LS–LSlLS–HSlHS–HS transition,

the presence of a low-lying intermediate-spin state for

most of the characterized [Fe^II(tmtaa)(L)(L%)] complexes

could allow the more appealing S=0lS=1lS=2

interpretation, and a correct interpretation of the mag-

netic behavior of 8a would require further calorimetric

and spectroscopic investigations. The magnetic behav-

ior of the monomeric cyanide complex 9 is illustrated in

Fig. 7(b) and is indicative of a S=1lS=2 spin

transition, therefore supporting the involvement of the

S=1 state also in the dimeric complex 8a. The different

nature of the spin states involved in the monomeric and

dimeric cyanide complexes is reﬂected in the remark-

able differences of structural parameters and spectro-

scopic properties. Indeed, at variance with the dimer 8,

which shows a little high-spin character even at room

temperature, the monomeric complex 9 has a predomi-

nant high-spin character above 100 K and is expected

to present longer metal–ligand bond distances and a

higher out-of-plane in agreement with the experimental

evidence (see Tables 2 and 3). Moreover, the depopula-

tion occurring in the high-spin state of the d_p-orbitals,

responsible of the p-back-donation to CN⁻, explains

the remarkable higher stretching vibration observed for

9 (2089 cm⁻¹vs. 1987 in 8). The slight ligand ﬁeld

changes induced by the rupture of the dimeric structure

probably lead to the destabilization of the low-spin

state showing that the spin-state ordering of these

iron(II) tmtaa complexes is very sensitive to subtle

structural effects.

In an attempt to rationalize these magnetic data, a

few explanations can be given to interpret the anoma-

lous dependence of v_effon the temperature: (1) a high-

spin quartet ground state with weak antiferromagnetic

exchange within the solid state structure; (2) a large

zero-ﬁeld splitting of a high-spin quartet ground state

into two M_s= 93/2 and M_s= 91/2 levels; (3) a spin

conversion between a doublet and a quartet state; (4) a

doublet ground state admixed with a low-lying ther-

mally non-populated quartet. Explanation (1) can al-

ready be eliminated on the basis of the X-ray structure,

which shows no contact between neighboring spin cen-

ters. When we consider explanation (2) we should ﬁrst

note that the temperature dependence of the magnetic

moment is not typical of systems in which zero-ﬁeld

effects are observed. Zero-ﬁeld interactions have usu-

ally a magnitude of few cm⁻¹and, as a consequence,

v_effremains constant from room temperature to about

30–60 cm⁻¹showing a decrease only at lower tempera-

tures [30]. Although a large zero-ﬁeld splitting could be

induced by the interaction of the quartet ground state

with a low-lying excited state, it was not possible to ﬁt

the magnetic data to either an axial or rhombic spin

Hamiltonian, so that we can rule out this explanation.

The remaining explanation of the magnetic behavior is

either the existence of a spin-conversion between an

S=1/2 and an S=3/2 state, or a spin-admixed S=1/

2–3/2 state.

It is noteworthy that the magnetic behavior of com-

plex 2 is similar to that observed for a series of ﬁve-co-

ordinated mono-nitrosyl complexes of iron, formally

iron(I), with salen (salen=N,N%-ethylenebissalicyl-

ideneiminato) [39,40], or TMC (TMC=1,4,8,11-te-

tramethyl-1,4,8,11-tetraazacyclotetradecane) [40] ring-

substituted derivatives. These complexes have a mag-

netic moment at room temperature consistent with

three unpaired electrons: one of them (with unsubsti-

tuted salen) shows a sharp transition in the v_effversus

temperature curve at ca. 180 K [39], while in the other

cases the decrease of the magnetic moment with de-

creasing temperature was found to be more gradual and

qualitatively similar to that observed for 2 [40,41]. The

nature of the states involved in the proposed explana-

tions (S=1/2lS=3/2 spin conversion or S=1/2–3/2

spin admixed state) can be discussed on the basis of the

56

A. Klose et al. / Journal of Organometallic Chemistry 591 (1999) 45–62

qualitative molecular orbital diagram for [Fe(tmtaa)] in

justed to a C₂₆symmetry. Such a simpliﬁed model has

already been employed in EH [43] and SCF-X-SW

[12d,e] calculations and showed a good correlation with

the results obtained with the whole ligand [12d,e].

The molecular orbitals of the [Fe(tmtaa)] fragment

are built in a step-by-step approach. We ﬁrst considered

the planar (tmtaa)²⁻ligand of D_2hsymmetry, which

was then deformed so as to reproduce the geometry of

the (tmtaa) skeleton in the ﬁnal [Fe(tmtaa)(CPh₂)] com-

plex. The molecular orbitals for the planar (tmtaa)²⁻

ligand are reported in the ﬁrst column of Fig. 9 while in

the second one we illustrate the effect of the bending up

of the two benzo groups (19.8°) and the bending down

of the two allyl groups (26.7°). We ﬁnally add Fe in the

middle of the (tmtaa) ring and the resulting MO dia-

gram is reported in the third column of Fig. 9. Upon

coordination, the d-orbitals mix strongly with the lig-

and frontier orbitals so that no pure d-orbitals can be

assigned. However, ﬁve MOs with large metal d charac-

ter can be identiﬁed. These are the four highest occu-

Fig. 9, see extended Hu¨ckel analysis below. The low-

1

spin state is probably a (d )²(d )²(d )²(d

)

2

z

x

y

x

−y

conﬁguration, while the high-spin state could be either

a (d )²(d )²(d )¹(d

₂)¹(d_xy)¹, of iron(I) character,

₂)¹(a )¹, or (d )²(d )²-

2

z

xz

yz

x

−y

or

a

(d )²(d )²(d )¹(d

2

z

xz

yz

x

−y

2

xz

(d_yz)¹(d_x

₂)¹(b₂)¹of iron(II)-p radical anion charac-

2

−y

ter. Although the low energy of the a₂and b₂orbitals

(of p*(tmtaa) character) favors the latter alternative,

only further spectroscopic investigations and a theoreti-

cal study could clarify this point.

Magnetic susceptibility data for complex 11a are

illustrated in Fig. 8(b). The temperature dependence of

the magnetic moment is qualitatively similar to that of

2, but shows a larger decrease at low temperatures,

reaching a value of 1.1v_Bat 1.9 K. This latter value is

not consistent with the spin only value for an S=1/2

low-spin state of a magnetically isolated spin center,

and would suggest a small antiferromagnetic interac-

tion between the two iron ions within the dimeric unit

mediated by the –CꢀO–Na⁺–OꢀC– bridge.

pied orbitals a (d ), b (d ), b (d ) and a (d

₂).

2

1

z

2

yz

1

xz

1

x

−y

The a₂(d_xy), pointing more closely towards the nitrogen

atoms of tmtaa, is pushed very high in energy. The two

LUMOs correspond closely to the two LUMOs of the

(tmtaa)²⁻distorted ligand, b_1gand a_u.

2.6. Extended Hu¨ckel analysis on ironꢁtmtaa

complexes

The molecular orbitals of the [Fe(porphyrin)] frag-

ment have been the subject of several extended Hu¨ckel

investigations [44], and its molecular fragment are re-

ported on the extreme right of Fig. 9 for purpose of

comparison. We see that both [Fe(porphyrin)] and

[Fe(tmtaa)] fragments have a very similar d-orbital

splitting pattern, the main differences being: (i) the d_xz

and d_yzare no longer degenerate in the tmtaa fragment

(due to the lower symmetry), although still very close in

Extended Hu¨ckel calculations [42] were performed to

gain a better understanding of the main electronic and

magnetic properties of the iron(I) and iron(II) com-

plexes supported by the dibenzotetramethyltetraazaan-

nulene ligand and to elucidate the nature of the

bonding between a carbene unit and the central metal

in complexes 12 and 13. The tmtaa ligand has been

simpliﬁed by replacing the benzene units by ethylene

and the methyl groups by hydrogens, and slightly ad-

energy; (ii) the d_xyand the d_x

₂-orbitals are signiﬁ-

2

−y

cantly higher in energy in the tmtaa fragment, essen-

tially as a consequence of the smaller ring size. In

particular, this latter difference is responsible of the

stabilization of the intermediate-spin state observed for

the tmtaa complexes (see above).

We then considered the bonding in the carbene com-

plexes 12 and 13, using a CH₂as a model for the

carbene unit. The interaction between the [Fe(tmtaa)]

fragment and the CH₂unit is illustrated by the molecu-

lar orbital diagram for the [Fe(tmtaa)(CH₂)] complex in

Fig. 10. The molecular orbitals of the [Fe(tmtaa)] frag-

ment described above are reported on the left, while on

the right we report the frontier orbitals of CH₂, i.e. the

s-donor a₁(sp²hybrid) and the p-acceptor b₁(p_y). The

X-ray structure of the [Fe(tmtaa)CPh₂)] complex 12

shows that the carbene CPh₂plane forms an angle of

about 57° with the yz symmetry plane bisecting the

benzo groups. However, in order to keep the C₂₆sym-

metry of the metal fragment and simplify the analysis

of the results we considered a model orientation with

the CH₂plane lying in the yz symmetry plane.

Fig. 9. Orbital interaction diagram for [Fe(tmtaa)(CH₂)].

A. Klose et al. / Journal of Organometallic Chemistry 591 (1999) 45–62

57

b₂, which would be more typical of Schrock alkylide-

nes. Therefore, compounds 12–13 can be actually de-

scribed as intermediate between a Fischer carbene and

a Schrock alkylidene without a deﬁnite electrophilic or

nucleophilic character of the carbene carbon. The negli-

gible electron donation of the [Fe(tmtaa)] fragment to

the CH₂moiety upon carbene formation (0.1 e) sup-

ports the attribution a formal oxidation state of +2

and a d⁶conﬁguration for iron, the same as [Fe(tm-

taa)], based on geometrical evidences (see above).

3. Conclusions

The [Fe(tmtaa)] fragments showed a great versatility

in binding p-acid ligands from CO to RNC, CN and

carbenes, displaying a wide spectrum of magnetic be-

haviors, which depend on the coordination number of

the metal and the axial ligand. A remarkable difference

has been observed between the monomeric and the

sodium-bridged dimeric cyano complexes in terms of

magnetic behavior. This result emphasizes the primary

role of alkali cations in determining not only the molec-

ular complexity, but also the physical properties. A

unique structural model of diamagnetic iron(II)–car-

bene complexes has been reported. A quite rare d⁷-

iron(I)-macrocycle occurring in an ion-pair form has

been synthesized and structurally characterized and its

reactivity with CO and CO₂explored. In the former

case, a quite unique carbonyl dimer bridged by sodium

cation was identiﬁed, while in case of CO₂, for the ﬁrst

time, the solvent-dependence of the reduction of CO₂

has been singled out.

Fig. 10. Molecular orbitals for [Fe(tmtaa)] and [Fe(porphyrin)].

In principle, the p-acceptor orbital of CH₂could

interact with both the doubly occupied b₁(d_xz), b₂(d_yz)

metal orbitals leading to different orientations of the

carbene moiety with the HCH unit lying in the yz and

xz planes, respectively. The former orientation has been

calculated ca. 0.3 eV more stable, probably due to the

more favorable interaction of the carbene p-acceptor

orbital with the higher-lying 1b₁(d_xz) donor orbital and

is that reported in Fig. 10. The apparent disagreement

with the experimental X-ray structure can be rational-

ized by noting that the electronically most favorable

orientation of the carbene moiety would lead to strong

steric repulsion between the two bent benzo groups and

the carbene phenyl substituents. The observed orienta-

tion is, therefore, a compromise between electronic

factors favoring the orientation of the CPh₂in the yz

plane and steric repulsion favoring the orientation of

the carbene in the xz plane.

Fig. 10 shows a strong interaction between the

1a₁(d_z2) and the s-donor 1a₁of CH₂, and between the

1b₁(d_xz), and the p-acceptor b₁. These interactions are

typical of most of the known metal carbene species [45].

In particular, frontier orbital criteria suggest for this

[Fe(tmtaa)(CH₂)] species a Fischer carbene nature

rather than a Schrock alkylidene, supporting the used

terminology. This is supported by the low energy of the

metal d_p-orbitals, lower than that of the p-acceptor

b₂-orbitals, and the presence of a low-lying LUMO (see

Fig. 10), which gives a signiﬁcant electrophilic character

to the carbene moiety, both effects being characteristic

of Fischer carbene systems. However, the carbene unit

shows a small, negative charge on the carbon atom

(−0.05) and a signiﬁcant population of the p-acceptor

4. Experimental

4.1. General procedures

All reactions were carried out under an atmosphere

of puriﬁed nitrogen. Solvents were dried and distilled

before use by standard methods. Infrared spectra were

recorded with a Perkin–Elmer FT 1600 spectrophoto-

meter, UV–vis spectra were recorded with a Hewlett–

Packard 8452A diode array spectrophotometer, NMR

spectra were recorded on AC-200E and DPX-400

Bruker instrument. The synthesis of tmtaaH₂ligand has

been carried out according to the literature [46].

Magnetic susceptibility measurements were made on

an MPMS5 SQUID susceptometer (Quantum Design),

operating at a magnetic ﬁeld strength of 1 kOe. Correc-

tions were applied for diamagnetism calculated from

Pascal constants [47]. Effective magnetic moments were

calculated as v_eff=2.828(_FeT)^1/2were is the mag-

Fe

netic susceptibility per iron. Fitting of the magnetic

data to the theoretical expression were performed by

minimizing the agreement factor, deﬁned as

58

A. Klose et al. / Journal of Organometallic Chemistry 591 (1999) 45–62

[^obsdT_i−^calcdT_i]²

i

4.5. Synthesis of 4

%

(^obsdT_i)²

i

Pyridine (1.0 ml, 12.4 mmol) was added to a stirred

THF solution (100 ml) of 1·THF (1.53 g, 3.25 mmol).

After 15 min, the blue solution was saturated with CO

with a concomitant color change to red. Stirring was

maintained overnight and the solution then concen-

trated (20 ml). n-Hexane was then added (30 ml) and

the crude product ﬁltered and dried in vacuo (1.31 g,

80%). It was recrystallized in acetonitrile. (Found: C,

66.48; H, 5.23; N, 13.29%. 4, C₂₈H₂₇FeN₅O requires: C,

66.54; H, 5.38; N, 13.86%.) IR (Nujol, w_max/cm⁻¹):

1927 (s, CO), 1599 (w), 1571 (w), 1518 (s), 1402 (s),

1284 (m), 1199 (s), 1034 (s), 762 (s), 733 (m), 695 (w),

by a Levenberg–Marquardt routine.

4.2. Synthesis of 1

A solution of butyllithium (63 ml, 1.63 M, 102 mmol)

in n-hexane was added to a THF solution (550 ml) of

tmtaaH₂(17.6 g, 51.1 mmol) at room temperature (r.t.).

The reaction was exothermic and became red. After

stirring for 3 h, [FeCl₂×THF_1.5] (12.0 g, 51.1 mmol)

was added and then stirred overnight at −20°C. The

resulting purple microcrystalline solid was ﬁltered and

dried in vacuo (16.9 g, 71%). No extraction was needed

to remove the salt. (Found: C, 66.45; H, 6.52; N,

12.3%. 1·THF, C₂₆H₃₀FeN₄O requires C, 66.39; H,

6.43; N, 11.91%.) IR (Nujol, w_max/cm⁻¹): 1578 (w),

1508 (s), 1463 (s), 1398 (s), 1278 (m), 1198 (s), 1053 (w),

1034 (m), 920 (w), 787 (m), 772 (m), 749 (s), 535 (w).

¹H-NMR (C₅D₅N, 200 MHz, 298 K, ppm): l 6.45 (bs,

4H, Ar); 6.18 (bs, 4H, Ar); 3.48 (bs, 2H, CH); 1.59 (bs,

12H, CH₃). v_eff=3.45v_Bat 298 K.

1

525 (w). H-NMR (C₅D₅N, 200 MHz, 298 K, ppm): l

6.59–6.56 (m, 4H, Ar); 6.47–6.42 (m, 4H, Ar); 6.08–

6.04 (m, 0.5H); 4.93 (s, 0.13H); 4.14 (s, 2H, CH); 1.79

(s, 12H, CH₃); 1.41 (s, 2H). ¹³C-NMR (C₆D₆, 100.6

MHz, 298 K, ppm): l 235.2 (CO), 159.8, 153.7, 124.3,

123.4, 113.4, 26.2.

4.6. Synthesis of 5

(o-Me₃Si)PhNC (0.95 g, 4.92 mmol) was added via

syringe to a cooled (−25°C) THF solution (130 ml) of

1·THF (2.32 g, 4.92 mmol). A rapid color change was

observed and the solution was allowed to warm to r.t.

After stirring overnight, the solvent was reduced to 5

ml and n-hexane (30 ml) slowly added. The microcrys-

talline product was ﬁltered and dried in vacuo (2.38 g,

83%). (Found: C, 65.38; H, 5.94; N, 11.88%. 5,

C₃₂H₃₅FeN₅Osi requires: C, 65.19; H, 5.98; N, 11.88%.)

IR (Nujol, w_max/cm⁻¹): 2037 (s, NC), 1588 (w), 1519

(m), 1487 (m), 1291 (m), 1251 (m), 1196 (s), 1105 (w),

4.3. Synthesis of 2

Sodium sand (0.478 g, 20.8 mmol) was added to a

THF suspension (240 ml) of 1·THF (8.1 g, 17.2 mmol)

and the mixture was stirred for 10 h. The green solution

was warmed (ca. 40°C) to complete solubilization and

then ﬁltered. The solvent was reduced in vacuo to 20 ml

and n-hexane (20 ml) was then added. The black crys-

talline solid was ﬁltered and dried in vacuo (8.6 g,

78%). (Found: C, 64.26; H, 7.30; N, 8.68%. 2,

C₃₄H₄₆FeN₄NaO₃requires C, 64.05; H, 7.27; N, 8.79%.)

IR (Nujol, w_max/cm⁻¹): 1385 (s), 1340 (m), 1269 (m),

1

1032 (m), 915 (s), 844 (s), 746 (s), 595 (w). H-NMR

(C₆D₆, 200 MHz, 298 K, ppm): l 7.01–6.51 (m, 12H,

Ar); 4.85 (s, 2H, CH); 2.21 (s, 12H, CH₃); 0.09 (s, 9H,

Me₃). ¹³C-NMR (C₅D₅N, 100.6 MHz, 298 K, ppm): l

205.3 (CN), 158.3, 151.6, 147.8, 130.0, 126.3, 126.0,

121.9, 121.4, 121.2, 120.9, 108.2, 24.6.

1182 (w), 1049 (m), 823 (w), 737 (m), 588 (w). v_eff

4.15v_Bat 298 K.

=

4.4. Synthesis of 3

4.7. Synthesis of 6, Method A

A THF solution (150 ml) of 1×THF (2.49 g, 5.29

mmol) was saturated with CO at r.t. A sudden reaction

was observed and the solution became red. The solvent

volume was reduced to 10 ml and n-hexane (25 ml)

added. The crystalline material was ﬁltered and dried in

vacuo (2.0 g, 75%). (Found: C, 64.99; H, 6.02; N,

11.34%. 3, C₂₇H₃₀FeN₄O₂requires C, 65.07; H, 6.07; N,

11.27%.) IR (Nujol, w_max/cm⁻¹): 1913 (s, CO), 1577

(w), 1531 (m), 1378 (s), 1282 (w), 1198 (m), 1034 (m),

t

A THF solution (40 ml) of BuNC (0.38 g, 4.54

mmol) was added dropwise to a cooled (−30°C) THF

solution (150 ml) of 1·THF (2.14 g, 4.54 mmol). The

solution was allowed to reach r.t. and then stirred

overnight. A microcrystalline product of low solubility

precipitated. The solvent was reduced to 25 ml and the

solid ﬁltered, washed with n-hexane (30 ml) and dried

in vacuo (1.66 g, 66%). (Found: C, 67.52; H, 7.22; N,

13.25. 6, C₃₁H₃₉FeN₅O requires: C, 67.27; H, 7.10; N,

12.65%.) IR (Nujol, w_max/cm⁻¹): 1989 (s, NC), 1570

(w), 1538 (s), 1515 (m), 1430 (s), 1283 (m), 1194 (s),

1

785 (w), 740 (m). H-NMR (C₅D₅N, 200 MHz, 298 K,

ppm): l 6.59–6.56 (m, 4H, Ar); 6.47–6.43 (m, 4H, Ar);

6.01–5.99 (m, 0.4H); 4.14 (s, 2H, CH); 3.69 (m, 4H,

THF); 1.79 (s, 12H, CH₃); 1.58 (m, 4H, THF); 1.30 (s,

1.6H). v_eff=3.10v_Bat 298 K.

1

1033 (s). H-NMR (C₅D₅N, 200 MHz, 298 K, ppm): l

6.52–6.45 (m, 8H, Ar); 6.18 (bs, 0.5H); 4.95 (s, 0.6H);

A. Klose et al. / Journal of Organometallic Chemistry 591 (1999) 45–62

59

3.90 (s, 2H, CH); 3.63 (m, 4H, THF); 1.69 (s, 12H,

4.11. Synthesis of 8, Method B (reaction of 6 with

sodium)

t

CH₃); 1.58 (m, 4H, THF); 1.39 (s, 9H, Bu). ¹³C-NMR

(C₅D₅N, 100.6 MHz, 298 K, ppm): l 194.6 (CN), 155.2,

153.3, 120.7, 120.1, 119.5, 110.6, 32.0, 23.6.

Sodium (0.062 g, 2.66 mmol) was added to a THF

suspension (150 ml) of 6 (1.28 g, 2.32 mmol). Stirring

overnight resulted in a brown–red solution, which was

then ﬁltered, concentrated to 10 ml, and n-hexane (30

ml) added. The microcrystalline solid was ﬁltered and

dried in vacuo (0.58 g, 37%). The analysis showed the

product to be 8a (EA, IR, NMR, see above).

4.8. Synthesis of 6, Method B (reaction of 3 with

^tBuNC)

A THF solution (30 ml) of ^tBuNC (0.13 g, 1.6

mmol)was added dropwise to a red THF solution (100

ml) of 3 (0.73 g, 1.5 mmol). The solution slowly became

purple and was stirred for 12 h. The solvent was

reduced to 30 ml and n-hexane (25 ml) added. The

resulting crystalline product was ﬁltered and dried in

vacuo (0.51 g, 63%). The analysis showed the product

to be 6. IR and NMR spectra were identical to those

above.

4.12. Synthesis of 9

Bu₄N⁺CN⁻(2.40 g, 8.90 mmol) was added to a

THF solution (250 ml) of 1·THF (3.56 g, 7.60 mmol).

The solution was stirred overnight, then the solvent was

evaporated. A red crystalline product was obtained,

which was ﬁltered and dried in vacuo (4.51 g, 89%).

(Found: C, 69.56; H, 8.98; N, 10.27%. 9·2THF,

C₄₇H₇₄FeN₆O₂requires C, 69.54; H, 9.12; N, 10.36%.)

IR (Nujol, w_max/cm⁻¹): 2089 (m), 2011 (w), 1535 (m),

1506 (m), 1400 (s), 1269 (m), 1185 (s), 1026 (m), 922

(w), 734 (m). v_eff=5.80v_Bat 298 K.

4.9. Synthesis of 7

t

A THF solution (50 ml) of BuNC (0.90 g, 11.0

mmol) was added dropwise to a THF solution (150 ml)

of 1·THF (2.37 g, 5.04 mmol). At approximately half

the addition the crystalline 6 precipitated, then redis-

solved and was stirred for 6 h. Solvent was reduced to

20 ml and n-hexane (30 ml) added. The crystalline

material was ﬁltered and dried in vacuo (2.23 g, 81%).

(Found: C, 67.89; H, 7.36; N, 14.49%. 7, C₃₂H₄₀FeN₆

requires C, 68.08; H, 7.14; N, 14.89%). IR (Nujol,

w_max/cm⁻¹): 2156 (m), 2089 (s, CN), 1564 (w), 1535 (m),

4.13. Reaction of 2 with CO₂: (A) reaction in THF

A cooled (−30°C) THF solution (100 ml) of 2 (0.93

g, 1.46 mmol) was saturated with CO₂. A sudden color

change to red was observed. The solution was allowed

to warm to r.t. and stirred for 3 h. The viscous Na₂CO₃

precipitate was ﬁltered and the solution concentrated to

5 ml. n-Hexane (20 ml) was added after which the

microcrystalline product was ﬁltered and dried in vacuo

(0.525 g). The ¹H-NMR showed a 1/1 mixture of

1·THF (a) and 3 (b) and the IR spectrum was identical

to the latter. IR (Nujol, w_max/cm⁻¹): 1913 (s), 1578 (w),

1531 (s), 1281 (m), 1198 (s), 1034 (s), 786 (w), 749 (s).

¹H-NMR (C₅D₅N, 200 MHz, 298 K, ppm): l 6.62–6.57

(m, 4H, Ar, b); 6.50–6.43 (m, 8H, Ar, a+b); 6.25–6.20

(m, 4H, Ar, a); 4.15 (s, 2H, CH, b); 3.76 (s, 2H, CH, a);

3.64 (m, 4H, THF); 1.80 (s, 12H, CH₃, b); 1.66 (s, 12H,

CH₃, a); 1.59 (m, 4H, THF). [a=Fe(tmtaa) and b=

Fe(tmtaa)CO].

1

1407 (s), 1283 (w), 1190 (s), 1032 (m), 746 (m). H-

NMR (C₆D₆, 400 MHz, 298 K, ppm): l 6.47–6.44 (m,

4H, Ar); 6.39–6.37 (m, 4H, Ar); 4.19 (s, 2H, CH); 1.71

t

(s, 12H, CH₃); 1.32 (s, 18H, Bu). ¹³C-NMR (C₆D₆,

100.6 MHz, 298 K, ppm): l 178.9 (CN), 155.4, 153.5,

120.3, 119.9, 110.8, 56.5, 31.3, 23.3. v_eff=2.55v_Bat 298

K.

4.10. Synthesis of 8, Method A

Degassed NaCN (0.14 g, 2.80 mmol) was added to a

THF solution (120 ml) of 1·THF (1.68 g, 2.72 mmol).

The solution was allowed to reach reﬂux conditions

during which time the NaCN suspension dissolved (ca.

15 h). The solution was then ﬁltered and solvent re-

moved in vacuo to 10 ml. n-Hexane (12 ml) was added

and the microcrystalline solid ﬁltered and dried in

vacuo (1.75 g, 97%). (Found: C, 63.02; H, 7.24; N,

10.10%. 8a, C₃₅H₄₆FeN₅NaO₃requires C, 63.35; H,

6.99; N, 10.55%.) IR (Nujol, w_max/cm⁻¹): 1987 (m, CN),

1531 (m), 1511 (m), 1401 (s), 1278 (m), 1189 (s), 1052

4.14. Reaction of 2 with CO₂: (B) reaction in toluene

A cooled (−30°C) toluene solution (120 ml) of 2

(1.39 g, 2.18 mmol) was saturated with CO₂. A sudden

color change to violet–red was observed. The solution

was allowed to warm to r.t. and stirred overnight.

Solvent was reduced to 30 ml and the microcrystalline

product collected and dried in vacuo (0.53 g, 61%). The

1

(m), 1030 (m), 918 (w), 744 (s). H-NMR (C₅D₅N, 200

MHz, 298 K, ppm): l 6.53–6.49 (m, 4H, Ar); 6.30–6.25

(m, 4H, Ar); 3.91 (s, 2H, CH); 3.67 (m, 8H, THF); 1.71

(s, 12H, CH₃); 1.6 (m, 8H, THF). v_eff=3.95v_Bat 298

K. Crystals suitable for X-ray analysis were grown in

DME, as 8b·DME.

1

IR and H-NMR spectra were identical to 1. (Found:

C, 66.59; H, 5.31; N, 14.01%. C₂₂H₂₂N₄Fe requires C,

66.34; H, 5.57; N, 14.07%.) IR (Nujol, w_max/cm⁻¹):

1577 (w), 1508 (m), 1278 (m), 1198 (m), 1034 (m), 787

60

A. Klose et al. / Journal of Organometallic Chemistry 591 (1999) 45–62

(w), 749 (s). ¹H-NMR (C₅D₅N, 200 MHz, 298 K, ppm):

l 6.46 (bs, 4H, Ar); 6.21 (bs, 4H, Ar); 3.50 (bs, 2H,

CH); 1.61 (s, 12H, CH₃). In order to detect the presence

of sodium oxalate, the mother liquor was extracted

with water (2×40 ml), separated and ﬁltered. Acti-

vated carbon was added to the brown solution and

stirred overnight. Filtration gave a pale brown solution,

which was titrated with KMnO₄(0.02 M).

4.18. Reaction of 12 with O₂

Dry oxygen was allowed to diffuse slowly from an

adjoining round-bottom ﬂask to a THF solution (100

ml) of 12 (0.99 g, 1.8 mmol). After 2 days a brown solid

precipitated (0.61 g, 86%), which analyzed as [Fe(tm-

taa)]₂O [11o]. A GC analysis of the mother liquor

showed the presence of benzophenone.

4.15. Synthesis of 11, Method A

4.19. Synthesis of 13

A cooled (−30°C) THF solution (350 ml) of 2 (6.19

g, 9.71 mmol) was saturated with CO. The solution

rapidly turned brown. After stirring overnight under

CO, the solution was concentrated (40 ml) and n-hex-

ane (30 ml) added. The crystalline product was then

ﬁltered and dried in vacuo (4.44 g, 77%). (Found: C,

62.62; H, 6.84; N, 9.07%. 11a, C₆₂H₇₆Fe₂N₈Na₂O₆re-

Freshly prepared phenyldiazomethane (0.63 g, 5.4

mmol) was added to a cooled (−30°C) THF solution

(150 ml) of 1·THF (2.14 g, 4.55 mmol). The solution

rapidly turned brown with the concomitant evolution

of nitrogen gas. The solution was allowed to warm to

r.t. and then stirred overnight. The solvent was re-

moved in vacuo (10 ml) and n-hexane (20 ml) was

added. The purple product (1.03 g, 49%) analyzed as

[Fe(tmtaa)] (IR, NMR). A GC analysis of the mother

liquor showed the presence of cis- and trans-stilbene

(19 and 81%, respectively).

quires C, 62.74; H, 6.45; N, 9.44%.) IR (Nujol, w_max

/

cm⁻¹): 1918 (w), 1736 (s, CO), 1536 (m), 1410 (s), 1277

(w), 1188 (s), 1048 (m), 1027 (m), 906 (w), 743 (m).

v_eff=3.90v_Bat 298 K. Crystals for X-ray analysis were

obtained from DME/toluene as 11b·6C₆H₆.

4.20. Reaction of 13 with O₂

4.16. Synthesis of 11, Method B (reaction of (3) with

sodium)

Freshly prepared phenyldiazomethane (0.15 g, 1.26

mmol) was added to a cooled (−50°C) THF solution

(80 ml) of 1·THF (0.49 g, 1.03 mmol). The solution

rapidly became brown with the concomitant evolution

of nitrogen gas. The solution was stirred for 10 min at

this temperature and was then saturated with oxygen.

After 10 min the solution was allowed to warm to r.t.

and stirred overnight. A brown solid precipitated,

which analyzed as [Fe(tmtaa)]₂O [11o]. A GC analysis

of the mother liquor showed the presence of benzalde-

hyde and no traces of stilbene.

Sodium sand (0.11 g, 4.7 mmol) was added to a THF

solution (150 ml) of 3 (1.9 g, 3.8 mmol). After 30 min,

the solution became brown. After stirring overnight, the

excess sodium was removed by ﬁltration, and the

ﬁltrate concentrated to 10 ml. n-Hexane (30 ml) was

then added dropwise, and the black microcrystalline

solid ﬁltered and dried in vacuo (1.49 g, 66%). The

product analyzed as 11a (EA, IR, NMR, see above).

4.17. Synthesis of 12

A THF solution (40 ml) of diphenyldiazomethane

(0.78 g, 4.0 mmol) was added dropwise to a cooled

(−30°C) THF solution (140 ml) of 1·THF (1.79 g, 3.8

mmol). When warmed to r.t., the solution became red.

Nitrogen gas was slowly given off. After stirring for 2

days, the green solution was concentrated in vacuo (20

ml) and then n-hexane (15 ml) was added. A green

crystalline material was obtained (1.7 g, 75%). (Found:

C, 74.53; H, 5.77; N, 9.91%. 12, C₃₅H₃₂FeN₄requires

C, 74.47; H, 5.71; N, 9.92%.) IR (Nujol, w_max/cm⁻¹):

1576 (w), 1515 (m), 1386 (s), 1307 (w), 1279 (w), 1194

(m), 1028 (m), 745 (s), 696 (m), 667 (w). ¹H-NMR

(C₆D₆, 200 MHz, 298 K, ppm): l 6.88–6.76 (m, 8H,

Ar); 6.35–6.33 (m, 10H, Ph); 5.13 (s, 2H, CH); 2.12 (s,

12H, CH₃). ¹³C-NMR (C₆D₆, 100.6 MHz, 298 K, ppm):

l 313.2 (C carbene), 166.6, 157.9, 149.3, 127.1, 124.8,

120.9, 120.5, 119.7, 107.3, 67.8, 25.8, 23.2. Crystals

suitable for X-ray analysis were grown in toluene and

contain one molecule of toluene.

4.21. X-ray experimental

Single crystals suitable for X-ray diffraction were

grown from common organic solvents (Table 1). Data

for 2 were collected on a MAR345 image plate, for 8, 9,

12 on a Rigaku AFC6S diffractometer, while those for

11 were collected on a Kuma CCD using Mo–K_a

radiation. The solutions and reﬁnements were carried

out using the programs SHELX-76 [48] and SHELXL-93

[49].

5. Supplementary material

Crystallographic data (excluding structure factors)

for the structures reported in this paper have been

deposited with the Cambridge Crystallographic Data

Centre as supplementary publication nos. CCDC-

125018 for 2, CCDC-125019 for 8, CCDC-125020 for 9,

A. Klose et al. / Journal of Organometallic Chemistry 591 (1999) 45–62

61

Villa, C. Rizzoli, J. Am. Chem. Soc. 120 (1998) 437. (f) L.

Giannini, E. Solari, C. Floriani, A. Chiesi-Villa, C. Rizzoli, J.

Am. Chem. Soc. 120 (1998) 823.

CCDC-125021 for 11, CCDC-125022 for 12. Copies of

the data can be obtained free of charge from The

Director, CCDC, 12 Union Road, Cambridge CB2

1EZ, UK (fax: +44-1223-336-033; e-mail: de-

posit@ccdc.cam.ac.uk or http://www.ccdc.cam.ac.uk).

[5] For metal–porphyrinogen: (a) C. Floriani, Pure Appl. Chem. 68

(1996) 1. (b) D. Jacoby, S. Isoz, C. Floriani, A. Chiesi-Villa, C.

Rizzoli, J. Am. Chem. Soc. 117 (1995) 2805. (c) S. Isoz, C.

Floriani, K. Schenk, A. Chiesi-Villa, C. Rizzoli, Organometallics

15 (1996) 337. (d) C.M. Kretz, E. Gallo, E. Solari, C. Floriani,

A. Chiesi-Villa, C. Rizzoli, J. Am. Chem. Soc. 116 (1994) 10775.

(e) S. De Angelis, E. Solari, C. Floriani, A. Chiesi-Villa, C.

Rizzoli, Angew. Chem. Int. Ed. Engl. 34 (1995) 1092. (f) S. De

Angelis, E. Solari, C. Floriani, A. Chiesi-Villa, C. Rizzoli,

Organometallics 14 (1995) 4505. (g) D. Jacoby, S. Isoz, C.

Floriani, K. Schenk, A. Chiesi-Villa, C. Rizzoli, Organometallics

14 (1995) 4816. (h) G. Solari, E. Solari, C. Floriani, A. Chiesi-

Villa, C. Rizzoli, Organometallics 16 (1997) 508.

Acknowledgements

We thank the ‘Fonds National Suisse de la

Recherche Scientiﬁque’ (Bern, Switzerland, grant no.

20-53336.98), and Fondation Herbette (University of

Lausanne, N.R.) for ﬁnancial support.

[6] For metal–Schiff base complexes: (a) S. Gambarotta, M. Maz-

zanti, C. Floriani, A. Chiesi-Villa, C. Guastini, J. Chem. Soc.

Chem. Commun. (1985) 829. (b) C. Floriani, E. Solari, F.

Corazza, A. Chiesi-Villa, C. Guastini, Angew. Chem. Int. Ed.

Engl. 28 (1989) 64. (c) F. Corazza, E. Solari, C. Floriani, A.

Chiesi-Villa, C. Guastini, J. Chem. Soc. Dalton Trans. (1990)

1335. (d) J.-M. Rosset, C. Floriani, M. Mazzanti, A. Chiesi-

Villa, C. Guastini, Inorg. Chem. 29 (1990) 3991. (e) E. Solari, C.

Floriani, A. Chiesi-Villa, C. Rizzoli, J. Chem. Soc. Dalton

Trans. (1992) 367.

[7] For metal tropocoronand complexes, see: (a) M.J. Scott, S.J.

Lippard, J. Am. Chem. Soc. 119 (1997) 3411. (b) M.J. Scott, S.J.

Lippard, Organometallics 16 (1997) 5857. (c) M.J. Scott, S.J.

Lippard, Organometallics 17 (1998) 466. (d) M.J. Scott, S.J.

Lippard, Organometallics 17 (1998) 1769. (e) K.J. Franz, S.J.

Lippard, J. Am. Chem. Soc. 120 (1998) 9034.

References

[1] For leading references on organometallic chemistry based on

macrocyclic ligands, see Refs. 2–7.

[2] For tmtaa [dibenzotetramethyltetraaza[14]annulene], see: (a)

V.L. Goedken, J.A. Ladd, J. Chem. Soc. Chem. Commun.

(1981) 910 and (1982) 142. (b) C. Floriani, S. Ciurli, A. Chiesi-

Villa, C. Guastini, Angew. Chem. Int. Ed. Engl. 26 (1987) 70. (c)

E. Solari, S. De Angelis, C. Floriani, A. Chiesi-Villa, C. Rizzoli,

Inorg. Chem. 31 (1992) 96. (d) S. De Angelis, E. Solari, E. Gallo,

C. Floriani, A. Chiesi-Villa, C. Rizzoli, Inorg. Chem. 31 (1992)

2520. (e) L. Giannini, E. Solari, C. Floriani, A. Chiesi-Villa, C.

Rizzoli, Angew. Chem. Int. Ed. Engl. 33 (1994) 2204. (f) L.

Giannini, E. Solari, S. De Angelis, T.R. Ward, C. Floriani, A.

Chiesi-Villa, C. Rizzoli, J. Am. Chem. Soc. 117 (1995) 5801. (g)

D.G. Black, D.C. Swenson, R.F. Jordan, Organometallics 14

(1995) 3539. (h) A.J. Blake, P. Mountford, G.I. Nikonov, D.

Swallow, Chem. Commun. (Cambridge) (1996) 1835. (i) H.

Schumann, Inorg. Chem. 35 (1996) 1808. (j) G.I. Nikonov, A.J.

Blake, P. Mountford, Inorg. Chem. 36 (1997) 1107. (k) D.G.

Black, R.F. Jordan, R.D. Rogers, Inorg. Chem. 36 (1997) 103.

(l) A. Martin, R. Uhrhammer, T.G. Gardner, R.F. Jordan,

Organometallics 17 (1998) 382. (m) P. Mountford, Chem. Soc.

Rev. 27 (1998) 105.

[3] For porphyrins, see: (a) H. Brand, J. Arnold, Angew. Chem. Int.

Ed. Engl. 33 (1994) 95. (b) H. Brand, J. Arnold, Organometallics

12 (1993) 3655. (c) H.-J. Kim, D. Whang, K. Kim, Y. Do, Inorg.

Chem. 32 (1993) 360. (d) J. Arnold, S.E. Johnson, C.B. Knobler,

M.F. Hawthorne, J. Am. Chem. Soc. 114 (1992) 3996. (e) H.

Brand, J. Arnold, J. Am. Chem. Soc. 114 (1992) 2266. (f) H.

Brand, J. Arnold, Coord. Chem. Rev. 140 (1995) 137. (g) T.

Aida, S. Inoue, Acc. Chem. Res. 29 (1996) 39. (h) J.P. Collman,

C.E. Barnes, P.N. Swepston, J.A. Ibers, J. Am. Chem. Soc. 106

(1984) 3500. (i) J.P. Collman, P.J. Brothers, L. McElwee-White,

E. Rose, L.J. Wright, J. Am. Chem. Soc. 107 (1985) 4570. (j) J.P.

Collman, P.J. Brothers, L. McElwee-White, E. Rose, J. Am.

Chem. Soc. 107 (1985) 6110.

[8] (a) R.D.W. Kemmitt, D.R. Russell, in: G. Wilkinson, F.G.A.

Stone, E.W. Abel (Eds.), Comprehensive Organometallic Chem-

istry, vol. 5, Pergamon, Oxford, UK, 1982, p. 81. (b) J.M. Pratt,

J.P. Craig, Adv. Organomet. Chem. 11 (1973) 331.

[9] (a) D. Mansuy, M. Lange, J.C. Chottard, J.F. Bartoli, B.

Chevrier, R. Weiss, Angew. Chem. 90 (1978) 828. (b) D. Man-

suy, Pure Appl. Chem. 52 (1980) 681. (c) L. Latos-Grazynski,

R.-J. Cheng, G.N. La Mar, A.L. Balch, J. Am. Chem. Soc. 103

(1981) 4270. (d) A.L. Balch, R.-J. Cheng, G.N. La Mar, L.

Latos-Grazynski, Inorg. Chem. 24 (1985) 2651. (e) P.J. Brothers,

J.P. Collman, Acc. Chem. Res. 19 (1986) 209. (f) D. Mansuy,

Pure Appl. Chem. 59 (1987) 759. (g) I. Artaud, N. Gregoire,

J.-P. Battioni, D. Dupre, D. Mansuy, J. Am. Chem. Soc. 110

(1988) 8714. (h) I. Artaud, N. Gregoire, P. Leduc, D. Mansuy, J.

Am. Chem. Soc. 112 (1990) 6899. (i) J.R. Wolf, C.G. Hamaker,

J.-P. Djukic, T. Kodadek, L.K. Woo, J. Am. Chem. Soc. 117

(1995) 9194. (j) C.J. Ziegler, K.S. Suslick, J. Am. Chem. Soc. 118

(1996) 5306. (k) D. Mansuy, J.P. Mahy, in: F. Montanari, L.

Casella (Eds.), Metalloporphyrins Catalyzed Oxidations,

Kluwer, Dordrecht, The Netherlands, 1994, pp. 175–206.

[10] G. Wilkinson, R.D. Gillard, J.A. McCleverty (Eds.), Compre-

hensive Coordination Chemistry, (a) Ch. 45, and (b) Ch. 48, vol.

4, Pergamon, Oxford, UK, 1987.

[11] (a) M.C. Weiss, B. Bursten, S.M. Peng, V.L. Goedken, J. Am.

Chem. Soc. 98 (1976) 8021. (b) M. Weiss, G. Gordon, V.L.

Goedken, Inorg. Chem. 16 (1977) 305. (c) L.F. Warren, V.L.

Goedken, J. Chem. Soc. Chem. Commun. (1978) 909. (d) G.

Gordon, P.W. De Haven, M.C. Weiss, V.L. Goedken, J. Am.

Chem. Soc. 100 (1978) 1003. (e) P.W. De Haven, V.L. Goedken,

Inorg. Chem. 18 (1979) 827. (f) M. Tsutsui, R.L. Bobstein, G.

Cash, R. Pettersen, Inorg. Chem. 18 (1979) 758. (g) V.L. Goed-

ken, J.J. Pluth, S.M. Peng, B. Bursten, J. Am. Chem. Soc. 98

(1976) 8014. (h) V.L. Goedken, S.M. Peng, J.A. Molin-Norris,

Y. Park, J. Am. Chem. Soc. 98 (1976) 8391. (i) M.C. Weiss, V.L.

Goedken, Inorg. Chem. 18 (1979) 819. (j) V.L. Goedken, S.M.

[4] For calix[4]arenes, see: (a) L. Giannini, E. Solari, A. Zanotti-

Gerosa, C. Floriani, A. Chiesi-Villa, C. Rizzoli, Angew. Chem.

Int. Ed. Engl. 35 (1996) 85. Angew. Chem. Int. Ed. Engl. 35

(1996) 2825; Angew. Chem., Int. Ed. Engl. 36 (1997) 753. (b) B.

Castellano, A. Zanotti-Gerosa, E. Solari, C. Floriani, A. Chiesi-

Villa, C. Rizzoli, Organometallics 15 (1996) 4894. (c) L. Gian-

nini, A. Caselli, E. Solari, C. Floriani, A. Chiesi-Villa, C. Rizzoli,

N. Re, A. Sgamellotti, J. Am. Chem. Soc. 119 (1997) 9198 and

9709. (d) A. Caselli, L. Giannini, E. Solari, C. Floriani, N. Re,

A. Chiesi-Villa, C. Rizzoli, Organometallics 16 (1997) 5457. (e)

A. Zanotti-Gerosa, E. Solari, L. Giannini, C. Floriani, A. Chiesi-

62

A. Klose et al. / Journal of Organometallic Chemistry 591 (1999) 45–62

Peng, Y. Park, J. Am. Chem. Soc. 96 (1974) 284. (k) V.L.

(1993) 9858. (f) M.A. Esteruelas, F.J. Lahoz, E. On˜ate, L.A.

Oro, B. Zeier, Organometallics 13 (1994) 4258. (g) A. Hafner,

P.A. van der Schaaf, A. Mu¨hlebach, Chimia 50 (1996) 131.

[28] C.G. Barraclough, R.L. Martin, S. Mitra, R.C. Sherwood, J.

Chem. Phys. 53 (1970) 1643.

[29] P.W.D. Boyd, D.A. Buckingham, R.F. McMeeking, S. Mitra,

Inorg. Chem. 18 (1979) 3585.

[30] (a) O. Kahn, Molecular Magnetism, VCH, New York, 1993. (b)

R.L. Carlin, Magnetochemistry, Springer, Berlin, 1986. (c) C.J.

O’Connor, Prog. Inorg. Chem. 29 (1982) 203.

[31] W.R. Scheidt, C.A. Reed, Chem. Rev. 81 (1981) 543.

[32] W.R. Scheidt, M. Gouterman, in: H.B. Gray, A.B.P. Lever

(Eds.), Iron Porphyrins, Addison-Wesley, Reading, MA, 1982, p.

89.

[33] (a) E.V. Ivanov, Thesis, Institute of Technical Physics, Moscow,

Russia, 1969. (b) V.V. Zelentsov, Sov. Sci. Rev. B Chem. 81

(1981) 543.

[34] H. Ko¨ppen, E.W. Mu¨ller, C.P. Ko¨ler, H. Spiering, E. Meissner,

P. Gu¨tlich, Chem. Phys. Lett. 91 (1982) 348.

[35] (a) V. Petrouleas, J.P. Tuchagues, Chem. Phys. Lett. 137 (1987)

21. (b) E. Ko¨nig, G. Ritter, J. Dengler, S.M. Nelson, Inorg.

Chem. 26 (1987) 3582. (c) J.-A. Real, J. Zarembowitch, O. Kahn,

X. Solans, Inorg. Chem. 26 (1987) 2939.

[36] (a) J.-A. Real, H. Bolvin, A. Bousseksou, A. Dvorkin, O. Kahn,

F. Varret, J. Zarembowitch, J. Am. Chem. Soc. 114 (1992) 4650.

(b) J.-A. Real, I. Castro, A. Bousseksou, M. Verdaguer, R.

Burriel, M. Castro, J. Linares, F. Varret, Inorg. Chem. 36 (1997)

455.

[37] D.C. Figg, R.H. Herber, I. Felner, Inorg. Chem. 30 (1991) 2535.

[38] (a) V.V. Zelentsov, G.I. Lapouchkin, S.S. Sobolev, V.I. Shipilov,

Dokl. Akad. Nauk. 289 (1986) 393. (b) N. Sasaki, T. Kambara,

Phys. Rev. B 40 (1989) 2442.

[39] R.A. Earnshaw, E.A. King, L.F. Larkworthy, J. Chem. Soc. A

(1959) 2459.

[40] K.D. Hodges, R.G. Woollmann, S.L. Kessel, D.N. Hendrickson,

D.G. Van Derveer, E.K. Bareﬁeld, J. Am. Chem. Soc. 101 (1979)

906.

[41] F.V. Wells, S.W. McCann, H.H. Wickman, S.L. Kessel, D.N.

Hendrickson, R.D. Feltham, Inorg. Chem. 21 (1982) 2306.

[42] (a) R. Hoffmann, W.N. Lipscomb, J. Chem. Phys. 36 (1962)

2179. (b) R. Hoffmann, J. Chem. Phys. 39 (1963) 1397.

[43] (a) K. Tatsumi, R. Hoffmann, Inorg. Chem. 20 (1981) 3771. (b)

L. Giannini, E. Solari, S. De Angelis, T.R. Ward, C. Floriani, A.

Chiesi-Villa, C. Rizzoli, J. Am. Chem. Soc. 117 (1995) 5801.

[44] K. Tatsumi, R. Hoffmann, J. Am. Chem. Soc. 103 (1981) 3328

and Refs. therein.

Goedken, Y.A. Park, J. Chem. Soc. Chem. Commun. (1975)

214. (l) V.L. Goedken, S.M. Peng, J. Chem. Soc. Chem. Com-

mun. (1975) 258. (m) P. Berno, C. Floriani, A. Chiesi-Villa, C.

Guastini, J. Chem. Soc. Dalton. Trans. (1989) 551. (n) P. Berno,

C. Floriani, A. Chiesi-Villa, C. Guastini, J. Chem. Soc. Dalton.

Trans. (1988) 1409. (o) M.C. Weiss, V.L. Goedken, Inorg.

Chem. 18 (1979) 819.

[12] (a) F.A. Cotton, J. Czuchajowska, Polyhedron 9 (1990) 2553. (b)

F.A. Cotton, J. Czuchajowska, Polyhedron 9 (1990) 1221. (c)

F.A. Cotton, J. Czuchajowska, J. Am. Chem. Soc. 113 (1991)

3427. (d) F.A. Cotton, J. Czuchajowska, X. Feng, Inorg. Chem.

29 (1990) 4329. (e) F.A. Cotton, J. Czuchajowska, L.R. Falvello,

X. Feng, Inorg. Chim. Acta 172 (1990) 135.

[13] S. Ciurli, E.M. Meyer, C. Floriani, A. Chiesi-Villa, C. Guastini,

J. Chem. Soc. Chem. Commun. (1987) 281.

[14] A. Klose, E. Solari, C. Floriani, N. Re, A. Chiesi-Villa, C.

Rizzoli, Chem. Commun. (Cambridge) (1997) 2297.

[15] (a) G. Fachinetti, C. Floriani, F. Zanazzi, A.R. Zanzari, Inorg.

Chem. 18 (1979) 3469. (b) S. Gambarotta, M. Mazzanti, C.

Floriani, M. Zehnder, J. Chem. Soc. Chem. Commun. (1984)

1116. (c) S. Gambarotta, F. Urso, C. Floriani, A. Chiesi-Villa,

C. Guastini, Inorg. Chem. 22 (1983) 3966.

[16] (a) T. Mashiko, C.A. Reed, K.J. Haller, W.R. Scheidt, Inorg.

Chem. 23 (1984) 3192. (b) S. Ciurli, S. Gambarotta, C. Floriani,

A. Chiesi-Villa, C. Guastini, Angew. Chem. Int. Ed. Engl. 25

(1986) 553. (c) G.S. Srivatsa, D.T. Sawyer, N.J. Boldt, D.F.

Bocian, Inorg. Chem. 24 (1985) 2123.

[17] G.R. Eaton, S.S. Eaton, J. Am. Chem. Soc. 97 (1975) 235.

[18] (a) A. Tullberg, N.G. Vannerberg, Acta Chem. Scand. Ser. A 28

(1974) 551. (b) H.J. Buser, D. Schwarzenbach, W. Petter, A.

Ludi, Inorg. Chem. 16 (1977) 2704.

[19] M. Pierrot, R. Kern, R. Weiss, Acta Crystallogr. 20 (1966) 425.

[20] P.T. Manoharan, W.C. Hamilton, Inorg. Chem. 2 (1963) 1043.

[21] (a) S.M. Peng, J.A. Ibers, J. Am. Chem. Soc. 98 (1976) 8032. (b)

W.R. Scheidt, K.J. Haller, M. Fons, T. Mashiko, C.A. Reed,

Biochemistry 20 (1981) 3653. (c) C. Caron, A. Mitschler, G.

Riviere, L. Ricard, M. Schappacher, R. Weiss, J. Am. Chem.

Soc. 101 (1979) 7401.

[22] (a) V.L. Goedken, S.-M. Peng, J. Chem. Soc., Chem. Commun.

(1974) 914. (b) G.N. Schrauzer, J.H. Weber, T.M. Beckham, J.

Am. Chem. Soc. 92 (1970) 7078. (c) G. Costa, G. Mestroni, G.

Tauzher, J. Chem. Soc. Dalton Trans. (1972) 450.

[23] (a) G. Fachinetti, C. Floriani, P.F. Zanazzi, A.R. Zanzari, Inorg.

Chem. 17 (1978) 3002. (b) C. Floriani, G. Fachinetti, Transition

Met. Chem. 1 (1976) 50. (c) A. Croisy, D. Lexa, M. Momenteau,

J.M. Saveant, Organometallics 4 (1985) 1574.

[45] P. Hofmann, Electronic structures of transition metal complexes,

in: K.H. Do¨tz, H. Fischer, P. Hofmann, F.R. Kreissl, U. Schu-

bert, K. Weiss (Eds.), Transition Metal Carbene Complexes,

VCH, Weinheim, 1983, p. 113.

[46] (a) E.G. Ja¨ger, Z. Chem. 4 (1964) 437. (b) E.G. Ja¨ger, Z. Anorg.

Allg. Chem. 364 (1969) 177.

[47] E.A. Boudreaux, L.N. Mulay, Theory and Applications of

Molecular Paramagnetism, Wiley, New York, 1976, pp. 491–

495.

[48] G.M. Sheldrick, ^SHELX-76, Program for crystal structure deter-

mination, University of Cambridge, Cambridge, UK, 1976.

[49] G.M. Sheldrick, ^SHELXL-93, Program for crystal structure reﬁne-

ment, University of Go¨ttingen, Go¨ttingen, Germany, 1993.

[24] D. Ramprasad, G.P. Pez, B.H. Toby, T.J. Markley, R.M. Pearl-

stein, J. Am. Chem. Soc. 117 (1995) 10694.

[25] (a) A. Behr, Carbon Dioxide Activation by Metal Complexes,

VCH, Weinheim, Germany, 1988. (b) D.H. Gibson, Chem. Rev.

96 (1996) 2063.

[26] J.-P. Djukic, D.A. Smith, V.G. Young Jr., L.K. Woo,

Organometallics 13 (1994) 3020.

[27] (a) P. Schwab, R.H. Grubbs, J.W. Ziller, J. Am. Chem. Soc. 118

(1996) 100. (b) S.J. Miller, H.E. Blackwell, R.H. Grubbs, J. Am.

Chem. Soc. 118 (1996) 9606. (c) B. Mohr, D.M. Lynn, R.H.

Grubbs, Organometallics 15 (1996) 4317. (d) R.H. Grubbs, S.J.

Miller, G.C. Fu, Acc. Chem. Res. 28 (1995) 446. (e) S.T.

Nguyen, R.H. Grubbs, J.W. Ziller, J. Am. Chem. Soc. 115

.

Article Doi

The metal-carbon multiple bond in iron(I)- and iron(II)-dibenzotetramethyltetra[14]azaannulene: Carbene, carbonyl, and isocyanide derivatives

DOI: 10.1016/S0022-328X(99)00354-X

Source and publish data:

Authors:

Article abstract of DOI:10.1016/S0022-328X(99)00354-X

Full text of DOI:10.1016/S0022-328X(99)00354-X

Products guided by the article

R&D Labs maybe for 256648-41-6

Relevant to this article

Hot Product