Dinuclear iridium complexes with unsupported Ir-Cl-Ir bridges

Article abstract of DOI:10.1016/j.ica.2012.04.014

Two unusual dinuclear iridium complexes are reported with unsupported bridging chloride ligands. The first comes from an aqueous solution of H ₂ClIr(PMe₃)₃ by the addition of KPF ₆. The second is formed in the r

Full text of DOI:10.1016/j.ica.2012.04.014

Inorganica Chimica Acta 390 (2012) 33–36
Contents lists available at SciVerse ScienceDirect
Inorganica Chimica Acta
journal homepage: www.elsevier.com/locate/ica
Dinuclear iridium complexes with unsupported Ir–Cl–Ir bridges
⇑
Joseph S. Merola , Trang Le Husebo, Henry E. Selnau
Department of Chemistry, Virginia Tech, Blacksburg, VA 24061, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
Two unusual dinuclear iridium complexes are reported with unsupported bridging chloride ligands. The
first comes from an aqueous solution of H2ClIr(PMe3)3 by the addition of KPF6.
Received 27 February 2012
Received in revised form 6 April 2012
Accepted 10 April 2012
Available online 19 April 2012
Keywords:
Iridium compounds
Bridging chloride
X-ray structures
The second is formed in the reaction between H(C6H5)ClIr(PMe3)3 and TlPF6 in dichloromethane.
Single crystal X-ray structures are provided for both
l
-chlorodiiridium complexes.
Ó 2012 Elsevier B.V. All rights reserved.
1. Introduction
[9]. In both cases, the characterization of these compounds as dinu-
clear complexes containing a single Ir–Cl–Ir came as a surprise gi-
ven the conditions of the reactions involved. In the following
sections we will describe the preparation of these compounds
and discuss their structures.
Bridging halides as structural features in transition metal di-
and polynuclear compounds are quite common. A search of the
Cambridge Crystallographic Database for structures that contain
any transition metals bridged by any halogen uncovered over
6800 structures [1]. In the great majority of cases, two metals are
bridged by two or three halide ligands, or a halide ligand in combi-
nation with another bridging ligand such as a bridging carbonyl or
a bridging phosphide. Compounds in which a single halide ion
bridges two metals without the support of another bridging ligand
or a metal–metal bond is very rare making up less than 1% of the
structures uncovered in the above-mentioned search.
2. Experimental
2.1. Syntheses
General: The water used in the following experiments was
deionized and degassed prior to use.
In the specific case of iridium chemistry, few examples of an
unsupported Ir–X–Ir moiety have been structurally characterized
[2–7]. In this paper, we report on two additions to the family of
compounds with unsupported Ir–Cl–Ir bridges. The first com-
2.1.1. Synthesis of 3
A
50.0 mL side-armed one-necked flask equipped with
a
magnetic stir bar and septum was charged with 0.300 g
a
(0.660 mmol) of (2) under nitrogen in the dry box. The flask was
then connected to a double manifold (vacuum/nitrogen) Schlenk
line and 14.0 mL of distilled water was added by syringe. Next,
0.134 g (0.730 mmol) of KPF6 was added quickly. White solids
precipitated out of the solution immediately. The solids were fil-
tered, washed three times with distilled water and dried in vacuo
to yield 0.250 g of [(Me3P)3(H)2Ir–Cl–Ir(H)2(PMe3)3]PF6, compound
3 (0.244 mmol, 37.0% yield).
pound, [{(Me3P)3IrHPh}2-
of the activation of aromatic C–H bonds with [Ir(COD)(PMe3)3]Cl
[8]. The second, [{(Me3P)3IrH2}2- -Cl]PF6, was isolated during our
l-Cl]PF6 was isolated during our studies
l
studies of the aqueous hydrogenation chemistry of IrH2Cl(PMe3)3
⇑
Corresponding author.
E-mail address: jmerola@vt.edu (J.S. Merola).
0020-1693/$ - see front matter Ó 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.ica.2012.04.014

34
J.S. Merola et al. / Inorganica Chimica Acta 390 (2012) 33–36
Anal. Calc. for C18H58ClF6Ir2P7: C, 21.08; H, 5.34. Found: C, 20.66;
H, 5.44%.
1H NMR (CD2Cl2): d ꢀ26.8 ppm (m, JH–P = 10.46 Hz, JH–H
=
4.9 Hz, 1H), ꢀ10.5 (dtd, JH–Ptrans = 133.4 Hz, JH–Pcis = 21.72 Hz,
JH–H = 4.9 Hz, 1H), 1.5 (d, JH–P = 7.9 Hz, 9H of cis PMe3), 1.63 (virtual
triplet, 18H of trans PMe3).
ð1Þ
31P NMR (CD2Cl2): d ꢀ44.75 (t, JP–P = 52.7 Hz, 1P of cis PMe3),
ꢀ40.5 (d, JP–P = 52.7 Hz, 2P of trans PMe3), ꢀ135.18 (septet,
JF–P = 710 Hz) 1P of PF6.
The dihydrido complex, 2, is formed in excellent yield by the
reaction between 1 and H2 (Eq. (2)). We found that 2 is very water
soluble and that aqueous solutions of 2 displayed two different sets
of hydride peaks both indicative of a meridional arrangement of
the trimethylphosphine ligands as well as a cis arrangement of
the hydrides. Further studies led to the conclusion that the two
species in aqueous solution were the starting chloride as well as
the cationic aquo complex in equilibrium as shown in Eq. (3). In
an attempt to isolate the theorized cationic aquo complex from
the solution, KPF6 was added to an aqueous solution of 2 and a
white precipitate immediately formed. The NMR spectra of the
precipitate are still consistent with a meridional arrangement of
phosphine ligands and a cis arrangement of the hydrides, but res-
onances for an aquo ligand were notably absent. NMR spectros-
copy offered no further information concerning the identity of
the sixth ligand in the iridium octahedral coordination spheres.
Single crystals suitable for X-ray crystallography were grown from
diglyme by slow diffusion of pentane. Diglyme as a crystallization
solvent was specifically chosen based on multiple indications in
the literature [16] and from our own experience [17,18] that aquo
complexes need a hydrogen bond acceptor in order to form stable
crystals.
2.1.2. Synthesis of 5
A 25 mL flask equipped with a stir bar and septum was charged
with (4) (200 mg, 0.375 mmol) under nitrogen in a dry box. The
flask was connected to a double manifold (vacuum/nitrogen)
Schlenk line. Dry dichloromethane (20 mL) was injected by syr-
inge. Thallium hexafluorophosphate (131.0 mg, 0.375 mmol) was
added as a solid against a counter-stream of nitrogen and the mix-
ture was stirred overnight. The thallium chloride was removed
using a cannula fitted with filter paper. The solvent was then re-
moved under vacuum. The crude product was redissolved in
dichloromethane (3 mL) and precipitated using pentane (10 mL).
A yield of 165 mg (0.14 mmol, 75% based on (4)) of a white crystal-
line solid of [(Me3P)3(Ph)HIr–Cl–Ir(Ph)H(PMe3)3]PF6, compound 5.
Anal. Calc. for C30H66Cl1Ir1P4F6: C, 40.37; H, 7.45; Cl, 3.97.
Found: C, 39.88; H, 7.66; Cl, 3.85%. 1H NMR (CD2Cl2):
d = ꢀ23.8 ppm (br s, 2H, Ir–H), d = 1.26 ppm (br s, 36H, PMe3), d =
1.62 ppm (br s, 18H, PMe3), d = 6.6–8.3 ppm (m, 10H aromatic).
31P NMR (CD2Cl2):d ꢀ42.3 (t, JP–P = 55 Hz, 1P of cis PMe3), ꢀ38.5
(d, JP–P = 55 Hz, 2P of trans PMe3), ꢀ139 (septet, JF–P = 700 Hz) 1P of
PF6.
2.2. Crystallographic studies
All crystallographic analyses were performed using a Bruker P4
diffractometer with the Bruker XSCANS software used for data collec-
tion and the Bruker version of SHELXTL plus for data workup and
structure solution and refinement. The program OLEX2 was used
for final generation of cif files and thermal ellipsoid plots. Table 1
summarizes experimental details and results for the single crystal
structures of 3 and 5.
ð2Þ
Table 1
Crystallographic data for compounds 3 and 5.
2.2.1. Crystals of compound 3 were recrystallized from a diglyme
solution by the slow diffusion of pentane vapor.
2.2.2. Crystals of compound 5 were grown by slow diffusion of
ether into a dichloromethane solution.
Compound
Formula
M
Crystal system
Space group
a (Å)
3
C
5
18H58ClIr2P7F6
C30H66ClIr2P7F6
1177.5
monoclinic
P2/1
10.323(2)
19.337(4)
12.204(2)
90
110.09(3)
90
1025.28
triclinic
P1
ꢀ
3. Results and discussion
9.812(2)
14.079(3)
14.816(3)
82.53(3)
71.30(3)
89.84(3)
1920.6(7)
2
b (Å)
c (Å)
For some time, we have been exploring the chemistry of
[Ir(COD)(PMe3)3]Cl [10], 1, especially with regard to oxidative
addition reactions of E–H bonds where E@H [9], B [11], C [8,12],
N [13] and O [14,15]. While oxidative addition reactions were ob-
served for each class of element compounds studied, the reaction
chemistry of the resulting E–Ir–H complexes varied greatly for
each element (Eq. (1)). Nevertheless, one common theme for reac-
tion chemistry of the octahedral iridium(III) compounds generated
in these oxidative addition reactions is the need for chloride disso-
ciation for any further reactions to take place. In one case (E@B),
the chloride dissociates spontaneously in polar solvents [11]. In
the case of E@H, dissociation occurs spontaneously only in water
[9]. For the case of E@C, chloride dissociation can be driven by
the addition of a chloride precipitator such as thallium(I) hexa-
fluorophosphate [12]. Attempts to preform and isolate the unsatu-
rated iridium(III) species, perhaps as an aquo or dichloromethane
solvate, resulted instead in dinuclear complexes with unsupported
Ir–Cl–Ir bridges.
a
(°)
b (°)
c
(°)
V (Å3)
2287.9(8)
2
Z
Dimensions (mm)
0.4 ꢁ 0.2 ꢁ 0.2
0.3 ꢁ 0.2 ꢁ 0.2
q
l
(g cmꢀ3
)
1.773
1.709
(Mo K
a)
7.321
6.133
998
295
4411
4169, 0.019
0 6 h 6 12
0 6 k 6 23
ꢀ14 6 l 6 13
flack 0.080(12)
99.8 (50)
1.026
F(000)
992
T (K)
295
Reflections (total)
Reflections (unique), Rint
Index ranges
5238
4895, 0.022
ꢀ10 6 h 6 0
ꢀ15 6 k 6 15
ꢀ15 6 l 6 15
n/a
Absolute Config
Completeness to 2h
97.4 (45)
1.107
Goodness-of-fit (GOF) on F2
R1, wR2 [I > 2
R1, wR2 (all data)
r
(I)]
0.0510, 0.1253
0.0714, 0.1389
0.039, 0.073
0.053, 0.078

J.S. Merola et al. / Inorganica Chimica Acta 390 (2012) 33–36
35
Table 2
Bond lengths and angle comparisons for compounds 2 and 3.
Compound
2 CCDC: 868870
3 CCDC: 868869
Ir–P1
Ir–P2
Ir–P3
Ir–Cl
P–Ir–P (cis)
P–Ir–P (trans)
Ir1–Cl–Ir2
2.291(2) Å
2.319(2) Å
2.287(2) Å
2.498(2) Å
98.8(1)°, 97.8(1)°
163.3(1)°
n/a
2.295(5) Å
2.340(6) Å
2.292(5) Å
2.563(6) Å, 2.578(5) Å
97.9(2)°, 97.3(3)°, 97.3(3)°, 97.7(3)°
161.9(2)°, 161.9(2)°
139.6(3)°
Fig. 1. Thermal ellipsoid plot of the
l-chloroiridium cation, 3. Hydrogen atoms
omitted for clarity.
ð3Þ
Instead of a cationic aquo complex, the precipitate, compound
3, turns out to be a chloro-bridged dinuclear iridium species and
the thermal ellipsoid plot of the structure of 3 is shown in Fig 1.
We compare several structural features of 3 with the same features
of the parent compound 2, the subject of a separate paper [19] in
Table 2.
Fig. 2. Thermal ellipsoid plot for the l-chlorodiiridium cation, 5. Hydrogen atoms
and phosphorus methyl carbons omitted for clarity.
ð4Þ
The only significant difference is the not surprising elongation
of the Ir–Cl bond distances in the dinuclear complex, 3, since both
iridium centers are competing for the chlorine’s electron density.
While it is a truism not to over-interpret the isolation of a crystal-
line compound from a chemical reaction, it is nevertheless surpris-
ing that the aquo complex was not isolated. We, and others, have
reported on aquo compounds of the later 2nd and 3rd row metals
and there is nothing inherently unstable about them. There are
numerous cases in the literature of structurally characterized aquo
complexes of iridium and the high concentration of water would
seem to favor its formation. Apparently, the dinuclear complex is
favored by its insolubility and the equilibrium to it is driven by
its precipitation.
not be an unexpected outcome. However, we again discovered that
the formation of a dinuclear Ir–Cl–Ir complex, 5, had occurred (Eq.
(6)).
O
ð5Þ
The phenyl hydrido iridium complex is formed via the oxidation
addition reaction of benzene with compound 1 (Eq. (5)) [8]. The
only isomer formed in the reaction appears to be one with a merid-
ional arrangement of the PMe3 groups, the phenyl and the hydro-
gen cis to each other with the hydrogen trans to chlorine. We have
reported on its reaction chemistry with alkynes previously and all
reactions require the removal of the chloride ligand with TlPF6
[12,20]. Typically, that reaction was carried out with the alkyne
present when the TlPF6 was added so that the reactant was avail-
able immediately upon formation of the coordinatively unsatu-
rated cation. In an attempt to pre-form the cation, perhaps as a
dichloromethane solvate, compound 4 was treated with TlPF6 in
dichloromethane. Reports of isolable chloromethane [21], dichloro-
methane and other haloalkane [22] complexes of iridium
suggested that a dichloromethane complex in this case would
ð6Þ
The crystal structure of 5 is shown in Fig. 2 and Table 3 com-
pares select bond distances and angles for compounds 4 and 5.
The Ir–Cl bond for 4 is already somewhat lengthened compared
with the dihydride, 2, and the dinuclear complex exhibits further
lengthening of the Ir–Cl bonds in the Ir–Cl–Ir bridge as the dinucle-
ar dihydride.
Given that there are higher concentrations of other ligands
known to be able to bond to an iridium(III) center available in
every case where Ir–Cl–Ir bridges are formed, the formation of
these dinuclear compounds is quite interesting. There are a few
other cases of Ir–Cl–Ir bridges reported in the literature and it is

36
J.S. Merola et al. / Inorganica Chimica Acta 390 (2012) 33–36
Table 3
4. Conclusion
Bond length and angle comparisons for compounds 4 and 5.
Compound
4 CCDC: 626064
5 CCDC: 868870
Based upon the data presented in this paper as well as that in
the literature, the bonding to unsupported bridging chlorides can
display a wide range of Ir–Cl bond distances (2.427–2.655 Å) and
Ir–Cl–Ir bond angles (113.4–173.3°). This is best interpreted as
the metal–chloride bonding very non-directional and determined
in the main by steric constraints about the metal systems. This is
clearest in the compounds presented here where the substitution
of a hydride by a phenyl ring opens up the Ir–Cl–Ir bond angle
by over 33°.
Ir–P1
Ir–P2
Ir–P3
Ir–Cl
P–Ir–P (cis)
P–Ir–P (trans)
Ir1–Cl–Ir2
2.314(3) Å
2.342(2) Å
2.317(3) Å
2.508(3) Å
94.10(9)°, 94.33(9)°
167.4(1)°
2.344(5) Å, 2.327(6) Å
2.323(7) Å, 2.345(5) Å
2.319(5) Å, 2.337(6) Å
2.6542(4) Å, 2.655(4) Å
96.4()°, 95.1(2)°, 97.3(2)°, 95.2(2)°
166.4(2)°, 166.1(2)°
n/a
173.3(2)°
Acknowledgment
Table 4
Ir–Cl and Ir–Cl–Ir comparisons for unsupported chloride bridges.
We gratefully acknowledge support of this work by the National
Science Foundation (CHE-9214027).
Compound
Ir1–Cl (Å)
Ir2–Cl (Å)
Ir1–Cl–Ir2 (°)
Goldman
2.6058(7)
2.543(2)
2.637(2)
2.419(2)
2.535(3)
2.5600(2)
2.563(6)
2.654(4)
2.6121(6)
2.521(3)
2.646(1)
2.427(3)
2.544(3)
2.5600(2)
2.578(5)
2.655(4)
136.78(3)
139.2(1)
113.44(6)
123.8(1)
130.3(1)
169.39(7)
139.6(3)
173.3(2)
Paredes CCDC ID: 691413
Tilley CCDC ID: 164006
Ikariya CCDC ID: 743360
Garralda CCDC ID: 606708
Roddick CCDC ID: 836506
HHIr CCDC ID: 868869
PhHIr CCDCID: 868870
Appendix A. Supplementary material
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.ica.2012.04.014.
References
[1] F. Allen, Acta Crystallogr., Sect. B 58 (2002) 380.
[2] S. Kundu, Y. Choliy, G. Zhuo, R. Ahuja, T.J. Emge, R. Warmuth, M. Brookhart, K.
Krogh-Jespersen, A.S. Goldman, Organometallics 28 (2009) 5432.
[3] P. Paredes, J. Diez, M.P. Gamasa, J. Organomet. Chem. 693 (2008) 3681.
[4] M. Stradiotto, K.L. Fujdala, T.D. Tilley, Chem. Commun. (2001) 1200.
[5] Y. Kashiwame, S. Kuwata, T. Ikariya, Chemistry – Eur. J. 16 (2010) 766.
[6] F. Acha, M.A. Garralda, R. Hernández, L. Ibarlucea, E. Pinilla, M.R. Torres, M.
Zarandona, Eur. J. Inorg. Chem. 2006 (2006) 3893.
[7] J.J. Adams, A. Lau, N. Arulsamy, D.M. Roddick, Organometallics 30 (2011) 689.
[8] H.E. Selnau, J.S. Merola, Organometallics 12 (1993) 1583.
[9] T.X. Le, J.S. Merola, Organometallics 12 (1993) 3798.
[10] J.F. Frazier, J.S. Merola, Polyhedron 11 (1992) 2917.
[11] J.R. Knorr, J.S. Merola, Organometallics 9 (1990) 3008.
[12] J.S. Merola, Organometallics 8 (1989) 2975.
[13] F.T. Ladipo, J.S. Merola, Inorg. Chem. 29 (1990) 4172.
[14] F.T. Ladipo, M. Kooti, J.S. Merola, Inorg. Chem. 32 (1993) 1681.
[15] F.T. Ladipo, J.S. Merola, Inorg. Chem. 32 (1993) 5201.
[16] G. Ferraris, M. Franchini-Angela, Acta Crystallogr., Sect. B 28 (1972) 3572.
[17] P.M. Boyer, C.P. Roy, J.M. Bielski, J.S. Merola, Inorg. Chim. Acta 245 (1996) 7.
[18] J.J. Bergmeister, B.E. Hanson, J.S. Merola, Inorg. Chem. 29 (1990) 4831.
[19] J.S. Merola, T.L. Husebo, K.E. Matthews, Organometallics, in press, http://
dx.doi.org/10.1021/om300140f.
[20] H.E. Selnau, J.S. Merola, Organometallics 12 (1993) 3800.
[21] D.M. Tellers, R.G. Bergman, J. Am. Chem. Soc. 123 (2001) 11508.
[22] R.J. Kulawiec, R.H. Crabtree, Coord. Chem. Rev. 99 (1990) 89.
instructive to view the structural data from those. Table 4 summa-
rizes Ir–Cl bond distances and Ir–Cl–Ir bond angles for those previ-
ously structurally characterized.
The two compounds characterized in this study display Ir–Cl
bond distances (especially the Ph compound) that are at the ex-
treme end of Ir–Cl distances as well as at the extreme end of Ir–
Cl–Ir bond angles. The Ir–Cl bond distances most affected by the
nature of the trans ligand with trans hydride or trans phosphine
inducing the greatest elongation. An examination of structurally
characterized iridium(III) complexes with Ir–H and Ir–Cl bonds
shows Ir–Cl bond distances in the range of 2.3–2.4 Å in most cases
but elongating to 2.5–2.6 Å when trans to H or P ligands. Attribut-
ing any significance to the Cl–Ir–Cl bonds in this study is problem-
atic given that steric constraints can result not only from
intramolecular contacts within the molecule but also intermolecu-
lar contacts within the crystal lattice. Any possibility of compari-
son is particularly problematic when there are solvent molecules
in the lattice that could profoundly affect that geometry.