Change in the spin state of an FeIII center upon one N-to-O switch in the coordination of a 2,6-pyridinedicarboxamido unit: The effect of methyl thioether and methyl ether appendages at the ligand periphery

Article abstract of DOI:10.1002/ejic.200390067

Two ligands bearing the 2,6-pyridinedicarboxamide unit, namely PyPSMeH₂ and PyPOMeH₂, have been used to prepare complexes of Fe^III which have been structurally characterised. The potentially pentadentate thioether ligand PyPSMeH₂ affords only the bis complex (Et₄N)[Fe(PyPSMe)₂] (1), irrespective of the metal to ligand ratio, due to the weak affinity of the thioether groups for the Fe^III center. In [Fe(PyPSMe)₂]^-, the two PyPSMe^2- ligands bind to the iron center in a mer fashion and steric crowding among the pendant thioether groups forces one of the four deprotonated carboxamido moieties to bind to the Fe^III center through the carbonyl oxygen, giving rise to an FeN₅O chromophore. The tridentate ligand PyPOMeH₂, in its deprotonated form, also gives rise to a bis complex, (Et₄N)[Fe(PyPOMe)₂] (2), in which the two PyPOMe^2- ligands bind to the iron center in a mer fashion. In this case, however, less steric crowding among the smaller pendant ether groups allows binding of all four carboxamido nitrogens to the Fe^III center in [Fe(PyPOMe)₂]^- giving rise to an FeN₆ chromophore. This one N-to-O switch in donor atom makes the two complexes 1 and 2 very different. While complex 2 is low spin (g = 2.18, 1.94) much like the other reported Fe^III bis complexes with two ligated 2,6-pyridinedicarboxamido units, complex 1 is high spin (g = 4.3). The spin flip due to a switch in only one N or O donor atom in these two complexes is unprecedented. The two complexes also exhibit different stability towards water. ( Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim, Germany, 2003).

Full text of DOI:10.1002/ejic.200390067

FULL PAPER
Change in the Spin State of an FeIII Center upon One N-to-O Switch in the
Coordination of a 2,6-Pyridinedicarboxamido Unit: The Effect of Methyl
Thioether and Methyl Ether Appendages at the Ligand Periphery
Todd C. Harrop,[a] Laurie A. Tyler,[a] Marilyn M. Olmstead,[b] and Pradip K. Mascharak*[a]
Keywords: Iron / N ligands / S ligands / Spin states / EPR spectroscopy
Two ligands bearing the 2,6-pyridinedicarboxamide unit,
namely PyPSMeH2 and PyPOMeH2, have been used to pre-
mer fashion. In this case, however, less steric crowding
among the smaller pendant ether groups allows binding of
pare complexes of FeIII which have been structurally charac- all four carboxamido nitrogens to the FeIII center in
terised. The potentially pentadentate thioether ligand
PyPSMeH2 affords only the bis complex (Et4N)[Fe(PyPSMe)2] one N-to-O switch in donor atom makes the two complexes
(1), irrespective of the metal to ligand ratio, due to the weak 1 and 2 very different. While complex 2 is low spin (g = 2.18,
affinity of the thioether groups for the FeIII center. In 1.94) much like the other reported FeIII bis complexes with
[Fe(PyPOMe)2]− giving rise to an FeN6 chromophore. This
[Fe(PyPSMe)2]−, the two PyPSMe2− ligands bind to the iron
center in a mer fashion and steric crowding among the pen-
dant thioether groups forces one of the four deprotonated
carboxamido moieties to bind to the FeIII center through the
carbonyl oxygen, giving rise to an FeN5O chromophore. The
tridentate ligand PyPOMeH2, in its deprotonated form, also
gives rise to a bis complex, (Et4N)[Fe(PyPOMe)2] (2), in
which the two PyPOMe2− ligands bind to the iron center in a
two ligated 2,6-pyridinedicarboxamido units, complex 1 is
high spin (g = 4.3). The spin flip due to a switch in only one
N or O donor atom in these two complexes is unprecedented.
The two complexes also exhibit different stability towards
water.
( Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2003)
Introduction
ters.[7,8,9] In addition, it appears that FeIII complexes with
FeIIIϪNamido bonds are resistant toward hydrolysis despite
the common belief that the highly basic carboxamido nitro-
gens could be readily attacked by water. Collectively, these
results indicate that carboxamido nitrogens have a high af-
finity for FeIII centers and impart stability towards hydro-
lysis of the resultant complexes.
In recent years, coordination of carboxamido nitrogen(s)
to iron centers and the spectroscopic, redox and reactivity
properties of the resulting complexes have received much
interest following the discovery of ironϪNamido (Namido
ϭ
carboxamido N) bonds in nitrogenase[1] and nitrile hy-
dratase.[2,3] Initially, this type of coordination was thought
to be highly unlikely since FeIII rapidly precipitates out as
its hydroxide at much lower pH values than that required
for the deprotonation of carboxamido nitrogens.[4] During
the past ten years, however, research by us,[5Ϫ10] and
others[11Ϫ15] has demonstrated that coordination of carbox-
amido nitrogens to FeIII centers can afford sufficiently
stable and isolable complexes. Indeed, coordination of this
group results in highly negative E1/2 values (up to Ϫ1.10 V)
indicating that carboxamido nitrogens provide significant
stability to iron in the ϩ3 oxidation state.[7] Furthermore,
the majority of FeIII complexes coordinated to one, two, or
four carboxamido nitrogens contain low spin FeIII cen-
Coordination of designed ligands with built-in 2,6-
pyridinedicarboxamido unit (Figure 1) to FeIII centers has
resulted in complexes with novel structures and proper-
ties.[5,6,9,11,15] Recent work in this laboratory has afforded a
variety of FeIII complexes based on this tridentate moiety
in the first coordination sphere.[5,6,10] For example, coor-
dination of the pentadentate ligand POPyH4 to FeIII (H
represents dissociable amide protons) results in high-spin,
seven-coordinate, pentagonal bipyramidal complexes with
the entire ligand in the basal plane and with two exogenous
axial ligands (structure i, Figure 2).[6] In contrast, the FeIII
complex of the similar pentadentate ligand PyPSH4 is high
spin, five-coordinate with a distorted trigonal bipyramidal
geometry due to the steric bulk of the thiolato sulfur (struc-
ture ii, Figure 2).[10] It is interesting to note that the poten-
tially pentadentate ligands Py3PH2 and MePy3PH2 invari-
ably afford six-coordinate bis complexes in which the two
ligands are coordinated to the FeIII center through the
2,6-pyridinedicarboxamido moiety with unbound pendant
[a]
Department of Chemistry and Biochemistry, University of
California,
Santa Cruz, CA 95064, USA
Fax: (internat.) ϩ1-831/459-2935
E-mail: pradip@chemistry.ucsc.edu
[b]
Department of Chemistry, University of California,
Davis, CA 95616, USA
Eur. J. Inorg. Chem. 2003, 475Ϫ481  2003 WILEY-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim 1434Ϫ1948/03/0203Ϫ0475 $ 20.00ϩ.50/0
475

T. C. Harrop, L. A. Tyler, M. M. Olmstead, P. K. Mascharak
FULL PAPER
PyPSH4, we report here that the deprotonated form,
PyPSMe2Ϫ, coordinates to FeIII in a bis fashion irrespective
of the ligand to Fe ratio and in the resulting complex,
(Et4N)[Fe(PyPSMe)2] (1), all the thioether groups remain
unbound. Clearly, thioether sulfur (like pyridine nitrogen)
is not a good donor for FeIII and hence PyPSMeH2 behaves
like Py3PH2. It is also interesting to note that due to the
steric repulsion among the pendant thioether groups in 1,
one of the carboxamido groups employs the carbonyl oxy-
gen instead of the carboxamido nitrogen, as a donor atom,
resulting in an FeN5O chromophore. As a result of this N-
to-O donor switch around the FeIII center in 1, the complex
is high spin, which is in contrast to other similar bis com-
plexes with two bound 2,6-pyridinedicarboxamido units
(FeN6 chromophore and all low spin).[5,11] In order to fur-
ther substantiate our hypothesis that the steric repulsion
among the pendant thioether groups in 1 is responsible for
this N-to-O donor switch, we have also synthesized the bis
FeIII complex of the tridentate ligand PyPOMeH2. The pen-
dant ether groups in (Et4N)[Fe(PyPOMe)2] (2) give rise to
steric hindrance around the FeIII center much like the
crowding in 1. As shown later, the structure of 2 is very
similar to that of 1 except for the crucial fact that the extent
of the steric repulsion among the pendant ether groups in
2 is slightly less than that among the pendant thioether
groups in 1. This allows the four carboxamido nitrogens of
the two PyPOMe2Ϫ ligands to bind to the FeIII center in 2.
Since there is no N-to-O switch, complex 2 is once again
low spin much like [FeL2]Ϫ and [Fe(Py3P)2]Ϫ.
Figure 1. Ligands with the 2,6-pyridinedicarboxamide moiety (P
stands for the peptide functionality in all abbreviations)
pyridine groups (structure iii, Figure 2).[5] The FeIII centers
in both of these complexes are low spin, much like the FeIII
complex of LH2, namely [FeL2]Ϫ, reported by Mukherjee
and co-workers.[11] The failure of the pendant pyridine units
to bind to the FeIII centers in these complexes is not related
to any steric effect, since Py3PH2 gives rise to many
[Co(Py3P)(L)]ϩ/0 complexes.[16] We believe that formation
of the bis complex [Fe(Py3P)2]Ϫ instead of [Fe(Py3P)(L)]ϩ/0
with weak FeIII-py interactions is a result of the high
stability derived from coordination of four carboxamido
nitrogens.
In this paper, we report (a) the syntheses of complexes 1
and 2, (b) comparisons of the structures of 1 and 2 with
other related FeIII complexes with bound 2,6-pyridinedicar-
boxamide moieties, and (c) studies on spin-flip due to the
N-to-O switch in 1 and 2.
Results and Discussion
In order to elucidate the effect(s) of steric hindrance on
the mode of coordination of carboxamido nitrogens to FeIII
centers and the spin-states of the resulting complexes, we
The potentially pentadentate ligand PyPSMeH2 was syn-
have synthesized additional ligands with the 2,6-pyridine- thesized by following a procedure previously reported by
dicarboxamide moiety. The two designed ligands employed this group.[17] PyPOMeH2 was prepared by reacting o-anisi-
in this study are the potentially pentadentate ligand dine with 2,6-pyridinedicarbonyl dichloride (prepared from
PyPSMeH2 and the tridentate ligand PyPOMeH2. Al- 2,6-pyridinedicarboxylic acid and thionyl chloride) in chlo-
though PyPSMeH2 is expected to bind to iron much like roform. Reaction of (Et4N)[FeCl4] with PyPSMe2Ϫ in DMF
Figure 2. Different structures of FeIII complexes of ligands with the 2,6-pyridinedicarboxamide moiety
476
Eur. J. Inorg. Chem. 2003, 475Ϫ481

Effect of Methyl Thioether and Methyl Ether Appendages at the Ligand Periphery
afforded the red-brown FeIII complex (Et4N)[Fe(PyPSMe)2]
FULL PAPER
Table 1. Summary of crystal data and intensity collection and
structure refinement parameters for (Et4N)[Fe-
(1) in good yield. It is important to note that the bis com-
plex 1 is the only product isolated from this reaction irre-
spective of the ligand to Fe ratio. Thus when a 1:1 ligand
to Fe ratio is used, only half of the iron is complexed by
the ligand while the rest remains in solution in the form of
[FeCl4]Ϫ. Clearly, the strong preference of the carboxamido
nitrogens for FeIII, and the weak affinity of the thioether
sulfur for FeIII lead to the formation of 1 as the sole prod-
uct. The bright red complex (Et4N)[Fe(PyPOMe)2] (2) was
synthesized in high yield by reacting (Et4N)[FeCl4] with two
(PyPSMe)2]·1.7CH3CN·Et2O (1·1.7CH3CN·Et2O) and (Et4N)-
[Fe(PyPOMe)2]·2H2O (2·2H2O)
Parameters
1·1.7CH3CN·Et2O
2·2H2O
Empirical formula
Molecular weight
C57.40H69.10FeN8.70O5S4
C50H58FeN7O10
972.88
1145.00
18.772(4)
14.548(3)
21.297(4)
90
90.787(8)
90
5816(2)
4
monoclinic
P21/n
1.308
91(2)
0.458
1.019
6.76
14.97
˚
a [A]
11.0430(13)
12.3769(17)
17.997(2)
89.039(3)
76.953(3)
75.930(2)
2322.5(5)
2
˚
b [A]
˚
c [A]
α [°]
equivalents of PyPOMeH2 and four equivalents of NaH β [°]
(both relative to iron) in DMF.
γ [°]
3
˚
V [A ]
Z
Molecular Structure of (Et4N)[Fe(PyPSMe)2] (1)
Crystal system
Space group
triclinic
¯
P1
ρ [g·cmϪ3
T (K)
]
1.391
91(2)
0.393
1.024
3.77
An X-ray crystallographic study of 1 (Table 1) revealed
that the coordination geometry around the FeIII center is
distorted octahedral, with two PyPSMe2Ϫ units ligated in a
mer fashion (Figure 3). Interestingly the FeIIIN5O chromo-
phore consists of two pyridine nitrogens, three depro-
tonated carboxamido nitrogens, and the carbonyl oxygen of
the fourth deprotonated carboxamido moiety. This ob-
served mode of binding of the two PyPSMe2Ϫ ligands in 1
is not accidental and has also been noted in the correspond-
ing CoIII complex (Et4N)[Co(PyPSMe)2] (3).[17] The last
carboxamido nitrogen moiety fails to bind through nitrogen
µ, [mmϪ1
]
GOF[a] on F2
[b]
R1
Rw2
[c]
9.04
[a]
2
2 2
GOF ϭ [Σ[w(F0 Ϫ Fc ) ]/(M Ϫ N)]1/2 (M ϭ number of
[b]
reflections, N ϭ number of parameters refined).
R1 ϭ Σ||F0| Ϫ
|Fc||/Σ|F0|. Rw2 ϭ {Σ[w(F0 Ϫ Fc ) ]/Σ[w(F0 ) ]}1/2
.
[c]
2
2 2
2 2
The FeIIIϪNamido [av. 2.093(3) A] and FeIIIϪNpy [av.
˚
˚
due to the steric repulsion that builds up between the pen- 2.081(3) A] bond lengths in 1 are significantly longer than
dant thioether groups as an increasing number of carbox- those reported for similar FeIII complexes containing two
amido nitrogens become ligated to the MIII center. To the coordinated 2,6-pyridinedicarboxamido moieties in their
best of our knowledge, there is no other example of such primary coordination sphere.[5,11] A close examination of
an N-to-O donor switch in FeIII complexes containing car- the literature reveals three bis complexes of FeIII that con-
boxamido group(s). A few examples of N-to-O linkage iso- tain the 2,6-pyridinedicarboxamido unit, namely
merization have been reported with ruthenium com- (Et4N)[FeL2],[11] Na[Fe(Py3P)2],[5] and Na[Fe(MePy3P)2].[5]
plexes[18] but these complexes contain neutral and not de- All three complexes consist of an FeIIIN6 chromophore
protonated carboxamide unit(s).
with four deprotonated carboxamido nitrogens in the basal
Figure 3. Thermal ellipsoid plot of [Fe(PyPSMe)2]Ϫ, the anion of 1 (at the 50% probability level); H atoms are omitted for the sake
of clarity
Eur. J. Inorg. Chem. 2003, 475Ϫ481
477

T. C. Harrop, L. A. Tyler, M. M. Olmstead, P. K. Mascharak
FULL PAPER
plane and two axial pyridine donors. As a result of the boxamido nitrogens in its coordination sphere and hence
strong ligand field of the carboxamido nitrogens, these FeIII resembles typical bis complexes with bound 2,6-pyridined-
complexes are all low spin (vide infra) and exhibit shorter icarboxamido units such as [Fe(Py3P)2]Ϫ. It is interesting to
FeIIIϪNamido [av. 1.961 A] and FeIIIϪNpy [av. 1.877 A] bond note that by simply decreasing the size of the pendant
lengths.[5,11] Interestingly, the bond lengths of 1 compare groups from SMe to OMe at the ligand periphery, a more
well with those noted for high spin (vide infra) FeIII com- sterically favorable situation can be created such that all six
plexes that contain one 2,6-pyridinedicarboxamido unit in nitrogens are able to bind to the FeIII center.
the primary coordination sphere.[6,10,15] The CϪO bond
˚
˚
˚
length of the O-bound carbonyl group [1.286(4) A] is signi-
ficantly longer than the CϪO bond of the N-bound car-
˚
bonyl groups [av. 1.236 A] as a result of coordination of the
carbonyl oxygen to the FeIII center. Similarly, the CϪN
˚
bond of the O-bound carboxamido unit [1.299(4) A] is, as
expected, shorter than the CϪN bond of the N-bound car-
˚
boxamido group [av. 1.336 A] (Table 2).
˚
Table 2. Selected bond lengths [A] and angles [°] for
(Et4N)[Fe(PyPSMe)2]·1.7CH3CN·Et2O (1·1.7CH3CN·Et2O) and
(Et4N)[Fe(PyPOMe)2]·2H2O (2·2H2O)
Complex 1
FeϪO(4)
FeϪN(5)
FeϪN(3)
FeϪN(2)
FeϪN(1)
O(4)ϪFeϪN(5) 75.73(10)
O(4)ϪFeϪN(3) 90.39(10)
N(5)ϪFeϪN(3) 112.90(11)
O(4)ϪFeϪN(2) 115.67(10)
N(5)ϪFeϪN(2) 166.85(10)
N(3)ϪFeϪN(2) 74.92(11)
O(4)ϪFeϪN(1) 92.37(10)
N(5)ϪFeϪN(1) 99.07(11)
N(3)ϪFeϪN(1) 147.56(12)
N(2)ϪFeϪN(1) 74.80(11)
O(4)ϪFeϪN(4) 149.29(11)
2.014(2)
2.072(3)
2.077(3)
2.090(3)
2.091(3)
FeϪN(4)
O(1)ϪC(8)
N(1)ϪC(8)
2.111(3)
1.236(4)
1.336(4)
1.286(4)
1.299(4)
74.50(11)
94.25(12)
94.81(11)
99.59(12)
O(4)ϪC(35)
N(6)ϪC(35)
N(5)ϪFeϪN(4)
N(3)ϪFeϪN(4)
N(2)ϪFeϪN(4)
N(1)ϪFeϪN(4)
O(4)ϪC(35)ϪN(6) 129.7(3)
O(4)ϪC(35)ϪC(34) 114.3(3)
N(6)ϪC(35)ϪC(34) 115.9(3)
Figure 4. Thermal ellipsoid plot of [Fe(PyPOMe)2]Ϫ, the anion of
2 (at the 50% probability level); H atoms are omitted for the sake
of clarity.
O(1)ϪC(8)ϪN(1)
O(1)ϪC(8)ϪC(9)
N(1)ϪC(8)ϪC(9)
129.0(3)
119.8(3)
111.3(3)
The FeIIIϪNamido [av. 1.979(3) A] and FeIIIϪNpy [av.
˚
˚
1.875(3) A] bond lengths of 2 are similar to those noted
in FeIII complexes with FeN6 chromophores arising from
coordination of two 2,6-pyridinedicarboxamido units.[5,11]
However, these bonds are consistently shorter than those in
1 due to the low spin nature of the FeIII center in 2 (vide
infra). Furthermore, the octahedral geometry of the N6
framework in 2 is less distorted than that formed by the
N5O framework in 2. For example, the NpyϪFeϪNpy and
Complex 2
FeϪN(2)
FeϪN(5)
FeϪN(3)
FeϪN(1)
N(2)ϪFeϪN(5) 178.14(5)
N(2)ϪFeϪN(3) 81.52(5)
N(5)ϪFeϪN(3) 96.98(5)
N(2)ϪFeϪN(1) 81.01(5)
N(5)ϪFeϪN(1) 100.49(5)
N(3)ϪFeϪN(1) 162.52(5)
N(2)ϪFeϪN(6) 99.89(5)
N(5)ϪFeϪN(6) 81.22(5)
N(3)ϪFeϪN(6) 92.00(5)
1.8743(12) FeϪN(2)
1.8757(12) FeϪN(4)
1.9591(11) N(1)ϪC(1)
1.9802(12) O(1)ϪC(1)
1.9842(12)
1.9924(12)
1.3430(18)
1.2438(17)
91.06(5)
97.86(5)
81.04(5)
91.12(5)
91.20(5)
162.23(5)
128.20(13)
120.41(13)
111.39(12)
N(1)ϪFeϪN(6)
N(2)ϪFeϪN(4)
N(5)ϪFeϪN(4)
N(3)ϪFeϪN(4)
N(1)ϪFeϪN(4)
N(6)ϪFeϪN(4)
O(1)ϪC(1)ϪN(1)
O(1)ϪC(1)ϪC(2)
N(1)ϪC(1)ϪC(2)
N
amidoϪFeϪNamido angles in 1 (166.85° and 147.56° re-
spectively) are substantially smaller than the corresponding
angles in 2 (178.14 ° and 162.23° respectively, Table 2).
Spectroscopic Properties
Coordination of PyPSMe2Ϫ and PyPOMe2Ϫ to the FeIII
centers is readily indicated by the IR spectra of the com-
plexes. Ligation of the deprotonated carboxamido nitrogens
to the metal centers results in a red-shift of the carbonyl
stretching frequency (νco) from 1684 cmϪ1 (in both
Molecular Structure of (Et4N)[Fe(PyPOMe)2] (2)
The coordination geometry around the FeIII center in PyPSMeH2 and PyPOMeH2) to 1593 cmϪ1 in 1 and 1603
[Fe(PyPOMe)2]Ϫ is distorted octahedral with two cmϪ1 in 2. Also, the NϪH stretching frequency of the free
PyPOMe2Ϫ ligands coordinated in a mer fashion (Figure 4). ligand (ഠ 3200 cmϪ1 for both ligands) is absent in the IR
Unlike 1, complex 2 consists of an FeIIIN6 chromophore spectra of the iron complexes confirming coordination of
with two axial pyridine nitrogens and four equatorial car- the ligands in their deprotonated forms.
478
Eur. J. Inorg. Chem. 2003, 475Ϫ481

Effect of Methyl Thioether and Methyl Ether Appendages at the Ligand Periphery
FULL PAPER
The EPR spectrum of 1 in DMF glass at 80K confirms
Complexes 1 and 2 are soluble in a variety of solvents
the high spin nature of the FeIII center, displaying a strong such as DMF and CH3CN giving deep red-orange solu-
signal at g ϭ 4.3 (Figure 5). This is contrary to other bis tions. This color arises from strong ligand-to-metal charge
FeIII complexes with coordinated 2,6-pyridinedicarboxam- transfer (LMCT) absorptions at 450 nm for 1 and 440 nm
ido units.[5,11] The FeIII centers in such complexes are ex- for 2 (both in DMF). Similar LMCT bands have been ob-
clusively low spin and exhibit axial EPR resonances with g served with analogous iron() complexes.[5,11] Although 1
values in the range of 2.19Ϫ1.94. This is also true for com- is stable in aprotic solvents, it decomposes readily in water.
plex 2. In the same DMF glass, complex 2 exhibits an axial The initial red solution loses its color within minutes and
EPR spectrum (g ϭ 2.18, 1.94) typical of a low spin FeIII free PyPSMeH2 precipitates from the solution as a white
center (Figure 5). The only difference between the two bis solid. This behavior contrasts sharply with that of other
complexes 1 and 2 is the fact that in 1, one of the four FeIII complexes with FeN6 chromophores, such as
carboxamido moieties employs the carbonyl oxygen to bind Na[Fe(Py3P)2],[5] Na[Fe(MePy3P)2],[5] and especially com-
the FeIII center giving rise to an FeN5O chromophore. This plex 2. It is therefore evident that the O-bonded FeIII com-
rather simple N-to-O switch in the donor set causes a spin plex 1 is inherently unstable, presumably due to its high
flip and gives rise to the high spin complex 1. In 2, all six spin nature.
nitrogens from the two 2,6-pyridinedicarboxamido units are
bonded to FeIII, and the FeN6 chromophore is low spin
much like the other complexes of a similar kind.[5,11] To the
best of our knowledge, change in the spin state due to only
one N-to-O donor switch in otherwise similar complexes of
FeIII with carboxamide ligands has not been observed be-
fore. It appears that the FeϪOcarbonyl interaction exerts a
crucial effect on the strong ligand field of the carboxamido
nitrogens in these species and just one such interaction can
change the overall spin state of the complex. A similar trend
in spin state has also been observed in other FeIII complexes
that contain FeϪNamido bonds in the presence of FeϪOphen
bonds (where phen ϭ phenolato oxygen).[8]
Conclusion
In conclusion, we have synthesized and structurally char-
acterized two new, six-coordinate FeIII complexes that con-
tain the 2,6-pyridinedicarboxamido unit. With the poten-
tially pentadentate thioether ligand PyPSMeH2, only the
bis complex (Et4N)[Fe(PyPSMe)2] (1) is obtained irrespect-
ive of the ligand to Fe ratio. This strongly suggests that
thioether sulfurs are not good donor groups for FeIII in the
presence of carboxamido groups. On inspection of the few
FeIII complexes with bound thioether groups, it becomes
evident that coordination of a thioether sulfur to FeIII oc-
curs only when the thioether group(s) exist between other
strong donor groups on the ligand frame.[19Ϫ21] The steric
repulsion among the pendant thioether groups in 1 gives
rise to an FeN5O coordination sphere in which one car-
bonyl oxygen binds to the FeIII center. It appears that this
subtle N-to-O switch in donor atom results in a high spin
FeIII center in 1. In the case of the other FeIII complex
(Et4N)[Fe(PyPOMe)2] (2), the smaller pendant ether groups
allow coordination of all four carboxamido nitrogens re-
sulting in a low spin complex with an FeN6 chromophore.
Collectively, these results indicate that the spin state of an
FeIII center coordinated by 2,6-pyridinedicarboxamido
moieties can be influenced by simple addition of bulky
groups to the ligand periphery. Further modifications to the
2,6-pyridinedicarboxamide ligand frame should enable the
preparation of other FeIII complexes with novel structural
and spectroscopic properties. Such investigations are cur-
rently being performed in this laboratory.
Experimental Section
Physical Measurements: Infrared spectra were recorded with a
PerkinϪElmer 1600 FTIR spectrophotometer. Absorption spectra
were recorded on a PerkinϪElmer Lambda 9 spectrophotometer.
1H and 13C NMR spectra were recorded at 25 °C on a 500 MHz
Varian Unity Plus spectrometer. EPR spectra at X-band frequen-
cies were obtained at 80K with a Bruker ESP-300 spectrometer.
Figure 5. X-band EPR spectra of 1 (bottom panel) and 2 (top
panel) in DMF glass at 80 K; instrument settings: microwave power
13 mW, microwave frequency 9.43 GHz, modulation frequency 100
KHz, modulation amplitude 2 G; selected g values are shown
Materials and Methods: All manipulations were carried out using
standard Schlenk techniques. 2,6-Pyridinedicarboxylic acid, 2-
Eur. J. Inorg. Chem. 2003, 475Ϫ481
479

T. C. Harrop, L. A. Tyler, M. M. Olmstead, P. K. Mascharak
FULL PAPER
(methylthio)aniline, o-anisidine, and sodium hydride were pur- X-ray Crystal Structure Analysis: Dark red/brown crystals of
chased from Aldrich and used without further purification. (Et4N)[Fe(PyPSMe)2]·1.7CH3CN·Et2O (1·1.7CH3CN·Et2O) were
were synthesized by following grown from CH3CN/Et2O (3:1) at Ϫ20 °C. Bright red X-ray quality
(Et4N)[FeCl4][22] and PyPSMeH2
[17]
published procedures. 2,6-Pyridinedicarbonyl dichloride was pre-
pared by refluxing 2,6-pyridinedicarboxylic acid in neat thionyl
blocks of (Et4N)[Fe(PyPOMe)2]·2H2O (2·2H2O) were grown from
CH3CN/Et2O/THF (3:1:1) at Ϫ20 °C. Diffraction data for both
chloride. All solvents were distilled from the appropriate drying complexes were collected at 90 K on a Bruker SMART 1000 dif-
˚
agents prior to use: DMF from BaO, CH3CN from CaH2, THF
from Na/benzophenone, Et2O from Na, and CHCl3 from CaCl2.
fractometer using Mo-Kα (0.71073 A) radiation, and an absorption
correction was applied in each case. The structures were solved by
direct
methods
(SHELXS-97).
In
the
structure
of
1·1.7CH3CN·Et2O, there is disorder of one of the terminal groups
of the ligand containing C22ϪS3ϪC28. This disorder has been
omitted in Figure 3 for sake of clarity. Instrument parameters, crys-
tal data, and data collection parameters for 1·1.7CH3CN·Et2O and
2·2H2O are summarized in Table 1. Selected bond lengths and
angles are listed in Table 2.
Preparation of PyPOMeH2: A solution of o-anisidine (3.79 g,
30 mmol) and Et3N (5.00 g, 50 mmol) in CHCl3 (40 mL) was
slowly added to a solution of 2,6-pyridinedicarbonyl dichloride
(3.11 g, 15 mmol) and Et3N (5.00 g, 50 mmol) in CHCl3 (50 mL)
at 0 °C. The mixture was stirred at room temperature for 24 h.
The resultant red solution was washed with aqueous NaHCO3 and
NaCl. The CHCl3 layer was then dried over MgSO4, filtered, and
the solvent was removed by rotary evaporation to yield a red oil.
The oil was triturated with Et2O (25 mL) to yield a light brown
solid (3.75 g, 65% yield). 1H NMR (500 MHz, CDCl3): δ ϭ 3.95
(s, 6 H, OCH3), 7.00 (d, 2 H, ArH), 7.10 (t, 2 H, ArH), 7.15 (t, 2
H, ArH), 8.14 (t, 1 H, ArH), 8.55 (d, 2 H, ArH), 8.61 (d, 2 H,
ArH), 10.36 (s, 2 H, NH) ppm. 13C NMR (500 MHz, CDCl3): δ ϭ
56.14 (OCH3), 110.54 (ArC), 120.24 (ArC), 121.50 (ArC), 124.55
(ArC), 125.34 (ArC), 127.44 (ArC), 139.56 (ArC), 148.97 (ArC),
149.41 (ArC), 161.15 (CϭO) ppm. Selected IR absorption bands
(KBr pellet): ν˜ ϭ 3371 cmϪ1 (s, NϪH), 3014 (w), 2979 (w), 1684
(vs, CϭO), 1602 (s), 1536 (s), 1488 (s), 1461 (s), 1334 (m), 1291
(m), 1248 (s), 1220 (m), 1175 (w), 1138 (w), 1110 (m), 1070 (m),
1047 (m), 1029 (s), 1000 (w), 947 (w), 925 (w), 890 (w), 845 (w),
790 (w), 758 (w), 702 (w), 671 (s), 594 (w).
CCDC-187840 (1·1.7CH3CN·Et2O) and CCDC-187841 (2·2H2O)
contain the supplementary crystallographic data for this
paper. These data may be obtained free of charge at
www.ccdc.cam.ac.uk/conts/retrieving.html [or from the Cambridge
Crystallographic Data Centre, 12 Union Road, Cambridge,
CB2 1EZ, UK (Fax: (internat.) ϩ44-1223-336-033; E-mail:
deposit@ccdc.cam.ac.uk].
Acknowledgments
This research was supported by a grant from the NIH (GM61636).
T.C.H. received support from the NIH IMSD grant GM 58903.
L. A. T. was a recipient of a GAAN fellowship from the U. S.
Department of Education.
(Et4N)[Fe(PyPSMe)2] (1): NaH (0.037 g, 1.54 mmol) was added to
a solution of PyPSMeH2 (0.310 g, 0.76 mmol) in DMF (15 mL)
and the mixture was stirred until all the NaH had reacted. To the
resultant yellow solution was added a solution of (Et4N)[FeCl4]
(0.120 g, 0.38 mmol) in DMF (2 mL). The resultant dark red solu-
tion was stirred for 2 h at 25 °C. The solvent was then removed
under vacuum and the residue dissolved in CH3CN (15 mL) and
filtered to remove NaCl. Diethyl ether (7 mL) was added to the
CH3CN filtrate and the resulting solution was stored at Ϫ20 °C
for 48 h. The dark red crystalline solid thus obtained was collected
by filtration and dried under vacuum (0.190 g, 51% yield). Selected
IR absorption bands: (KBr pellet): ν˜ ϭ 3049 cmϪ1 (w), 2983 (w),
2918 (w), 1593 (vs, νCO), 1461 (m), 1438 (w), 1377 (m), 1342 (w),
1144 (w), 1068 (w), 968 (w), 749 (w). Absorption spectrum in
DMF: λmax (ε, Ϫ1·cmϪ1) ϭ 550 nm sh, 450 (7000), 365 (12 000).
[1]
J. W. Peters, M.H. Stowell, M. Soltis, M. G. Finnegan, M. K.
Johnson, D. C. Rees, Biochemistry 1997, 36, 1181Ϫ1187.
S. Nagashima, M. Nakasako, N. Dohmae, M. Tsujimura, K.
Takio, M. Odaka, M. Yohda, N. Kamiya, I. Endo, Nature
Struct. Biol. 1998, 5, 347Ϫ351.
W. Huang, J. Jia, J. Cummings, M. Nelson, G. Schneider, Y.
[2]
[3]
Lindqvist, Structure 1997, 5, 691Ϫ699.
[4]
H. Sigel, R. B. Martin, Chem. Rev. 1982, 82, 385Ϫ426.
[5]
D. S. Marlin, M. M. Olmstead, P. K. Mascharak, Inorg. Chem.
1999, 38, 3258Ϫ3260.
D. S. Marlin, M. M. Olmstead, P. K. Mascharak, Inorg. Chim.
[6]
Acta 2000, 297, 106Ϫ114.
[7]
D. S. Marlin, P. K. Mascharak, Chem. Soc. Rev. 2000, 29,
69Ϫ74.
[8]
D. S. Marlin, M. M. Olmstead, P. K. Mascharak, Eur. J. Inorg.
Chem. 2002, 38, 859Ϫ865.
J. M. Rowland, M. M. Olmstead, P. K. Mascharak, Inorg.
[9]
Chem. 2001, 40, 2810Ϫ2817.
(Et4N)[Fe(PyPOMe)2] (2): PyPOMeH2 (0.310 g, 0.82 mmol) was
dissolved in DMF (10 mL) and to this was added solid NaH
(0.046 g, 1.90 mmol). The mixture was stirred for 20 min and
(Et4N)[FeCl4] (0.140 g, 0.43 mmol) in DMF (2 mL) was then ad-
ded. The resultant bright red solution was stirred at 25 °C for 2 h.
The solvent was then removed in vacuo and the residue dissolved
in CH3CN (15 mL) and filtered to remove NaCl. To the CH3CN
filtrate was added Et2O (5 mL) and THF (5 mL) and the solution
was stored at Ϫ20 °C for three days. The bright red crystalline
solid thus obtained was collected by filtration (0.290 g, 73% yield).
Selected IR absorption bands: (KBr pellet): ν˜ ϭ 3072 cmϪ1 (w),
2989 (w), 2943 (w), 1603 (vs, C ϭ O), 1578 (s), 1491 (s), 1458 (m),
1374 (m), 1249 (m), 1156 (w), 1112 (m), 1026 (m), 760 (m), 684
(w). Absorption spectrum in DMF: λmax (ε, Ϫ1·cmϪ1) ϭ 440 nm
(6600), 305 (18 000).
[10]
J. C. Noveron, M. M. Olmstead, P. K. Mascharak, J. Am.
Chem. Soc. 2001, 123, 3247Ϫ3259.
M. Ray, D. Ghosh, Z. Shirin, R. N. Mukherjee, Inorg. Chem.
1997, 36, 3568Ϫ3572.
C.-M. Che, W.-H. Leung, C.-K. Li, H.-Y. Cheng, S.-M. Peng,
Inorg. Chim. Acta 1992, 196, 43Ϫ48.
M. Ray, R. N. Mukherjee, J. F. Richardson, R. M. Buchannan,
J. Chem. Soc., Dalton Trans. 1993, 36, 2451Ϫ2457.
M. J. Bartos, C. Kidwell, K. E. Kauffmann, S. W. Gordon-
Wylie, T. J. Collins, G. C. Clark, E. Munck, S. T. Weintraub,
Angew. Chem. Intl. Ed. Engl. 1995, 34, 1216Ϫ1219.
R. W. Saalfrank, S. Trummer, H. Krautscheid, V. Schunemann,
A. X. Trautwein, S. Hien, C. Stadler, J. Daub, Angew. Chem.
Intl. Ed. Engl. 1996, 35, 2206Ϫ2208.
F. A. Chavez, J. M. Rowland, M. M. Olmstead, P. K. Masch-
[11]
[12]
[13]
[14]
[15]
[16]
arak, J. Am. Chem. Soc. 1998, 120, 9015Ϫ9027.
480
Eur. J. Inorg. Chem. 2003, 475Ϫ481

Effect of Methyl Thioether and Methyl Ether Appendages at the Ligand Periphery
FULL PAPER
[17]
[18]
[19]
[20]
[21]
[22]
L. A. Tyler, M. M. Olmstead, P. K. Mascharak, Inorg. Chim.
Acta 2001, 321, 135Ϫ141.
D. P. Farlie, Y. Illan, H. Taube, Inorg. Chem. 1997, 36,
1029Ϫ1037.
S. Karmakar, S. B. Choudhury, A. Chakravorty, Inorg. Chem.
1994, 33, 6148Ϫ6153.
P. Chakraborty, S. K. Chandra, A. Chakravorty, Inorg. Chem.
1994, 33, 6429Ϫ6431.
C. A. Grapperhaus, A. K. Patra, M. S. Mashuta, Inorg. Chem.
2002, 41, 1039Ϫ1041.
N. S. Gill, F. B. Taylor, Inorg. Synth. 1967, 9, 136Ϫ142.
Received June 19, 2002
[I02324]
Eur. J. Inorg. Chem. 2003, 475Ϫ481
481