Ansa-chromocene complexes. 2. Isocyanide derivatives of Cr(II) and Cr(III), their syntheses, X-ray crystal structures, and physical properties

Article abstract of DOI:10.1021/om050710j

A variety of 18e ansa-chromocene isocyanide complexes were prepared via ligand substitution on the carbonyl complexes. The relative π-accepting character of the different isocyanide ligands was evaluated from the structural, spectroscopic, and electrochemical properties of the complexes. One-electron chemical oxidation of the complexes generates low-spin, 17e unsa-chromocene isocyanide cations, the structural and spectroscopic properties of which were compared with their neutral precursors. Density functional theory (DFT) calculations were performed on model complexes in order to elucidate the nature of the bonding between the chromium and the isocyanide ligands and to explain the effects of one-electron oxidation of the complexes.

Full text of DOI:10.1021/om050710j

Organometallics 2006, 25, 719-732
719
ansa-Chromocene Complexes. 2. Isocyanide Derivatives of Cr(II)
and Cr(III), Their Syntheses, X-ray Crystal Structures, and
Physical Properties
Pamela J. Shapiro,*,† Ralph Zehnder,†,‡ David M. Foo,† Philippe Perrotin,†
Peter H. M. Budzelaar,§ Sharon Leitch,† and Brendan Twamley†
Departments of Chemistry, UniVersity of Idaho, Moscow, Idaho 83844-2343, and UniVersity of Manitoba,
Winnipeg, Manitoba R3T 2N2, Canada
ReceiVed August 17, 2005
A variety of 18e ansa-chromocene isocyanide complexes were prepared via ligand substitution on the
carbonyl complexes. The relative π-accepting character of the different isocyanide ligands was evaluated
from the structural, spectroscopic, and electrochemical properties of the complexes. One-electron chemical
oxidation of the complexes generates low-spin, 17e ansa-chromocene isocyanide cations, the structural
and spectroscopic properties of which were compared with their neutral precursors. Density functional
theory (DFT) calculations were performed on model complexes in order to elucidate the nature of the
bonding between the chromium and the isocyanide ligands and to explain the effects of one-electron
oxidation of the complexes.
In our efforts to explore the reactivity of bent-sandwich
chromocene complexes we have found isocyanide ligands to
be useful alternatives to CO for stabilizing 16e ansa-chromocene
fragments, which are destabilized relative to parent Cp2Cr by
the enforced bending of the cyclopentadienyl rings by the ansa-
bridge.18 Schwemlein et al. recognized the need to trap the ansa-
chromocene as it formed in situ in their synthesis of the complex
Me4C2(C5H4)2CrCO.19 In the absence of CO, an ill-defined,
THF-insoluble, brick-red material is obtained instead. We later
found that tert-butyl isocyanide could be used instead of CO to
form the corresponding 18e ansa-chromocene isocyanide de-
rivative according to the reaction shown in eq 1.20,21
Introduction
Isocyanide ligands are often interchangeable with CO in
transition metal complexes, although they are, in general, better
σ-donors and poorer π-acceptors than CO.1,2 They are also
similar to CO in their usefulness for probing the electron-
donating/accepting properties of the metal, which is manifested
in the energy of the C-N stretching vibration, the bond angle
about nitrogen, and the 13C and 15N NMR chemical shifts of
the ligands.3,4 Isocyanide ligands are more versatile ligands than
CO in the sense that the substituent on nitrogen can be varied
to influence the donor/acceptor properties of the ligand and to
manipulate the architectures of metal complexes that are
constructed with the ligand. For example, bridging diisocyanide
ligands have been used to support a variety of dinuclear metal-
metal bonded species5-10 as well as metal-containing macro-
cycles and polymers.10-17
* Corresponding author. E-mail: shapiro@uidaho.edu.
† University of Idaho.
Other isocyanide ligands are unsuccessful at trapping the
ansa-chromocene fragment, affording only insoluble, brick-red
materials. A more general approach to isocyanide derivatives
involves reversible ligand substitution on the carbonyl species.
‡ Current address: Department of Chemistry, University of Louisiana
at Monroe, Monroe, LA 71209.
§ University of Manitoba.
(1) Trechel, P. M. In AdV. Organomet. Chem; Stone, F. G. A., West, R.,
Eds.; Academic Press: New York, 1973; Vol. 11, pp 21-81.
(2) Singleton, E.; Oosthuizen, H. In AdV. Organomet. Chem; Stone, F.
G. A., West, R., Eds.; Academic Press: New York, 1983; Vol. 22, pp 209-
310.
(3) Becker, W.; Beck, W.; Rieck, R. Z. Naturforsch. 1970, 25b, 1332-
1337.
(12) Irwin, M. J.; Jia, G.; Vittal, J. J.; Puddephatt, R. J. Organometallics
1996, 15, 5321-5329.
(13) Su, Z. M.; Wang, R. S.; Che, C.-M. Mater. Res. Soc. Symp. Proc.
1999, 576, 389-394.
(4) Guy, M. P.; Coffer, J. L.; Rommel, J. S.; Bennett, D. W. Inorg. Chem.
1988, 27, 2942-2945.
(14) Hanack, M.; Deger, S.; Lange, A. Coord. Chem. ReV. 1988, 83,
115-136.
(5) Bartolomaes, T.; Lentz, D.; Neubert, I.; Roettger, M. Z. Anorg. Allg.
Chem. 2002, 628, 863-871.
(15) Feinstein-Jaffe, I.; Maisuls, S. E. J. Organomet. Chem. 1988, 350,
57-75.
(6) Mixa, M. M.; Sykes, A.; Mann, K. R. Inorg. Chim. Acta 1989, 160,
159-65.
(16) Schroelkamp, S.; Sperber, W.; Lentz, D.; Fehlhammer, W. P. Chem.
Ber. 1994, 127, 621-629.
(7) Ohtani, Y.; Miya, S.; Yamamoto, Y.; Hiroshi, Y. Inorg. Chim. Acta
1981, 53, L53-L55.
(17) Fournier, E.; Sicard, S.; Decken, A.; Harvey, P. D. Inorg. Chem.
2004, 43, 1491-1501.
(8) Dartiguinave, M.; Dartiguinave, Y.; Guitard, A.; Mari, A. Polyhedron
1989, 8, 317-323.
(18) Green, J. C.; Jardine, C. N. J. Chem. Soc., Dalton Trans. 1999,
3767-3770.
(9) Harvey, P. D. Coord. Chem. ReV. 2001, 219-221, 17-52.
(10) Siemeling, U.; Rother, D.; Ruhr, C.; Fink, H.; Weidner, T.; Tra¨ger,
F.; Rothenberger, A.; Fenske, D.; Priebe, A.; Maurer, J.; Winter, R. J. Am.
Chem. Soc. 2005, 127, 1102-1103.
(19) Schwemlein, H.; Zsolnai, L.; Huttner, G.; Brintzinger, H.-H. J.
Organomet. Chem. 1983, 256, 285-289.
(20) Foo, D. M. J.; Shapiro, P. J. Organometallics 1995, 14, 4957-
4959.
(11) MacDonald, M.; Puddephatt, R. J. Organometallics 2000, 19, 2194-
2199.
(21) Matare, G. J.; Foo, D. M.; Kane, K. M.; Zehnder, R.; Wagener,
M.; Shapiro, P. J. Organometallics 2000, 19, 1534-1539.
10.1021/om050710j CCC: $33.50 © 2006 American Chemical Society
Publication on Web 01/05/2006

720 Organometallics, Vol. 25, No. 3, 2006
Shapiro et al.
Scheme 1
This method has enabled us to prepare a series of ansa-
chromocene isocyanide complexes containing alkyl, aryl, and
ferrocenyl isocyanide ligands. With the exception of the
ferrocenyl isocyanide derivatives, the redox potentials, molecular
structures, isocyanide carbon chemical shifts, and CN stretching
frequencies of these complexes exhibit congruous trends that
reflect the degree of π-backbonding from the chromium center.
One-electron oxidation of the 18e ansa-chromocene isocyanide
species affords stable, low-spin, 17e cations and is accompanied
by significant changes in bonding between the chromium and
the isocyanide ligand. Density functional theory (DFT) calcula-
tions were performed on model complexes in order to elucidate
the nature of the bonding between the chromium and the
isocyanide ligands and to explain the effects of one-electron
oxidation of the complexes.
The thermal and photochemical stability of the other isocya-
nide derivatives was not examined in detail; however, these
complexes appear to have a longer shelf life than the tert-butyl
isocyanide derivatives. Nevertheless, all compounds were stored
at -30 °C in the absence of light in order to minimize their
decomposition.
X-ray Crystallographic Characterization of ansa-Chro-
mocene Isocyanide Complexes. The molecular structures of
complexes 2a, 3, 4, 5, 6, and 7a were determined by single-
crystal X-ray diffraction. Crystallographic data are listed in
Tables 1 and 2. Selected geometrical parameters for the
complexes are compared in Tables 3 and 4. The data in Table
3 reveal little variation among the complexes with respect to
the geometries of their metallocene fragments. The most
distinctive features of the structures are the bonding parameters
associated with the isocyanide ligands (compared in Table 4),
which reflect the degree of π-backbonding from the chromium
to the π* orbital of the isocyanide ligand. The greater the
π-backbonding from the metal, the more carbenoid tautomer B
contributes to the structure (Figure 1).
Results and Discussion
Preparation of ansa-Chromocene Isocyanide Complexes.
A variety of ansa-chromocene isocyanide complexes were
prepared according to the reactions shown in Scheme 1. The
visible evolution of CO gas indicated the progress of the
reaction. Addition of excess isocyanide ligand leads to the
formation of unidentified paramagnetic material. As reported
previously,20 the reactions can be reversed by exposing the
isocyanide derivatives to an atmosphere of carbon monoxide.
All of the 18e ansa-chromocene isocyanide complexes were
isolated as air-sensitive, red-brown, microcrystalline solids or
powders. The tert-butyl isocyanide derivatives, 2a and 2b, are
both thermally and photochemically sensitive. Over months of
storage in a nitrogen atmosphere at room temperature, they form
black, shiny paramagnetic materials, the identities of which have
not been determined. Both thermolytic and photolytic decom-
position of the complexes produces an equivalent of isobutene,
presumably from the decomposition of the tert-butyl isocyanide
ligand. Dealkylation of coordinated isocyanide ligands, particu-
larly tert-butyl isocyanide, to form a metal-cyanide linkage
has been observed in other complexes.22-24
The molecular structure of 2a and its crystal packing diagram
are shown in Figure 2. The C17-N1-C18 angle of 131.1(3)°
(Table 4) is smaller than the corresponding angle of 139.5(4)°
for the related nonbridged molybdocene complex (C5H5)2-
MoCN-t-Bu.25 The crystal packing diagram for 2a exhibits a
C-H‚‚‚N interaction between the isocyanide nitrogen of one
molecule and a cyclopentadienyl C-H of another. The inter-
molecular C9-N1 distance of 3.423(4) Å and C9-H9A-N1
angle of 144.2° are characteristic of a weak hydrogen-bonding
(22) Gambarotta, S.; Floriani, C.; Chiesi-Villa, A.; Guastani, C. Inorg.
Chem. 1984, 23, 1739-1747.
(23) Giandomenico, C. M.; Hanau, L. H.; Lippard, S. J. Organometallics
1982, 1, 142-148.
(24) Dewan, J. C.; Giandomenico, C. M.; Lippard, S. J. Inorg. Chem.
1981, 20, 4069-4074.
(25) Martins, A. M.; Calhorda, M. J.; Romao, C. C. J. Organomet. Chem.
1992, 423, 367-390.

ansa-Chromocene Complexes
Organometallics, Vol. 25, No. 3, 2006 721
Table 1. Crystallographic Data for 2a, [2b][B(C6F5)4], 3, and 4
2a
[2b][B(C6F5)4]
3
4
formula
mol wt
cryst syst
space group
a (Å)
b (Å)
c (Å)
R (deg)
â (deg)
γ (deg)
V (Å3)
Z
C
21H29CrN
C53H29BCrF20
1122.58
triclinic
P1h
12.8489(7)
13.4768(7)
15.3169(8)
74.943(1)
70.386(1)
73.237(1)
2352.9(2)
2
N
C25H29CrNO2S
459.55
triclinic
C25H29CrN
395.49
monoclinic
C2/c
10.025(9)
27.72(3)
7.709(7)
90
112.64(2)
90
347.45
monoclinic
P2(1)/c
9.6236(8)
10.3948(9)
18.3517(16)
90
93.680(2)
90
P1h
8.2361(6)
11.5531(8)
12.3836(8)
75.863(2)
74.716(2)
82.416(2)
1099.35(13)
2
1832.0(3)
4
1977(3)
4
T (K)
λ (Å)
203(2)
0.71073
1.260
0.624
0.14 × 0.09 × 0.06
2.12 to 25.25
-11ehe11,
-12eke7,
-22ele21
10607
203(2)
0.71073
1.585
203(2)
0.71073
1.388
0.637
0.16 × 0.09 × 0.04
1.75 to 28.31
-10ehe10,
-15eke15,
-15ele16
11680
203(2)
0.71073
1.329
F
calc (Mg/m3)
µ (mm-1
)
0.362
0.588
cryst size (mm3)
θ range (deg)
index ranges
0.35 × 0.17 × 0.09
1.60 to 25.33
-15ehe15,
-16eke16,
-18ele17
22340
0.27 × 0.07 × 0.03
2.32 to 24.99
-11ehe11,
-32eke27,
-9ele6
8234
no. reflns collected
no. indep reflns
3316 [R(int) )
0.0730]
8594 [R(int) )
0.0296]
5439 [R(int) )
0.0421]
1739 [R(int) )
0.1467]
no. data/restraints/params
GOF
3316/0/215
0.988
0.0498
0.0938
0.323, -0.246
8594/78/675
1.023
0.0501
0.1163
0.718, -0.405
5439/0/276
1.085
0.0581
1199
0.458, -0.313
1739/0/128
0.934
0.0621
0.1262
0.441, -0.390
R1 [I>2σ(I)]a
wR2 [I>2σ(I)]b
largest diff peak, hole (e Å-3
)
b
2
2
2
a R1 ) ∑||Fo| - |Fc||/∑|Fo|. wR2 ) {∑[w(Fo - Fc )2]/∑[w(Fo )2]}1/2
.
Table 2. Crystallographic Data for [4][B(C6F5)4], 5, 6, and 7a
[4][B(C6F5)4]
5
6
7a
formula
mol wt
cryst syst
space group
a (Å)
C56H37BCrF20N
1166.68
monoclinic
Cc
11.5318(8)
27.2656(19)
15.4718(11)
90
93.05(2)
90
4857.8(6)
4
C35H29CrFeN
571.44
monoclinic
P2(1)/c
7.641(3)
10.176(7)
34.17(2
90
95.51(5)
90
2645(3)
4
C40H44Cr2N2
656.77
monoclinic
P2(1)/n
10.551(9)
9.168(8)
17.063(18)
90
107.79(4)
90
C50H54Cr2FeN2
842.80
triclinic
P1h
12.8945(5)
13.3213(5)
13.4444(5)
69.17(1)
b (Å)
c (Å)
R (deg)
â (deg)
γ (deg)
V (Å3)
Z
68.65(1)
71.84(1)
1965.8(2)
2
1572(3)
2
T (K)
203(2)
203(2)
81(2)
81(2)
λ (Å)
0.71073
1.595
0.354
0.71073
1.435
0.985
0.71073
1.388
0.724
0.71073
1.424
0.946
F
calc (Mg/m3)
µ (mm-1
)
cryst size (mm3)
θ range (deg)
index ranges
0.26 × 0.23 × 0.04
1.92 to 25.25
-13ehe13,
-32eke30,
-15ele18
21 193
7472 [R(int) ) 0.0480]
7472/2/717
1.014
0.29 × 0.21 × 0.05
2.09 to 27.49
-9ehe9,
-8eke13,
-35ele44
23 445
6072 [R(int) ) 0.0655]
6072/0/349
1.073
0.23 × 0.08 × 0.03
2.03 to 25.47
-12ehe10,
-11eke10,
-20ele20
10 327
2873 [R(int) ) 0.1740]
2873/12/199
0.894
0.19 × 0.10 × 0.04
1.67 to 28.29
-17ehe17,
-17eke17,
-17ele17
24 182
9731 [R(int) ) 0.0868]
9731/0/504
0.967
no. reflns collected
no. indep reflns
no. data/restraints/params
GOF
R1 [I>2σ(I)]a
0.0438
0.0798
0.268, -0.251
0.0609
0.1390
0.789, -0.494
0.0640
0.1251
0.927, -0.773
0.0625
0.1115
0.565, -0.523
wR2 [I>2σ(I)]b
largest diff peak, hole (e Å-3
)
b
2
2
2
a R1 ) ∑||Fo| - |Fc||/∑|Fo|. wR2 ) {∑[w(Fo - Fc )2]/∑[w(Fo )2]}1/2
.
interaction.26 Similar C-H‚‚‚N interactions were observed in
the X-ray crystal structure of cyanoferrocene, (C5H5)Fe(C5H4-
CN), with slightly shorter C- - -N distances of 3.384(6) and
3.381(6) Å.27 The geometry of the metallocene fragment of 2a
(Table 3) is effectively the same as that of the carbonyl
derivative 1a19 as shown by the similar angles between the
cyclopentadienyl ring planes (1, 38.5(5)°; 2a, 38.7(4)°), the
centroid-Cr-centroid angles (1a, 143.3(5)°; 2a, 143.4(2)°), and
the average Cr-centroid distances (1a, 1.78 Å; 2a, 1.791 Å).
The molecular structure of 3 and its crystal packing diagram
are shown in Figure 3. This complex has an even smaller C17-
N-C18 bond angle (123.1(3)°) than 2a (Table 4). A search of
the 2004 Cambridge Data Base showed that only 6 out of 1439
structurally characterized terminal isocyanide-containing transi-
tion metal complexes have bending angles in the range 120-
130°. The greater π-backbonding from the metal in 3 vs 2a is
also manifested in a shortening of the chromium-isocyanide
carbon bond distance (Cr-C) and a lengthening of the isocya-
(26) Desiraju, G. R.; Steiner, T. The Weak Hydrogen Bond; Oxford
University Press: Oxford, 1999.
(27) Bell, W.; Ferguson, G.; Glidewell, C. Acta Crystallogr. 1996, C52,
1928-1930.

722 Organometallics, Vol. 25, No. 3, 2006
Shapiro et al.
Table 3. Comparison of Selected Bond Lengths and Angles
for Metallocene Portion of Neutral and Cationic
ansa-Chromocene Isocyanide Complexesa
The molecular structure of the isocyanoferrocene derivative
5 is shown in Figure 6. No intermolecular hydrogen-bonding
interactions are evident in the crystal packing diagram of 5.
The C25-N1-C26-C30 (4.5(7)°) and C25-N1-C26-C27
(178.1(4)°) torsion angles show that the isocyano-substituted
cyclopentadienyl ring of the ferrocene is nearly parallel to the
equatorial plane of the chromocene, in contrast to the perpen-
dicular orientation of the xylyl group in 4. The ferrocenyl C-C
bond distances in the isocyano-substituted ring are comparable
to each other and to those in the unsubstituted ring. There are
relatively few examples of the coordination of isocyanoferrocene
to other metals.30-32 Of these, only the homoleptic [Cr(CNFc)6]
complex of Barybin and co-workers30 has been structurally
characterized. The C25-N1-C27 bond angle in 5 of 134.4-
(4)° is considerably narrower than the C-N-C(Fc) angles of
[Cr(CNFc)6] (av 162°) and quite similar to the corresponding
bond angle in 2a.
cpd
Cp Cp (deg)
Centa-Cr-Cent (deg)
av Cent-Cr (Å)
2a
3
4
5
7
38.7
40.1
38.8
40.0
39.1 (Cr1)
39.9 (Cr2)
39.1
143.4
143.0
143.1
143.2
143.4
142.5
141.2
141.8
1.792
1.797
1.782
1.802
1.792
1.799
1.821
1.818
2b+
4+
38.9
a Cent ) ring centroid; Cp Cp ) dihedral angle between cyclopenta-
dienyl ring planes.
Figure 1. Resonance tautomers of a chromium isocyanide complex.
The molecular structure of the homodinuclear complex 6 is
shown in Figure 7. The two halves of the molecule are related
by a crystallographic inversion center. The phenylene diiso-
cyanide ligand is oriented parallel to the equatorial planes of
the ansa-chromocene moieties with a dihedral angle of 65.7°
between the arene and centroid-Cr-centroid ring planes.
Considerable π-backbonding beween the chromium centers and
the isocyanide groups is indicated by the small bond angle
(128.6°) about the nitrogens and the lengthening of the C-N
bonds and shortening of the Cr-C bonds relative to complexes
2a, 5, and 4.
nide carbon-nitrogen bond (C-N). Three different examples
of weak, nonclassical, intermolecular C-H hydrogen bonding
are found in the crystal packing diagram of this complex: C12-
H12‚‚‚O1A; C23-H23A‚‚‚O2B; and C18-H18D‚‚‚N1C.
Unlike the isocyanide ligands in 2a and 3, the terminal xylyl
isocyanide ligand in the molecular structure of complex 4
(Figure 4) is perfectly linear because it lies on a crystallographic
C2 axis. The Cr-C9 bond in 4 is longer and the C9-N1 bond
is shorter than the corresponding bond lengths in 2a and 3,
consistent with a lower degree of π-backbonding from the metal.
No intermolecular hydrogen-bonding interactions were evident
in the crystal packing diagram of this species, consistent with
the absence of lone pair electron density on nitrogen and with
the reduced accessibility of the nitrogen atom in this crowded
molecule. The xylyl moiety is nearly perpendicular to the
equatorial wedge of the metallocene, with an 18.4° dihedral
angle between the arene ring plane and the centroid-Cr-
centroid plane. A similar orientation of the xylyl moiety is seen
in the molecular structure of the half-open chromocene complex
Cp(C5H7)Cr{CN-2,6-(CH3)2C6H3};28 however, in that case the
isocyanide is bent with an angle of 166.1(4)°. In the ansa-
chromocene complex Me2Si(C5Me4)2CrCNXyl, reported by
Schaper et al.,29 the xylyl ring plane is parallel to the equatorial
plane of the metallocene, probably due to greater steric
interference from the methyl groups on the cyclopentadienyl
rings. In this complex, the isocyanide ligand is bent at nitrogen
with an angle of 153.2(3)°. The N-Caryl bond distance (1.401-
(4) Å) in Schaper’s complex is significantly longer than that in
4 (1.351(8) Å), which is quite short for a single N-C bond
and suggests additional π bonding between the nitrogen and
the arene ring. Thus, delocalization of electron density onto the
xylyl ring of the isocyanide ligand, as represented by the
resonance tautomers in Figure 5, could be partially responsible
for the linearity of the isocyanide ligand.
Complexes 7a and 7b are rare examples of transition metal
complexes containing the diisocyanoferrocene ligand.33 The only
other examples, to our knowledge, are Fc[(NC)Cr(CO)5]2 and
polymeric [Fc(NC)2(AuCl)2]n, which were recently reported by
Siemeling and co-workers.10 Repeated efforts to grow crystals
of 7b were unsuccessful. However, 7a did afford X-ray quality
crystals. Figure 8 shows top (a) and side (b) views of the
molecular structure of the trinuclear heterometallic complex.
The crystal packing diagram in Figure 8c shows the weak C-H‚
‚‚Cr interactions between ansa-chromocene fragments of dif-
ferent molecules. The cyclopentadienyl rings of the diisocy-
anoferrocene are eclipsed, and their rotational conformation
places the two ansa-chromocene fragments in a syn orientation.
This arrangement does not appear to be dictated by intra- or
intermolecular hydrogen-bonding interactions. Interestingly, a
similar conformation was identified for Fc[(NC)Cr(CO)5]2.10 The
Fc(C)-N-C bond angles for that complex are 174.9° and
170.2°, considerably wider than the corresponding bond angles
in 7a (126.4° and 125.8°).
NMR and IR Spectroscopic Characterization of ansa-
Chromocene Isocyanide Complexes. Previous studies have
shown that there is a correlation between the degree of
π-backbonding from the metal to the isocyanide ligand and the
C-N stretching frequency, 13C NMR chemical shift, and l4N
Table 4. Comparison of Selected Crystallographic, IR, NMR, and Electrochemical Data for the Cr-CNR Portion of Neutral
ansa-Chromocene Isocyanide Complexes
ν
C-N (cm-1
Nujol mull (THF)
)
CrII/III (mV)
(Ea +Ec)/2
13C δ (ppm)
CtNR
cpd
Ra
t-Bu
CH2Tos
Xyl
Fc
-Ph-
-Fc-
Cr-C (Å)
CrC-N (Å)
N-CR (Å)
C-N-R (deg)
2a
3
4
5
6
1.873(3)
1.859(3)
1.892(7)
1.879(4)
1.861(5)
1.862(4)
1.852(4)
1.214(4)
1.226(4)
1.195(7)
1.205(5)
1.232(6)
1.224(4)
1.230(4)
1.482(4)
1.462(4)
1.351(8)
1.411(5)
1.427(6)
1.409(4)
1.415(4)
131.1(3)
123.1(3)
180.0(1)
134.4(4)
128.6(5)
126.4(3)
125.8(3)
1835
-1154
-1058
-1181
-1153
-1096
-1149
261.2
290.6
239.6
274.0
278.3
278.6
1752 (1774)
2006 (1980)
1829
1766
1751
7a
a Tos ) p-tolylsulfonyl; Fc ) 1-ferrocenyl; -Fc- ) 1,1′-ferrocenyl; -Ph- ) 1,4-phenylene.

ansa-Chromocene Complexes
Organometallics, Vol. 25, No. 3, 2006 723
Figure 2. (a) Thermal ellipsoid (30%) drawing of 2a. Hydrogen atoms are omitted for clarity. (b) Crystal packing diagram illustrating
intermolecular hydrogen bonding; C9‚‚‚N1 3.423(4) Å, C9-H9A‚‚‚N1 144.2°.
Figure 3. (a) Thermal ellipsoid (30%) drawing of 3. Hydrogen atoms are omitted for clarity. (b) Crystal packing diagram illustrating weak
intermolecular hydrogen bonding. Only hydrogen atoms involved in bonding are shown: C12‚‚‚O1A 3.267(4) Å, C12-H12‚‚‚O1A 150.0°;
C18‚‚‚N1B 3.344(4) Å, C18-H18B‚‚‚N1B 140.7°, C23‚‚‚O2C 3.314(4) Å, C23-H23‚‚‚O2C 155.3°.
complexes listed in Table 4 roughly exhibit the expected trends.
We were not able to detect 14N NMR signals for these
complexes, probably due to the low-symmetry environments
about the nitrogen atoms.
Since π-backbonding from the metal to the π* orbital of the
isocyanide ligand causes a reduction in the C-N bond order,
the magnitude of νC-N should decrease with increasing π-elec-
tron donation from the metal. A plot of C-N stretching
frequencies against the crystallographically determined C-N
bond lengths (average bond length for 7a) is shown in Figure
9 (data from Table 4). Deviations from a perfect correlation
between the parameters can be attributed at least in part to the
fact that the different masses of the substituents on the
isocyanide ligands must also influence νC-N. Furthermore, νC-N
for 3 shifted to 22 cm-1 higher energy and νC-N for 4 shifted
to 26 cm-1 lower energy when the complexes were analyzed
in THF solution instead of a Nujol mull. This shows that the
bonding between the metal and the isocyanide ligand is
somewhat flexible and is probably influenced by crystal packing
forces.
Figure 4. Thermal ellipsoid (30%) drawing of 4. Hydrogen atoms
are omitted for clarity.
NMR chemical shift of the isocyanide ligand.4,34,35 The IR and
13C NMR spectroscopic data for the neutral 18e isocyanide
(28) Freeman, J. W.; Hallinan, N. C.; Arif, A. M.; Gedridge, R. W.;
Ernst, R. D.; Basolo, F. J. Am. Chem. Soc. 1991, 113, 6509-6520.
(29) Schaper, F.; Wrobel, O.; Schwoerer, R.; Brintzinger, H.-H. Orga-
nometallics 2004, 23, 3552-3555.
(30) Holovics, T. C.; Deplazes, S. F.; Toriyama, M.; Powell, D. R.;
Lushington, G. H.; Barybin, M. V. Organometallics 2004, 23, 2927-2938.
(31) Knox, G. R.; Pauson, P. L.; Willison, D. Organometallics 1990, 9,
301-306.
(32) El-Shihi, T.; Siglmueller, F.; Herrmann, R.; Fernanda, M.; Carvalho,
N. N.; Pombeiro, A. J. L. J. Organomet. Chem. 1987, 335, 239-47.
(33) van Leusen, D.; Hessen, B. Organometallics 2001, 20, 224-226.
(34) Grubisha, D. S.; Rommel, J. S.; Lane, T. M.; Tysoe, W. T.; Bennett,
D. W. Inorg. Chem. 1992, 31, 5022-5027.
(35) Rommel, J. S.; Weinrach, J. B.; Grubisha, D. S.; Bennett, D. W.
Inorg. Chem. 1988, 27, 2945-2949.

724 Organometallics, Vol. 25, No. 3, 2006
Shapiro et al.
Figure 5. Resonance tautomers that could contribute to the linearity of the isocyanide ligand in 4.
is comparable to that of a tert-butyl group in the ansa-
chromocene system. The similar ν(CN) and C-N-R values
for 2b and 5 (Table 4) also support this conclusion. This
contrasts with the findings for Cr(CNR)6 and (CO)5CrCNR
complexes, in which the ferrocenyl moiety behaves more like
an aryl group than an alkyl group.1,10,30
Cyclic voltammograms of complexes 6 and 7a are shown in
Figure 11. The appearance of only one CrII/III redox couple for
these complexes indicates that there is little if any electronic
communication between the chromium centers via the diisocy-
anoferrocene and phenylene diisocyanide bridges. The cyclic
voltammogram for 7b is similar to that of 7a. Each of the
voltammograms exhibits an electrochemically irreversible CrIII/IV
couple. The redox potentials of the coordinated isocyano- and
diisocyanoferrocene ligands in 5, 7a, and 7b are more positive
than even the anodic wave for the oxidation of CrIII to CrIV.
The iron-centered redox potential in 5 (282 mV) is almost 100
mV higher than that of the free isocyanoferrocene ligand (186
mV). It can be argued that this positive shift in the redox
potential is due to the electron-withdrawing properties of CrIV.
On the other hand, the iron-centered redox potential in 7a (589
mV) is 50 mV lower than that of free diisocyanoferrocene (639
mV). It is possible that the diisocyanoferrocene ligand is
dissociated from the ansa-chromocene dications at this point,
since there is evidence of free isocyanoferrocene in the IR
spectrum of [7b][PF6]2, in which the ansa-chromocenes are only
monocations. Efforts to attach only one ansa-chromocene moiety
to diisocyanoferrocene for comparison were unsuccessful. The
1:1 reaction between 1a and diisocyanoferrocene afforded only
7a and half an equivalent of unreacted diisocyanoferrocene.38
Chemical Oxidation of the ansa-Chromocene Complexes.
One-electron oxidation of the isocyanide complexes was ac-
complished using either trityl or ferrocenium salts to form low-
spin, 17e cations that are red-brown in color, like the neutral,
18e species (Scheme 2). Magnetic susceptibility measurements
on the monocations gave spin-only magnetic moments consistent
with one unpaired electron on the metal: [2b][B(C6F5)4] (µeff
) 1.8 µB), [4][B(C6F5)4] (µeff ) 1.9 µB), [5]PF6 (µeff ) 1.9
µB). Since selective oxidation of only one ansa-chromocene
moiety in the diisocyanide-bridged complexes was not feasible,
Figure 6. Thermal ellipsoid (30%) drawing of 5. Only the ansa
bridge hydrogen atoms are shown for clarity. Selected bond lengths
(Å) and angles (deg): C26-C27 1.411(5), C26-C30 1.436(6),
C27-C28 1.422(6), C29-C30 1.412(6), C29-C28 1.399(6), C31-
C32 1.393(7), C31-C35 1.420(0), C32-C33 1.404(7), C33-C34
1.403(7), C34-C35 1.409(7), C25-N1-C26 134.4(4), C25-N1-
C26-C30 4.5(7), C25-N1-C26-C27 178.1(4).
Figure 7. Thermal ellipsoid (30%) drawing of 6. Hydrogen atoms
are omitted for clarity. Selected bond lengths (Å): C18-C19 1.386-
(7), C19-C20 1.395(6), C20-C18A 1.399(7).
The trend in 13C chemical shifts for the complexes is
consistent with that observed for other transition metal isocya-
nide complexes. The isocyanide carbon resonance shifts down-
field with increasing M-C π bonding because the M-C π
bonding increases the magnitude of the negative-valued para-
magnetic shielding constant (σp). This constant dominates the
chemical shift for atoms, such as carbon, with 2pπ elec-
trons.4,34,35 Plots of redox potential (Figure 10a) and νC-N
(Figure 10b) against 13C chemical shift show the expected
general trends; the 13C chemical shifts increase with increasing
redox potential and decreasing νC-N
.
both were oxidized to form the dications [6][B(ArF)4]2 (µeff
)
Electrochemical Measurements. The isocyanide complexes
exhibit a quasi-reversible CrII/III couple and an anodic peak at
higher potential for the oxidation of CrIII to CrIV, the electro-
chemical irreversibility of which has been attributed to the
coordination of the THF solvent by CrIV.36 The CrII/III redox
potentials for the complexes (referenced to (C5H5)2Fe0/1+) are
listed in Table 4. With the exception of 7a, there is good
correlation between the redox potentials of the complexes and
the π-accepting character of the isocyanide ligands. In general,
the redox potentials of the complexes increase with increasing
M-C π-backbonding (as reflected in the structural and spec-
troscopic parameters). This behavior has been observed in a
variety of other metal isocyanide complexes, and the redox
potential of the metal complex has even been used to param-
etrize the electron donor/acceptor character of the isocyanide
ligand.37
2.3 µB; ArF ) 3,5-(CF3)2C6H3) and [7b][PF6]2 (µeff ) 2.6 µB).
The magnetic moments of these polynuclear complexes cor-
respond to magnetic moments of 1.6 µB (7b2+) and 1.8 µB (62+
)
per chromium center.39 This supports our conclusion from the
cyclic voltammetry that there is no electronic interaction between
chromium centers via the phenylene or ferrocenylene bridge.
X-ray quality crystals of [2b][B(C6F5)4] and [4][B(C6F5)4]
were obtained for molecular structure determinations. Their
(36) van Raaij, E. U.; Brintzinger, H. H. J. Organomet. Chem. 1988,
356, 315-323.
(37) Pombeiro, A. J. L. Inorg. Chim. Acta 1985, 103, 95-103.
(38) A reviewer suggested that the selectivity for the heterotrinuclear
species over the heterodinuclear species in the 1:1 reaction between 1a and
diisocyanoferrocene may be entropic in nature since the first coordinated
ansa-chromocene should impede the rotation of the noncoordinated fer-
rocene ring, facilitating the coordination of a second ansa-chromocene
fragment.
Interestingly, the CrII/III potential for 2a is similar to that of
5. This indicates that the electronic effect of a ferrocenyl moiety
(39) Since µeff is proportional to xøM, µeff for each ansa-chromocene
moiety was estimated by dividing øM for the entire molecule by x2.

ansa-Chromocene Complexes
Organometallics, Vol. 25, No. 3, 2006 725
Figure 8. Thermal ellipsoid (30%) drawing of 7a: (a) side view, (b) top view, and (c) crystal packing diagram illustrating intermolecular
hydrogen bonding. Only the hydrogen atoms involved in bonding are shown.
and only slightly lower than that of [Cp3ZrCN-t-Bu]+ (2209
cm-1),40 indicating that the interaction between the isocyanide
ligand and chromium is primarily σ in nature.
Since the xylyl isocyanide ligand does not lie on an axis of
symmetry in [4][B(C6F5)4)], as it does in 4, it is not perfectly
linear but nearly so. The Cr-C17 bond is longer and the C17-
N1 bond is shorter than the corresponding bonds in the neutral
complex, due to reduced π-backbonding from the metal. The
N-C(aryl) bond distance (N1-C18, 1.396(5) Å) is longer than
that in 4 (1.351(8) Å), indicating less π e- delocalization from
the Cr to the arene ring. The orientation of the xylyl moiety is
nearly perpendicular to the equatorial wedge, similar to the
neutral complex, with a 19.3° dihedral angle between the arene
ring plane and the centroid-Cr-centroid plane. The energy of
the C-N stretch for this compound is 2119 cm-1, which is
Figure 9. Plot of C-N stretching frequencies (measured in Nujol
mull unless otherwise indicated) vs Cr-N bond length (an average
between the two C-N bond distances of 7a was used).
comparable to that of the free ligand (2115 cm-1).
We were unable to obtain X-ray quality crystals of either
[5][PF6] and [7b][PF6]2. The energy of the C-N stretch for
[5][PF6] of 2118 cm-1 is similar to that of the free isocyanide
(ν ) 2121 cm-1). Two C-N stretches were observed for [7b]-
[PF6]2 at 2113 and 2137 cm-1. We attribute the lower energy
band to free diisocyanoferrocene (ν ) 2114 cm-1) and the
higher energy band to the dication.
An X-ray crystal structure determination for [6][B(ArF)4]2
was performed, but the quality of the crystal was so poor that
we could not refine the structure beyond an R value of 12.5%.
A preliminary thermal ellipsoid drawing of the dication is shown
in Figure 14. The phenylene diisocyanide bridge is clearly linear
in [6][B(ArF)4]2, in contrast to its zigzag geometry in the neutral
complex. The C-N stretch for this complex occurs at 2088
crystallographic data are listed in Table 1. Thermal ellipsoid
drawings of the cations 2b+ and 4+ are shown in Figures 12
and 13, respectively, along with selected bond lengths and
angles.
Removal of an electron from 2b results in a straightening of
the tert-butyl isocyanide ligand, which is close to linear in [2b]-
[B(C6F5)4], with a C13-N1-C14 bond angle of 173.9(4)°. The
Cr-C13 bond is significantly longer and the C13-N1 bond is
significantly shorter than the corresponding bond lengths in 2b.
All of this is a reflection of the poorer π-backbonding ability
of the cationic chromium center. Two IR bands at 2199 and
2139 cm-1 (Nujol mull) were observed for this complex. Neither
band corresponds to the neutral complex, 2b. We attribute the
second band to free t-BuNC (ν NC ) 2140 cm-1). The 2199
cm-1 band is higher in energy than that of the free isocyanide
(40) Brackemeyer, T.; Erker, G.; Fro¨hlich, R. Organometallics 1997,
16, 531-536.

726 Organometallics, Vol. 25, No. 3, 2006
Shapiro et al.
Figure 10. Plot of (left) redox potential vs isocyanide 13C chemical shift and (right) νC-N vs isocyanide 13C chemical shift for the
ansa-chromocene isocyanide complexes.
Figure 11. Cyclic voltammograms of complexes 6 (left) and 7a (right) (scan rate 25 mV/s).
Scheme 2
cm-1, somewhat lower in energy than that of the free phenylene
diisocyanide ligand (ν ) 2118 cm-1) but substantially higher
than that of 6 (1766 cm-1).
The bonding between the chromium and its cyclopentadienyl
rings appears to be unaffected by one-electron oxidation of the
metal since the structural parameters of the metallocene frag-
ments of the cations are similar to those of the neutral complexes
(Table 3). Thus, the change in the oxidation state of the metal
is mostly manifested in its interaction with the isocyanide ligand.
Calculations. The structures of models for complexes 2-6
(containing a CH2CH2 bridge instead of the CMe2CMe2 and
CHPhCHPh bridges used experimentally) were studied using
DFT methods. Geometries of all neutral 18e and cationic 17e
species were fully optimized at the RIBP86/SV(P) level, and

ansa-Chromocene Complexes
Organometallics, Vol. 25, No. 3, 2006 727
calculated data is revealing. In contrast to the neutral complexes,
the cations show virtually no changes in the CdN bond length,
even though the Cr-C bond length changes appreciably (Figure
15b). The difference between neutral and cationic species must
be due to a difference in bonding interaction. In the neutral
complexes, π-backbonding dominates. Small, electron-with-
drawing substituents on the ligand (e.g. CH2Tos) increase
π-backbonding and, hence, produce shorter Cr-C and longer
CdN bonds. For the cation, π-backbonding appears to be
relatively unimportant, and the isocyanide functions mainly as
a σ-donor. The CdN bond lengths are nearly identical to those
calculated for the free ligands, and the Cr-C bond length
appears to be determined mainly by the steric bulk of the
substituent at nitrogen, t-Bu and Xyl giving the largest distances.
We have calculated the structures of both “parallel” and
“perpendicular” conformations of the PhNC and XylNC com-
plexes. These conformations are very similar in energy for the
neutral as well as the cationic complexes. However, the CdN
bond lengths are significantly different for the neutral complexes
only. On going from the planar to the perpendicular conforma-
tion, the C-N-C angle opens up (by ca. 20°), the Cr-C bond
becomes longer, and the CdN distance becomes smaller, all of
which indicates decreased π-backbonding. The fact that this is
not accompanied by a large change in energy demonstrates that
the bonding situation in these molecules is rather fluid and can
easily adapt itself to electronic or steric variations of the ligand.
The calculated monocation of the phenylene-bridged dinuclear
system shows clearly localized CrII and CrIII centers character-
ized by very different C-N-C angles.
Figure 12. Thermal ellipsoid (30%) drawing of [2b][B(C6F5)4].
Only the cation is shown. Hydrogen atoms are omitted for clarity.
Selected bond lengths (Å) and angles (deg): Cr1-C13 1.956(3),
C13-N1 1.155(4), av Cr-centroid 1.821, C13-N1-C14 173.9-
(4). Only part of the disordered ansa-bridge is displayed.
Figure 13. Thermal ellipsoid (30%) drawing of [4][B(C6F5)4]. Only
the cation is shown. Hydrogen atoms are omitted for clarity.
Selected bond lengths (Å) and angles (deg): Cr1-C17 1.933(5),
C17-N1 1.174(5), N1-C18 1.396(5), C18-C19 1.404(6), C18-
C23 1.390(6), C19-C20 1.378(6), C20-C21 1.378(7), C21-C22
1.383(8), C22-C23 (1.393(6), C17-N1-C18 178.8(4).
ii. Electronic Structures of the Cations. The highest
occupied orbitals of the cationic complexes are not well-
separated, making it difficult to interpret the nature of the
oxidized state. However, the orbital plot of the â LUMO (shown
in Figure 16 for the analogue of 5+) clearly shows that oxidation
2
has removed an electron from a dz -like orbital lying in the
wedge between the two Cp rings, perpendicular to the Cr-N
bond. This orbital is similar to the highest occupied molecular
orbital of Cp2CrCO determined by Lauher and Hoffman in 1976
using extended Hu¨ckel theory41 and more recently by Green et
al. using density functional theory.18 The spin-density plot (not
shown) shows that there is a small amount of spin polarization
on the Cr-bound Cp rings, but virtually no spin density on the
ferrocenyl group. This is because the oxidation occurs essentially
in the σ system, so the effects are not communicated efficiently
via the cyclopentadienyl π system of the ferrocenyl moiety.
Presumably, this is also part of the reason we see virtually no
communication between the oxidized centers in the bridged
dinuclear systems. The nonbonding nature of this orbital explains
why removal of an electron from it does not result in large
geometrical changes in the chromocene fragment. This orbital
cannot have been involved directly in Cr-isocyanide π-back-
bonding, because it is of the wrong symmetry. However,
removing an electron from it lowers the levels of all remaining
occupied chromium d orbitals and so makes them less available
for electron back-donation.
Figure 14. Preliminary thermal ellipsoid (30%) drawing of [6]-
[B(ArF)4]2. Only the dication is shown.
improved energies were obtained from single-point b3-lyp/TZVP
calculations including a COSMO solvation correction (ꢀ ) 7.5);
this correction is particularly important for the cationic species.
The resulting geometric parameters are collected and compared
with experimental values in Table 5; energy-derived data are
listed in Table 6.
i. Geometries. One of the advantages of using calculated
structures is the lack of “random” variations found in X-ray
structures due to packing effects and noise. Systematic errors
can still be important, but trends often appear more clearly from
calculated data. For the neutral complexes, for which we have
collected a fair amount of structural data, the calculated Cr-
Cisocyanide distance is typically too short by 0.01-0.02 Å, while
the calculated CdN bond is usually too long by roughly the
same amount (this bond length is also overestimated for the
free ligands). The spread in both parameters induced by ligand
variation is larger than this, and the experimental and calculated
data both show the inverse correlation between the two bond
lengths, as can be seen in the plot of both experimental and
calculated Cr-Cisocyanide bond lengths vs CdN bond lengths for
the neutral complexes in Figure 15a. For the cationic complexes,
only three X-ray structures are available, one of which is of
insufficient quality to trust the bond lengths. Here, use of the
iii. Energies. To facilitate interpretation of the results, we
have included the CO complex as a reference compound. Table
6 shows relative binding energies of CO and isocyanides to the
ansa-chromocene fragment. Apparently, all isocyanides bind
somewhat more weakly than CO. This is not very surprising:
in the neutral systems, π-backbonding dominates, and this will
be stronger for CO than for any isocyanide ligand. Of the
(41) Lauher, J. W.; Hoffman, R. J. Am. Chem. Soc. 1976, 98, 1729-
1741.

728 Organometallics, Vol. 25, No. 3, 2006
Shapiro et al.
Table 5. Geometrical Parameters (Å, deg) for Calculated and Observed Structures
free ligand complex
CdN
ligand
C-N
N-C
Cr-C
C-N
N-C
CNC
elongation
Neutral Complexes
1.855
CO
CNH
CNtBu
calc
calc
X-ray
calc
X-ray
calc
calc
calc (par)
X-ray
calc
calc (par)
X-ray
calc
X-ray
calc (C2H4)
calc (C2Me4)
calc (C2H4)
calc (C2Me4)
X-ray
(1.142)
1.185
1.145b
1.188
1.155c
1.190
1.191
(1.174)
1.232
1.214(4)
1.223
1.226(4)
1.238
1.211
1.232
1.195(7)
1.213
1.224
1.205(5)
1.231
0.032
0.047
0.069
0.035
0.048
0.048
0.020
1.851
1.873(3)
1.869
1.859(3)
1.842
1.875
1.849
1.892(7)
1.880
1.869
1.879(4)
1.852
1.861(5)
1.849
1.843
1.848
1.841
122.0
131.1(3)
136.5
123.1(3)
130.5
166.2
136.1
180
163.4
148.8
134.4(4)
135.0
128.6(5)
134.6
132.2
138.1
134.6
125.8(3)
126.4(3)
1.469
1.444
1.430
1.399
1.385
1.482(4)
1.474
1.462(4)
1.423
1.366
1.402
1.351(8)
1.373
1.392
1.411(5)
1.393
1.427(6)
1.401
CNTs
CNPh
CNXyl
1.160d
1.192
1.400
1.386
0.035
0.021
0.032
0.048
0.040
0.069
0.041
0.042
CNFc
1.157e
1.191
1.163f
1.192
1.389
1.375
1.393
1.382
CNPhNC
1.232(6)
1.233
1.234
1.227
1.229
1.402
1.395
1.397
1.415(4)
CNPhNC
(monocation)
CNFcNC
1.852(4)
1.862(4)
1.230(4)
1.224(4)
Cations
1.867
1.886
1.956(3)
1.931
1.891
CO
CNH
CNtBu
calc
calc
X-ray
calc
calc
calc
(1.142)
1.185
1.145b
1.188
1.190
1.191
(1.159)
1.198
1.155(4)
1.186
1.197
1.191
0.017
0.013
0.014
-0.002
0.007
147.5
173.9(4)
178.7
158.8
177.1
1.469
1.444
1.399
1.385
1.455(4)
1.450
1.401
CNTs
CNPh
1.915
1.381
0.000
calc (par)
X-ray
calc
calc (par)
calc
calc (C2H4)
calc (C2Me4)
X-raya
calc (C2H4)
calc (C2Me4)
1.914
1.933(5)
1.927
1.919
1.918
1.888
1.882
1.942(8)
1.899
1.896
1.192
1.174(5)
1.192
1.192
1.192
1.201
1.200
1.156(10)
1.197
1.199
1.384
1.396(5)
1.382
1.384
1.371
1.376
1.373
1.411(10)
1.376
1.375
178.2
178.8(4)
179.4
176.9
173.2
176.9
179.5
177.9(9)
179.3
179.5
CNXyl
0.014
0.000
0.000
0.001
0.009
0.008
-0.007
0.005
0.007
1.192
1.386
CNFc
1.191
1.192
1.375
1.382
CNC6H4NC
(monocation)
CNC6H4NC
1.163e
1.192
1.393
1.382
a Poor quality structure. b Tada, H.; Tozyo, T.; Shiro, M. J. Org. Chem. 1988, 53, 3366 (for a t-alkylNC). c Weener, J.-W.; Versleijen, J. P. G.; Meetsma,
A.; ten Hoeve, W.; van Leusen, A. M. Eur. J. Org. Chem. 1998, 1511, 1 (for a CNCH2-phosphonate). d Mathieson, T.; Schier, A.; Schmidbaur, H. J. Chem.
Soc., Dalton Trans. 2001, 1196. e Wrackmeyer, B.; Maisel, H. E.; Tok, O. L.; Milius,W.; Herberhold, M. Z. Anorg. Allg. Chem. 2004, 630, 2106-2109.
f Colapietro, M.; Domenicano, A.; Portalone, G.; Torrini, I.; Hargittai, I.; Schultz, G. J. Mol. Struct. 1984, 125, 19.
Table 6. Calculated Complexation Energies and Oxidation
Potentials
Summary and Conclusions
Ligand substitution on ansa-chromocene carbonyl precursors
is a general route to a variety of ansa-chromocene isocyanide
derivatives. We have yet to determine the mechanism for this
ligand substitution reaction; however, we suspect that the
substitution of CO by isocyanide ligands is associative in nature
for the following reasons: (a) Brintzinger and co-workers have
identified ansa-chromocene dicarbonyl complexes,36,42 showing
that ring slippage can occur to accommodate a second 2e-
donating ligand on the metal. Furthermore, this ring slippage is
catalyzed by small amounts of oxidants, indicating that one-
electron oxidation of the ansa-chromocene carbonyl complex
facilitates ring slippage;36 (b) it is difficult to remove CO from
the neutral and cationic complexes by either thermal or
photochemical means.43
Ecompl (kcal/mol)a
neutral cation
(0) (0)
Eox (mV)b
ligand
CO
CNH
CNtBu
CNPh
CNXyl
CNFc
CNTs
calc
obs
-880
-1106
-1413
-1215
-1269
-1208
(-1058)
-1128
3.54
10.21
3.96
6.85
5.23
0.62
2.91
-1.66
-2.07
-3.75
-2.11
-2.32
-3.47
-1154
-1181
-1153
-1058
-1096
CNC6H4NC
a Relative to the CO complexes. b For calculated data, Eox for CNTs
arbitrarily fixed at its experimental value.
isocyanide complexes studied, the tosyl-substituted derivative
is the best π-acceptor and binds most strongly.
There is fairly good agreement among the crystallographic,
IR, NMR, and electrochemical data with respect to the π-ac-
ceptor strength of the isocyanide ligands, giving the ordering
CNCH2Tos > FcNC ≈ t-BuNC > CNXyl.
In the cationic complexes, all ligands bind more strongly than
CO, but the differences between ligands are very small,
consistent with the interpretation that σ-donation dominates and
is relatively insensitive to electronic effects. The correlation
between calculated and experimental redox potentials (Figure
17) is fair, except for the t-BuNC derivative, for which the
calculated value is much too negative. At present we do not
have an explanation for this discrepancy.
Consistent with the experimental data, DFT calculations
indicate that CrCNR bonding in the neutral complexes is
dominated by π-back-donation. The amount of backbonding is
(42) Schaper, F.; Rentzsch, M.; Prosec, M.-H.; Rief, U.; Schmidt, K.;
Brintzinger, H. H. J. Organomet. Chem. 1997, 534, 67-69.

ansa-Chromocene Complexes
Organometallics, Vol. 25, No. 3, 2006 729
Figure 15. C-N vs Cr-C bond lengths (Å) for all calculated and observed neutral (left) and cationic (right) structures.
tetrahydrofuran (Fisher) was used as received for electrochemical
measurements.
NMR spectra were recorded on Bruker Avance 300MB (300
MHz 1H; 75.4 MHz, 13C; 78 MHz, 27Al) and a Bruker Avance 500
(500 MHz 1H; 125 MHz 13C; 470 MHz 19F; 202 MHz 31P)
spectrometers. 1H and 13C chemical shifts were referenced to
solvent, and 19F and 13P chemical shifts were referenced to external
standards (BF3(OEt2), δ 0; H3PO4(80% aq), δ 0).
Cyclic voltammetry measurements were performed on a BAS
CV50W voltammetric analyzer in a nitrogen-filled glovebox.
Potentials were measured, without IR compensation, against a
reference electrode consisting of a silver wire immersed in a solution
Figure 16. â-LUMO for model of cation 5+.
of AgNO3 (0.01 M) in acetonitrile, using a glassy carbon working
electrode and a platinum wire as the counter electrode. Measure-
ments were performed on THF solutions of the sample complex
(<1 mM) containing 0.3 M [NBu4][PF6] electrolyte. All potentials
are referenced to the ferrocene0/1+ couple (Ep ≈ 200 mV), which
was measured separately prior to each sample measurement.
IR spectra were obtained on an ATI Mattson Genesis Series FTIR
spectrophotometer and on a Bio Rad FT-IR spectrometer FTS 175C.
NIR spectra were obtained on a Perkin-Elmer 330 spectrophotom-
eter. The samples were measured as THF solutions of the complexes
under nitrogen in sealed 1 cm quartz cuvettes.
Magnetic susceptibilities were determined on either solid samples
of the complexes using a Johnson Matthey magnetic susceptibility
balance type MSB, model MK I, or on solution samples according
to the Evans NMR method.44 Molar magnetic susceptibilities were
corrected for diamagnetism using Pascal constants.45
Figure 17. Calculated vs observed oxidation potentials (triangular
Elemental analyses were determined by Desert Analytics (Tus-
con, AZ). In some cases the carbon values were 1-3% below
theoretical. In one case (compound 7b) addition of WO3 to the
sample prior to combustion improved the carbon value. Therefore,
the low carbon values could be partly due to incomplete combustion.
With the exception of paramagnetic [7b][PF6]2, H NMR spectra
have been deposited in the Supporting Information for compounds
whose analyses were out of range.
Complexes 1a, 1b, 2a, and 2b were prepared as described
previously.20,21 Isocyanoferrocene was prepared in six steps starting
from ferrocene according to literature preparations. Aminoferrocene
was prepared as described by Bildstein et al.46 This was converted
to isocyanoferrocene in a slightly modified version of the synthesis
described by Knox et al.31 A more efficient method for preparing
isocyanoferrocene from aminoferrocene was recently reported by
Holovics et al.30 Diisocyanoferrocene was prepared as described
data point is for t-Bu isocyanide derivative).
sensitive to the steric and electronic properties of the ligands,
but changes in backbonding are associated with relatively small
energy changes. In the 17e cationic species, back-donation is
greatly diminished, and σ-donation is the dominant interaction.
Since oxidation occurs from a chromocene orbital that does
not interact with any π orbital on the isocyanide ligand, there
is little communication between the oxidized Cr center and any
π system attached to the isocyano group. This provides a
rationale for the lack of communication between the metal
centers in the phenylene- and ferrocenylene-bridged systems.
1
Experimental Section
General Procedures. All manipulations were performed under
dinitrogen, argon, or vacuum using a combination of glovebox,
high-vacuum, and Schlenk techniques. HPLC grade solvents were
dried by passage over alumina and then stored in line pots from
which they were vacuum transferred from sodium/benzophenone.
Argon was purified by passage over an oxy tower BASF catalyst
(Aldrich) and 4A molecular sieves. Deuterated solvents were dried
over 4A molecular sieves and stored in the glovebox. HPLC grade
(43) Perrotin, P.; Shapiro, P. J., unreported results.
(44) Evans, D. F. J. Am. Chem. Soc. 1959, 92, 2003-2005.
(45) Ko¨nig, E.; Ko¨nig, G. K. Magnetic Properties of Coordination and
Organometallic Transition Metal Compounds; Hellwege, K.-H., Hellweg,
A. M., Eds.; Springer-Verlag: Berlin, 1981; Vol. 11.
(46) Bildstein, B.; Malaun, M.; Kopacka, H.; Wurst, K.; Mitterboeck,
M.; Ongania, K. H.; Opromeolla, G.; Zanello, P. Organometallics 1999,
18, 4325-4336.

730 Organometallics, Vol. 25, No. 3, 2006
Shapiro et al.
by van Leusen and Hessen.33 [(C5H5)2Fe][B(ArF)4] (ArF ) 3,5-
(CF3)C6H3) was prepared according to a literature procedure47 and
recrystallized from CH2Cl2 at -78 °C. 1,4-Phenylene diisocyanide
was purchased from Aldrich and used as received.
product. Separation of 5 from the side product was achieved by
extracting the solid with ca. 300 mL of Et2O. Compound 5
precipitated upon reducing the volume of the ethyl ether extract to
30 mL under vacuum, and 0.67 g (54% yield) was isolated after
filtration. Single crystals of 5 were obtained by slow evaporation
of a solution of the compound in benzene-d6 under a nitrogen
Preparation of [rac-Ph2H2C2(η5-C5H4)2}CrCN-t-Bu][B(C6F5)4]
([2b][B(C6F5)4)]. Solutions of [Ph3C][B(C6F5)4] (0.52 g, 0.56 mmol)
in 30 mL of toluene and 2b (0.23 g, 0.52 mmol) in 30 mL of toluene
were combined in a round-bottom flask in the glovebox and stirred
for 6 h. Upon standing for an hour unstirred, the reaction mixture
separated into two distinct layers, a yellow layer atop a red oil.
The yellow layer was removed with a pipet and the volatiles from
the red oil were removed under reduced pressure with gentle
heating, leaving a red solid, which was recrystallized from toluene
by storing the solution for several days at -30 °C (yield: 0.31 g,
53%). IR (Nujol mull): 2139, 2199 cm-1 (ν NC). Anal. Calcd for
C53H29CrBF20N: C, 56.72; H, 2.58; N, 1.25. Found: C, 57.07; H,
2.61; N, 0.90. µeff ) 1.8 µB (Evans NMR method).
1
atmosphere. H NMR (300 MHz, 298 K, C6D6): δ 7.13 (m, 4H,
C6H5), 7.07-7.00 (m, 6H, C6H5), 4.94 (m, 2H, (C5H4)2Cr), 4.54
(m, 2H, (C5H4)2Cr), 4.45 (m, 1H, CN(C5H4)Fe), 4.42 (m, 1H, NC-
(C5H4)Fe), 4.26 (m, 2H, (C5H4)2Cr), 4.17 (s, 5H, (C5H5)Fe), 3.90
(s, 2H, Ph2C2H2), 3.91 (m, 2H, NC(C5H4)Fe), 3.87 (m, 2H, (C5H4)2-
Cr). 13C{1H} NMR (125 MHz, 298 K, C6D6): δ 274.0 (NCCr),
126.7, 127.5, 128.3, 142.4 (C6H5), 82.0, 80.0, 78.6, 77.9 ((C5H4)2-
Cr), 69.7 ((C5H5)Fe), 64.0, 65.9 ((CNC5H4)Fe), 54.8 (Ph2C2H2).
IR (Nujol mull): 1829 cm-1 (ν NC). Anal. Calcd for C35H29CrFeN:
C, 73.56; H, 5.10; N, 2.45. Found: C, 73.95; H, 5.03; N, 2.52.
Preparation of [[rac-{Ph2H2C2(η5-C5H4)2}CrCN-(η5-C5H4)]-
(η5-C5H5)Fe][PF6] ([5][PF6]). Complex 5 was prepared in situ by
reacting 1a (800 mg, 2.1 mmol) with isocyanoferrocene (400 mg,
1.9 mmol) in 50 mL of THF. The solution was cooled to 0 °C, and
[(C5H5)2Fe][PF6] (626 mg, 1.9 mmol) was added via a solid
dispenser sidearm. This mixture was stirred for 30 min at 0 °C
under an atmosphere of argon. Cooling the solution to -78 °C
produced a brown precipitate, which was collected by filtration to
Preparation of {(CH3)4C2(η5-C5H4)2}CrCN(CH2-p-Tosyl) (3).
Diethyl ether (50 mL) was transferred under vacuum into a flask
containing 1a (0.75 g, 2.6 mmol) and CN(CH2-p-Tosyl) (0.51 g,
2.6 mmol) cooled at -78 °C. The reaction was allowed to warm
gradually to room temperature. At room temperature, a rapid
evolution of CO gas was observed. The reaction was stirred for an
additional hour after all gas evolution had ceased. All volatiles were
removed from the reaction product under reduced pressure, and
the red-brown powder was washed with 25 mL of petroleum ether.
Crystallization of the material from toluene at -30 °C afforded
1
afford 0.91 mg of product (68% yield). H NMR (300 MHz, 298
K, acetone-d6): 7.88, 7.08 (broad, C6H5), 4.76 (broad, CN(C5H4)-
Fe), 4.13 (broad, (C5H5)Fe), 3.27 (broad, CN(C5H4)Fe). 31P NMR
1
0.75 g of 3 (yield: 63%) as brown plates. H NMR (300 MHz,
1
(202 MHz, 298 K, acetone-d6): δ -142.7(hept., JPF ) 707 Hz).
3
3
298 K, C6D6): δ 8.01(d, JHH ) 7.8 Hz, 2H, C6H4), 6.75 (d, JHH
) 7.7 Hz, 2H, C6H4), 4.98 (m, 4H, C5H4), 4.53 (s, 2H, SCH2),
3.97 (m, 4H, C5H4), 1.83 (s, 3H, CH3C6H4S), 0.88 (s, 12H,
(CH3)4C2). 13C{1H} NMR (125 MHz, 298 K, C6D6): δ 290.6
(CrCN), 144.1, 136.1, 129.4 128.3 (C6H4), 112.7, 81.4, 76.5 (C5H4),
71.4 (NCCH2S), 44.3 (C2(CH3)4), 27.1 ((CH3)4C2), 21.0 (CH3C6H4).
IR (Nujol mull): 1752 cm-1; THF solution, 1774 cm-1 (ν CN).
Anal. Calcd for C25H29CrNO2S: C, 65.33; H, 6.37; N, 3.04.
Found: C, 63.03; H, 6.31; N, 2.32. Carbon was consistently low
upon reanalysis.
1
19F NMR, (470 MHz, 298 K, acetone-d6): δ -69.85 (d, JPF
)
707 Hz). IR (Nujol mull): 2118 cm-1 (ν NC). Anal. Calcd for
C35H29CrFeNPF6: C, 59.06; H, 4.42; N, 1.86. Found: C, 59.12; H,
4.35; N, 1.66. µeff ) 1.9 µB (magnetic susceptibility balance and
Evans NMR method).
Preparation of 1,4-[Me2C2(η5-C5H4)2}CrCN]2C6H4 (6). The
synthetic procedure is similar to the one described for 5. After
heating the 50 mL THF solution of reactants 1a (1.3 g, 4.6 mmol)
and 1,4-phenylene diisocyanide (295 mg, 2.3 mmol) at 40 °C for
20 min, the solution was concentrated to 25 mL and cooled to -78
°C to afford crystals of deep red 6 (0.98 g, 65% yield), which were
isolated by filtration. Single crystals of 6 were obtained by slow
Preparation of {(CH3)4C2(η5-C5H4)2}CrCN(Xyl) (4). The
synthetic procedure is identical to that described for 3. The reaction
between 1a (0.50 g, 1.7 mmol) and CNXyl (0.22 g, 1.7 mmol)
1
evaporation of a THF/C6D6 solution of the compound. H NMR
1
afforded 0.45 g of 4 (72% yield) as orange plates. H NMR (300
(500 MHz, 25 °C, C6D6): δ 7.53 (s, 4H, CNC6H4NC) 4.57 (m,
8H, (C5H4)2Cr), 3.99 (m, 8H, (C5H4)2Cr), 0.95 (s, 24H, C2(CH3)4).
13C NMR, (500 MHz, 25 °C, C6D6): δ 278.3 (NC), 124.7, 140.4
(CNC6H4NC), 76.2, 80.6, ((C5H4)2Cr), 44.5 (C2(CH3)4), 27.4 (C2-
(CH3)4). IR (Nujol mull): 1766 cm-1 (ν NC)). Anal. Calcd for
C40H44Cr2N2: C, 73.15; H, 6.75; N, 4.27. Found: C, 70.36; H, 6.37;
N, 3.81. Carbon was consistently low on reanalysis.
MHz, 298 K, C6D6): δ 6.90-6.80 (m, 3H, C6H3), 4.70 (m, 4H,
C5H4), 3.84 (m, 4H, C5H4), 2.20 (s, 6H, 2,6-(CH3)2C6H3), 0.87 (s,
12H, (CH3)4C2). 13C{1H} NMR (125 MHz, 298 K, C6D6): δ 239.7
(CrCN), 134.7, 132.4, 128.4, 123.1 (C6H3), 108.4, 80.3, 75.3 (C5H4),
44.2 (C2(CH3)4), 27.3 ((CH3)4C2), 18.5 ((CH3)2C6H3). IR (Nujol
mull): 2006 cm-1; (THF) 1980 cm-1 (ν CN). Anal. Calcd for
C25H29CrN: C, 75.92; H, 7.41; N, 3.53. Found: C, 74.76; H, 6.69;
N, 3.31. Carbon was consistently low on reanalysis.
Preparation of [{(CH3)4C2(η5-C5H4)2}CrCN(Xyl)][B(C6F5)4]
([4][B(C6F5)4)]). The synthesis is the same as that described above
for [2b+][B(C6F5)4)]. The reaction between [Ph3C][B(C6F5)4] (0.42
g, 0.46 mmol) and 4 (0.16 g, 0.40 mmol) in 30 mL of toluene
afforded 0.24 g (49% yield) of the product as a red crystalline solid.
IR (Nujol mull): 2119 cm-1 (ν NC). µeff ) 1.9 µB (Evans NMR
method). Anal. Calcd for C56H37CrBF20N: C, 54.75; H, 2.73; N,
1.30. Found: C, 55.27; H, 3.21; N, 1.31.
Preparation of [rac-{Ph2H2C2(η5-C5H4)2}CrCN-(η5-C5H4)]-
(η5-C5H5)Fe (5). The synthetic procedure is similar to the ones
above except that the reaction between 1a (0.99 g, 2.5 mmol) and
isocyanoferrocene (0.46 g, 2.2 mmol) was performed in 50 mL of
THF instead of Et2O, and the reaction mixture was heated at 40
°C for 20 min. Removal of the solvent under reduced pressure left
a reddish-brown solid consisting of 5 and a paramagnetic side
Preparation of [1,4-[Me2C2(η5-C5H4)2}CrCN]2C6H4][B(3,5-
(CF3)2C6H3)4]2 ([6][B(ArF)4]2). A mixture of complex 6 (0.145 g,
0.22 mmol) and [Cp2Fe][B(ArF)4] (0.462 g, 0.44 mmol) was
dissolved in 30 mL of THF at -78 °C. The reaction mixture was
allowed to warm to 0 °C and stirred for 45 min. The solvent of the
resulting dark brown solution was removed under vacuum, and the
residue was washed with warm benzene (4 × 15 mL) to remove
the ferrocene coproduct. The red-brown powder was dissolved in
methylene chloride, and the solution was filtered and then concen-
trated to 15 mL and stored for 20 h at -35 °C, forming large,
brown crystals, which were isolated by filtration and dried under
vaccum, affording 0.365 g of 62+ (70% yield). 1H NMR (300 MHz,
298 K, DMSO-d6): δ 7.71 (s, 8H, B(ArF)4 p-H), 7.60 (s, 16H,
B(ArF)4 o-H), -2.16 (bs, C2(CH3)4). 19F NMR (280 MHz, DMSO-
d6): δ -61.58 (s, CF3). IR (Nujol mull): 2088 cm-1 (ν CN). µeff
) 2.3 µB (magnetic susceptibility balance and Evans NMR method).
The molecule cocrystallized with 0.225 equiv of CH2Cl2. Anal.
Calcd for C104.225H68.500B2Cl0.500F48N2Cr2: C, 52.11; H, 2.87; N, 1.17.
Found: C, 51.72; H, 3.08; N, 1.21.
(47) Heinekey, D. M.; Radzewich, C. E. Organometallics 1988, 7, 51-
58.

ansa-Chromocene Complexes
Organometallics, Vol. 25, No. 3, 2006 731
Preparation of [Me2C2(η5-C5H4)2}CrCN-(η5-C5H4)]2Fe (7a).
The synthetic procedure is similar to the one described for 5. After
heating the 50 mL THF solution of reactants 1a (1.0 g, 3.4 mmol)
and diisocyanoferrocene (0.40 g, 1.7 mmol) at 40 °C for 20 min,
the solution was concentrated to 20 mL and 60 mL of hexane was
layered on top of the solution. The reddish-brown product was
crystallized by cooling this mixture at -78 °C, and 0.64 g (49%
yield) was collected by cold filtration. Single crystals of 7a were
obtained by slow evaporation of a solution of the compound in
C6D6. 1H NMR (500 MHz, 298 K, THF-d8): δ 4.76 (m, 8H, (C5H4)-
Cr), 4.30 (m, 4H, (C5H4)Fe), 4.04 (m, 4H, (C5H4)Fe) 3.92 (m, 8H,
(C5H4)Cr), 1.19 (s, 24H, C2(CH3)4). 13C{1H} NMR (125 MHz, 298
K, C6D6): δ 278.6 (NCCr), 80.8, 76.1, (C5H4)Cr), 65.2, 66.8,
((C5H4)Fe), 44.5 (C2(CH3)4), 27.5 (C2(CH3)4). IR (Nujol mull):
1751 cm-1 (ν NC). Anal. Calcd for C50H54Cr2FeN2: C, 69.1; H,
6.37; N, 3.63. Found: C, 68.72; H, 6.21; N, 3.49.
1471 frames); [2b][B(C6F5)4] (20 s, 2132 frames); 3 (20 s, 2132
frames); 4 (30 s, 1471 frames); [4][B(C6F5)4] (20 s, 1471 frames);
5 (20 s, 1471 frames); 6 (40 s, 1471 frames); 7a (5 s, 2132 frames).
The first 50 frames were re-collected at the end of data collection
to monitor for decay. Cell parameters were retrieved using
SMART49 software and refined using SAINTPlus50 on all observed
reflections. Data reduction and correction for Lp and decay were
performed using the SAINTPlus software. Absorption corrections
were applied using SADABS.51 The structure was solved by direct
methods and refined by least-squares method on F2 using the
SHELXTL program package.52 All non-hydrogen atoms were
refined anisotropically except for [2b][B(C6F5)4]. In this compound,
the ansa backbone and phenyl are disordered over two positions
(60% major fraction). The disordered carbon atoms were held
isotropic. No decomposition was observed during data collection.
Details of the data collection and refinement are given in Table
1.
Preparation of [rac-Ph2H2C2(η5-C5H4)2}CrCN-(η5-C5H4)]2Fe
(7b). The synthetic procedure is similar to the one described for 5.
After heating the 50 mL THF solution of reactants 1b (1.0 g, 2.6
mmol) and diisocyanoferrocene (0.30 g, 1.3 mmol) at 40 °C for 20
min, the volatiles were removed under reduced pressure and the
product was extracted from the reddish-brown residue with 300
mL of ethyl ether. The product was precipitated by concentrating
the ether solution to 30 mL and was collected by filtration (0.78 g,
Calculations. All calculations were carried out with the Turbo-
mole program53,54 coupled to the PQS Baker optimizer.55,56 Calcula-
tions on open-shell systems used the spin-unrestricted formalism.
Geometries were optimized without constraints at the bp8657,58
/
RIDFT59 level using the Turbomole SV(P) basis set53,60 on all atoms.
Improved energies were obtained from single-point calculations at
the b3-lyp level61-64 using the TZVP basis.53,65 Solvent corrections
were calculated at the b3-lyp/TZVP level using the COSMO
solvation model66 with a dielectric constant of 7.5 (THF). For these
relatively large systems, vibrational analyses (and hence thermal
corrections) were not feasible.
1
63% yield). H NMR (500 MHz, 298 K, THF-d8): δ 7.08, 7.16,
7.28 (m, 20H, C6H5), 5.30 (m, 4H, (C5H4)Cr), 4.75 (m, 4H, (C5H4)-
Cr), 4.43 (m, 1H, (C5H4)Fe), 4.41 (m, 2H, (C5H4)Fe), 4.38 (m, 1H,
(C5H4)Fe), 4.27 (m 4H, (C5H4)Cr), 4.20 (s, 4H, [Ph2C2H2]), 4.12
(m, 4H, (C5H4)Fe), 3.86 (m, 4H, (C5H4)Cr). 13C NMR, (125 MHz,
298 K, THF-d8): δ 276.2 (NCCr). 127.2, 128.5, 128.9, 143.7
(C6H5), 82.6, 82.5, 80.5, 79.1, 78.1, ((C5H4)Cr), 67.0 65.4, 65.3,
65.2, (C5H4)Fe), 55.3 (Ph2C2H2). IR (Nujol mull): 1763 cm-1 (ν
NC). Anal. Calcd for C60H48Cr2FeN2: C, 75.31; H, 5.06; N, 2.93.
Found: C, 73.95; H, 5.03; N, 2.52. Carbon was consistently low
on reanalysis.
CCDC 265602-265609 contain the supplementary crystal-
lographic data for this paper. These data can be obtained free of
charge via www.ccdc.cam.ac.uk/data_request/cif, by e-mailing
data_request@ccdc.cam.ac.uk, or by contacting The Cambridge
Crystallographic Data Centre, 12 Union Road, Cambridge CB2
1EZ, UK; fax: +44 1223 336033.
Preparation of [[rac-Ph2H2C2(η5-C5H4)2}CrCN-(η5-C5H4)]2Fe]-
[PF6]2 ([7b][PF6]2). The same procedure as described for [5][PF6]
was used. Complex 7b was formed in situ from 1b (0.90 g, 2.3
mmol) and diisocyanoferrocene (0.27 g,1.16 mmol) in 50 mL of
THF and reacted with [(C5H5)2Fe][PF6] (0.77 g, 2.3 mmol) at 0
°C. The product was precipitated from solution at -78 °C and
collected by filtration, affording 0.95 g of [7b][PF6]2 (65% yield).
1H NMR (300 MHz, 298 K, acetone-d6): δ 8.03 (broad, C6H5).
7.37 (broad, C6H5), 7.07 (broad, C6H5), 4.96 (broad, CN(C5H4)-
Fe), 3.13 (broad, CN(C5H4)Fe). 31P NMR, (202 MHz, 298 K,
acetone-d6): δ -142.2 hept., 1JPF ) 707 Hz). 19F NMR (470 MHz,
25 °C, acetone-d6): δ -69.83 (d, 1JPF ) 707 Hz). IR (Nujol mull):
Acknowledgment. The authors are grateful to the donors
of the Petroleum Research Fund, administered by the American
Chemical Society, the National Science Foundation (grant no.
(49) SMART: Bruker Molecular Analysis Research Tool; Bruker AXS:
Madison, WI, 2002.
(50) SAINTPlus: Data Reduction and Correction Program; Bruker
AXS: Madison, WI, 2001.
(51) SADABS: an empirical absorption correction program; Bruker AXS
Inc.: Madison, WI, 2001.
(52) Sheldrick, G. M. SHELXTL: Structure Determination Software
Suite; Bruker AXS Inc.: Madison, WI, 2001.
(53) Ahlrichs, R.; Ba¨r, M.; Baron, H.-P.; Bauernschmitt, R.; Bo¨cker, S.;
Ehrig, M.; Eichkorn, K.; Elliott, S.; Furche, F.; Haase, F.; Ha¨ser, M.; Ha¨ttig,
C.; Horn, H.; Huber, C.; Huniar, U.; Kattannek, M.; Ko¨hn, A.; Ko¨lmel, C.;
Kollwitz, M.; May, K.; Ochsenfeld, C.; O¨ hm, H.; Scha¨fer, A.; Schneider,
U.; Treutler, O.; Tsereteli, K.; Unterreiner, B.; von Arnim, M.; Weigend,
F.; Weis, P.; Weiss, H. Turbomole Version 5; Theoretical Chemistry Group,
University of Karlsruhe, January 2002.
2113 cm-1 (free diisocyanoferrocene), 2137 cm-1 (ν NC). µeff
)
2.6 µB (magnetic susceptibility balance and Evans NMR method).
Anal. Calcd for C60H48Cr2FeN2P2F12: C, 57.80; H, 3.88; N, 2.25.
Found: C, 54.98; H, 3.84; N, 2.21. Carbon was consistently low
on reanalysis. The IR band for free diisocyanoferrocene at 2113
cm-1 indicates that this ligand dissociates readily from the complex.
Partial ligand loss during purification could account for the low
carbon value, as could incomplete combustion.
(54) Treutler, O.; Ahlrichs, R. J. Chem. Phys. 1995, 102, 346-354.
(55) PQS version 2.4; Parallel Quantum Solutions: Fayetteville, AK,
2001 (the Baker optimizer is available separately from PQS upon request).
(56) Baker, J. J. Comput. Chem. 1986, 7, 385-395.
(57) Becke, A. D. Phys. ReV. A 1988, 38, 3098-3100.
(58) Perdew, J. P. Phys. ReV. B 1986, 33, 8822-8824.
(59) Eichkorn, K.; Weigend, F.; Treutler, O.; Ahlrichs, R. Theor. Chem.
Acc. 1997, 97, 119-124.
X-ray Structure Determinations. Crystals of each compound
were removed from the flask and covered with a layer of
hydrocarbon oil. A suitable crystal was selected from each, attached
to a glass fiber, and placed in the low-temperature nitrogen stream.48
Data for 2, [2b][B(C6F5)4], [4][B(C6F5)4], and 5 were collected at
203(2) K using a Bruker/Siemens SMART 1K and Siemens LT-
2A low-temperature device; 3, 4 at 203(2) K (LT-2A) and 6, 7b at
81(2) K using a Bruker APEX instrument and a Cryocool NeverIce
low-temperature device. Data were measured with Mo KR radiation
(λ ) 0.71073 Å) using omega scans of 0.3° per frame for 2 (40 s,
(60) Scha¨fer, A.; Horn, H.; Ahlrichs, R. J. Chem. Phys. 1992, 97, 2571-
2577.
(61) Lee, C.; Yang, W.; Parr, R. G. Phys. ReV. B 1988, 37, 785-789.
(62) Becke, A. D. J. Chem. Phys. 1993, 98, 1372-1377.
(63) Becke, A. D. J. Chem. Phys. 1993, 98, 5648-5652.
(64) All calculations were performed using the Turbomole functional
“b3-lyp”, which is not identical to the Gaussian “B3LYP” functional.
(65) Scha¨fer, A.; Huber, C.; Ahlrichs, R. J. Chem. Phys. 1994, 100,
5829-5835.
(66) Klamt, A.; Schu¨rmann, G. J. Chem. Soc., Perkin Trans. 2 1993,
799-805.
(48) Hope, H. Prog. Inorg. Chem. 1995, 41, 1-19.

732 Organometallics, Vol. 25, No. 3, 2006
Shapiro et al.
Supporting Information Available: CIF files with X-ray
crystallographic information for compounds 2a, [2b][B(C6F5)4], 3,
4, [4][B(C6F5)4], 5, 6, and 7a and 1H NMR spectra for compounds
3, 4, 6, and 7b. This material is available free of charge via the
Internet at http://pubs.acs.org.
CHE-9816730), and the Department of Energy (grant no. DE-
FG0298ER45709) for their generous financial support. The
establishment of a Single-Crystal X-ray Diffraction Laboratory
and the purchase of a 500 MHz NMR spectrometer were
supported by the M. J. Murdock Charitable Trust of Vancouver,
WA, the National Science Foundation, and the NSF Idaho
EPSCoR Program.
OM050710J