Lanthanide nitrate complexes of the redox active ligand (η5C5H4P(O)Ph2)2Fe

Article abstract of DOI:10.1016/j.poly.2009.02.028

The synthesis of complexes of (η⁵C₅H₄P(O)Ph₂)₂Fe = L with lanthanide nitrates is described. The single crystal X-ray structures for La(NO₃)₃L(μ-L)La(NO₃)₃L

Full text of DOI:10.1016/j.poly.2009.02.028

Polyhedron 28 (2009) 1497–1503
Contents lists available at ScienceDirect
Polyhedron
journal homepage: www.elsevier.com/locate/poly
5
Lanthanide nitrate complexes of the redox active ligand (
g
C H P(O)Ph ) Fe
5
4
2 2
a
b
a
c,*
Eileen Buckley-Dhoot , John Fawcett , Roman A. Kresinski , Andrew W.G. Platt
a
School of Chemical and Pharmaceutical Sciences, Kingston University, Penhryn Road, Kingston-upon-Thames KT1 2EE, UK
Department of Chemistry, The University, Leicester LE1 7RH, UK
Faculty of Sciences, Staffordshire University, College Road, Stoke-on-Trent ST4 2DE, UK
b
c
a r t i c l e i n f o
a b s t r a c t
5
Article history:
The synthesis of complexes of (
g
C H P(O)Ph ) Fe = L with lanthanide nitrates is described. The single
5
4
2 2
Received 22 October 2008
Accepted 24 February 2009
Available online 6 March 2009
crystal X-ray structures for La(NO
Ho(NO ] (3) and [Lu(NO ] NO
by the lanthanide contraction alone. The cyclic-voltammetric (CV) oxidation–reduction behaviour of 1, 2,
and Dy(NO O is described. This was reversible on a timescale of a few seconds in all cases. In
Á 2H
our hands the CV behaviour of L also seemed reversible on this timescale, although attempted chemical
)
3 3
L(l
-L)La(NO
3 3 3 2 2 2 3 5 3 2 2 2
) L (1), [Eu(NO ) L ] [Eu(NO ) ] (2), [Ho(NO ) L ] [-
3
)
5
3
)
2
L
2
3
(4) are reported. Trends in Ln–O bond distances cannot be explained
4
3
)
3
L
2
2
Keywords:
Lanthanide nitrate
Ferrocene phosphine oxide
Structure
oxidation of
L
led to the isolation of [FeL
2
3
(NO )
2
]NO
3
(5) which was characterised by X-ray
crystallography.
Ó 2009 Elsevier Ltd. All rights reserved.
Electrochemistry
1
. Introduction
The complexes, which are all air and moisture stable, are sparingly
soluble in ethanol but readily dissolve in CH Cl and CHCl . Solu-
2
2
3
Several complexes of (
g
5C
5
H
4
P(O)Ph
2
)
2
Fe = L with transition
tions are stable over prolonged periods unlike those of the ligand
itself which, although stable over periods of months in the dark,
deposit insoluble brown solids on standing in sunlight for a few
hours. Crystals of the complexes suitable for X-ray crystallography
were obtained by slow diffusion of toluene into chloroform solu-
tions of the complexes. Crystals were transferred to and handled
under a dried mineral oil to minimise the deterioration of the crys-
tals prior to data collection. The rapid solvent loss on exposure to
normal laboratory conditions is probably responsible for elemental
metals and their redox properties have been reported. These in-
clude complexes wherein L acts as a chelating ligand such as
+
2+
LPdCl
2
[1], LCoI
Fe] [4] and an example in which L bridges between metal centres
in the polymeric [CoI2 L] [2]. In its oxidised form a complex with
], has been reported [5]. There are no reports of
2
[2], [CuL
2
EtOH] , [CuL
2
]
5 4 2 2
[3] and CuL[(C H PPh )
+
l
n
+
2 8
H , [LH][Sb Cl
complexes between lanthanide metals and the L ligand. The elec-
trochemistry of substituted ferrocenes has been extensively stud-
ied [6–8], but there appear to be fewer studies on phosphine
oxide substituted ferrocenes. Phosphine substituted ferrocene
derivatives undergo ferrocene-based reversible oxidation followed
by formation of the phosphine oxide [9–11]. Those such as L can be
analyses which are consistent with the presence of fewer CHCl
3
molecules than found in the crystal structures. Attempted chemi-
+
cal oxidation of the ligand was investigated by reaction with Ag
+
À
3
. Stirring L with silver nitrate
with the aim of producing [L] NO
at room temperature in ethanol gave what appeared to be (by
infrared spectroscopic evidence) an AgNO complex of L with no
redox reaction. Overnight reflux with AgNO led to the formation
of a dark brown solution from which was isolated the iron(III) com-
plex [FeL
(NO ]NO
5 4 2
reversibly oxidised [10], whilst others such as (CpFeC H ) P(O)Ph
are irreversibly oxidised [12]. Similarly studies of the redox
behaviour of phosphino and PO substituted ferrocene-based
coordination complexes are sparsely reported. Complexes of
3
3
(
(
C
C
5
H
H
4
PPh
2
)
2
Fe with Re are irreversibly oxidised [13] and
undergoes oxidation at a potential very
2
5
4
P(O)(OEt)
2
)
2
Fe Á ZnCl
2
3
)
2
3
, characterised by single crystal X-ray crystallography.
similar to the free ligand [14]. There being no reports of complexes
between lanthanide metals and the L ligand, we were interested in
their synthesis and structural and electrochemical properties.
3
. Infrared spectroscopy
The infrared spectra show the expected features with two in-
2
. Synthesis and properties
tense bands for the bidentate nitrate ligands and a shift to lower
wavenumber of the PO stretching on complex formation. For the
Lu complex the presence of the ionic nitrate is indicated by three
medium intensity absorptions between those assigned to the che-
The complexes are readily prepared by direct reaction of the li-
gand with the appropriate lanthanide nitrate in ethanol solutions.
lated nitrate ligands. The absence of the expected band at
À1
ꢀ
1390 cm [15] is not without precedent; we have found that
*
Corresponding author.
E-mail address: A.Platt@staffs.ac.uk (A.W.G. Platt).
2 2 2
in lanthanide nitrate complexes of Ph P(O)CH P(O)Ph [16] the
0
277-5387/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.poly.2009.02.028

Table 1
Crystal data and structure refinement for the complexes 1–5.
Complex
1
2
3
4
5
Empirical formula
C108 H90 Cl18 Fe3
La2 N6 O24 P6
3125.16
150(2)
0.71073
C148 H124 Cl36 Eu3
Fe4 N9 O35 P8
4791.8
150(2)
0.71073
C150 H126 Cl42 Fe4
Ho3 N9 O35 P8
5069.45
150(2)
0.71073
C68 H56 Fe2Lu N2
O10 P4
1471.7
150(2)
0.71073
C68 H56 Fe3 N2 O10
P4
1352.58
150(2)
0.71073
Formula weight
Temperature (K)
Wavelength (Å)
Crystal system
Space group
triclinic
monoclinic
C2/c
monoclinic
C2/c
triclinic
triclinic
ꢀ
ꢀ
ꢀ
P1
P1
P1
Unit cell dimensions
a = 12.982(7) Å
b = 16.820(9) Å
c = 19.634(11) Å
3721(4)
a
= 70.047(9)° a = 31.509(3) Å
a
= 90°
a = 31.356(3) Å
a
= 90°
a = 12.682(2) Å
a
= 64.028(2)° a = 12.888(3) Å
a
= 76.119(4)°
b = 75.644(4)°
= 80.810(4)°
b = 73.561(9)° b = 28.599(3) Å
c
b = 110.530(3)° b = 28.608(3) Å
c
b = 110.697(3)° b = 18.445(3) Å
c
b = 73.205(3)° b = 14.574(3) Å
c
= 70.107(7)° c = 23.677(2) Å
= 90°
c = 23.647(2) Å
19843(3)
4
1.697
2.164
= 90°
c = 19.542(3) Å
3900.0(12)
2
1.253
1.754
= 76.209(3)° c = 18.439(4) Å
c
Volume (Å3)
19981(3)
4
1.593
1.821
3238.8(12)
2
1.387
0.821
Z
D
1
calc (Mg/m3)
1.394
1.293
Absorption
coefficient
À1
(
mm
)
F(000)
1560
9520
10032
1482
1392
Crystal size (mm)
h range for data
collection (°)
0.28 Â 0.19 Â 0.15
0.23 Â 0.12 Â 0.10
0.35 Â 0.14 Â 0.11
0.25 Â 0.21 Â 0.10
0.29 Â 0.22 Â 0.20
1.12–23.33
0.99–27.00
0.99–27.00
1.33–23.29
1.64–23.29
Index ranges
À14 6 h 6 14,
À18 6 k 6 18,
À21 6 l 6 21
22911
À40 6 h 6 38,
À36 6 k 6 36,
À30 6 l 6 30
83615
À39 6 h 6 39,
À36 6 k 6 36,
À29 6 l 6 30
82304
À14 6 h 6 14,
À20 6 k 6 20,
À21 6 l 6 21
24421
À14 6 h 6 14,
À16 6 k 6 16,
À20 6 l 6 20
20229
Reflections collected
Independent
10688 [0.0535]
21769 [0.0996]
21618 [0.1364]
11201 [0.0228]
9296 [0.0225]
reflections [Rint
Completeness to h
%)
]
98.9
99.8
99.8
99.90
99.4
(
Absorption
correction
Max. and min.
transmission
Refinement method
empirical
empirical
0.862 and 0.724
empirical
0.862 and 0.652
empirical
empirical
0.93 and 0.81
0.86 and 0.66
full-matrix least-
squares on F
10688/0/754
0.745 and 0.652
full-matrix least-
squares on F2
21769/0/1096
full-matrix least-
squares on F2
21618/0/1096
full-matrix least-
squares on F2
11201/0/784
full-matrix least-
squares on F2
9296/0/784
2
Data/restraints/
parameters
Goodness-of-fit on F2 1.043
0.861
0.890
0.993
1.059
Final R indices
I > 2 (I)]
R indices (all data)
R
1
= 0.0672,
wR = 0.1685
= 0.0836,
wR = 0.1760
R
1
= 0.0572,
wR = 0.1160
= 0.1090,
wR = 0.1283
R
1
= 0.0613,
wR = 0.1181
= 0.1265,
wR = 0.1377
R
1
= 0.0248,
wR = 0.0590
= 0.0280,
wR = 0.0559
R
1
= 0.0345,
wR = 0.0955
= 0.0412,
wR = 0.0987
[
r
2
2
2
2
2
R
1
R
1
R
1
R
1
R
1
2
2
2
2
2
Largest difference in
1.883 and À1.513
1.683 and À0.842
1.457 and À1.126
0.940 and À0.646
0.366 and À0.267
peak and hole
À3
(
e Å
)
Solvent accessible
void (Å )
Estimated solvent
electron count
858.2
222e
1805.4
476e
1795
516e
1129.9
290e
460
3
125e

E. Buckley-Dhoot et al. / Polyhedron 28 (2009) 1497–1503
265
1499
interaction of ‘‘ionic” nitrate with H-bonding molecules can lead to
2
R
= 0.9997
infrared spectra which are essentially the same as for a coordinated
nitrate.
2
2
2
2
2
2
2
60
55
50
45
40
35
30
Ln-O(N)
4
. Structures
2
4.1. General points
R = 0.9951
Details of the data collection and refinement for the crystal
Ln-O(P)
structures are given in Table 1 and selected bond lengths in Table
2.
225
The Ln–O distances show the effect of the lanthanide contrac-
2
2
20
15
tion showing a linear decrease in both Ln–O(N) and Ln–O(P) with
decreasing ionic radius of the lanthanide as shown in Fig. 1. The
Ln–O(N) distances are consistently longer than the Ln–O(P) which
is a general feature of PO complexes with lanthanide nitrates. The
110
115
120
125
130
135
Ionic radius / pm
Fig. 1. The relationship between the Ln–O(P) and Ln–O(N) distances and the ionic
radius of the lanthanide ion for complexes 1–4.
magnitude of the difference in Ln–O(N) À Ln–O(P) (=
D) appears to
reflect the degree of strain in the bonding region. Due to the
strongly ionic nature of the interaction between the PO oxygen
and the lanthanide ion a linear Ln3 –O –P arrangement is fa-
+
À
+
paired t-test assuming a null hypothesis of no difference between
the means and unequal variance at a 95% confidence level. The re-
sults of such an analysis show that there are significant differences
between the La–O(N) and other Ln–O distances, with La–O(N)
being shorted than expected. The differences between the
Eu–O(N), Ho–O(N) and Lu–O(N) are not significant once the effect
of the lanthanide contraction is taken into account. The average
Ln–O distances corrected for the lanthanide ion size are signifi-
cantly larger for the Eu and Ho complexes. In the Eu and Ho com-
plexes the larger residual differences with averages of 1.314(2) Å,
compared with 1.256(2) Å for the La and Lu complexes, can be
ascribed to the hydrogen bonding between chloroform and the
coordinated nitrate ions. In the La complex this interaction is con-
voured on electrostatic grounds. Values of
ature structures show that for non-chelated PO such as in
Ln(NO (Ph PO) [17] and Y(NO (Ph PO) [18] values of in
the region of 0.21 Å are found. Constraining the PO in a six mem-
D calculated from liter-
3
)
3
3
n
3
)
3
3
n
D
i
bered chelate ring in complexes such as Ln(NO
3
)
3
[( PrO)
[20] and
[21] lengthens the
2
P(O)
C
5
H
4
N(O)]
Ln(NO [(MeO)
Ln–O(P) distance and hence reduces
esting exceptions to this are complexes of Ph
where a smaller of 0.06 Å is observed [16]). Increasing the size
2
[19], Ln(NO
3
)
3
[(EtO)
2
P(O)CH
2
P(O)(OEt)
2
]
2
3
)
3
2
P(O)CMe(OH)P(O)(OMe)
2
]
D
to about 0.13–0.14 Å (inter-
P(O)CH P(O)Ph
2
2
2
D
of the chelate ring might be expected to increase the flexibility of
the structure and allow Ln–O–P angles to approach linearity and
siderably weaker with a shortest (N)OÁÁÁHCCl
3
of 2.547 Å compared
thus values of
Complexes with 10-membered chelate rings 1,3-[Ph
22] do indeed show this effect with of 0.21 Å. In this study, a va-
lue of = 0.22 Å is found, again implying relatively unconstrained
D
approaching those of an ‘‘unconstrained” ligand.
with 2.246 and 2.232 Å, respectively, for the Eu and Ho complexes
and this may in part account for the stronger La–O interaction.
There are short contacts between one of the H atoms on the cyclo-
pentadiene ring and one of the O(N) atoms in all the complexes.
These distances, in the region of 2.3 Å, are shorter than the sum
of van der Waals radii of O and H (2.60 Å), but, as such interactions
occur in all complexes, they cannot account for the variations in
distances observed.
2 2 6 4
P(O)] C H
[
D
D
coordination of the PO groups. This is further seen in the values of
the Ln–O–P angles which average 161.8° (range 152.3–177.3°) over
all the lanthanide complexes reported here.
It has been previously observed that small bite angle ligands
such as nitrate should preferentially bind to larger metal ions
A similar analysis for the Ln–O(P) distances for the chelate rings
reveals significant differences between the La–O(P) distances and
those for the other complexes. The values of Ln–O(P) corrected
for the size of the lanthanide ion, again indicate a stronger La–O
interaction with residual distances of 1.048 Å for La and 1.085,
[
23]. Although it should be noted that all the metals here fall in
to the category of large ions it is still interesting in this respect
to examine the changes in Ln–O bond distances. When the Ln–
O(N) distances are corrected for the lanthanide contraction, by
subtracting the ionic radius of the lanthanide ion with the appro-
priate coordination number [24], the difference should be constant
if no other effects are significant. Single factor Anova analysis of the
data reveals that significant differences occur between the sets of
distances. Post hoc analysis of the results was carried out by un-
1
.094 and 1.087 Å for the Eu, Ho and Lu complexes respectively.
It is difficult to see a reason for this. The non-bonded (P)OÁÁÁO(P)
distances within the chelate rings are larger for the Eu and Ho com-
plexes (3.262 (Eu) and 3.212 Å (Ho) compared with 3.160 and
Table 2
Selected bond lengths for the lanthanide complexes.
La
Eu
Ho
Lu
Fe
La(1)–O(2)
La(1)–O(1)
La(1)–O(3)
La(1)–O(11)
La(1)–O(8)
La(1)–O(5)
La(1)–O(7)
La(1)–O(4)
La(1)–O(10)
2.403(5)
2.406(5)
2.425(5)
2.591(5)
2.592(6)
2.593(5)
2.632(6)
2.632(5)
2.654(5)
Ln(1)–O(2)
Ln(1)–O(1)
Ln(1)–O(4)
Ln(1)–O(3)
Ln(1)–O(9)
Ln(1)–O(5)
Ln(1)–O(6)
Ln(1)–O(8)
2.276(4)
2.283(4)
2.303(4)
2.302(4)
2.490(4)
2.523(4)
2.523(4)
2.542(4)
2.232(4)
2.240(4)
2.258(4)
2.266(4)
2.428(5)
2.467(5)
2.468(4)
2.497(5)
2.2043(18)
2.2152(19)
2.2008(19)
2.1962(19)
2.4158(19)
2.4183(19)
2.434(2)
1.9753(19)
1.9713(19)
1.9749(18)
1.9814(19)
2.0458(19)
2.0120(19)
2.423(2)
P(1)–O(1)
P(2)–O(2)
P(3)–O(3)
P(4)–O(4)
1.495(5)
1.496(5)
1.503(5)
1.504(4)
1.501(4)
1.492(4)
1.499(4)
1.492(5)
1.501(5)
1.488(4)
1.493(5)
1.496(2)
1.4935(19)
1.502(2)
1.496(2)
1.517(2)
1.502(2)
1.500(2)
1.4987(19)

1
500
E. Buckley-Dhoot et al. / Polyhedron 28 (2009) 1497–1503
conform to any simple polyhedron, but can be considered as a
somewhat distorted octahedron if the nitrates are visualised as
monodentate ligands bonded via the N-atoms. The octahedron
thus conceptualised is a mer-isomer with ‘‘cis” angles at La averag-
ing 92.4 ± 9.3° whilst the corresponding ‘‘trans” angles average
1
62.1 ± 9.4°. The La–O(P) distances for the chelating phosphine
oxide are essentially identical (2.404(5) Å) and shorter than that
from the bridging PO group (2.425(5) Å). These differences are
not reflected in the PO distances themselves which are apparently
equal at 1.4978(5) Å regardless of whether the PO groups are in the
chelating or bridging mode. The nitrate groups bond asymmetri-
cally to the lanthanum ion with one short and one long La–O dis-
tance, with averages of 2.592(8) and 2.639(10) Å, respectively.
These differences are reflected in the bond distances within each
nitrate ion itself, with a short O–N distance (1.254(6) Å)corre-
sponding to the short La–O and the longer N–O associated with
the longer La–O (1.276(9) Å). The P–O–La angles approach the
idealised linear geometry for all the phosphoryl groups, as ex-
pected for predominantly ionic bonding where the structure im-
poses relatively few constraints on ligand architecture.
Fig. 2. The structure of [LaL(NO
symmetry. Symmetry related atoms generated by 1 À x,1 À y,Àz.
3 3 2
) ] l-L. The Fe2 atom is located on a centre of
3
.175 Å for the La and Lu complexes, respectively) implying smal-
ler electrostatic repulsions, which in turn suggests that stronger
binding to the metal might be possible. The other obvious candi-
date for repulsive interactions which might lead to a weakening
of the Ln–O(P) bonds are the non-bonded interactions between
the (P)O and (N)O atoms. Significant differences are also indicated
by single factor Anova analysis for the non-bonded P(O)ÁÁÁO(N) dis-
tances. A post hoc analysis of the non-bonded P(O)ÁÁÁO(N) distances
by t-tests between the pairs of complexes shows a trend of increas-
ing significance in the differences of the means. The differences be-
tween the La and Eu, complex is only significant at an 88.3%
confidence level, whilst this increases to 98.9% and 99.9% for the
differences between the La complex and the Ho and Lu complexes,
respectively. For the La complex the shortest of these non-bonded
interactions is over 3 Å whilst for all the others it is lower, showing
a trend in values of 2.885 (Eu), 2.814 (Ho) and 2.768 Å (Lu). This
trend mirrors the confidence levels at which the differences in
the P(O)ÁÁÁO(N) means between the La complex and the respective
sets of distances for the other complexes become significant.
The structures of the complexes have a number of features in
common. The twist angles of the cyclopentadiene rings, measured
from the torsional angles between the vectors from the ring cen-
troid and ring carbon atoms on corresponding positions of the
two rings, vary depending on whether the ligand is chelating or
bridging. In the bridging mode observed in the lanthanum com-
plex, the ligand is able to adopt a similar structure to that of the
free ligand itself, namely with staggered rings and the diphenyl-
phosphine oxide groups in a trans arrangement. Here the twist an-
gle between the rings averages at 38.0 ± 0.7°, which is in good
agreement with the average values of 37.7 ± 0.5° in the free ligand
4.3. The structures of [Eu(NO
3
)
2
L
2
]
2
[Eu(NO
3
)
5
] (2),
][NO
[Ho(NO [Ho(NO ] (3) and [Lu(NO
)
3 2
L ]
2 2
)
3 5
3
)
2
L
2
3
] (4)
In general, the reduction of the ionic radius tends to lower the
coordination number of the metal. The cations in these complexes
have eight-coordinate lanthanide ions and are rather similar in
motif; the structure of the cation in 2 is shown as a representative
example in Fig. 3. The coordination geometries can be visualised as
distorted octahedra if the nitrate ions are considered as pseudomo-
nodentate ligands attached via the N-atom. The trans-N–Ln–N
angles are 176.21(15)° (Eu), 176.63(17)° (Ho) and 177.53(7)° (Lu)
whilst the average corresponding trans-O–Ln–O angles are
163.6 ± 0.3°, 162.9 ± 0.4° and 164.9 ± 4.1°. The cis-N–Ln–O angles
give averages close to 90°, at 90.1 ± 8.5° (Eu), 90.0 ± 8.7° (Ho) and
90.0 ± 10.1° (Lu) whilst the cis-O–Ln–O angles are rather more uni-
form with averages of 91.2 ± 1.0°, 91.3 ± 0.8° and 91.7 ± 0.8° for Eu,
2À
Ho and Lu, respectively. The Ln(NO
lanthanide ions but are readily envisaged as trigonal bipyramids,
3
)
5
ions contain 10-coordinate
[
2
25], 36.1 ± 1.5° found in the polymeric form of CoI
6.6 ± 0.9° found for the bridging ligands in the zinc chloride com-
plex of [C P(O)(OEt) Fe [14]. The formation of chelate rings im-
2
L [2], and
5
H
4
2 2
]
poses a more eclipsed conformation on the ferrocenyl nucleus. The
twist angles for the chelating ligands in the complexes are similar,
with those in the lanthanum complex averaging at 8.6 ± 0.4° with
averages of 12.8 ± 0.6° and 5.9 ± 0.5° (Eu), 12.1 ± 0.8° and 5.6 ± 0.3°
(
Ho) and 6.8 ± 0.3° and 4.5 ± 0.3° (Lu). These values are similar to
that found in PdCl L (7.3 ± 1.4°) but larger than observed in mono-
meric form of CoI L which has an average twist angle of 1.8 ± 0.6°.
2
2
3 3 3 3
4.2. The structure of LLa(NO ) -lL–La(NO ) L (1)
The lanthanum nitrate complex (1) crystallised as a dimer,
LLa(NO L–La(NO L in which each La is nine coordinate. The
structure is shown in Fig. 2. The coordination geometry does not
3
3
) -
l
3 3
)
+
2À
Fig. 3. The cation in [EuL
2
3
(NO )
2
]
2
[Eu(NO
3
)
5
]
.

E. Buckley-Dhoot et al. / Polyhedron 28 (2009) 1497–1503
1501
Table 3
a
Peak potentials for selected complexes.
1
2
Dy(NO )
3 3
L
2
Á 2H
2
O
4
E
D
D
½
(complex) (mV)b
E (ferrocene) (mV)
E (complex) (mV)
628
70
57
661
77
66
681
70
75
683
78
83
a
Relative to ferrocene as internal reference.
All peak potentials are measured with respect to ferrocene using a pseudore-
b
ference Ag wire.
3 3
plex. Similarly the intensity of the peak due to [LnL (NO
)]2+
decreases from 40% (La) to below 5% (Yb) reflecting the increased
steric interactions as the ionic radius decreases.
The negative ion spectra of the complexes showed a base peak
À
at m/z = 62.1 assigned as NO
3
and no other peaks of significant
intensity. A much lower intensity signal (<1%) at m/z = 648 which
À
is assigned as [L+NO
3
] . There were no observed peaks which could
À
2À
be assigned to species such as Ln(NO
3
)
4
or Ln(NO
3
)
5
. Such spe-
cies might have been anticipated given the solid state structures of
the Ho and Eu complexes.
6
. Cyclic voltammetry
Fig. 4. The structure of [FeL
2
(NO
3 2
) ]NO
3
.
The oxidation behaviour of L, 1, 2, Dy(NO
)
L
Á 2H
O and 4 was
solution in aceto-
3
3
2
2
studied by cyclic voltammetry in 0.1 M Bu
4
NClO
4
nitrile, with ferrocene as internal standard. Ligand L and each com-
plex display a single reversible oxidation event corresponding to
each {L} unit; in no case is any other wave evident, such as might
suggest inequivalent L environments, or any degree of electronic
communication between equivalent environments. Nevertheless,
in the case of 1, we clearly have a case of a complex which pos-
sesses inequivalent L environments in the solid state. We can offer
no explanation for the single redox wave, other than a solution
fluxionality or lability of the complex over the approximate
again taking the nitrate ions as pseudomonodentate ligands. The
equatorial nitrogen atoms and the lanthanide ion are coplanar in
both structures, the Nax–Ln–Nax angles are 177.9°, the Neq–Ln–
N
N
eq average to 120.0° (range 118.77–122.45°) whilst the Neq–Ln–
ax angles average at 91.7 ± 5.3° and 90.0 ± 6.2° for the Eu and
Ho complexes, respectively. There are many examples of
2À
Ln(NO
.4. The structure of [FeL
The structure of [FeL
3
)
5
counter-ions in the literature [26].
(NO ]NO
3
(NO ]NO (5) is shown in Fig. 4 and
7
00 ms duration of the oxidation and re-reduction waves. All com-
4
2
3
)
2
3
plexes, as written above, display a continuous shift toward more
positive oxidation potentials with atomic number (Table 3). Ligand
L itself is shifted some +458 mV with respect to ferrocene in good
agreement with the literature value of +430 mV afforded by a pre-
vious determination in a different solvent system [10]. The Ln com-
plexes follow a clear pattern which offers evidence for a closely
associated molecular complex between Ln and L, albeit fluxional,
being present in solution in each case. The trend is consistent with
increasing polarising power of the smaller lanthanide metals draw-
ing electron density onto the O atoms of the L ligand and away
from the Fe centre. We attempted to reduce the Eu complex 2 in
the same electrolyte solution after deoxygenation by nitrogen
sparging, but no reduction peaks were observed in the potential
window of the solvent/electrolyte system used.
2
3
)
2
shows interesting differences with the Lu complex with which it
is superficially analogous. The iron is six-coordinate with the ni-
trate ions bound in a monodentate manner. The geometry about
the metal is essentially octahedral with the trans-O–Fe–O angles
averaging at 177.1 ± 1.1° and the cis -O–Fe–O at 90.0 ± 3.3°. The
cyclopentadiene rings are eclipsed with twist angles of 2.6 ± 0.5°
and 0.3 ± 0.1° for the two ligands. The P–O–Fe angles have a
slightly larger range than the corresponding angles in the lantha-
nide complexes with one ligand having angles approaching linear-
ity, whilst the other is significantly bent. This seems to have little
effect on the Fe–O or P–O distances which are in essence indepen-
dent of the angle at the oxygen atoms.
7
. Experimental
5
. Mass spectrometry
7.1. Crystallography
Electrospray mass spectra were recorded for all the complexes
+
in CH
nitrate to give [LnL
2
Cl
2
. The base peak in all cases is [L+H] . The loss of a single
Crystals were transferred to and handled under a dried mineral
+
2
3
(NO )
2
] is observed for all the complexes gen-
oil to minimise the deterioration of the crystals prior to data collec-
tion. All X-ray data for the complexes were measured on a Bruker
SMART diffractometer with a 2 K CCD area detector using graphite
erally as a low intensity signal. The most notable feature of the
mass spectra is the presence of ions formed as the result of ligand
redistribution reactions. Thus signals of significant intensity are
monochromated Mo K
a radiation (k = 0.71073 Å). Absorption cor-
3+
3+
observed for ions such as [LnL
4
]
(5–10%), [LnL
3
]
(70–10%),
rections, based on comparison of Laue equivalents, were applied
to the data sets. The structures were solved by Patterson methods
and refined by full-matrix least-squares cycles on F2 for all data,
using SHELXTL [27]. All non-hydrogen atoms of the metal complexes
were refined with anisotropic displacement parameters and all
2
+
2+
[
LnL (NO )] (40–5%) and [LnL(NO )] (10–30%). Some trends in
the relative intensities of the signals from the ions are explicable
by the lanthanide contraction. Thus the intensity of the peak as-
3 3 3
3
+
signed as [LnL
3
]
decreases from 70% for La to 10% for the Yb com-

1
502
E. Buckley-Dhoot et al. / Polyhedron 28 (2009) 1497–1503
hydrogen atoms were included in refinement cycles riding on
bonded atoms. For all structures the metal complex was unambig-
3
grown by slow diffusion of toluene into a CHCl solution rapidly
lost solvent on exposure to air.
À1
uous and refined satisfactorily. For 1, 2 and 3 lattice CHCl
located and refined. However, all the structures were found to have
regions of unresolved disordered solvent, thought to be CHCl , that
3
was
Infrared/cm (ATR)
m
NO 1473(s) 1282(s),
[Eu(NO ]2.5CHCl : C, 45.48; H, 3.16; N, 3.45.
Found: C, 45.11; H, 3.66; N, 3.81%.
[Ho(NO [Ho(NO ]2CHCl
mPO 1142(s). Anal. Calc.
for [Eu(NO
3
)
2
L ]
2 2
3
)
5
3
3
could not be modelled satisfactorily. The SQUEEZE option of the
programme PLATON [28] was used to omit the data for the disor-
dered regions for each of the structures. The solvent accessible void
volume and estimated solvent electron count for each structure is
given in Table 1.
)
3 3
L
2
]
2
3
)
5
3 3 3 2
: Ho(NO ) 6H O (0.32 g, 0.70
mmol) in 2 ml EtOH was added in portions to a solution of L
(0.30 g, 0.50 mmol) in 3 ml hot EtOH as for the La complex above
gave 0.24 g yellow-brown powder. Crystals for X-ray analysis
3
grown by slow diffusion of toluene into a CHCl solution rapidly
lost solvent on exposure to air.
À1
7
.2. Infrared spectroscopy
Infrared/cm (ATR)
m
NO 1473(s) 1287(s),
[Ho(NO ]2CHCl : C, 45.57; H, 3.16; N, 3.47.
Found: C, 45.37; H, 3.24; N, 3.63%.
Dy(NO O: A solution of Dy(NO
Á 2H
mPO 1146(s). Anal. Calc.
for [Ho(NO
3
)
2
L
2
]
2
)
3 5
3
Infrared spectra were recorded on a Thermo Nicolet Avatar 370
FT-IR spectrometer operating in ATR mode. Samples were com-
pressed onto the optical window and spectra recorded without fur-
ther sample pre-treatment.
)
3 3
L
2
2
3 3 2
) 6H O (0.12 g) in
1.0 ml EtOH was added to a solution of L (0.36 g) in 2 ml EtOH. A
yellow-brown precipitate formed on each addition. On complete
addition the solid was filtered, washed with EtOH and dried at
the pump (0.25 g).
7
.3. Electrospray mass spectrometry
Infrared (ATR)
52.45; H, 3.88; N, 2.70. Found: C, 52.08; H, 3.79; N, 2.54%.
Lu(NO O: A solution of Lu(NO 6H O (0.10 g) in 1.5 ml
EtOH was added to a solution of L (0.27 g) in 10 ml EtOH. On reduc-
tion of the volume of the solution to about 5 ml brown crystals
were deposited which were filtered washed with EtOH and dried
mNO 1477(s) 1283(s), mPO 1142(s). Anal. Calc.: C,
Electrospray mass spectra were obtained by the EPSRC National
Mass Spectrometry Service Centre at Swansea University as de-
scribed previously [29]. The spectra were recorded on a VG Quattro
II triple quadrupole mass spectrometer. Samples dissolved in
)
3 3
L
2
Á 2H
2
3
)
3
2
2 2
CH Cl were loop injected into a stream of MeOH passing through
a steel capillary held at high voltage (+3.5 kV for positive mode and
À3.0 kV for negative mode). Nebulisation of the resulting spray
was pneumatically assisted by a flow of nitrogen bath gas and
heated source (70 °C). Declustering and molecular fragmentation
were promoted by increasing the cone voltage from 8 to 50 V.
at the pump to give the product (0.25 g).
À1
Infrared/cm (ATR)
m
OH 3400(w)
PO 1152(s). Anal. Calc.: C, 52.03; H,
3.85; N, 2.68. Found: C, 52.30; H, 3.81; N, 2.74%.
Fe(NO NO O: A solution of L (2.01 g) and silver nitrate
Á 3.5H
1.03 g) in 40 ml ethanol was heated under reflux for 8 h. The solu-
mNO 1499(s) 1482(s) 1359(m)
1337(m) 1307(m) 1288(s),
m
)
3 2
L
2
3
2
(
7
.4. Electrochemistry
tion was cooled, filtered and allowed slowly to evaporate to 25 ml
when dark brown crystals formed. These were filtered, washed
with a small quantity of ethanol, and dried at the pump to give
The oxidation behaviour of L, 1, 2, Dy(NO
3
)
3
L
2
2
Á 2H O and 4 was
studied using a PC-driven Bioanalytical Systems BAS100B Electro-
chemical Analyzer/BAS PA1 Preamplifier combination operating in
cyclic voltammetry mode with a sweep width of 1400 mV and
dark brown crystals (0.59 g).
À1
Infrared/cm
1280(s),
mOH 3389(w) mNO 1489(m), 1488(m), 1481(m),
m
PO 1135(s), 1120(s). Anal. Calc.: C, 55.25; H, 4.30; N,
À1
sweep rate of 200 mV s . The working electrode was polished Pt
2.84. Found: C, 55.30; H, 4.50; N, 2.94%.
disc, the auxiliary electrode Pt wire and the pseudoreference elec-
trode Ag wire. Approximately 1–5 mg samples were dissolved in
approximately 20 ml of 0.1 M Bu NClO in acetonitrile solution,
4 4
Supplementary data
with ferrocene as internal standard. Commercial white-spot nitro-
gen was used as spargent for reduction experiments.
CCDC 620324–620328 contains the supplementary crystallo-
graphic data for Fe, La, Eu, Ho and Lu structures, respectively. These
data can be obtained free of charge via http://www.ccdc.cam.ac.uk/
conts/retrieving.html, or from the Cambridge Crystallographic
Data Centre, 12 Union Road, Cambridge CB2 1EZ, UK; fax: (+44)
7.5. Synthesis
(
C
5
H
4
P(O)Ph PPh
2
)
2
Fe: (C
5
H
4
2 2
) Fe (5.17 g) was suspended in
1
223-336-033; or e-mail: deposit@ccdc.cam.ac.uk.
5
0 ml acetone and H O (3.00 g 30% aqueous solution) slowly
2 2
added at such a rate as to maintain a gentle reflux. The suspension
was stirred overnight, filtered and the resulting yellow solid
washed with a little acetone and diethyl ether and dried at the
Acknowledgements
We are grateful to the Royal Society of Chemistry for support
from the research fund and the EPSRC for use of the National Mass
Spectrometry Service at Swansea University.
pump to give 5.31 g (93% as the hydrate) yellow powder. Infra-
À1
red/cm (ATR)
m
PO 1165(s)
1.5CHCl : La(NO
EtOH was added in portions to a solution of L (0.25 g 0.42 mmol) in
ml hot EtOH. A yellow-brown precipitate was formed on each
m
OH 3409(m).
[
La(NO
3 3
) ]
L
2 3
3
3
)
3
6H O (0.38 g 0.87 mmol) in 3 ml
2
References
5
addition. On cooling, the yellow-brown solid formed, and was fil-
tered, washed with a little EtOH and dried at the pump to yield
[
[
1] J.S.L. Yeo, J.J. Vittal, T.S.A. Hor, J. Chem. Soc., Chem. Commun. (1999) 1477.
2] T. Aviles, A. Dinis, J. Orlando GonÇalves, V. Felix, M.J. Calhorda, A. Prazeres,
M.G.B. Drew, H. Alves, R.T. Henriques, V. da Gama, P. Zanello, M. Fontana, J.
Chem. Soc., Dalton Trans. (2002) 4595.
0
.23 g. Crystals suitable for X-ray analysis were grown by slow dif-
fusion of toluene into a CHCl solution to give dark brown crystals.
Infrared (ATR) NO 1436(s) 1295(s), PO 1138(s). Anal. Calc.: C,
8.03; H, 3.33; N, 3.25. Found: C, 47.88; H, 3.39; N, 3.51%.
Eu(NO [Eu(NO ]2.5CHCl Eu(NO 6H (0.29 g, 0.65
mmol) in 2 ml EtOH was added in portions to a solution of L
0.32 g, 0.55 mmol) in 3 ml hot EtOH as for the La complex above
and gave 0.41 g yellow-brown powder. Crystals for X-ray analysis
3
[
3] G. Pilloni, G. Valle, C. Corvaja, B. Longato, B. Corain, Inorg. Chem. 34 (1995)
m
m
5910.
4
[4] G. Pilloni, B. Corain, M. Degano, B. Longato, G. Zanotti, J. Chem. Soc., Dalton
Trans. (1993) 1777.
[
[
3 3 2
) L ]
2
)
3 5
3
:
3
)
3
2
O
5] I.A. Razak, U. Usman, H.-K. Fun, B.M. Yamin, N.A.M. Kasim, Acta Crystallogr.,
Sect. C 58 (2002) 225.
(
[6] M. Emília, N.P.R.A. Silva, A.J.L. Pombeiro, J.J.R. Fraústo da Silva, R. Herrmann, N.
Deus, T.J. Castilho, M. Fátima, C.G. Silva, J. Organomet. Chem. 421 (1991) 75.

E. Buckley-Dhoot et al. / Polyhedron 28 (2009) 1497–1503
1503
[
7] M. Emília, N.P.R.A. Silva, A.J.L. Pombeiro, J.J.R. Fraústo da Silva, R. Herrmann, N.
Deus, R.E. Bozak, J. Organomet. Chem. 480 (1994) 81.
[19] G.S. Conary, A.A. Russell, R.T. Paine, J.H. Hall, R.R. Ryan, Inorg. Chem. 27 (1988)
3242.
[
[
8] W.L. Davis, R.F. Shago, E.H.G. Langner, J.C. Swart, Polyhedron 24 (2005) 1611.
9] J. Podlaha, P. Stepnicka, J. Ludvik, I. Cisarova, Organometallics 15 (1996) 543.
[20] A.M.J. Lees, R.A. Kresinski, A.W.G. Platt, New J. Chem. 28 (2004) 1457.
[21] A.M.J. Lees, J.M. Charnock, R.A. Kresinski, A.W.G. Platt, Inorg. Chim. Acta 312
(2001) 170.
[
[
10] G. Pilloni, B. Longato, B. Corain, J. Organomet. Chem. 420 (1991) 57.
11] P. Zanello, G. Opromolla, G. Giorgi, G. Sasso, A. Togni, J. Organomet. Chem. 506
[22] Y.-C. Tan, X.-M. Gan, J.C. Stanchfield, E.N. Duesler, R.T. Paine, Inorg. Chem. 40
(2001) 2910.
(
1996) 61.
[
[
[
12] D.A. Durfey, R.U. Kirss, C. Frommen, W. Feighery, Inorg. Chem. 39 (2000) 3506.
13] T.M. Miller, K.J. Ahmed, M.S. Wrightson, Inorg. Chem. 28 (1989) 2347.
14] O. Oms, F. Maurel, F. Carré, J. le Bideau, A. Vioux, D. Leclercq, J. Organomet.
Chem. 689 (2004) 2654.
[23] R.D. Hancock, A.E. Martell, Chem. Rev. 89 (1989) 1875.
[24] R.D. Shannon, Acta Crystallogr. A32 (1972) 751.
[25] Z.-G. Fang, T.S.A. Hor, Y.-S. Wen, L.-K. Liu, T.C.W. Mak, Polyhedron 14 (1995) 2403.
[26] F.H. Allen, Acta Crystallogr. B58 (2002) 380.
[
15] K. Nakamoto, Infrared and Raman Spectra of Inorganic and Coordination
Compounds, 5th Ed., Wiley, 1997.
16] A.M.J. Lees, A.W.G. Platt, Inorg. Chem. 42 (2003) 4673.
17] W. Levason, E.H. Newman, M. Webster, Polyhedron 19 (2000) 2697.
18] L. Deakin, W. Levason, M.C. Popham, G. Reid, M. Walker, J. Chem. Soc., Dalton
Trans. (2000) 2439.
[27] SHELXTL, An Integrated System for Solving, Refining and Displaying Crystal
Structures, Version 5.10, Bruker Analytical X-ray Systems, Madison, WI, USA,
1997.
[28] A.L. Spek, PLATON, Version 1.5, A Multi-Purpose Crystallographic Tool, Utrecht
University, Utrecht, The Netherlands, 2002.
[
[
[
[29] J. Fawcett, A.W.G. Platt, D.R. Russell, Polyhedron 21 (2002) 287.