The relationship between the structure and magnetic properties of bioinspired iron(II/III) complexes with Schiff-base-like chelate ligands, part I: Complexes with dianionic [N4] macrocycles

FULL PAPER

The Relationship between the Structure and Magnetic Properties of Bioinspired

Iron(^II/III) Complexes with Schiff-Base-Like Chelate Ligands, Part I:

Complexes with Dianionic [N₄] Macrocycles

Birgit Weber^[a][‡]Indira Käpplinger,^[a]Helmar Görls,^[a]and Ernst-G. Jäger*^[a]

Keywords: Bioinorganic chemistry / Iron / Macrocycles / Magnetic properties / Schiff bases

The molecular structures and the spin ground-state of the

iron(II/III) complexes Fe1, Fe2 and Fe3, which contain dian-

ionic macrocyclic [N₄] ligands derived from substituted ace-

tylacetaldehydes and 1,2-diamines, have been investigated

by X-ray analysis and temperature-dependent susceptibility

measurements, respectively. The iron(II) complexes Fe^II1

(which is dimeric by intermolecular coordination of one pe-

ripheral carbonyl group), Fe^II1MeOH, (Fe^IIL)₂dabco (L = 1,

antiferromagnetic interactions (shortest Fe–Fe distance of

3.4 Å). A new pair of octahedral iron(II/III) complexes with an

N-heterocyclic axial ligand, Fe^II3(Py)₂/[Fe^III3(Py)₂]PF₆, could

be crystallised. In contrast to the previously described pairs

Fe^II1(Him)₂/[Fe^III1(Him)₂]PF₆and Fe^II2Py₂/[Fe^III2Py₂]ClO₄,

the orientation of the planes of the axial ligands is nearly

independent of the oxidation step of the central atom. Octa-

hedral derivatives with biologically relevant anions as axial

ligands have been crystallised for the first time, namely

2, 3; dabco = 1,4-diazabicyclo[2.2.2]octane) and the iron(III

)

complex Fe^III1Cl are pentacoordinate and have an interme-

diate-spin ground-state at room temperature (S = 1 for Fe^II;

S = 3/2 for Fe^III). The intermediate-spin state was confirmed

by DFT-MO calculations for Fe^II1MeOH and Fe^III1Cl and is

also in agreement with the Mössbauer data of Fe^II1 and ESR

measurements of Fe^III1Cl. The iron centre in Fe^II2 is nearly

square-planar and shows a strong decrease of the magnetic

moment below T Ϸ 250 K. This is probably due to an S =

0 to S = 1 spin-crossover or, more likely, to intermolecular

[BzEt₃N][Fe^III1(NCS)₂], [Et₄N][Fe^III2(CN)₂](H₂O)_0.5

,

Na-

[Fe^III2(NO₂)₂](MeOH)₂(H₂O)_0.5and [Fe^III2(NO₂)OH₂](MeOH).

The latter shows a magnetic moment of μ_effϷ 2.85 μ_Bat room

temperature, which decreases to 1.85 μ_Bat lower tempera-

ture. This indicates an incomplete S = 1/2 to S = 3/2 spin-

crossover, which is probably due to the presence of H₂O as

a weak axial ligand.

Germany, 2005)

complexes with tetradentate [N₂O₂]-coordinated dianions

of N,N-bridged salicylaldimines (e.g. “salen”) have been in-

vestigated extensively, and in part quite successfully, as

models for the active sites of haem proteins. For instance,

the reversible uptake of dioxygen − the main function of

haemoglobin − was realized with simple “(salen)Co” much

earlier^[3]than with (porphyrin)iron compounds detached

from the protein; the first catalysts of technical relevance

for cytochrome-P450-like oxygenation reactions were de-

rived from (salen)Mn.^[4]

During the last decades we have created a pool of chelate

complexes^[5]with ligand systems that can be subdivided

into the three basic structures given in Scheme 1. All these

compounds are structural or functional models of active

sites of metalloenzymes: Type I, which is formally most re-

lated to porphyrins, contains tetradentate twofold nega-

tively charged macrocyclic ligands with an [N₄] donor set

and a more or less extended π-system. [The special ligand

Me₄TAA, first published in 1969 as its nickel complex

(Me₄TAA)Ni^[6]and today known in complexes with nearly

all transition elements and many main group metals,^[7]has

been characterised by F. A. Cotton: “The latter bears many

similarities to porphyrinato systems, but its chemical behav-

Introduction

Iron is known to be a ubiquitous biometal with a great

variety of metabolic functions.^[1]Many iron-containing pro-

teins are derived from haem with the porphyrinato macro-

cycle as a typical dianionic equatorial [N₄] ligand. The

unique biological role of (porphyrin)iron compounds may

be one reason for the unbroken interest in this important

class of molecules over nearly a century. This is reflected in

the countless number of papers and many extensive reviews,

monographs and handbooks.^[2]Compared with the porphy-

rins, including such variations as expanded porphyrins, con-

fused porphyrins etc., there are only a few other examples

of similar [N₄] macrocycles that can act as dianionic equa-

torial ligands. Besides the synthetic porphyrins and their

tetrabenzotetraaza derivatives, the phthalocyanines and

[a] Institute of Inorganic and Analytical Chemistry, University of

Jena,

August-Bebel-Strasse 2, 07743 Jena, Germany

Fax: +49-3641-442716

E-mail:cej@uni-jena.de.

[‡] Present address: Ludwig-Maximilians-Universität München,

Department Chemie,

Butenandtstraße 5–13 (Haus D), 81377 München, Germany

2794

DOI: 10.1002/ejic.200400759

Eur. J. Inorg. Chem. 2005, 2794–2811

Complexes with Dianionic [N₄] Macrocycles

FULL PAPER

iour seems to be even more diverse.”^[7b]] Type II is derived characterised as a square-planar intermediate-spin (S = 1)

from dianionic open-chain ligands with an [N₂O₂] coordi- complex^[15a]that gives an octahedral diadduct, probably

nation sphere (“salen” is a special case with a fused benzene with a high-spin (S = 2) ground state, in pyridine solution.

ring for R¹+ R²). Type III is derived from tridentate di- No addition of imidazole − a very characteristic reaction

anions; the coordinatively highly unsaturated formula unit of (porphyrin)iron compounds − could be observed. Even

promotes the formation of oligonuclear structures^[8]similar in our earlier reports^[17]we accentuated the high tendency

to those found in biological systems. Previous investigations of the carbonyl-substituted derivative Fe1 to give octahedral

on cobalt, copper and nickel complexes of types I, II and low-spin diadducts, especially with imidazole. Many such

III have already shown that the variation of the N,NЈ-brid- derivatives have been isolated as solids and could be charac-

ges X, Y and the substituents R^1–3can be used to control terised by X-ray structure analysis and susceptibility mea-

properties such as the electronic state of the central atom,^[9]surements.^[18]The existence of others (also with biologically

the redox potentials,^{[10,11a,11d,11e]}the affinity of vacant axial relevant anions such as nitrite and sulfite^[19b]) could be

coordination sites towards additional ligands,^[11]as well as proved in solution by spectrophotometric titration.^[19]In

the binding and activation of dioxygen,^[5b,12]carbon diox- this respect, the carbonyl-substituted derivative Fe1 is more

ide^[13]or nitric oxide.^[14]These results indicated that the closely related to porphyrins than the parent compound.

macrocycles I and, in some respect, also the complexes II On the other hand, several pentacoordinate iron(iii) halides

share many common characteristics with porphyrins, but of type I have been described as intermediate-spin (S = 3/2)

marginal ligand variations can dramatically change the be- complexes,^[15,20]whereas the derivative Fe(Me₄TAA)Cl has

haviour. This situation is particularly typical of the chemis- a high-spin (S = 5/2) ground state, just as most halides of

try of the iron complexes.

(porphyrin)iron(iii) compounds. The formation of various

pentacoordinate halides and pseudo-halides Fe1X (X = F,

–

Cl, Br, I, N₃, SCN^–) has been monitored by spectrophoto-

metric equilibrium studies in two-phase systems,^[21]but

their spin state is uncertain in some cases. In contrast, all

the pentacoordinate nitrosyl derivatives FeLNO (L = 1, 2,

3) have an S = 1/2 ground state, just as the analogous (por-

phyrin)iron compounds.^[22]

The existence of coordinatively unsaturated species that

are able to reversibly bind additional axial ligands, as well

as the possibilities of changing the oxidation number and

the spin state of the central atom, are key features of many

haem proteins. For instance, the iron(ii) atom in deoxy-

genated haemoglobin has an effective coordination number

of five as the bonding to the nitrogen atom of the imidazole

ring of the “proximal” histidine group causes a strong dis-

placement from the [N₄] plane of the equatorial porphyrin;

the ground state is high spin (S = 2). After oxygenation, the

central atom is best described as six-coordinate low-spin (S

= 1/2) iron(iii) in a weakly distorted octahedral environ-

ment, antiferromagnetically coupled with the unpaired elec-

tron of an axially coordinated superoxide radical ion. For

comparison with our macrocycles we are therefore mainly

interested in structural and magnetic information as a func-

tion of the coordination number and the oxidation state of

iron in these complexes.

Scheme 1. Formulae of basic chelate complexes studied in our labo-

ratory (I–III) and formulae of macrocyclic iron complexes used in

this work (Fe1–Fe3). The arabic numbers represent the dianion of

the macrocyclic ligand H₂L (L^2–= 1, 2, 3). Fe1a denotes the more

hydrophilic variant of Fe1 with acetyl (COMe) instead of the ester

groups COOEt. The abbreviation Me₄TAA (“tetramethyldibenzo-

tetraaza[14]annulene”) is used for L^2–in type I with X = Y = o-

phenylene, R¹= R³= Me, and R²= H. The basic macrocycles FeL

and the axial ligands [MeOH, dabco (μ-1,4-diazabicyclo[2.2.2]-

octane), Py (pyridine), N-MeIm (N-methylimidazole), I^–etc.] are

given in bold (with the exception of the tables); all other compo-

nents of the formula units or within the lattice are given in plain

characters.

This paper is focused on (i) the spin state of complexes

with coordination number less than 6, (ii) the dependence

of the orientation of axial N-heterocycles in octahedral di-

adducts on the oxidation state of the central atom, and (iii)

attempts to isolate adducts with biologically relevant anions

as solids for structural and magnetic studies.

Results and Discussion

Syntheses and General Characterisation

Busch et al.,^[15a,15b]Holm et al.^[15c,15d]and Goedken et

al.^[16]have described iron(ii⁄iii) complexes of type I with R²

In contrast to the analogous nickel and copper com-

= H or Ph. The parent compound of Fe1 (R²= H) has been pounds, the iron complexes of type I cannot be synthesised

Eur. J. Inorg. Chem. 2005, 2794–2811

www.eurjic.org

2795

B. Weber, I. Käpplinger, H. Görls, E.-G. Jäger

FULL PAPER

by a direct metal-templated condensation of the precursor

The iron(iii) complexes are best prepared starting from

complexes of type II with diamines. The best synthetic the iron(ii) derivatives by in-situ oxidation with iodine in

method is the reaction of an iron(ii) salt, chiefly the acetate, the presence of an excess of sodium iodide to promote the

with the free ligands H₂1, H₂2 and H₂3, respectively, which precipitation of the iodide FeLI.^[20,21]The latter is a good

can be obtained by demetallation of the copper(ii) com- precursor for a large range of iron(iii) derivatives with dif-

plexes. The synthesis of the basic iron(ii) complexes ferent axial ligands.^[5c,18]Carbon tetrachloride is a suitable

Fe1,^[17b,17c,21]Fe2^[10d]and Fe3^[23]has been described pre- Cl donor to yield the chloride derivatives FeLCl. All the

viously. However, since all these publications are in German iron(iii) complexes are relatively air-stable but water-sensi-

in mainly non-international, and in part no longer available, tive, especially in the presence of primary or secondary

journals, and also because some steps of the ligand synthe- amines (piperidine!) or phosphonic acid esters as axial li-

ses have been modified to prevent the use of gaseous hydro- gands. Here, hydrolysis and/or radical redox processes oc-

gen sulfide, we will give a summary of the syntheses in the cur. All complexes are therefore best handled under dry ar-

Experimental Section.

gon.

Adducts of the iron(ii) complexes with axial ligands were

obtained by adding an excess of the corresponding ligand

to the reaction mixture during the synthesis of the iron(ii)

complex. The products precipitated upon slow cooling of

the reaction mixture. In some cases mono- and/or diadducts

were formed depending on the concentrations and the tem-

Intermediate-Spin Complexes with Coordination Numbers

4, 4+1 and 5

Crystals suitable for X-ray structure analysis were ob-

perature regime.^[17b,17c]Diffusion procedures are better tained for the solvent-free iron(ii) complexes Fe1 and Fe2.

suited for diadduct formation starting from the isolated Only one stable solvate with an O-donor, Fe1MeOH, could

parent complexes. The air sensitivity of the iron(ii) com- be crystallised. The previously described crystalline 1:1 sol-

plexes is strongly dependent on the type of additional axial vates of Fe1 with the N-donors ammonia or pyridine^[17b]

ligands and decreases in general from Fe1 to Fe3. This se- decompose even at room temperature under dry argon.

quence is in agreement with the redox potentials of the Attempts were made with the alternative N-donor 1,4-diaza-

complexes.^[5c,10d]

bicyclo[2.2.2]octane (dabco) and, in all cases, good crystals

Table 1. Spin state (S), room-temperature magnetic moment (μ_eff) and bond lengths [Å] and angles [°] within the first coordination sphere

of macrocyclic iron(ii/iii) complexes with coordination numbers (CN) 4, 4+1 and 5.

[a]

[b]

Complex

CN

S

Fe–N_1,2

Fe–N_3,4

Average

L_eq

Fe–L_ax

Fe–pl_eq

Average

Shortest

Fe–Fe

μ_eff[μ_B]

N–Fe–N^[c]N–Fe–N^[d]

Fe^II1

4+1

5

1

3.7

1

3.3

1

1.877(2)

1.879(3)

1.879(2)

1.906(2)

1.87

1.86

1.93

1.907(2)

1.908(2)

1.915(2)

1.906(2)

1.90

1.93

1.95

1.893

1.902

1.89

2.702

2.329(2)

–

0.077

0.110

0.03

93.1(1)

92.8(1)

94

87.0(1)

86.8(1)

86

6.502

Fe^II1MeOH

Fe^II2^[e]Fe(1)

Fe(2)

7.098

4

ca. 3.0

3.345

4.503

1.91

–

0.04

93

87

1.86

1.92

(Fe^II1)₂dabco

(Fe^II2)₂dabco

(Fe^II3)₂dabco

Fe^III1Cl

5

4

5

1

3.3

1

3.0

1

3.1

3/2

3.8

3/2

4.2

1/2

1.8

1

1.886(2)

1.888(2)

1.895(2)

1.899(2)

1.894(5)

1.896(5)

1.909(5)

1.912(5)

1.892(3)

1.884(3)

1,928(2)

1.895(2)

1.916

1.913(2)

1.914(2)

1.917(2)

1.912(2)

1.933(5)

1.927(5)

1.920(5)

1.948(5)

1.925(3)

1.924(3)

1.920(2)

1.954(2)

1.922

1.900

1.906

1.913

1.922

1.906

1.924

1.918

1.927

2.002

2.338(2)

2.326(2)

2.327(5)

2.262(4)

2.637(1)

1.732(2)

–

0.197

0.158

0.181

0.405

0.34

92.8(1)

93.4(1)

94.3

86.1(1)

85.8(1)

84.7(2)

85.2(2)

85.4(2)

5.244^[f]

7.285^[g]

5.279^[f]

7.249^[g]

4.211^[f]

7.274^[g]

6.217

90.2(2)

90.8(2)

Fe^III1I

7.172

Fe^III1NO

0.376

Fe^IIMe₄TAA

Fe^II(Me₄TAA)CO

Fe^III(Me₄TAA)Cl

–0.114

0.292

0.596

95.5

94.8

89.8

84.1

82.6

80.0

5.909

5.917

7.485

1.917

1.918

1.933

2.003

1.916

1.926

1.931

2.006

1.693

5/2

6.0

2.250

1.995

2.005

[a] Designation according to Figure 1. [b] Displacement of the iron atom from the main plane of the equatorial N-donor atoms. [c] Five-

membered rings. [d] Six-membered rings. [e] The quality of the data is not sufficient and we are only publishing the conformation of the

molecule, although distances and angles are similar to those of the other complexes. [f] Shortest distance between dinuclear units in the

lattice. [g] Distance between dabco-bridged Fe atoms.

2796

www.eurjic.org

Eur. J. Inorg. Chem. 2005, 2794–2811

Complexes with Dianionic [N₄] Macrocycles

FULL PAPER

of the bridged complexes (FeL)₂dabco (L = 1, 2, 3) were data of Fe1I,^[20]Fe1NO,^[22]Fe(Me₄TAA),^[16a]Fe(Me₄TAA)

obtained. With imidazole and its derivatives the octahedral CO^[16c]and Fe(Me₄TAA)Cl^[16b]are included for compari-

diadducts FeL(HIm)₂are strongly favoured. Sub-stoichio- son. The atom numbering in the table corresponds to that

metric amounts of imidazole result in the formation of given in Figure 1. Figure 2 shows the molecular structures

polymeric species.^[17b,17c]In contrast, several iron(ii) com- of selected examples, if possible approximately in the same

plexes of the general type II give mononuclear mono- and projection of the basic chelate ring.

diadducts with imidazole.^[5a]Additionally to the known

At first sight, the new data show few characteristic differ-

pentacoordinate iron(iii) complex Fe^III1I with axial io- ences. The room-temperature magnetic moments are in sat-

dide,^[20]all the chlorides Fe^IIILCl were isolated as solids, isfactory agreement with former measurements using a

but only Fe^III1Cl gave crystals of sufficient quality. Selected Gouy balance^[17b]for Fe1 (3.38 μ_B), Fe1MeOH (3.37 μ_B),

parameters of the molecular structures and the magnetic Fe1NH₃(3.12 μ_B), Fe1Py (3.08 μ_B), and Fe1I (4.07 μ_B).^[17c]

moments at room temperature are listed in Table 1. The These values are all higher to a greater or lesser degree than

the spin-only value of 2.83 μ_Bfor two (S = 1) and 3.87 μ_B

for three (S = 3/2) unpaired electrons, but all of them are

within the region expected for intermediate-spin iron(ii) and

iron(iii) complexes.^[15a](The reason for the increased values

is probably a contribution of an orbital magnetic moment,

due to an unsymmetrical population of the d_xzand d_yzor-

bitals, see also Figure 7.)

The iron atom has an approximately square-planar coor-

dination sphere in Fe2. The central atom in all the other

complexes is pentacoordinate with an approximately

square-pyramidal geometry − in Fe1 by intermolecular co-

ordination of one of the peripheral carbonyl groups (CN =

Figure 1. Unified atom numbering scheme used in Tables 1 and 3

4+1).The bond lengths and angles within the first coordina-

for the equatorial donor atoms.

Figure 2. Molecule structures and coordination numbers CN of macrocyclic iron complexes: (a) Fe^II2 (CN = 4), (b) Fe^II1 (CN = 4+1),

(c) Fe^II1MeOH (CN = 5), (d) (Fe^II3)₂dabco (CN = 5; the open lines represent the bonds of the 1:1 disorder), (e) Fe^III1Cl (CN = 5).

Ethoxy groups are omitted for clarity.

Eur. J. Inorg. Chem. 2005, 2794–2811

www.eurjic.org

2797

B. Weber, I. Käpplinger, H. Görls, E.-G. Jäger

FULL PAPER

tion sphere, including those of the octahedral complexes de-

scribed in the next sections (cf. Table 3), reflect the rules

formerly elaborated for a large number of our macrocyclic

iron complexes:^[5a]the Fe–N distances within the equatorial

plane are, for low-spin and intermediate-spin complexes, ne-

arly independent of the type of equatorial ligand L^2–, the

coordination number and the oxidation state of the central

atom [average of 16 significant mean values from Tables 1

and 3 of 191(1) pm]. Larger differences exist between the

Fe–N distances on the non-methylated side (N1, N2) and

the methylated side (N3, N4) of the six-membered chelate

rings [average of 32 significant values in each case: 189(1)

and 192(1) pm, respectively]. The distances to the axial li-

gands are always longer than those within the equatorial

plane. This is in contrast to the conditions in the case of

many derivatives of (porphyrin)iron compounds and is ob-

viously due to the higher effective field strength, which is

accompanied by shorter equatorial Fe–N distances, of the

ligands L^2–(L = 1, 2, 3). The displacement of the central

atom from the best equatorial [N₄] plane in pentacoordinate

complexes is more obvious in the presence of relatively

strong axial ligands with short distances to the iron atom

(Cl^–Ͼ dabco Ͼ MeOH Ͼ COOEt) but there is no clear

correlation. Interestingly, a small but significant displace-

ment is also observed for the iron atom in Fe^II2, which is

directed towards the neighbouring iron atom in the asym-

metric unit (Fe–Fe distance of 335 pm).

Figure 3. Temperature dependence of the effective magnetic mo-

ment of selected macrocyclic iron complexes: Fe2 (CN = 4), Fe1

(CN = 4+1), Fe1MeOH, Fe1Cl, (Fe2)₂dabco, (Fe3)₂dabco (all CN

= 5). All measurements were carried out at 0.2 and 0.5 T.

Two selected prototypes of complexes with a quite dif-

ferent behaviour are included in Table 1. The first one, the

nitrosyl derivative Fe1NO, which we interpreted as

(Fe^III)⁺NO^–,^[22a]has a temperature-independent S = 1/2

spin ground state. The axial Fe–N distance of 173 pm is

shorter than the equatorial Fe–N bonds, which are approxi- magnetic behaviour. Selected examples are depicted in Fig-

mately in the same range as for the other complexes. These ure 3.

deviations can be explained by assuming a very strong li-

With the exception of Fe1Cl, the magnetic moment de-

gand field of the axial NO^–ligand. The other one, the creases at lower temperature. Fe2 shows a characteristic sig-

pentacoordinate chloride complex Fe(Me₄TAA)Cl, which moid curve progression which is apparent even at room

has a magnetic moment of 5.95 μ_B,^[16b]is undoubtedly a temperature (Figure 3). This peculiarity can be explained as

high-spin iron(iii) complex and resembles the behaviour of an incomplete and smooth transition into the low-spin S

the pentacoordinate halides of (porphyrin)iron compounds. = 0 state, accompanied by antiferromagnetic interactions

This is also reflected in an average Fe–N distance of over between two iron atoms (as mentioned above, the lattice of

200 pm that is typical for high-spin iron(ii⁄iii) complexes Fe2 consists of pairs of two molecules with an iron–iron

such as those of the general type II.^[5a,22]Taking into ac- distance of 335 pm, which is short enough for direct interac-

count that the similar complex Fe3^IIICl is clearly an inter- tions). The relatively steep decrease of the magnetic mo-

mediate-spin complex (μ_eff= 4.20 μ_Bat room temperature; ment of several complexes at T Ͻ 20 K (left side of Fig-

structure data unfortunately not available), the difference ure 3) is not quite clear because the shortest Fe–Fe distance

indicates a lower field strength of the ligand Me₄TAA. The in the lattice is quite long (for Fe1 and Fe1MeOH Ͼ

reason for this could be the stronger “saddle-shaped” dis- 650 pm). In the case of the dinuclear iron(ii) complex

tortion of the macrocycle, caused by the two additional (Fe3)₂dabco the decrease of the magnetic moment starts al-

methyl groups in R³, and/or the absence of the peripheral ready at T Ͻ 150 K. The Fe–Fe distances within the dinu-

carbonyl group in R², which increases the π-acceptor clear unit are nearly identical for all (FeL)₂dabco derivatives

strength of the ligand and the effective charge of the central (729, 725, and 727 pm for L = 1, 2, and 3, respectively), but

atom in the other complexes. In agreement with this idea, the Fe–Fe distances between the dinuclear units in the lat-

Fe(Me₄TAA) forms derivatives with carbon monoxide and tice are significantly shorter, especially for (Fe3)₂dabco (524,

other strong π-acceptors as axial ligand much more easily 528, and 421 pm for L = 1, 2, and 3, respectively). This

than Fe3.^[16c,16d]

could explain the different behaviour of the latter.

On closer inspection, the structural nuances of the com-

The strong influence of the equatorial ligand is reflected

plexes are reflected in the temperature dependence of the in the zero-field Mössbauer spectra of solid samples of the

2798

www.eurjic.org Eur. J. Inorg. Chem. 2005, 2794–2811

Complexes with Dianionic [N₄] Macrocycles

FULL PAPER

solvent-free complex Fe1. The spectrum at 4.2 K is given

in Figure 4. The observed quadrupole splitting of ΔE_Q

=

3.37 mms^–1(isomer shift δ = 0.32 mms^–1) is very large for

an iron(ii) complex, even if a square-planar coordination

sphere is considered. The data for similar complexes are

given in Table 2. ΔE_Qappears to increase with increasing

σ-donating ability (basicity) of the equatorial ligand. This

is in agreement with earlier observations revealing a

stronger σ-donor ability of the ligands 1–3 in comparison

to similar porphyrin ligands.

Figure 5. ESR spectrum of Fe^III1Cl in frozen toluene solution at

10 K.

Fe1Cl in toluene at 45 °C is given in Figure 6. With the help

of predictions about the signal shifts based on literature

proposals^[25]and the signal intensities a clear assignment is

possible. The observed paramagnetic shift can be interpre-

ted by taking only the contact shift into account. The tem-

perature dependence of the shifts obeys the Curie law.

Molecular orbital calculations based on the molecular

structure were performed for Fe1MeOH and Fe1Cl to con-

firm the intermediate-spin ground state. Similar calcula-

tions were also performed for Fe1I and, for better compari-

son, the coordinate system was chosen the same way. The z

axis of the molecular coordinate system was chosen in the

direction of the axial ligand bond with the iron atom lo-

cated at the origin of the coordinate system. The x and y

axes are directed between the equatorial nitrogen atoms so

that the σ-bonds with the equatorial ligand are formed by

the 3d_xyorbital of the iron atom. The results of the spin-

polarized calculations for Fe1MeOH are given in Figure 7.

The MO diagram of Fe1Cl is similar to that of Fe1I dis-

cussed in detail previously,^[22]and is therefore not displayed

here. Small differences are found in the bonding conditions

between the iron atom and the axial ligand. In the case of

the iodide adduct, covalent interactions between the 3d_z₂

(σ-bonding) and 3d_xz,yz(weakly π-antibonding) orbitals of

the iron atom and the 5p_zorbital of the axial ligand are

present, while the 5p_x,yorbitals of the iodide ion are nearly

Figure 4. Mössbauer spectrum of Fe^II1 at 4.2 K. Solid line: simula-

tion of the spectrum with Lorentzian doublets.

Table 2. Mössbauer data of square-planar intermediate-spin (S =

1) iron(ii) complexes.

Com-

pound

Condi-

tions

δ(Fe)

[mms^–1

ΔE_Q

[mms^–1

μ_eff

[μ_B]^[a]

Ref.

]

Fe^II1

solid,

4.2 K

solid,

77 K

solid,

83 K

0.32

0.50

0.59

3.37

3.7

4.4

4.7

this

work

Fe^II(TPP)

+1.51

+1.60

[24a]

Fe^II(-

OEP)

[24a]

[a] At room temperature.

The ESR spectrum of a frozen solution of Fe1Cl in tolu- non-bonding. In case of the chloride adduct π-interactions

ene at 10 K is given in Figure 5. The g values (g_y^eff= 4.594, between the 3p_x,yorbitals of the chloride ion and the 3d_xz,yz

eff

g_x= 3.486 and g_z= 2.039) can be assigned to the pop- orbitals of the iron atom are present and are more pro-

ulation of the ground-state (S = 1/2) Kramers doublet of nounced than in the case of the iodide adduct. This explains

the S = 3/2 system. The signal at 6.203 is due to small (μ- the hypsochromic shift of the typical LTMCT (equatorial

oxo?) impurities. The signal intensity of the impurity in- ligand-to-metal charge transfer) band in the UV/Vis spec-

creases with decreasing temperature. The spectrum is sim- trum of the chloride complex relative to the iodide one.^[21]

ilar to those reported for Fe1I,^[20]although with slightly

The intermediate-spin state of Fe1MeOH is clearly mir-

shifted g values, which can be assigned to a different rhom- rored in the molecular orbital diagrams. The unoccupied

bicity parameter E/D (D = axial, E = rhombic zero-field d_xyorbital responsible for the σ-bond with the equatorial

splitting parameter) of around 0.1 for the chloride complex ligand is clearly separated from the other four d orbitals

(0.22 for the iodide adduct). In contrast to Fe1I, the chlo- and the strong interaction with the macrocycle leads to a

ride complex Fe1Cl is not ESR-silent in the solid state. A reduced iron contribution. The interactions with the axial

broad signal is observed at around g = 4.7, which can be ligand (methanol) are not reflected in the MO diagrams,

explained by differences in the molecular packing in the thereby indicating a very weak bond. The pattern of the

crystal.

MO energies is actually more similar to the typical pattern

The intermediate-spin state of the iron(iii) complexes is of a square-planar coordination than a square-pyramidal

also reflected in the characteristic shifts of the resonances one, therefore the coordination number of the iron atom

1

in the H NMR spectra. As an example, the spectrum of can also be described as 4+1.

Eur. J. Inorg. Chem. 2005, 2794–2811

www.eurjic.org

2799

B. Weber, I. Käpplinger, H. Görls, E.-G. Jäger

FULL PAPER

1

Figure 6. H NMR spectrum and signal assignment of Fe^III1Cl in toluene solution at 45 °C. “S” denotes the solvent.

Orientation of Axial N-Heterocycles in Octahedral

Complexes

One of the biological functions of b-type cytochromes is

to switch the direction of charge-transfer processes de-

pending on the potential. The active site contains haem

with two axial imidazole rings of histidine. The voltage-de-

pendent change of the orientation of the heterocyclic planes

could be a possible mechanism for the switching. Former

X-ray studies on the pair Fe^II1(HIm)₂and [Fe^III1(HIm)₂]-

PF₆gave evidence for such a charge-dependent reorienta-

tion of the imadazole planes,^[18a]but it was not possible to

clarify the role of the intermolecular hydrogen bonds in this

case. Later on, an overview of the orientation of axial li-

gand planes in octahedral homo- and heteroligand diad-

ducts of fifteen iron(ii⁄iii) complexes was given.^[18b]We have

now successfully crystallised Fe^II3(Py)₂and [Fe^III3(Py)₂]-

PF₆, the first pair of complexes derived from the basic com-

plex Fe3. The structures of both complexes are depicted in

Figure 8 (right side), together with those of Fe^II2Py₂and

[Fe^III2Py₂]ClO₄(left side) for comparison. The correspond-

ing bond lengths and angles are summarised in Table 3.

Addition of two neutral nitrogen ligands at the axial co-

ordination sites of the macrocyclic iron complexes leads to

low-spin complexes. The spin state was confirmed by ESR

investigations in the case of the iron(iii) complexes.^[18b,19c]

Figure 7. Orbital energies between 0 and –8 eV for the intermedi-

The iron atom is surrounded by six nitrogen atoms, re-

ate-spin state of Fe1MeOH for spin-polarised calculations. The ap-

sulting in a relatively symmetric close coordination sphere

with an approximate D_4hsymmetry. (Porphyrin)iron com-

proximate assignment for each level is given (L donates the equato-

rial ligand).

2800

www.eurjic.org

Eur. J. Inorg. Chem. 2005, 2794–2811

Complexes with Dianionic [N₄] Macrocycles

FULL PAPER

Figure 8. Molecule structures of (a) Fe^II2(Py)₂, (b) [Fe^III2(Py)₂]ClO₄, (c) Fe^II3(Py)₂and (d) [Fe^III3(Py)₂]PF₆. Ethoxy groups are omitted.

Table 3. Bond lengths [Å] and angles [°] within the first coordination sphere of macrocyclic iron(ii⁄iii) complexes with coordination

number (CN) 6 and spin state S.

[a]

[b]

[c]

Complex

[Fe^III1(Py)₂]PF₆

CN

S

Fe–N₁,₂

Fe–N₃,₄

Average FeL_eq

1.899

Fe–L_ax

Fe–pl_eq

0.000

Angle [L_ax]_{1_2}

15

6

1/2

1.882(5)

1.893(6)

1.917(2)

1.904(2)

1.893(2)

1.887(2)

1.898(2)

1.900(2)

1.89

1.926(6)

1.925(6)

1.925(2)

1.932(2)

1.909(2)

1.918(2)

1.934(2)

1.929(2)

1.93

2.045(3)

2.063(3)

2.018(2)

2.037(2)

2.022(2)

2.040(2)

2.035(2)

2.006(2)

2.00

Fe^II2(Py)₂

6

0

1.920

1.902

1.915

1.90

0.017

0.015

0.010

–

9

[d]

[Fe^III2(Py)₂]^+[d]

Fe^II3(Py)₂

1/2

0

79

90

80

88

[Fe^III3(Py)₂]^+[e]

[Fe^III1(N-MeIm)₂]⁺

[Fe^III1(NCS)₂]^–

[Fe^III2(CN)₂]^–

1/2

0

1.90

2.04

1.893(6)

1.882(5)

1.896(2)

1.892(2)

1.910(2)

1.902(2)

1.887(3)

1.895(2)

1.904(3)

1.85

1.925(6)

1.929(6)

1.923(2)

1.926(2)

1.931(2)

1.930(2)

1.907(3)

1.916(3)

1.925(3)

1.927(3)

1.90

1.907

1.909

1.918

1.901

1.913

1.89

2.017(7)

2.001(6)

1.947(3)

1.954(3)

2.002(3)

2.004(3)

1.982(3)

2.000(3)

1.963(3)

2.055(4)

2.00

0.003

–

[Fe^III2(NO₂)₂]^–

[Fe^III2(NO₂)OH₂]

[Fe^III3(NO₂)NO] (?)^[e,f]

–

13

0.018

0.032

1.91

1.96

[a] Designation according to Figure 1. [b] Displacement of the iron atom from the main plane of the equatorial donor atoms. [c] Dihedral

angle between the planes of the two axial ligands. [d] See ref.^[18b][e] The quality of the data is not sufficient and we are only publishing

the conformation of the molecule, although distances and angles are similar to those of the other complexes. [f] Because of the disorder

of the second axial NO_xligand and the diamagnetism of the compound, the correct composition is uncertain.

Eur. J. Inorg. Chem. 2005, 2794–2811

www.eurjic.org

2801

B. Weber, I. Käpplinger, H. Görls, E.-G. Jäger

FULL PAPER

pounds with similar axial ligands are also low-spin com- [Fe(CN)₆]^3–.^[19a]In several cases, for example X = HIm, Y

plexes, while complexes with sterically demanding axial li- = CN^–, strong stabilizing interactions between both axial

gands such as benzimidazole or 2-methylimidazole are high ligands have been detected.

spin due to the small energetic splitting of the iron 3d orbit-

als.^[26,27]Similar high-spin diadducts of our macrocycles

have not been obtained up to now.

The iron atom is situated in the [N₄] plane of the equato-

rial ligand. The average Fe–N_eqbond lengths (190–192 pm)

for both pairs in Figure 8 are in the range typical for all of

our low-spin and intermediate-spin complexes, but signifi-

cantly shorter than in similar (porphyrin)iron complexes

(196–200 pm).^[28]The change of the oxidation state of the

iron atom from +2 to +3 leads to an almost negligible

shortening of the equatorial bond lengths. A slight length-

ening of these bonds in octahedral complexes of Fe1 in

comparison with the square-planar and pentacoordinate

complexes can be explained by the increased electron den-

sity at the central atom. The bond lengths between the iron

atom and the axial ligands are in the region 200–204 pm,

similar to those in low-spin (porphyrin)iron complexes, and

are independent of the oxidation state of the central atom.

The most characteristic difference between the structures

of both pairs in Figure 8 is the orientation of the planes of

the axial pyridine rings. In the case of Fe^II2(Py)₂both pyri-

dine rings are nearly coplanar and are directed at the centre

carbon atom of the six-membered chelate rings, similar to

the case of Fe^II1(Him)₂. The angle between their planes

amounts to only 9°. In the iron(iii) derivative [Fe^III2(Py)₂]-

ClO₄, however, both planes of the pyridine rings are twisted

more in the direction at the equatorial N-atoms and the

five-membered rings, and the angle between them is now

79°. This finding suggests that the similar behaviour of the

pair Fe^II1(Him)₂/[Fe^III1(Him)₂]PF₆is not necessarily a re-

sult of reorganisation of intermolecular hydrogen bonds. In

contrast, the planes of the pyridine rings in the pair

Fe^II3(Py)₂and [Fe^III3(Py)₂]PF₆remain in a nearly un-

changed orientation with an angle of 90° and 80°, respec-

tively; they are independent of the oxidation state of the

central atom. This is a consequence of the “saddle-shaped”

distortion of the twofold phenylene-bridged equatorial li-

gand that allows only one energetically favoured orientation

of the planes of the axial pyridine rings (perpendicular to

each other) along the “trough” on each side.

(1)

(2)

In view of the fact that anionic and mixed axial ligands

play a key role in the chemistry of haem enzymes, we were

interested in obtaining structural information for such com-

plexes, especially to determine whether interactions between

axial ligands through the central atom are reflected in the

structural parameters. Unfortunately, attempts to isolate

pure crystals failed in most cases, therefore only a few ex-

amples are presented here.

One of the many functions of haem enzymes in biological

systems is the activation of small molecules and ions such

as nitric oxide (NO), nitrite (NO₂) or sulfite (SO₃^2–). There

–

has been a renewed interest in the biological nitrogen cycle

with the discoveries of the biomedical functions of nitric

oxide^[29,30]and the clarification of the X-ray structure of

the haem enzyme cytochrome c nitrite reductase.^[31]This

development increased the interest in model compounds

with manageable electronic interactions to obtain a better

understanding of the coherences of the biological activity

of these small molecules. An overview of the properties of

nitrosyliron complexes of ligand types I and II has been

published.^[22]A dinitrite adduct and several mononitrite ad-

ducts (with occupation of the sixth coordination site by

other ligands) of “picket-fence” porphyrins have been pre-

pared and several attempts have been made to describe the

iron–nitrite bond.^[32a–32d]All of these compounds are low-

spin iron(iii) complexes.

The reaction of our macrocyclic iron(ii) complexes with

nitrite under inert conditions yields the corresponding ni-

trosyliron complex as the main product, according to Equa-

tion (3), in all three cases. This was proved by ESR spec-

troscopy and by preparing Fe1¹⁵NO using ¹⁵N-labelled

NaNO₂.

All attempts to crystallise the iron(ii) counterparts of

[Fe^III1(Py)₂]PF₆and [Fe^III1(N-Meim)₂]ClO₄(Table 3) were

unsuccessful.

Octahedral Complexes with Biologically Relevant Anions as

Axial Ligands

–

2Fe1 + 3NO₂+ H₂O Ǟ Fe1NO + [Fe1(NO₂)₂]^–+ 2OH^–

(3)

Extended spectrophotometric equilibrium studies of the

The reaction of the iron(iii) complexes with nitrite in or-

equilibria (1) and (2),^{[5a,5c,18b,19]}particularly with the more ganic solvents is even more complex. Although the spectro-

hydrophilic complex Fe1a, have provided conclusive proof photometric titration of [Fe^III1a]⁺clearly indicates the exis-

of the existence of octahedral iron(iii) complexes tence of a dinitrito and a mixed hydroxo-nitrito derivative

[Fe^III1aXY] in solution with a large number of axial ligands, in alkaline aqueous solution,^[19b]all attempts to crystallise

including anions as hydroxide, sulfite, nitrite, cyanide, thio- such derivatives of Fe1 failed. Upon exclusion of air, the

cyanate, and even such “exotic” ligands as [Fe(CN)₆]^4–and isolated reaction product of Fe1Cl or Fe1I in methanol is

2802

www.eurjic.org

Eur. J. Inorg. Chem. 2005, 2794–2811

Complexes with Dianionic [N₄] Macrocycles

FULL PAPER

the corresponding nitrosyl complex according to Equa-

tion (4). On the other hand, Fe1NO reacts in methanol or

pyridine with air to give the nitrite adduct [Equation (5)].

The existence of a mono- and dinitrite adduct has been

proved by UV/Vis spectroscopy.^[22]Similar reactions have

been found for ferric porphyrin complexes.^[34]

Fe1⁺+ 2NO₂Ǟ Fe1NO + NO₃

(4)

–

Fe1NO + 1/2O₂+ D Ǟ Fe1(NO₂)D; D = Py, MeOH

(5)

For Fe2 it was possible, starting from the corresponding

iodide or chloride complex, to isolate a nitrite diadduct and

a few crystals of a mixed adduct with one nitrite ion and

one water molecule as axial ligand trans to the nitrite ion.

The molecule structures are given in Figure 9a, b. In both

complexes the iron atom has a nearly octahedral coordina-

tion sphere and is located in the plane of the macrocycle.

The Fe–N_eqand Fe–N

distances are similar to those of

NO₂

other diadducts. The oxygen atoms of the nitrite ligands

point towards the six-membered chelate ring and, in the

case of the diadduct, both nitrite groups are coplanar. The

coordination of water in the case of the monoadduct is

interesting. This has been postulated in many quantum-

chemical investigations but has never been described in de-

tail. The relatively large distance of 206 pm between the

iron atom and the bound water molecule suggests a weak

coordination.

While the nitrite diadduct is a pure low-spin complex

over the whole temperature range, the nitrite monoadduct

performs a spin transition between the S = 1/2 and S =

3/2 states. Such spin transition behaviour has already been

reported for other six-coordinate iron(iii) complexes^[35a,35b]Figure 9. Molecule structures of (a) Na[Fe^III2(NO₂)₂](MeOH)₂-

(H₂O)_0.5, (b) [Fe^III2(NO₂)OH₂] and (c) [Fe^III3(NO₂)NO_x]^–. Ethoxy

groups are omitted.

and is also typical of some nitrosyliron complexes of type

II.^[22,35c,35d]The effective moments plotted against tempera-

ture are given for both compounds in Figure 10.

Calculations based on the SCC-Xα method did confirm

the S = 1/2 configuration as the ground state for both com-

plexes. The energy difference to the S = 3/2 first excited

state of [Fe2(NO₂)OH₂](MeOH) is, at around 20 kJmol^–1,

small for this program. (The expected value at room tem-

perature is significantly lower, but experience with the pro-

gram used shows that the obtained values are within this

region for this program.^[40]) Therefore an admixture of this

state and the ground state at room temperature is plausible.

Another possibility is an interaction between both spin

Figure 10. Temperature dependence of the effective magnetic mo-

states that would explain the room-temperature effective

ment of Na[Fe^III2(NO₂)₂](MeOH)₂(H₂O)_0.5(left) and [Fe^III2(NO₂)

moment of 2.82 μ_B, which is significantly lower than the

expected spin-only value of 3.87 μ_B. Similar effects have

been described in the literature for other iron(iii) com-

plexes.^[36]Measurements of the effective magnetic moment

at higher temperatures are not possible as the complex de-

OH₂] (right). The measurements were carried out at 0.2 and 0.5 T.

For Fe3 − the complex most similar to (porphyrin)iron

composes with loss of water. On the basis of these calcula- compounds − a derivative with the approximate composi-

tions it could be shown that the overlap population between tion (nBu₄N)₂[Fe3(NO₂)NO_x](NO₂)(H₂O), with one axial

Fe and O_{H O}decreases by 60% during the spin transition. nitrite and one other axial NO_xligand has been crystallised.

2

The resulting pentacoordinate complex should be an inter- Unfortunately, the latter is disordered in such a manner that

mediate-spin complex similar to the corresponding iron(iii) it was impossible to decide between an NO₂with two and

halide complexes.

an NO with four orientations. Additionally, the presence of

Eur. J. Inorg. Chem. 2005, 2794–2811

www.eurjic.org

2803

B. Weber, I. Käpplinger, H. Görls, E.-G. Jäger

FULL PAPER

potential cations and anions in the lattice makes it more

difficult to clearly define the charge of the central unit. The

diamagnetism of the crystals suggests a neutral complex

[Fe^III3(NO₂)NO] with low-spin iron(iii), antiferromag-

netically coupled with the nitrosyl radical. This interpret-

ation agrees with the distances and angles within the central

unit. Mixed axial nitrite-nitrosyl coordination has also been

found in iron(ii)^[32e]and iron(iii)^[32d]complexes with

“picket-fence” porphyrins.

The thiocyanate anion is of interest not only because of

its biological relevance but also because it is, along with

azide, one of the pseudo-halides that give heterogeneous

equilibria with [Fe^III1a]⁺[Equation (6)] between aqueous

and lipophilic phases accompanied by a “structural mim-

icry”:

(6)

Figure 11. Molecule structures of (a) [Fe^III1(NCS)₂]^–and (b)

[Fe^III2(CN)₂]^–(H₂O)_0.5(cations omitted for clarity).

While the green product, in agreement with its EPR spec-

trum, is doubtless an octahedral low-spin iron(iii) complex,

the orange one, which is stable only in lipophilic solvents

such as dichloromethane, benzene or toluene, has not been

An octahedral anionic complex with two cyanide ions as

unambiguously characterised to date. Unfortunately, all axial ligands, [Et₄N][Fe^III2(CN)₂] could be crystallised with

attempts to obtain pure crystals of this product also failed. the unsymmmetrically bridged macrocycle.

Crystals suitable for X-ray structure determination were

Dicyanide complexes are well documented and charac-

only obtained for the diadduct [BzEt₃N][Fe^III1(NCS)₂].

terised in the literature.^[41,42]The question as to whether the

The only examples of thiocyanato adducts of (porphyrin)- cyanide ion is bound to the iron atom through the carbon

iron compounds described in the literature are a pentacoor- or the nitrogen atom was decided in favour of the carbon

dinate monoadduct of high-spin iron(iii)^[37]and a hexacoor- atom (the difference of the R1 values, the resulting C–N

dinate low-spin complex with pyridine as second axial li- bond length and the remaining electron density were con-

gand.^[38]Cationic macrocyclic iron(ii⁄iii) complexes with sidered). This result fits with results in the literature. The

neutral equatorial ligands easily form thiocyanato diad- distances between the iron atom and the equatorial nitrogen

ducts.^[39]Thus, the complex [BzEt₃N][Fe1(NCS)₂] is the first atoms are, with an average of 192 pm, slightly longer than

macrocyclic iron complex with an anionic [N₄] ligand and those of other octahedral iron(iii) complexes and are more

two axially bound thiocyanate ligands (Figure 11).

comparable with those found for iron(ii) complexes. Com-

For a proper description, the adduct has to be referred pared with the isothiocyanate diadduct, the axial bond

to as a diisothiocyanato complex, which is also the case lengths are also long and more similar to pyridine adducts.

for the corresponding porphyrin monoadducts. The linear These differences in the bond lengths lead to the conclusion

coordination of the thiocyanate ion is confirmed by the Fe– that in the case of cyanide as axial ligand, additional nega-

N_ax–C_axangles of 171° and 179°. The Fe–N distances are tive charge is localised at the iron atom (and therefore delo-

comparable with those in cationic complexes described pre- calised over the macrocyclic ligand), resulting in a lengthen-

viously. The only conspicuous feature of the complex is the ing of the bonds. In the case of the larger thiocyanate ion,

arrangement of the free ester groups of the equatorial li- the negative charge is localised on the easily polarizable sul-

gand, where both carbonyl oxygen atoms point to the un- fur rather than the iron atom, resulting in shorter bond.

substituted site of the ligand. For all other complexes of The low-spin state of the iron(iii) atom was confirmed by

this ligand type characterised by X-ray structure analysis, ESR spectroscopy. In contrast to neutral octahedral com-

at least one, but mostly both, of the terminal carbonyl oxy- plexes the ground-state electron configuration of the cya-

gen atoms are directed towards the methyl group. The low- nide diadduct is (d_xz,d_yz)⁴(d_xy)¹. The same results can be

spin state of the iron(iii) atom was confirmed by ESR spec- found for comparable cyanide diadducts in the litera-

troscopy (g values: 1.973, 2.125, 2.162).^[18]

ture.^[41,42]

2804

www.eurjic.org

Eur. J. Inorg. Chem. 2005, 2794–2811

Complexes with Dianionic [N₄] Macrocycles

FULL PAPER

[Fe^III1(Py)₂]PF₆and [Fe^III1(N-MeIm)₂]ClO₄, but unfortu-

nately their iron(ii) counterparts could not be obtained as

Conclusions

The results described above confirm the close relation of suitable crystals.

our macrocycles with porphyrins, but they also indicate

Octahedral low-spin derivatives of the macrocyclic iron

some special features. In general, the anions L^2–(L = 1, 2, complexes FeL with biologically relevant anions as axial li-

3) seem to be stronger ligands for iron than most porphy- gands have been crystallised for the first time, although

rinato systems. The reasons for this could be: (i) the their existence in solution was proved many years ago. Re-

stronger basicity and σ-donor ability of the four nitrogen actions with nitrite resulted in the anionic diadduct

donor atoms, especially in the case of L = 1; (ii) the pres- [Fe^III2(NO₂)₂]^–with pentacoordinate sodium as counterion,

ence of the electron-withdrawing carbonyl groups in conju- and, more interestingly, in the neutral mixed-ligand com-

gation with the π-system of the six-membered chelate rings, plex [Fe^III2(NO₂)OH₂]. The presence of water as a relatively

which promote potential π-backbonding and increase the weak axial ligand results in a temperature-dependent ad-

effective charge of the central atom, especially when L = 2 mixing of low-spin (favoured at low temperature) and, poss-

and 3, where the π-system is extended over one phenylene ibly, intermediate-spin species. DFT-MO calculations agree

ring (the second phenylene in 3 is hardly included in the with this interpretation. With cyanide and thiocyanate ions

conjugation); (iii) the smaller hollow within the cavity of the octahedral low-spin diadducts [Et₄N][Fe2(CN)₂] and

the 14-membered macrocycle, which is better suited for the [BzEt₃N][Fe1(NCS)₂] could be crystallised. Unfortunately,

radii of iron(ii⁄iii) in lower spin states.

it was not possible to obtain pure crystals of the pentacoor-

The pronounced tendency to stabilize the intermediate- dinate neutral counterpart of the latter, Fe^III1(NCS), which

spin ground state in iron(ii⁄iii) complexes with coordination exists together with the anionic diadduct in an equilibrium

numbers below six, even when similar compounds are in of distribution between a lipophilic and an aqueous phase.

a high-spin state [e.g. Fe^III(Me₄TAA)Cl or the halides of

With regard to the existence of high-spin iron(ii⁄iii) com-

(porphyrin)iron compounds], is a typical feature of the li- plexes and to the processes of spin conversion, some of the

gands L^2–. Intermediate spin (S = 1) was found in the iron complexes of type II are more closely related to (por-

iron(ii) complexes Fe2 (CN = 4, planar), Fe1 (CN = 4+1 phyrin)iron compounds than the macrocycles I. This will be

by weak intermolecular coordination of the ester carbonyl the subject of a following paper.

group), Fe1MeOH and the bridged complexes (FeL)₂dabco

(L = 1, 2, 3; all CN = 5, square-pyramidal). The preference

to bind axial O-ligands − MeOH or even such a weak O-

donor as COOEt − indicates that the iron atom in Fe1 is

the “hardest” central atom in the series FeL. The magnetic

behaviour of the iron(ii) complexes at lower temperature is

often determined by intermolecular interactions in the lat-

tice: Fe2, with the shortest Fe–Fe distance of 335 pm, shows

a sigmoid decrease of the magnetic moment, whereas

(Fe3)₂dabco, where the intermolecular Fe–Fe distance

(421 pm) is much shorter than the intramolecular distance

(727 pm), shows a steep decrease of the moment starting

already at around 150 K. The complexes FeLCl (L = 1, 2,

3) and the iodide Fe1I represent the iron(iii) case of inter-

mediate-spin (S = 3/2) behaviour. The intermediate-spin

state could be confirmed by Mössbauer spectrometry for

Fe1, by ESR and NMR spectroscopy for Fe1Cl, and by

DFT-MO calculations for Fe1MeOH, Fe1Cl and, pre-

viously, for Fe1I (see refs.^[22b,22c]for more calculations).

The pair Fe^II3(Py)₂/[Fe^III3(Py)₂]PF₆was characterised as

a third example of potential “voltage-dependent molecular

switches” based on octahedral low-spin iron(ii⁄iii) com-

plexes with N-heterocycles as axial ligands. Comparison

Experimental Section

General Procedures and Instrumentation: If not stated differently,

all syntheses were carried out under argon using Schlenk tube tech-

niques. All solvents were purified as described in the literature^[43]

and distilled under argon. Magnetic measurements of pulverized

samples were performed with a Quantum Design MPMSR-5s

SQUID magnetometer in a temperature range from 1.7 to 400 K,

with the exception of Fe2Cl and Fe3Cl. All measurements were

carried out at two field strengths (0.2 and 0.5 T). Diamagnetic cor-

rections were made using values calculated with tabulated Pascal

constants (as a rule, suitable values can be estimated as χ_dia

Ϸ

–6

0.6×10

M

complex

). The measurements of the highly air-sensitive

iron(ii) compounds were repeated several times under strongly inert

conditions to ensure that the samples remain in the iron(ii) state

during preparation. For comparison, an additional sample was

measured after exposure to air. ESR investigations were carried out

with an ESP 300E (Bruker) instrument at liquid-nitrogen tempera-

ture with the solid or in solution (methanol, toluene) or equipped

with an N₂/He-flow cryostat from 4.2 K up to room temperature.

Molecular orbital calculations were carried out in a local density

approximation (LDA) by the spin-polarized self-consistent-charge

(SCC-)Xα method.^[44,45]

with the pairs Fe^II1(HIm)₂/[Fe^III1(HIm)₂]PF₆and Fe^II2- Syntheses: The syntheses of the free ligands are summarised in

(Py)₂/[Fe^III2(Py₂)]ClO₄resulted in the conclusions that: (i)

Scheme 2 for H₂1 and H₂2 and in Scheme 3 for H₂3. All steps of

the ligand syntheses were carried out in air. From a thermodynamic

the charge-dependent change of the orientation of the axial

point of view, the driving force of the complex formation starting

planes is not necessarily accompanied by reorganisation of

from the free ligand, as well as of the reverse protolysis of the

intermolecular hydrogen bonds, and (ii) the “saddle-

complex, depends on the competition between two protons and the

shaped” distortion of the ligand 3 allows only orientation

divalent metal ion to bind the anion of the deprotonated ligand:

of the axial planes perpendicular to each other; no charge-

dependent twist of these planes is possible. Other new octa-

hedral low-spin complexes with axial N-heterocycles are M²⁺+ 2HL h ML + 2H⁺

(7)

Eur. J. Inorg. Chem. 2005, 2794–2811

www.eurjic.org

2805

B. Weber, I. Käpplinger, H. Görls, E.-G. Jäger

FULL PAPER

Scheme 3. Synthesis of the free macrocyclic ligand H₂3. The Cu-

driven condensation of two molecules D is expected to give the anti

isomer of the copper complex E with respect to the methyl groups.

The X-ray analysis of E confirms, however, the depicted syn

structure.^[14b]

Syntheses of the Macrocyclic Ligands

6,13-Bis(ethoxycarbonyl)-7,12-dimethyl-1,4,8,11-tetraazacyclotetra-

deca-5,7,12,14-tetraene (H₂1):^[17b,46b]The preparation of this ligand

by a template synthesis of the copper complex Cu1, followed by

demetallation with hydrogen sulfide in dilute sulfuric acid and pre-

cipitation with sodium hydroxide, gave the free ligand in nearly

quantitative yield. However, to prevent the use of gaseous hydrogen

sulfide, the latter can be replaced by precipitation from homogen-

eous solution by in situ hydrolysis with thioacetamide. In this case

the yield of the last step drops to about 45%.

Compound A:^[47]Ethyl acetoacetate (130 g, 1 mol), acetic anhydride

(204 g, 2 mol) and triethyl orthoformate (148 g, 1 mol) were re-

fluxed together for 40 min. The low-boiling side-products were dis-

tilled off at normal pressure until foggy clouds appeared in the

reaction vessel. The main product was then obtained by frac-

tionatal vacuum distillation (b.p. 145–150 °C at 2 kPa). The purity

of the oily, colourless to bright greenish-yellow product was deter-

mined by GC analysis. Yield: 140.1 g (75%).

Compound B:^[33]Compound A (140.1 g, 0.75 mol) was dissolved in

methanol (130 mL) and freshly distilled ethylenediamine (22.8 g,

0.379 mol), dissolved in methanol (130 mL), was then slowly added

to the solution. The reaction mixture was stirred vigorously and

the ligand precipitated immediately. The colourless product was fil-

tered off and recrystallised from ethanol or dioxane. Yield: 94 g

(77%).

Scheme 2. Syntheses of the free macrocyclic ligands H₂1 and H₂2.

The key step of the syntheses of the free ligands is a template reac-

tion using the copper(ii) ion as template, followed by the demetalla-

tion of the macrocyclic copper complex. The different conditions

of the demetallation reflect the different properties of the ligands:

the release of the protonated macrocycle H₂3 from Cu3 is possible

by dissolving it in sulfuric acid without any auxiliary reagent. This

is obviously due to the relatively low stability of the copper com-

plex Cu3, where the copper(ii) ion is too large for the small hollow

of the rigid anion 3^2–. To obtain the ligands H₂1 and H₂2, addition

of the auxiliary reagent hydrogen sulfide is necessary to remove the

copper ion. Different reaction conditions are also typical for the

formation of iron complexes. As in the case of porphyrins, iron(ii)

is easier to incorporate into the macrocycles than iron(iii). The

complex Fe1 was formed nearly quantitatively in a fast reaction

using a 1:1 ratio of ligand and iron(ii) acetate, even under moderate

basic conditions. For the complexes Fe2 and Fe3 an excess of the

iron(ii) salt is necessary to prevent contamination by the free li-

gand. Additionally, the complex formation requires an increase of

the reaction time from a few minutes (Fe1) to 2–3 h (Fe2) or 10–

11 h (Fe3). This increase reflects the kinetic effect of the increasing

rigidity of the ligand structures.

Compound C:^[33]Compound B (10 g, 0.029 mol) and copper acetate

monohydrate (5.9 g, 0.035 mol, finely pulverized) were mixed and

2806

www.eurjic.org

Eur. J. Inorg. Chem. 2005, 2794–2811

Complexes with Dianionic [N₄] Macrocycles

FULL PAPER

heated in DMF or ethanol (500 mL) to reflux for 30 min. After

cooling, violet crystals precipitated that were filtered off, washed

with methanol and dried. Yield: 10.1 g (85%).

H₂3:^[23a]Cu3 (2 g, 0.0038 mol) was suspended in cooled concen-

trated H₂SO₄(50 mL) where it underwent a colour change from

brown to yellow. The suspension was added dropwise to a water/

ice mixture, taking care to ensure the solution did not warm up.

The free ligand precipitated as yellow needles that were collected

on a paper filter, washed with water until a neutral pH was ob-

tained, and dried at room temperature. The crude material was

dissolved in a little dioxane and reprecipitated with distilled water

to obtain yellow crystals. Yield: 1.2–1.4 g (70–80%).

Cu1:^[46a]Compound C (50 g, 0.12 mol) was dissolved in ethylenedi-

amine (100 mL, dried and freshly distilled from solid potassium

hydroxide) and heated to reflux until the colour had changed from

deep blue to brownish-violet (15–30 min). After cooling, the pro-

duct was filtered off and washed with water. Yield: 38.5 g (72%).

H₂1:^[17b,46b]Cu1 (2.0 g, 0.0047 mol) was suspended in methanol

(10 mL) and stirred for 15 min. Ice-cold (0 °C) dilute sulfuric acid

(ca. 2 m, 40 mL) was then added to the suspension. After 1 min,

thioacetamide (4.2 g, 0.056 mol) was added to the suspension and

the reaction mixture was stirred for a further 3 min. The suspension

was filtered through a fluted filter paper into a calculated volume of

ice-cold sodium hydroxide solution (ca. 4 m) such that the mixture

became neutral or weakly basic after addition of all of the acidic

filtrate. The free ligand was filtered off and washed with water until

a nearly neutral pH was obtained. The crude material was recrys-

tallised from methanol as fine colourless needles. Yield: 0.8–1 g

(47–58%).

Syntheses of the Iron(II) Complexes

[6,13-Bis(ethoxycarbonyl)-7,12-dimethyl-1,4,8,11-tetraazacyclote-

tradeca-5,7,12,14-tetraenato(2–)]iron(II) (Fe1):^[17b]Iron(ii) acetate

(2.08 g, 11.9 mmol) and H₂1 (4.30 g, 11.8 mmol) were dissolved in

methanol (60 mL) and refluxed for 1 h. After standing overnight,

the red precipitate was collected and dried in vacuo. Yield: 2.27 g

(46%). C₁₈H₂₆FeN₄O₄(418.27): calcd. C 51.68, H 6.27, N 13.40;

found C 48.34, H 6.09, N 12.22; the discrepancy between the calcd.

and found values is due to the high air sensitivity of the complex.

IR (Nujol): ν_C=O= 1671 cm^–1. MS (DEI): m/z (%) = 418 (100)

[M⁺], 373 (15) [M⁺– OC₂H₅], 344 (5) [M⁺– COOEt].

[6,13-Bis(ethoxycarbonyl)-7,12-dimethylbenzo[b]-1,4,8,11-tetraaza-

cyclotetradeca-5,7,12,14-tetraenato(2–)]iron(II) (Fe2):^[10d]Iron(ii)

acetate (0.76 g, 4.4 mmol) and H₂2 (1.27 g, 3.1 mmol) were sus-

pended in methanol (20 mL) and refluxed for 6 h. The black pro-

duct was collected, washed with a small volume of methanol and

dried in vacuo. Yield: 1.14 g (77%). C₂₂H₂₆FeN₄O₄(466.31): calcd.

C 56.66, H 5.62, Fe 11.98, N 12.02; found C 55.23, H 5.52, Fe 12.1,

N 12.24; the discrepancy between the calcd. and found values is

6,13-Bis(ethoxycarbonyl)-7,12-dimethylbenzo[b]-1,4,8,11-tetraaza-

cyclotetradeca-5,7,12,14-tetraene (H₂2): Cu2 was synthesised in

analogy to Cu1 using 1,2-phenylenediamine instead of ethylenedi-

amine in step B.^[10d,33]

H₂2:^[10d]Cu2 (2.0 g, 0.0042 mol) was suspended in methanol

(10 mL) and stirred for 15 min. Ice-cold (0 °C) sulfuric acid (5 m,

40 mL) was then added to the suspension. After 1 min, thioacetam-

ide (4.2 g, 0.056 mol) was added to the suspension and the reaction

mixture was stirred for a further 3 min. The suspension was then

filtered into ice-cold sodium hydroxide (4 m, 100 mL) taking care

to ensure that the solution did not warm up. The free ligand was

filtered off and washed with water until a nearly neutral pH was

obtained. The crude material was dissolved in a little dioxane and

reprecipitated with distilled water to obtain yellow crystals. Yield:

0.8–1 g (45–60%).

due to the high air sensitivity of the complex. IR (Nujol): ν_C=O

=

1676 cm^–1. MS (DEI): m/z (%) = 43 (100), 466 (50) [M⁺].

[6,13-Bis(ethoxycarbonyl)-7,12-dimethyldibenzo[b,i]-1,4,8,11-tetra-

azacyclotetradeca-5,7,12,14-tetraenato(2–)]iron(II) (Fe3):^[23a]Iron(ii)

acetate (1.33 g, 7.7 mmol) and H₂3 (2.29 g, 5.0 mmol) were sus-

pended in methanol (20 mL) and refluxed for 12 h. The black pro-

duct was collected, washed with a small volume of methanol and

dried in vacuo. Yield: 2.24 g (88%). C₂₆H₂₆FeN₄O₄(514.35): calcd.

C 60.71, H 5.09, N 10.89; found C 59.95, H 5.09, N 11.01. IR

(Nujol): ν_C=O= 1686 cm^–1. MS (DEI): m/z (%) = 514 (100) [M⁺],

486 (20) [M⁺– C₂H₄], 468 (15) [M⁺– OC₂H₅].

6,13-Bis(ethoxycarbonyl)-7,12-dimethyldibenzo[b,i]-1,4,8,11-tetra-

azacyclotetradeca-5,7,12,14-tetraene (H₂3):^[23a]In earlier reports,

the complexes M3 (M = Cu, Co)^[23c]and the intermediates E^[23b]

were postulated as isomeric structures with an anti configuration

of the methyl groups attached to the six-membered chelate rings.

However, the X-ray structural analysis of the free ligand H₂3 and

the nitrosylcobalt complex Co3NO proved the syn structure.^[14b]

Dinuclear Adducts with Bridging 1,4-Diazabicyclo[2.2.2]octane

(dabco)

(Fe1)₂dabco: Iron(ii) acetate (0.31 g, 1.8 mmol) and 0.65 g of H₂1

were dissolved in 20 mL of methanol and refluxed for 20 min. After

cooling, a solution of dabco (0.2 g, 1.8 mmol) in 5 mL of methanol

was slowly added to the solution. Slow crystallisation overnight

yielded red needles suitable for X-ray structure analysis. The pro-

duct was collected, washed with a little ice-cold methanol and dried

in vacuo. Yield: 0.33 g (39%). C₃₆H₅₂Fe₂N₈O₈·C₆H₁₂N₂(948.73):

calcd. C 53.71, H 6.80, N 14.77; found C 52.48, H 6.59, N 14.78.

IR (Nujol): ν_C=O= 1673, 1652 cm^–1. MS (DEI): m/z (%) = 112

(100) [M⁺(dabco)], 418 (90) [M⁺(Fe1)]. DTG: Up to 250 °C loss of

11.3% (calcd. for 1 dabco: 11.8%); decomposition at 300 °C.

Compound D:^[23b]A suspension of o-phenylenediamine (12 g,

0.11 mol) in ethanol (100 mL) was heated to reflux. Compound A

(18.6 g, 0.1 mol) was slowly added to the solution and the mixture

was refluxed for 20 min. After cooling, compound D precipitated

as yellow needles that were collected and recrystallised twice from

methanol or ethanol. Yield: 13.14 g (53%). M.p. 84 °C.

Compound E:^[23b]Compound D (24.8 g, 0.1 mol) was heated to re-

flux in methanol (200 mL) and a suspension of finely ground cop-

per acetate monohydrate (9.98 g, 0.05 mol) in methanol (150 mL)

was slowly added to the solution. The suspension was refluxed for

approximately 6 h. After cooling, the product was filtered off im-

mediately, washed with methanol and dried. Yield: 23.56 g (87%).

(Fe2)₂dabco: Fe2 (0.5 g, 1.1 mmol) was suspended in 20 mL of

methanol. Then, 1.3 mL of a 1 m solution of dabco in methanol

was added to the suspension and left standing for slow crystallisa-

tion. After 2 d, red-black needles were obtained. The product was

collected, washed with a little methanol and dried in vacuo. Yield:

0.3 g (55%). C₄₄H₅₂Fe₂N₈O₈·C₆H₁₂N₂(1044.8): calcd. C 57.48, H

6.18, N 13.42; found C 56.60, H 6.12, N 13.41. IR (Nujol): ν_C=O

= 1674 cm^–1. MS (DEI): m/z (%) = 112 (100) [M⁺(dabco)], 466 (10)

[M⁺(Fe2)]. DTG: Up to 100 °C loss of 11.3% (solvent in the crys-

Cu3:^[23c]Compound E (6.68 g, 0.012 mol) was heated to reflux in

dry ethanol (150 mL). Metallic sodium (0.31 g, 0.013 mol), dis-

solved in dry ethanol (50 mL), was slowly added to the suspension

and the mixture refluxed for 6 h. After cooling, the product was

filtered off immediately, washed with cold methanol and dried.

Yield: 5.99 g (96%).

Eur. J. Inorg. Chem. 2005, 2794–2811

www.eurjic.org

2807

B. Weber, I. Käpplinger, H. Görls, E.-G. Jäger

FULL PAPER

tal), up to 300 °C loss of 12.4% (calcd. for 1 dabco: 10.7%); decom-

position at 350 °C.

H 5.0, N 11.7; found C 46.6, H 4.9, N 11.5. MS (DEI): m/z (%) =

79 (100) [pyridine], 418 (19) [M⁺(Fe1)].

(Fe3)₂dabco: Fe3 (0.65 g, 1.3 mmol) was suspended in 40 mL of

methanol. Then, 1.8 mL of a 1 m solution of dabco in methanol

was added to the suspension and left standing for slow crystallisa-

tion. After 2 d, black needles were obtained. The product was col-

lected, washed with a little methanol and dried in vacuo. Yield:

0.45 g (61%). C₅₂H₅₂Fe₂N₈O₈·C₆H₁₂N₂(1140.9): calcd. C 61.06, H

5.65, N 12.28; found C 58.96, H 5.90, N 12.03. IR (Nujol): ν_C=O

= 1686 cm^–1. MS (DEI): m/z (%) = 42 [fragment], 112 (90)

[M⁺(dabco)], 514 (10) [M⁺(Fe3)]. DTG: Up to 250 °C loss of 9.7%

(calcd. for 1 dabco: 9.8%); at 350 °C decomposition.

[Fe1(N-MeIm)₂]ClO₄(H₂O): Fe1I (30 mg) was dissolved in a solu-

tion of N-methylimidazole in methanol (1.4 m, 15 mL). A solution

of NaClO₄in water was added dropwise to the solution until a

precipitate was obtained. C₁₈H₂₆FeN₄O₄·2C₄H₆N₂·ClO₄·H₂O

(693.90): calcd. C 44.6, H 5.7, N 16.0; found C 61.4, H 5.9, N

13.5. MS (DEI): m/z (%) = 82 (100) [N-methylimidazole], 418 (25)

[M⁺(Fe1)].

Na[Fe2(NO₂)₂](MeOH)₂(H₂O)_0.5: Fe2Cl (0.032 g, 0.05 mmol) was

suspended in methanol (5 mL). A solution of sodium nitrite in

methanol/water (1 m, 5 mL) was then added. After a few days,

black crystals were obtained. The product was collected, washed

with a little methanol and dried in vacuo. Yield: 0.01 g (40%). IR

(Nujol): ν_C=O= 1665 cm^–1. MS (ESI neg.): m/z (%) = 512 (100)

[M^–(Fe2NO₂)], 558.1 (50) [M^–{Fe2(NO₂)₂}].

Syntheses of the Iron(III) Complexes: The syntheses of the iron(iii)

iodide complexes Fe1I^[19c,21]and Fe2I^[10d]have been described pre-

viously.

[6,13-Bis(ethoxycarbonyl)-7,12-dimethyl-1,4,8,11-tetraazacyclotetra-

deca-5,7,12,14-tetraenato(2–)]iron(III) Chloride (Fe1Cl): Fe1 (1.36 g, The following procedures were carried out only with the aim of

3.3 mmol) was suspended in tetrachloromethane (20 mL) for 2 d.

The CCl₄was then removed by cold vacuum distillation and the

obtained solid was recrystallised from methanol. The precipitate

was filtered off and dried in vacuo. The mother solution was al-

lowed to stand overnight to produce single crystals of Fe1Cl suit-

abl e for X -ray cr ys t a l l og rap hy. Yi e l d: 1 . 0 4 g (6 9 % ).

obtaining crystals suitable for X-ray investigations. No yield is

given and the main product in the solution may have another ident-

ity. Crystal growth was accomplished by using a temperature gradi-

ent or different solubility in different solvents. The choice of the

counterion may be crucial for crystallisation of the compound.

Dropwise addition of nonpolar solvents (n-heptane and n-hexane

C₁₈H₂₆ClFeN₄O₄(453.72): calcd. C 47.65, H 5.78, N 12.35; found are not suitable) can also be used for crystallisation. The addition

C 44.81, H 5.37, N 11.50. IR (Nujol): ν_C=O= 1680 cm^–1. MS

of water may be used as a last alternative in the case of water-

(DEI): m/z (%) = 28 (100), 453 (10) [M⁺], 418 (30) [M⁺– Cl]. DTG: miscible solvents.

Decomposition at 170 °C.

(BzEt₃N)[Fe1(SCN)₂]: Fe1I (30 mg) was dissolved in 15 mL of a

[6,13-Bis(ethoxycarbonyl)-7,12-dimethylbenzo[b]-1,4,8,11-tetraaza-

cyclotetradeca-5,7,12,14-tetraenato(2–)]iron(III) Chloride (Fe2Cl):

Fe2 (0.67 g, 1.4 mmol) was suspended in tetrachloromethane

(20 mL) for 2 d. The CCl₄was then removed by cold vacuum distil-

lation and the solid product was recrystallised from methanol. The

precipitate was filtered off and dried in vacuo. Yield: 0.59 g (84%).

C₂₂H₂₆ClFeN₄O₄(501.764): calcd. C 52.66, H 5.22, N 11.17; found

C 52.52, H 4.89, N 11.45. IR (Nujol): ν_C=O= 1685 cm^–1. MS

(DEI): m/z (%) = 36 (100), 501 (50) [M⁺], 466 (20) [M⁺– Cl].

saturated solution of (BzEt₃N)SCN in water/methanol (1:1).

Et₄N[Fe2(CN)₂](H₂O)_0.5: Fe2I (30 mg) was dissolved in a saturated

solution of Et₄NCN in water (15 mL). The mixture was heated and

filtered while still hot.

[Fe2(NO₂)OH₂](MeOH): Fe2I (30 mg) was dissolved in methanol

(15 mL). The solution was added dropwise to a heated saturated

solution of nBu₄NNO₂in water/methanol (4:1).

(nBu₄N)₂[Fe3(NO₂)NO_x](NO₂)(H₂O): Fe3I (40 mg) was dissolved

in dichloromethane (30 mL) and 5 mL of this solution was added

to a saturated solution of nBu₄NNO₂in methanol.

[6,13-Bis(ethoxycarbonyl)-7,12-dimethyldibenzo[b,i]-1,4,8,11-tetra-

azacyclotetradeca-5,7,12,14-tetraenato(2–)]iron(III) Chloride

(Fe3Cl): Fe3 (0.96 g, 1.9 mmol) was suspended in tetrachlorometh-

ane (30 mL) for 2 d. The CCl₄was then removed by cold vacuum

distillation and the solid was recrystallised from methanol. The pre-

cipitate was filtered off and dried in vacuo. Yield: 0.87 g (83%).

C₂₆H₂₆ClFeN₄O₄(549.80): calcd. C 56.79, H 4.77, N 10.19; found

C 55.60, H 4.61, N 10.48. IR (Nujol): ν_C=O= 1697 cm^–1. MS

(DEI): m/z (%) = 31 (100), 549 (90) [M⁺], 514 (15) [M⁺– Cl].

Crystal Structure Determination: The intensity data for the com-

pounds were collected with a Nonius KappaCCD diffractometer,

using graphite-monochromated Mo-K_αradiation. Data were cor-

rected for Lorentz and polarisation effects, but not for absorp-

tion.^[48a,48b]The structures were solved by direct methods

(SHELXS^[49]) and refined by full-matrix least-squares techniques

against F_o(SHELXL-97^[50]). The hydrogen atoms were included

2

at calculated positions with fixed thermal parameters. All non-

hydrogen atoms were refined anisotropically.^[50]Na[Fe2-

(NO₂)₂](MeOH)₂(H₂O)_0.5shows a disordered methanol and water

molecule with the same oxygen position. This disorder could be

solved. The data for compound Fe1Cl are not of adequate quality,

although the esd’s are good enough to be discussed in this paper.

For Fe2, [Fe3(Py)₂]PF₆and (nBu₄N)₂[Fe3(NO₂)NO_x](NO₂)(H₂O)

we are only publishing the conformation of the central complex

unit and the crystallographic data; we will not deposit the data

Synthesis of Diadducts of Iron(II/III) Complexes

Fe3(Py)₂: Fe3 (40 mg) was dissolved in pyridine (7 mL) and a layer

of heptane (10 mL) was added over the pyridine solution. After

a few days, a fine black-red crystalline precipitate was obtained.

C₂₆H₂₆FeN₄O₄·2C₅H₅N (672.55): calcd. C 64.3, H 5.4, N 12.5;

found C 64.4, H 5.2, N 12.4.

[Fe3(Py)₂]PF₆: Fe3I (30 mg) was dissolved in pyridine (15 mL) and

then a solution of NH₄PF₆in methanol (2 m) was added dropwise

until a precipitate was obtained. C₂₆H₂₆FeN₄O₄·2C₅H₅N·PF₆with the Cambridge Crystallographic Data Centre. XP (SIEMENS

(817.52): calcd. C 52.9, H 4.4, N 10.3; found C 52.8, H 4.6, N 10.1.

Analytical X-ray Instruments, Inc.) was used for structure represen-

tations.

MS (DEI): m/z (%) = 79 (100) [pyridine], 514 (Ͻ1) [M⁺(Fe3)].

[Fe1(Py)₂]PF₆: Fe1I (30 mg) was dissolved in a solution of pyridine

in benzene (1 m, 15 mL) and then a solution of NH₄PF₆in meth-

anol (2 m) was added dropwise to the solution until a precipitate

was obtained. C₁₈H₂₆FeN₄O₄·2C₅H₅N·PF₆(721.44): calcd. C 46.6,

Crystal Data for Fe1:^[51]C₁₈H₂₆FeN₄O₄, M_r= 418.28 g mol^–1

,

brown prism, size 0.32×0.20×0.12 mm, monoclinic, space group

P2₁/c, a = 12.5031(4), b = 11.0732(4), c = 27.4969(6) Å, β =

92.798(2)°, V = 3802.4(2) Å³, T = –90 °C, Z = 8, ρ_calcd.

=

2808

www.eurjic.org

Eur. J. Inorg. Chem. 2005, 2794–2811

Complexes with Dianionic [N₄] Macrocycles

FULL PAPER

1.461 gcm^–3, μ(Mo-K_α) = 8.25 cm^–1, F(000) = 1760, 11872 reflec-

tions in h(–16/16), k(–14/13), l(–34/33), measured in the range 1.63°

Յ θ Յ 27.47°, completeness for θ_max90.8%, 7906 independent re-

16.096(1) Å, β = 95.950(7)°, V = 3246.6(4) Å³, T = –90 °C, Z =

4, F(000) = 1468, 12274 measured reflections, 7233 independent

reflections, R_int= 0.082, 4759 reflections with I Ͼ 2σ(I), 425

flections, R_int= 0.028, 5880 reflections with F_oϾ 4σ(F_o), 508 parameters, 0 restraints, R1 = 0.131, wR²= 0.238, GOOF = 1.140,

parameters, 0 restraints, R1_obs= 0.050, wR²

= 0.127, R1_all

=

largest residual electron density: 0.552 eÅ^–3.

obs

0.075, wR²_all= 0.142, GOOF = 0.988, largest difference peak/hole:

Crystal Data for (BzEt₃N)[Fe1(NCS)₂]:^[51]C₃₃H₄₈FeN₇O₄S₂, mono-

clinic, space group P2₁/n, a = 11.7266(2), b = 24.7177(6), c =

12.7940(2) Å, β = 96.551(1)°, V = 3684.19(12) Å³, T = –90 °C, Z

= 4, F(000) = 1540, 7450 measured reflections, 7450 independent

reflections, 5659 reflections with I Ͼ 2σ(I), 487 parameters, 0 re-

straints, R1 = 0.056, wR²= 0.118, GOOF = 1.055, largest residual

electron density: 0.415 eÅ^–3.

Crystal Data for Et₄N[Fe2(CN)₂](H₂O):^[51]C₃₂H₄₈FeN₇O₅, mono-

clinic, space group P2₁/n, a = 11.1860(2), b = 11.8490(4), c =

25.0733(9) Å, β = 92.355(2)°, V = 3320.5(5) Å³, T = –90 °C, Z =

4, F(000) = 1420, 13220 measured reflections, 7482 independent

reflections, R_int= 0.062, 4899 reflections with I Ͼ 2σ(I), 446

parameters, 0 restraints, R1 = 0.063, wR²= 0.119, GOOF = 1.010,

largest residual electron density: 0.436 eÅ^–3.

0.446/–0.409 eÅ^–3.

Crystal Data for Fe1MeOH:^{[ 5 1 ]}C_{1 9}H_{3 0}FeN₄O₅, M_r

=

450.32 g mol^–1, brown prism, size 0.10 ×0.09 × 0.08 mm, mono-

clinic, space group P2₁/n, a = 7.3249(2), b = 29.5441(9), c =

9.7511(3) Å, β = 94.912(2)°, V = 2102.46(11) Å³, T = –90 °C, Z =

4, ρ_calcd.= 1.423 gcm^–3, μ(Mo-K_α) = 7.55 cm^–1, F(000) = 952, 8074

reflections in h(–9/9), k(–38/38), l(–12/12), measured in the range

2.76° Յ θ Յ 27.49°, completeness for θ_max97.2%, 4694 indepen-

dent reflections, R_int= 0.033, 3559 reflections with F_oϾ 4σ(F_o),

271 parameters, 0 restraints, R1_obs= 0.038, wR²= 0.093, R1_all

obs

= 0.060, wR²= 0.105, GOOF = 0.838, largest difference peak/

all

hole: 0.349/–0.283 eÅ^–3.

¯

Crystal Data for Fe2: C₂₂H₂₆FeN₄O₄, triclinic, space group P1, a

= 13.019(3), b = 13.753(3), c = 13.757(3) Å, α = 103.28(3), β =

117.00(3), γ = 97.04(3)°, V = 2061.7(8) Å³, T = –90 °C, Z = 4,

F(000) = 976, 5550 measured reflections, 5192 independent reflec-

tions, R_int= 0.045, 3598 reflections with I Ͼ 2σ(I), 535 parameters,

0 restraints, R1 = 0.162, wR²= 0.338, GOOF = 1.052, largest resid-

ual electron density: 1.129 eÅ^–3. Due to technical problems, the

measurement stopped too early and reflections at high θ angles are

missing. Therefore, the anisotropy refinement and hydrogen atoms

were set aside resulting in a dissatisfactory R1 value and a high

remaining maximum electron density.

Crystal Data for [Fe2(NO₂)OH₂](MeOH):^[51]C₂₃H₂₉FeN₆O₁₀, or-

thorhombic, space group Pna2₁, a = 16.273(1), b = 11.6454(7), c =

13.8026(7) Å, V = 2615.7(3) Å³, T = –90 °C, Z = 4, F(000) = 1180,

5725 measured reflections, 5725 independent reflections, 4004 re-

flections with I Ͼ 2σ(I), 344 parameters, 1 restraint (origin of the

unit cell), R1 = 0.077, wR²= 0.138, GOOF = 1.031, largest residual

electron density: 0.507 eÅ^–3.

[ 5 1 ]

Crystal Data for Na[Fe2(NO₂)₂](MeOH)₂(H₂O)_{0 . 5}

:

C_46.4H_67.2Fe₂N₁₂Na₂O₂₂(CH₄O)₂, M_r= 1366.89 gmol^–1, red-brown

prism, size 0.12×0.10×0.09 mm, monoclinic, space group P2₁/n, a

= 11.3381(3), b = 13.3795(6), c = 21.1046(8) Å, β = 102.522(2)°, V

= 3125.4(2) Å³, T = –90 °C, Z = 2, ρ_calcd.= 1.452 gcm^–3, μ(Mo-

K_α) = 5.65 cm^–1, F(000) = 1431, 11116 reflections in h(–14/14),

k(–14/17), l(–27/27), measured in the range 1.89° Յ θ Յ 27.48°,

Crystal Data for Fe1Cl:^[51]C₁₈H₂₆ClFeN₄O₄, M_r= 453.73 gmol^–1,

brown prism, size 0.20×0.18×0.12 mm³, monoclinic, space group

P2₁/n, a = 13.5466(6), b = 7.8215(4), c = 18.5515(9) Å, β =

96.539(3)°, V = 1952.83(16) Å³, T = –90 °C, Z = 4, ρ_calcd.

=

1.543 gcm^–3, μ(Mo-K_α) = 9.42 cm^–1, F(000) = 948, 7129 reflections

in h(–17/15), k(–10/9), l(–23/24), measured in the range 3.42° Յ θ

Յ 27.52°, completeness for θ_max96.2%, 4320 independent reflec-

tions, R_int= 0.065, 3701 reflections with F_oϾ 4σ(F_o), 257 parame-

completeness for θ_max98.8%, 7082 independent reflections, R_int

0.033, 5071 reflections with F_oϾ 4σ(F_o), 419 parameters, 0 re-

straints, R1_obs= 0.0546, wR²= 0.1280, R1_all= 0.0886, wR²

=

all

obs

ters, 0 restraints, R1_obs= 0.130, wR²

= 0.207, R1_all= 0.156,

0 . 1 4 6 , G O O F = 0 . 9 6 2 , l a rg e s t d i ff e re n c e p e ak / h ol e:

obs

0.845/–0.619 eÅ^–3.

wR²= 0.216, GOOF = 1.248, largest difference peak and hole:

all

1.009/–1.123 eÅ^–3.

Crystal Data for (nBu₄N)₂[Fe3(NO₂)NO_x](NO₂)(H₂O):

C₅₈H₁₀₀FeN₉O₁₁(x = 2), M_r= 1161.32 gmol^–1, green prism, size

Crystal Data for Fe3(Py)₂:^[51]C₃₆H₃₆Fe N₆O₄, triclinic, space

¯

0.34×0.32×0.20 mm, triclinic, space group P1, a = 12.961(3), b =

group P1, a = 10.6327(3), b = 12.0421(4), c = 14.7871(5) Å, α =

66.015(2), β = 71.582(2), γ = 66.368(2)°, V = 1558.09(9) Å³, T =

−90 °C, Z = 2, F(000) = 704, 12217 measured reflections, 7108

independent reflections, R_int= 0.028, 6385 reflections with I Ͼ

2σ(I), 424 parameters, 0 restraints, R1 = 0.044, wR²= 0.103,

GOOF = 1.016, largest residual electron density: 0.421 eÅ^–3.

15.203(4), c = 17.985(4) Å, α = 94.779(9), β = 106.047(9), γ =

107.84(1)°, V = 3186.9(13) Å³, T = –90 °C, measured in the range

3.03° Յ θ Յ 27.59°.

Crystal Data for (Fe1)₂dabco:^[51]C₄₂H₆₄Fe₂N₁₀O₈, M_r

=

948.73 gmol^–1, brown prism, size 0.10×0.09×0.08 mm, triclinic,

¯

Crystal Data for [Fe3(Py)₂]PF₆: C₃₆H₃₆F₆FeN₆O₄P, orthorhombic,

space group Pbca, a = 16.840(3), b = 20.335(4), c = 20.857(4) Å, V

= 7142(3) Å³, T = –90 °C, Z = 8, μ(Mo-K_α) = 5.65 cm^–1, F(000) =

3368. The data are not of the best quality and so we are publishing

only the conformation of the molecule, although the distances and

angles are of plausible size.

Crystal Data for [Fe1(Py)₂]PF₆:^[51]C₂₈H₃₆F₆FeN₆O₄P, monoclinic,

space group P2₁/n, a = 9.211(2), b = 14.861(3), c = 24.180(5) Å, β

= 100.51(3)°, V = 3254.2(11) Å³, T = 20 °C, Z = 4, F(000) = 1492,

space group P1, a = 8.3623(2), b = 11.8244(3), c = 12.6364(3) Å, α

= 110.604(2), β = 100.490(2), γ = 100.238(2)°, V = 1109.61(5) Å³,

T = –90 °C, Z = 1, ρ_calcd.= 1.420 gcm^–3, μ(Mo-K_α) = 7.17 cm^–1,

F(000) = 502, 8736 reflections in h(–10/10), k(–14/15), l(–16/16),

measured in the range 2.03° Յ θ Յ 27.46°, completeness for θ_max

99.6%, 5057 independent reflections, R_int= 0.0221, 4346 reflections

with F_oϾ 4σ(F_o), 307 parameters, 0 restraints, R1_obs= 0.0490,

wR²= 0.1398, R1_all= 0.0586, wR²= 0.1478, GOOF = 1.018,

obs

all

largest difference peak/hole: 1.059/–0.673 eÅ^–3.

Crystal Data for (Fe2)₂dabco:^[51]C₅₀H₆₄Fe₂N₁₀O₈, M_r

=

8631 measured reflections, 4697 independent reflections, R_int

=

1044.81 g mol^–1, black prism, size 0.10 ×0.09 × 0.08 mm, mono-

clinic, space group P2₁/n, a = 11.8119(2), b = 13.1486(3), c =

16.7867(3) Å, α = 90.00, β = 108.6320(10), γ = 90.00°, V =

0.028, 3596 reflections with I Ͼ 2σ(I), 442 parameters, 0 restraints,

R1 = 0.052, wR²= 0.138, GOOF = 1.067, largest residual electron

density: 0.436 eÅ^–3.

Crystal Data for [Fe1(N-MeIm)₂]ClO₄(H₂O):^[51]C₂₆H₄₀ClFeN₈O₉, 2470.50(8) Å³, T = –90 °C, Z = 2, ρ_calcd.= 1.405 gcm^–3, μ(Mo-K_α)

monoclinic, space group P2₁/n, a = 14.897(1), b = 13.613(1), c =

= 6.52 cm^–1, F(000) = 1100, 18857 reflections in h(–14/15), k(–17/

Eur. J. Inorg. Chem. 2005, 2794–2811

www.eurjic.org

2809

B. Weber, I. Käpplinger, H. Görls, E.-G. Jäger

FULL PAPER

Chem. 1980, 19, 3148; c) Yu. E. Kovalenko, Ya. D. Lampeka,

I. Ya. Levitin, K. B. Yatsimirkii, K. Müller, D. Seidel, E.-G.

Jäger, Russ. Chem. Bull. 1993, 42, 981 (Engl. Transl. from Izv.

Akad. Nauk Ser. Khim. 1993, no. 6, 1092); d) E.-G. Jäger, H.

Keutel, M. Rudolph, B. Krebs, F. Wiesemann, Chem. Ber.

1995, 128, 503.

a) E.-G. Jäger, M. Rudolph, R. Müller, Z. Chem. 1978, 18, 229;

b) E.-G. Jäger, P. Renner, R. Schmidt, Z. Chem. 1978, 18, 193;

c) E.-G. Jäger, B. Kirchhof, E. Schmidt, B. Remde, A. Kipke,

R. Müller, Z. Anorg. Allg. Chem. 1982, 485, 141; d) E.-G. Jäger,

G. Schlenvoigt, B. Kirchhof, M. Rudolph, R. Müller, Z. Anorg.

Allg. Chem. 1982, 485, 173; e) E.-G. Jäger, K. Schuhmann, H.

Görls, Inorg. Chim. Acta 1997, 255, 295; E.-G. Jäger, K.

Schuhmann, H. Görls, Chem. Ber./Recueil 1997, 130, 1643; f)

K. Schuhmann, E.-G. Jäger, Eur. J. Inorg. Chem. 1998, 2051.

a) K. Kubokura, H. Okawa, S. Kida, Bull. Chem. Soc. Jpn.

1978, 51, 2036; b) G. M. Larin, V. V. Minin, E.-G. Jäger, P.

Renner, Zh. Neorg. Khim. 1980, 25, 183; Chem. Abstr. 1980, 92,

139940z; c) E.-G. Jäger, K. Müller, Z. Anorg. Allg. Chem. 1983,

501, 40; d) E.-G. Jäger, J. Knaudt, M. Rudolph, M. Rost,

Chem. Ber. 1996, 129, 1041; e) J. Knaudt, St. Förster, U.

Bartsch, A. Rieker, E.-G. Jäger, Z. Naturforsch., Teil B 2000,

55, 86; f) R. Wegner, M. Gottschaldt, W. Poppitz, E.-G. Jäger,

D. Klemm, J. Mol. Catal. A: Chem. 2003, 201, 91.

M. Rudolph, S. Dautz, E.-G. Jäger, J. Am. Chem. Soc. 2000,

122, 10821.

a) E.-G. Jäger, D. Seidel, M. Rudolph, A. Schneider, Z. Chem.

1986, 26, 76; b) H. Görls, G. Reck, E.-G. Jäger, K. Müller, D.

Seidel, Cryst. Res. Technol. 1990, 25, 1277.

a) D. P. Riley, J. A. Stone, D. H. Busch, J. Am. Chem. Soc.

1977, 99, 767; b) D. P. Riley, D. H. Busch, Inorg. Chem. 1984,

23, 3235; c) T. J. Truex, R. H. Holm, J. Am. Chem. Soc. 1972,

94, 4529; d) S. Koch, R. H. Holm, R. B. Frankel, J. Am. Chem.

Soc. 1975, 97, 6714.

a) V. L. Goedken, J. J. Pluth, S.-M. Peng, B. Bursten, J. Am.

Chem. Soc. 1976, 98, 8014; b) M. C. Weiss, B. Bursten, S.-M.

Peng, V. L. Goedken, J. Am. Chem. Soc. 1976, 98, 8021; c) V. L.

Goedken, S. M. Peng, J. Molin-Norris, Y. Park, J. Am. Chem.

Soc. 1976, 98, 8391; d) V. L. Goedken, Y.-A. Park, J. Chem.

Soc., Chem. Commun. 1975, 214.

a) E.-G. Jäger, Vortragsberichte zum Symposium “Koordination-

schemie der Übergangselemente”, Sektion C, Jena 1969, p. 96;

b) E.-G. Jäger, E. Stein, F. Gräfe, W. Schade, Z. Anorg. Allg.

Chem. 1985, 526, 15; c) E.-G. Jäger, F. Gräfe, Z. Anorg. Allg.

Chem. 1988, 561, 25.

a) F. Wiesemann, R. Wonnemann, B. Krebs, H. Keutel, E.-G.

Jäger, Angew. Chem. Int. Ed. Engl. 1994, 33, 1363; b) I.

Käpplinger, H. Keutel, E.-G. Jäger, Inorg. Chim. Acta 1999,

291, 190; c) I. Käpplinger, M. Grodzicki, V. Schünemann, H.

Görls, A. X. Trautwein, E.-G. Jäger, private communication,

Cambridge Crystallographic Data Centre, CSD 408794/5, Oc-

tober 1998.

a) E.-G. Jäger, Z. Chem. 1985, 25, 446; b) E.-G. Jäger, B.

Schweder, S. Radzuweit, Z. Chem. 1988, 28, 152; c) E.-G. Jäger,

H. Keutel, Inorg. Chem. 1997, 36, 3512.

H. Keutel, I. Käpplinger, E.-G. Jäger, M. Grodzicki, V. Schüne-

mann, A. X. Trautwein, Inorg. Chem. 1999, 38, 2320.

17), l(–21/20), measured in the range 2.39° Յ θ Յ 27.47°, complete-

ness for θ_max99.8%, 5644 independent reflections, R_int= 0.0422,

4402 reflections with F_oϾ 4σ(F_o), 308 parameters, 0 restraints,

R1_obs= 0.0516, wR²= 0.1412, R1_all= 0.0710, wR²= 0.1535,

obs

all

GOOF = 1.026, largest difference peak/hole: 1.094/–0.833 eÅ^–3.

Crystal Data for (Fe3)₂dabco:^[51]C₅₈H₆₄Fe₂N₁₀O₈·CH₃OH, M_r=

[11]

[12]

1172.93 gmol^–1, dark-green prism, size 0.10×0.09×0.08 mm, tri-

¯

clinic, space group P1, a = 11.3944(6), b = 12.2186(8), c =

13.1020(8) Å, α = 116.067(4), β = 113.938(3), γ = 93.166(4)°, V =

1435.76(15) Å³, T = –90 °C, Z = 1, ρ_calcd.= 1.357 gcm^–3, μ(Mo-

K_α) = 5.7 cm^–1, F(000) = 616, 10088 reflections in h(–14/14), k(–15/

14), l(–16/16), measured in the range 2.04° Յ θ Յ 27.40°, complete-

ness for θ_max98.3%, 6431 independent reflections, R_int= 0.0396,

5119 reflections with F_oϾ 4σ(F_o), 387 parameters, 0 restraints,

R1_obs= 0.1018, wR²= 0.2585, R1_all= 0.1240, wR²= 0.2702,

obs

all

GOOF = 1.160, largest difference peak/hole: 1.090/–0.734 eÅ^–3.

Acknowledgments

We gratefully acknowledge financial support from the Deutsche

Forschungsgemeinschaft and the Fonds der Chemischen Industrie.

The authors thank Dr. V. Schünemann and P. Wegner (Lübeck) for

the Mössbauer spectrum, Dr. W. Poppitz and S. Schönau for the

MS measurements, Dr. M. Friedrich and B. Rambach for measur-

ing the ESR spectra, and Dr. K. Karaghiosoff for measuring the

¹H NMR spectrum of Fe1Cl.

[13]

[14]

[15]

[16]

[1] W. Kaim, B. Schwederski, “Bioinorganic Chemistry: Inorganic

Elements”, in The Chemistry of Life, John Wiley & Sons,

Chichester, 1994.

[2] The Porphyrin Handbook (Eds.: K. M. Kadish, K. M. Smith,

R. Guilard), Elsevier–Academic Press, 1999, vol. 1–10 and

2003, vol. 11–20.

[3] T. Tsumaki, Bull. Chem. Soc. Jpn. 1938, 13, 252.

[4] a) E. N. Jacobsen, Comprehensive Organometallic Chemistry II

(Eds.: G. Wilkinson, F. G. A. Stone, E. W. Abel, L. S. Heg-

edus), Pergamon, New York, 1995, vol. 12 chapter 11.1; b) T.

Katsuki, Coord. Chem. Rev. 1995, 140, 189.

[5] a) E.-G. Jäger, in Chemistry at the Beginning of the Third Mil-

lennium (Eds.: L. Fabbrizzi, A. Poggi), Springer-Verlag, Berlin,

Heidelberg, New York, 2000, p. 103; b) E.-G. Jäger, J. Knaudt,

K. Schuhmann, A. Guba, in Peroxide Chemistry – Mechanistic

and Preparative Aspects of Oxygen Transfer (Ed.: W. Adam),

Wiley-VCH, Weinheim, 2000, p. 249; c) E.-G. Jäger, St. Barth,

H. Keutel, F. Wiesemann, U. Gaudig, in Bioinorganic Chemis-

try – Transition Metals in Biology and their Coordination Chem-

istry (Ed.: A. X. Trautwein), Wiley-VCH, Weinheim, 1997, p.

584.

[17]

[18]

[6] a) E.-G. Jäger, Z. Anorg. Allg. Chem. 1969, 364, 177; b) E.-G.

[19]

Jäger, Z. Chem. 1964, 4, 437.

[7] a) P. Mountford, Chem. Soc. Rev. 1998, 27, 105; b) F. A. Cot-

ton, J. Czuchajowska, Polyhedron 1990, 9, 2553.

[20]

[21]

[22]

[8] a) R. Wegner, M. Gottschaldt, H. Görls, E.-G. Jäger, D.

Klemm, Chem. Eur. J. 2001, 7, 2143; Angew. Chem. 2000, 112,

608; Angew. Chem. Int. Ed. 2000, 39, 595; b) H. Keutel, H.

Görls, W. Poppitz, A. Schütz, E.-G. Jäger, J. Prakt. Chem.

1999, 341, 785; c) H. Elias, D. Hess, H. Paulus, E.-G. Jäger, F.

Gräfe, Z. Anorg. Allg. Chem. 1990, 101, 101; d) E.-G. Jäger, K.

Müller, Z. Anorg. Allg. Chem. 1985, 526, 29; e) E.-G. Jäger, K.

Müller, Z. Anorg. Allg. Chem. 1981, 482, 201.

[9] a) Y. Nishida, K. Hayashida, N. Oishi, S. Kida, Inorg. Chim.

Acta 1980, 38, 213; b) V. V. Minin, Yu. V. Rakitin, G. M. Larin,

E.-G. Jäger, Zh. Neorg. Khim. 1981, 26, 650; Chem. Abstr. 1981,

94, 200292z.

E.-G. Jäger, H. Hähnel, H.-F. Klein, A. Schmidt, J. Prakt.

Chem. 1991, 333, 423.

a) B. Weber, H. Görls, M. Rudolf, E.-G. Jäger, Inorg. Chim.

Acta 2002, 337, 247; b) B. Weber, PhD Thesis, University of

Jena, 2002; c) B. Weber, A Contribution to a Deeper Under-

standing of the Nature of the Iron–Nitrosyl Bond in Bioanalogue

Iron Complexes, Der Andere Verlag, Osnabrück, 2003.

a) K. Müller, E.-G. Jäger, Z. Anorg. Allg. Chem. 1989, 577, 195;

b) E.-G. Jäger, D. Seidel, W. Schade, Z. Chem. 1983, 23, 32; c)

E.-G. Jäger, D. Seidel, Z. Chem. 1983, 23, 261; d) W. Schade,

E.-G. Jäger, K. Müller, D. Seidel, J. Prakt. Chem. 1989, 331,

559.

[23]

[10] a) D. G. Pillsbury, D. H. Busch, J. Am. Chem. Soc. 1976, 98,

7836; b) J. A. Streeky, D. G. Pillsbury, D. H. Busch, Inorg.

2810

www.eurjic.org

Eur. J. Inorg. Chem. 2005, 2794–2811

Complexes with Dianionic [N₄] Macrocycles

FULL PAPER

[24]

[25]

Anderson, S. H. Strauss, Inorg. Chem. 1995, 34, 3953; b) H.

Nasri, M. Debbabi, Polyhedron 1998, 17, 3607.

a) O. K. Medhi, J. Silver, J. Chem. Soc., Chem. Commun. 1989,

1199; b) A. Hudson, H. J. Whitfield, Inorg. Chem. 1967, 6,

1120.

A. Walker, Proton NMR and EPR spectroscopy of Paramag-

netic Metalloporphyrins, in The Porphyrin Handbook (Eds.:

K. M. Kadish, K. M. Smith, R. Guilard), 2000, Vol. 5, Ch. 36,

p. 81.

K. R. Levan, C. E. Strouse, Abstract of paper, American Crys-

tallographic Association Summer meeting, Snowman, 1983.

a) E. Elkaim, K. Tanaka, P. Coppens, W. R. Scheidt, Acta

Crystallogr., Sect. B. 1987, 43; 457; b) D. K. Geiger, Y. J. Lee,

W. R. Scheidt, J. Am. Chem. Soc. 1984, 106, 6339.

W. R. Scheidt, Y. J. Lee, Struct. Bonding (Berlin) 1987, 64, 1.

Nitric Oxide. Principles and Actions (Ed.: J. Lancaster), Aca-

demic Press, San Diego, 1996.

P. L. Feldman, O. W. Griffith, D. J. Stuehr, Chem. Eng. News

1993, 71, 26.

[38] a) W. R. Scheidt, Y. J. Lee, D. K. Geiger, K. Taylor, K. Hatano,

J. Am. Chem. Soc. 1982, 104, 3367; b) D. K. Geiger, V. Chum-

plang, W. R. Scheidt, Inorg. Chem. 1985, 24, 4736.

[39] a) M. J. Maroney, E. O. Fey, D. A. Baldwin, R. E. Stenkamp,

L. H. Jensen, N. J. Rose, Inorg. Chem. 1986, 25, 1409; b) L.

Ballester, A. Gutierrez, M. F. Perpinan, S. Rico, M. T. Az-

condo, C. Belletto, Inorg. Chem. 1999, 38, 4430.

[40] I. Käpplinger, PhD Thesis, University of Jena, 2002.

[41] a) W. R. Scheidt, K. J. Haller, K. Hatano, J. Am. Chem. Soc.

1980, 102, 3017; b) M. Schappacher, J. Fischer, R. Weiss, Inorg.

Chem. 1989, 28, 389; c) T. J. Bartczak, S. Wolowiec, L. Latos-

Grazynski, Inorg. Chim. Acta 1998, 277, 242.

[42] M. Nakamura, T. Ikeue, H. Fujii, T. Toshimura, J. Am. Chem.

Soc. 1997, 119, 6284.

[43] Autorenkollektiv, Organikum, Johann Ambrosius Barth Ver-

lagsgesellschaft mbH, Leipzig, Berlin, Heidelberg, 1993.

[44] M. Grodzicki, J. Phys. B 1980, 13, 2683.

[26]

[27]

[28]

[29]

[30]

[31] O. Einsle, A. Messerschmidt, P. Stach, G. P. Bourenkov, H. D.

Bartunik, R. Huber, P. M. H. Kroneck, Nature 1999, 400, 476.

[32] a) H. Nasri, J. A. Goodwin, W. R. Scheidt, Inorg. Chem. 1990,

29, 185; b) H. Nasri, Y. Wang, B. H. Huynh, A. F. Walker,

[45] M. Grodzicki, “Theory and Application of the Self-Consistent-

Charge Xα Method”, habilitation thesis, Hamburg, Germany,

1985.

W. R. Scheidt, Inorg. Chem. 1991, 30, 1483; c) H. Nasri, K. J. [46] a) E.-G. Jäger, Z. Chem. 1968, 8, 30; b) E.-G. Jäger, Z. Chem.

Haller, Y. Wang, B. H. Huynh, W. R. Scheidt, Inorg. Chem.

1992, 31, 3459; d) M. K. Ellison, C. E. Schulz, W. R. Scheidt,

Inorg. Chem. 1999, 38, 100; e) H. Nasri, M. K. Ellison, S. Chen,

B. H. Huynh, W. R. Scheidt, J. Am. Chem. Soc. 1997, 119,

6724.

1968, 8, 470.

[47] L. Claisen, Justus Liebigs Ann. Chem. 1897, 279, 1.

[48] a) COLLECT, Data Collection Software, B. V. Nonius, The Ne-

therlands, 1998; b) Z. Otwinowski, W. Minor, Processing of X-

ray Diffraction Data Collected in Oscillation Mode, in Methods

in Enzymology, Macromolecular Crystallography, Part A (Eds.:

C. W. Carter, R. M. Sweet), Academic Press, San Diego, 1997,

vol. 276, p. 307.

[33] L. Wolf, E.-G. Jäger, Z. Anorg. Allg. Chem. 1966, 346, 76.

[34] a) A. R. Newman, A. N. French, Inorg. Chim. Acta 1987, 129,

L37; b) M. G. Finnegan, A. G. Lappin, W. R. Scheidt, Inorg.

Chem. 1990, 29, 181; c) M. Frangione, J. Port, M. Baldiwala,

A. Judd, J. Galley, M. DeVega, K. Linna, L. Caron, E. An-

derson, J. A. Goodwin, Inorg. Chem. 1997, 36, 1904.

[35] a) T. Ikeue, Y. Ohgo, T. Yamaguchi, M. Takahashi, M. Takeda,

M. Nakamura, Angew. Chem. Int. Ed. 2001, 40, 2617; b) W. O.

Koch, V. Schünemann, M. Gerdan, A. X. Trautwein, H.-J.

Krüger, Chem. Eur. J. 1998, 4, 686; c) F. V. Wells, S. W.

McCann, H. H. Wickman, S. L. Kessel, D. N. Hendrickson,

R. D. Feltham, Inorg. Chem. 1982, 21, 2306; d) E. König, G.

Ritter, J. Dengler, L. F. Larkworthy, Inorg. Chem. 1992, 31,

1196.

[49] G. M. Sheldrick, Acta Crystallogr., Sect. A 1990, 46, 467.

[50] G. M. Sheldrick, SHELXL-97, University of Göttingen, Ger-

many, 1993.

[51] CCDC-240934 (Fe1), -240945 (Fe1MeOH), -240936 (Fe1Cl),

-240937 {Na[Fe2(NO₂)₂](MeOH)₂(H₂O)_0.5}, -241784 [Fe3-

(Py)₂], -241781 {[Fe1(Py)₂]PF₆}, -241782 {[Fe1(N-MeIm)₂]

ClO₄(H₂O)}, -241785 {(BzEt₃N)[Fe1(NCS)₂]}, -241783

{Et₄N[Fe2(CN)₂](H₂O)}, -241780 {[Fe2(NO₂)OH₂]MeOH},

-266535 [(Fe1)₂dabco], -266536 [(Fe2)₂dabco], and -266537

[(Fe3)₂dabco·MeOH] contain the supplementary crystallo-

graphic data for this paper. These data can be obtained free of

charge from The Cambridge Crystallographic Data Centre via

www.ccdc.cam.ac.uk/data_request/cif.

[36] P. Gütlich, A. Hauser, H. Spiering, Angew. Chem. 1994, 106,

2109; Angew. Chem. Int. Ed. Engl. 1994, 33, 2024.

[37] a) M. Mylrajan, L. A. Andersson, J. Sun, T. M. Loer, C. S.

Thomas, E. P. Sullivan Jr., M. A. Thomson, K. M. Long, O. P.

Received: September 13, 2004

Published Online: June 14, 2005

Eur. J. Inorg. Chem. 2005, 2794–2811

www.eurjic.org

2811

Article Doi

The relationship between the structure and magnetic properties of bioinspired iron(II/III) complexes with Schiff-base-like chelate ligands, part I: Complexes with dianionic [N4] macrocycles

DOI: 10.1002/ejic.200400759

Source and publish data:

Authors:

Article abstract of DOI:10.1002/ejic.200400759

Full text of DOI:10.1002/ejic.200400759

Products guided by the article

R&D Labs maybe for 85413-24-7

Relevant to this article

Hot Product