Synthesis and coordination chemistry of organoiridium complexes supported by an anionic tridentate ligand

Article abstract of DOI:10.1039/c1dt10037h

Treatment of deprotonated N-(dimethylaminoethyl)-2-diphenylphosphinoaniline with bis(cyclooctene)iridium chloride dimer affords a thermally stable iridium(i) olefin complex. Infrared analysis of the corresponding monocarbonyl iridium(i) compound indicates

Full text of DOI:10.1039/c1dt10037h

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Dalton
Transactions
Cite this: Dalton Trans., 2011, 40, 4300
www.rsc.org/dalton
PAPER
Synthesis and coordination chemistry of organoiridium complexes supported
by an anionic tridentate ligand†
Nadia G. Leonard, Paul G. Williard and Wesley H. Bernskoetter*
Received 10th January 2011, Accepted 15th February 2011
DOI: 10.1039/c1dt10037h
Treatment of deprotonated N-(dimethylaminoethyl)-2-diphenylphosphinoaniline with
bis(cyclooctene)iridium chloride dimer affords a thermally stable iridium(I) olefin complex. Infrared
analysis of the corresponding monocarbonyl iridium(I) compound indicates a relatively electron rich
metal center. Reaction of the iridium(I) cyclooctene complex with iodomethane effects oxidation of the
metal yielding a five-coordinate iridium(III) methyl iodide complex which reversibly coordinates
tetrahydrofuran. X-ray crystallography confirms coordination of ether to the iridium(III) methyl iodide
complex and NMR spectroscopic experiments establish an equilibrium constant of 1.66(9) M for
tetrahydrofuran binding. A five-coordinate iridium(III) dimethyl complex has also been prepared and
characterized by X-ray diffraction. Hydrogenolysis of the dialkyl species permits identification of a
short-lived classical iridium(III) dihydride complex.
intriguing features in relation to the vast array of tridentate
Introduction
ligands previously reported,1,2 including an asymmetric combi-
Platinum group metal complexes supported by tridentate ligands
nation of ethane and aryl bridging groups linking a pair of hard
have demonstrated a rich chemistry, promoting such difficult
amino/amide nitrogens to a soft diphenyl phosphine group. In
transformations as C–H bond oxidative addition, hydrodesulfu-
addition, the small dimethyl amino substituents may be expected
rization, carbon dioxide activation, and C–C bond cleavage.1 A
to present a more sterically vulnerable metal center than the bulky
i
tBu or Pr groups commonly used to protect other tridentate
transition metal complexes.1,2 These characteristics motivated our
efforts to explore the coordination chemistry of the PNN ligand
on a platinum group transition metal and assess the ligand’s
ability to support reactive transition metal alkyl and hydride
species. Herein we present the synthesis and characterization of
a series of organoiridium complexes bearing the PNN ligand,
and investigations into the coordination chemistry and stability
of these compounds.
combination of hard and soft donor sites in these chelating ligands
has become a common motif which is believed to contribute to
their desirable reactivity by attenuating the electronic properties
of the metal and, in some cases, offering a hemilabile site for
substrate binding.2 For example, Milstein and coworkers have
demonstrated a range of ruthenium catalyzed transformations
of alcohols which rely on both a reversible dearomatization
of the chelate ligand and the dechelation of a hard amine or
pyridine donor trans to a soft pendant phosphine ligand.3 The
reactivity of tridentate supported transition metal complexes may
also be influenced by the rigidity of the hydrocarbon groups
which often link the hard and soft donor sites.4 Ligands bearing
more geometrically restricted aryl bridges typically enforce tight
binding of the donor sites, while those with ethane bridges
permit more facile dissociation or distortion.5 These attributes can
significantly impact the coordination chemistry of the transition
metal complexes bearing tridentate ligands and, in turn, their
ability to mediate desirable catalytic processes.
Experimental
General considerations
All manipulations were carried out using standard vacuum,
Schlenk, cannula or glovebox techniques. Argon, nitrogen, and
hydrogen were purchased from Corp Brothers and used as
received. N-(Dimethylaminoethyl)-2-diphenylphosphinoaniline
Recently Lee and Liang have described the prepa-
ration of the tridentate ligand N-(dimethylaminoethyl)-2-
diphenylphosphinoaniline (PNN) (Fig. 1) and its coordination
to dialkyl aluminium centers.6 The PNN ligand offers several
(PNN)6 and [Ir(COE)2Cl]2 were prepared according to literature
7
procedures. All other chemicals were purchased from Aldrich,
VWR, Strem, Fisher Scientific or Cambridge Isotope Laborato-
˚
ries. Volatile, liquid chemicals were dried over 4 A molecular sieves
and distilled prior to use. Solvents were dried and deoxygenated
using literature procedures.8
1H, 13C, and 31P NMR spectra were recorded on Bruker DRX
400 Avance and 300 Avance MHz spectrometers. 1H and 13C
Department of Chemistry, Brown University, Providence, RI 02912, USA.
E-mail: wb36@brown.edu
† CCDC reference numbers 806967 and 806968. For crystallographic data
in CIF or other electronic format see DOI: 10.1039/c1dt10037h
4300 | Dalton Trans., 2011, 40, 4300–4306
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
Fig. 1 Lee and Liang’s preparation of the PNN ligand.
Table 1 Crystal structure refinement data for (PNN)Ir(CH3)I(THF) and
d, 2H, COE-CH), 2.37 (t, 2H, N-CH2), 3.03 (t, 2H, N–CH2), 6.44
(m, ArH), 8.13 (m, ArH). 31P { H} NMR (benzene-d6): d 15.21.
(PNN)Ir(CH3)2
1
1
13C { H} NMR (benzene-d6): d 27.65 (COE-CH2), 32.47 (COE-
Compound
(PNN)Ir(CH3)I(THF)
(PNN)Ir(CH3)2
CH2), 33.02 (COE-CH2), 47.35 (N–CH3), 47.43 (COE-CH), 49.34
(N–CH2), 69.39 (N–CH2), 108.66, 115.49, 130.00, 131.10, 131.68,
132.34, 132.72, 134.80, 135.45, 135.95 (Ar).
Formula
C27H35IIrN2OP
753.67
1.915
C24H29IrN2P
568.70
1.726
M
D
calc/g cm-3
Preparation of {(j3-N,N,P)Me2N(CH2CH2)N(o-C6H4PPh2)}-
IrCO (1-CO). A 50 mL heavy walled reaction vessel was charged
with 0.035 g (0.054 mmol) of 1-COE dissolved in 10 mL toluene.
Crystal system
Space group
Monoclinic
P21/n
10.255(5)
16.564(8)
15.548(8)
90
98.213(6)
90
2614(2)
4
Orthorhombic
Pbca
15.389(5)
16.679(5)
17.057(5)
90
90
90
4378(2)
8
296
Mo Ka 0.71073
29.040
2232
0.0477
0.1315
˚
a/A
˚
b/A
◦
˚
The solution was frozen at -196 C, and one atmosphere of CO
c/A
a/◦
b/◦
g /◦
gas was added to the reaction vessel on a high vacuum line. The
reaction was warmed to ambient temperature and left to stir for
25 min during which time the color of the solution changed to
yellow. The volatiles were removed in vacuo and the resulting
residue washed with pentane to afford 0.023 g (74%) of 1-CO as
a yellow powder. 1H NMR (benzene-d6): d 2.26 (bs, 2H, N-CH2),
2.51 (bs, 5 Hz, 6H, N–CH3), 3.00 (bs, 2H, N–CH2), 6.42 (m, ArH),
6.52 (m, ArH), 6.99 (m, ArH), 7.01 (m, ArH), 7.26 (m, ArH),
3
˚
V/A
Z
T/K
173
Radiation
Mo Ka 0.71073
23.250
1456
0.0814
◦
2qmax
/
F(000)
R1 (I > 2s(I))
wR2 (all data)
0.2393
1
7.98 (m, ArH). 31P { H} NMR (benzene-d6): 26.04. 13C NMR
(benzene-d6): 49.91 (N–CH2), 52.76 (N–CH3), 68.01 (N–CH2),
chemical shifts are referenced to residual solvent signals; 31P
chemical shifts are referenced to an external standard of H3PO4.
Probe temperatures were calibrated using ethylene glycol and
methanol as previously described.9◦ Unless otherwise noted, all
NMR spectra were recorded at 23 C. IR spectra were recorded
on Jasco 4100 FTIR and Metler Toledo React IR spectrometers.
Elemental analyses were performed at Robertson Microlit Labo-
ratories, Inc., in Madison, NJ. X-ray crystallographic data were
collected on a Bruker Smart Apex I diffractometer. The structures
were solved from direct methods and Fourier syntheses and
refined by full-matrix least-squares procedures with anisotropic
thermal parameters for all non-hydrogen atoms. Crystallographic
calculations were carried out using SHELXTL. A summary of
crystallographic data is presented in Table 1.
110.30, 114.72, 120.86, 130.48, 133.26, 133.75, 134.60, 135.57,
136.17, 166.50 (Ar), 188.65 (CO). IR (KBr): uC 1916 cm-1. IR
O
(benzene): uC 1922 cm-1. The presence of only one carbonyl
O
ligand was also confirmed by 13CO labeling.
Preparation of {(j3-N,N,P)Me2N(CH2CH2)N(o-C6H4PPh2)}-
Ir(PPh3) (1-PPh3). A heavy walled glass reaction vessel was
charged with 0.050 g (0.077 mmol) of 1-COE, 0.018 g (0.069 mmol)
of triphenylphosphine and approximately 10 mL of toluene. The
orange solution was heated in the sealed tube at 80 ◦C for five day.
The volatiles were then removed in vacuo resulting in a red-
orange residue which was washed with pentane and extracted into
tetrahydrofuran. The solution was concentrated to approximately
2 mL and chilled to -35 ◦C overnight to afford 0.048 g (77%) of
1-PPh3 as a red-orange microcrystalline powder. Anal. Calcd. for
C40H39IrP2N2: C, 59.91; H, 4.90; N, 3.49. Found: C; 59.66; H, 4.83;
N, 3.27. 1H NMR (benzene-d6): d 2.00 (br, 6H, N–CH3), 2.55 (t,
2H, N–CH2), 3.23 (t, 2H, N–CH2), 6.40 (m, 1H ArH), 6.59 (m,
1H ArH), 6.84–7.01 (m, 16H, ArH), 7.27 (m, 1H ArH), 7.64–7.75
Preparation of {(j3-N,N,P)Me2N(CH2CH2)N(o-C6H4PPh2)}-
Ir(COE) (1-COE). A 20 mL scintillation vial was charged with
0.145 g (0.161 mmol) of [Ir(COE)2Cl]2 and 10 mL toluene.
To this orange solution, 0.114 g (0.322 mmol) of PNNLi
in diethyl ether was added (generated from mixing 1 eq. of
N-(dimethylaminoethyl)-2-diphenylphosphinoaniline and 1eq. n-
butyl lithium). The resulting orange-brown reaction mixture was
stirred for 2 days and subsequently filtered and washed with 5 mL
of toluene. The filtrate was collected into a 20 mL scintillation
vial and concentrated to a volume of approximately 5 mL toluene.
Then 2 mL of pent◦ane was slowly layered on top of the solution
and chilled at -35 C to afford 0.113 g (54%) of orange powder
identified as 1-COE. Anal. Calcd. for C30H38IrPN2: C, 55.45;
1
(m, 10H ArH). 31P { H} NMR (benzene-d6): d 22.62 (d, 14.6 Hz),
1
25.15 (d, 14.6 Hz) 13C { H} NMR (benzene-d6): d 48.76 (N–CH2),
50.62 (N–CH3), 69.88 (N–CH2), 108.07, 112.05, 126.78, 127.85,
128.15, 131.46, 133.13, 133.31, 134.71, 134.21 (Ar). Four Ar peaks
could not be located or discerned from other signals.
Preparation of {(j3-N,N,P)Me2N(CH2CH2)N(o-C6H4PPh2)}-
Ir(CH3)I (1-MeI). A round bottom flask fitted with a needle
valve was charged with 0.103 g (0.159 mmol) of 1-COE and 15 mL
of toluene. On a vacuum line, the solution was frozen at -196 ◦C
and 38 Torr (0.190 mmol) of methyl iodide in a 101 mL calibrated
gas bulb was admitted to the vessel. The solution was warmed to
room temperature and its color subsequently changed from orange
1
H, 5.89; N, 4.31. Found: C, 55.61; H, 6.13; N, 4.08. H NMR
(benzene-d6): d 0.99 (m, 4H, COE-CH2), 1.51 (m, 4H, COE-CH2),
1.74 (m, 4H, COE-CH2), 1.94 (d, 6 Hz, 3H, N-CH3), 2.20 (broad
This journal is
The Royal Society of Chemistry 2011
Dalton Trans., 2011, 40, 4300–4306 | 4301
©

View Article Online
to deep green. The volatiles were removed in vacuo after one hour
and the solid residue washed with pentane to afford 0.0667 g (62%)
of 1-MeI as a deep green powder. Anal. Calcd. for C23H27IrPN2I:
C, 40.53; H, 3.99; N, 4.11. Found: C, 40.80; H, 3.85; N, 3.83. 1H
NMR (benzene-d6): d 1.64 (d, 11 Hz, 3H, IrCH3), 2.10 (d, 8 Hz,
3H, NCH3), 2.68 (d, 8 Hz, 3H, NCH3), 1.82 (m, 1H, N–CH2),
2.35 (m, 1H, N–CH2), 2.85 (m, 1H, N–CH2), 3.01 (m, 1H, N–
CH2), 6.49–6.57 (m, ArH), 7.30 (m, ArH), 7.93 (m, ArH), 8.11
with 0.023 g (0.040 mmol) of 1-Me2 dissolved in benzene-d6. On
a high vacuum line, the sample was frozen at -196 ◦C and one
atmosphere of H2 gas was admitted to the tube. Upon warming
the sample, an immediate color change from dark purple to yellow
occurred. The product was identified by NMR spectroscopy
as 1-H2 formed in virtually quantitative yield. Removal of the
excess dihydrogen following hydrogenolysis generally produced
sharper signals in the 1H NMR spectrum. 1H NMR (benzene-d6):
d -22.76 (d, 16 Hz, 2H, Ir–H), 2.26 (t, 2H, N–CH2), 2.53 (d,
5 Hz, 6H, N–CH3), 3.19 (t, 2H, N–CH2), 6.58 (m, ArH), 6.98
1
1
(m, ArH). 31P { H} NMR (benzene-d6): d 8.53. 13C { H} NMR
(benzene-d6): d -29.30 (Ir–CH3), 44.52, 49.10 (N–CH3), 48.56,
66.34 (N–CH2), 111.63, 114.90, 126.03, 129.67, 130.19, 130.49,
130.77, 131.04, 132.85, 134.38, 135.05, 135.14, 136.09, 136.69 (Ar).
1
(m, ArH), 7.94 (m, ArH). 31P { H} NMR (benzene-d6): d 21.35.
1
13C { H} NMR (benzene-d6): d 51.93 (N–CH2), 54.05 (N–CH3),
69.12 (N–CH2), 110.56, 116.82, 130.06, 132.66, 133.90, 134.61,
Procedure for determination of equilibrium binding constant
of tetrahydrofuran to 1-MeI. In a representative experiment,
a J. Young NMR tube was charged with 414 mg of a 0.032 M
solution of 1-MeI in benzene-d6 and a ferrocene capillary of known
concentration. Then 20 equiv. of tetrahydrofuran were added via
calibrated gas bulb ◦on a high vacuum line at -196 ◦C. The sample
136.13 (Ar). Three Ar peaks could not be located.
Results and discussion
Coordination
of
N-(dimethylaminoethyl)-2-diphenylphos-
phinoaniline to iridium was accomplished by initial deprotonation
of the aniline N–H bond with n-butyllithium followed by treatment
with bis(cyclooctene)iridium chloride dimer. After stirring for
two days at ambient temperature, recrystallization from toluene
at -35 ◦C afforded (PNN)IrCOE (1-COE; COE = cyclooctene) in
moderate yield (eqn (1)). The orange material was characterized
by multinuclear 1D and 2D NMR spectroscopy as well as
combustion analysis. The 1H NMR spectrum of 1-COE revealed
the number of peaks expected for a Cs symmetric complex,
including a broad peak at 2.22 ppm and a doublet at 1.94 ppm
(4JP–H = 6.0 Hz) assigned to the vinylic and NMe2 proton,
respectively. Additionally two triplet resonances observed at 2.37
and 3.03 ppm were assigned as the methylene sites bridging the
nitrogen atoms along the ligand backbone. These assignments, as
well as those listed in the Experimental section, were confirmed by
1
1
was warmed to 23 C and monitored by H and 31P { H} NMR
spectroscopy using sufficient delay times between scans to permit
1
reliable integration of the ferrocene standard and all other H
NMR signals. The spectra were monitored over 2 days, but showed
no deviation from the data collected directly after THF addition.
The relative concentrations of 1-MeI and (PNN)Ir(CH3)I(THF)
were determined by the position of 31P signal relative to the limiting
values of 8.54 and -0.71 ppm, respectively. These were converted
to absolute concentrations by accounting for the total moles of
iridium complex and the total volume of solvent (benzene-d6 +
tetrahydrofuran) added to the tube. The concentration of free
THF was determined by integration of the 1H NMR signals,
and corrected for the amount of (PNN)Ir(CH3)I(THF) (free
and bound THF signals were not resolved). The procedure was
repeated for five concentrations of THF.
1
a 1H–13C HSQC NMR experiment. The 31P { H} NMR spectrum
of 1-COE displays a single resonance at 15.21 ppm, substantially
upfield of the -19.3 ppm chemical shift previously reported for
the free ligand, confirming coordination the of chelate.6
Preparation of {(j3-N,N,P)Me2N(CH2CH2)N(o-C6H4PPh2)}-
Ir(CH3)2 (1-Me2). A 20 mL scintillation vial was charged with
0.068 g (0.100 mmol) of 1-MeI and approximately 10 mL
tetrahydrofuran. The green solid quickly produced an orange-
yellow solution. Then 7.5 mg (0.140 mmol) of (CH3)2Mg was
added. The resulting bright orange reaction mixture was stirred
overnight, turning to a deep yellow-brown color. The volatiles were
subsequently removed in vacuo to give an orange solid. The solid
was extracted with pentane and filtered through a Pasteur pipette
to give a mauve colored solution. Recrystallization at -35 ◦C
afforded 0.027 g (46%) of deep purple crystals identified as 1-
Me2. Further fractions of 1-Me2 may be extracted from the crude
reaction residue using diethyl ether, although this material is of
diminished purity. Anal. Calcd. for C24H30IrPN2: C, 50.60; H, 5.31;
N, 4.92. Found: C, 50.34; H, 5.24; N, 4.78. 1H NMR (benzene-d6):
d 1.15 (d, 4 Hz, 6H, Ir–CH3), 2.08 (d, 5 Hz, 6H, N–CH3), 2.32
(t, 2H, N–CH2), 2.99 (t, 2H, N–CH2), 6.55 (m, ArH), 7.35 (m,
(1)
Cyclooctene complexes of iridium are fairly common given the
widespread use of bis(cyclooctene)iridium chloride dimer as a
metalation agent.10 Although such species are most often straight-
forward olefin adducts with no significant chemical perturbation
of the cyclooctene, a few examples of activated iridium–COE
compounds have been reported.11 For 1-COE, the high symmetry
indicated by 1H and 13C NMR spectra and the absence of a
metal–hydride signal are most consistent with simple binding
of the alkene within a four-coordinate iridium(I) environment.
With a successful protocol for ligation of PNN to iridium in
hand, the electrophilicity of the metal center was compared to
related complexes by preparation of the iridium(I) monocarbonyl
complex, (PNN)IrCO (1-CO) (eqn (2)). Addition of one atmo-
sphere of carbon monoxide to 1-COE resulted in immediate
1
ArH), 7.19 (m, ArH), 7.69 (m, ArH). 31P { H} NMR (benzene-
1
d6): d 12.01. 13C { H} NMR (benzene-d6): d -17.22 (Ir–CH3),
47.42 (N–CH2), 50.09 (N–CH3), 70.30 (N–CH2), 110.73, 117.45,
130.02, 131.42, 131.97, 132.49, 133.56, 134.21 (Ar). Two Ar peaks
could not be located.
Characterization
C6H4PPh2)}Ir(H)2 (1-H2). A J-Young NMR tube was charged
of
{(j3-N,N,P)
Me2N(CH2CH2)N(o-
4302 | Dalton Trans., 2011, 40, 4300–4306
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
Table 2 Infrared CO stretching frequencies of tridentate ligated irid-
benzene-d6 displayed the number of signals consistent with a
C1 symmetric molecule including doublet resonances at 2.68,
2.10, and 1.64 ppm for the two N–CH3 sites and the Ir–CH3
position, respectively. 1H–13C HSQC NMR spectroscopy indicated
a correlation between the proton resonance at 1.64 ppm and an
upfield 13C NMR signal at -29.30 ppm, confirming its assignment
ium(I) carbonyl species
Molecule
n(CO) (cm-1)
Ref.
1-CO
1922a
1916b
1928c
1949c
1930d
1930a
1999.5b,e
This work
This work
14
14
15
16
17
[(C6H3)-1,3-(CH2PtBu2)2]IrCO
[(C6H3)-1,3-(OPtBu2)2]IrCO
[(Ph2PCH2SiMe2)2N]IrCO
[(iPr2P(C6H3Me))2N]IrCO
[HB(C3N2Me2H)3]Ir(CO)2
1
as the methyl bound to iridium. Additionally, the 31P{ H} NMR
spectrum exhibited a singlet at 8.54 ppm, shifted upfield relative
to the olefin complex.
a Recorded in benzene. b Recorded in KBr. c Recorded in pentane.
d Recorded in dichloromethane. e Average frequency.
(3)
substitution of the bound olefin, and afforded 1-CO in good yield
upon removal of all volatile material. The monocarbonyl complex
was identified by NMR and IR spectroscopy. Significantly, the
solution infrared spectrum of 1-CO exhibits an intense band
at 1922 cm-1 indicative of a bound carbonyl ligand. This band
is slightly lower in energy than carbonyl stretching frequencies
observed for similar iridium(I) complexes bearing monoanionic
tridentate ligands (Table 2) and suggests the PNN ligand affords a
relatively less electrophilic metal center.12 This trend was somewhat
unanticipated as the PNN ligand contains an aryl substituted
phosphine group instead of typically more electron donating alkyl
substituents. The origin of the red shifted infrared band in 1-CO
may be due to significant p-donation from the anilido nitrogen
donor on the chelating ligand (vide infra) or alternatively may be
an artifact of electrostatic influences on the C–O bond stretching
frequency as proposed by Goldman and Krogh–Jespersen for
related tridentate iridium carbonyl complexes.13 In either case,
the band at 1922 cm-1 for 1-CO clearly indicates substantial
p-backbonding between the iridium center and CO ligand and
implies a significant binding energy may be expected for other
p-acidic ligands.
Five-coordinate iridium(III) methyl iodide complexes supported
by tridentate ligands have been described previously, and have
proven to be valuable precursors for the study of hydrocarbyl
reductive elimination reactions, C–P bond formation, and het-
erolytic dihydrogen activation.15,19 Fryzuk and coworkers have
reported the X-ray structure for a variant supported by the
[N(SiMe2CH2PiPr2)2]- ligand which exhibits a generalized square
pyramidal geometry.15 As was observed for the other steri-
cally encumbered tridentate iridium(III) methyl iodide complexes,
[N(SiMe2CH2PiPr2)2]Ir(CH3)I appears to require addition of
strong, small ligands such as carbon monoxide or primary
phosphines, to achieve coordinative saturation.15,19b In contrast,
the relatively open steric environment of 1-MeI permits binding
of a considerably weaker ligand.
Due to the modest solubility of 1-MeI in hydrocarbon sol-
vents, initial experiments to assay reactivity were attempted in
tetrahydrofuran (THF). Notably, when bright green samples of
1-MeI were added to THF the solutions immediately became
orange. Removal of the THF solvent and redissolving the residue
in toluene or benzene resulted in reversal of the solution to the
original green color. This behavior suggests a reversible interaction
between the cyclic ether and the iridium center. Coordination of
THF has been reported in numerous cationic iridium complexes,20
but isolated examples are rare for neutral compounds21 presumable
due to the reduced electrophilic character of the metal center.
Tetrahydrofuran coordination to the iridum(III) methyl iodide
species was confirmed by an X-ray diffraction experiment. An
orange single crystal sample was obtained by chilling a concen-
trated THF solution of 1-MeI at -35 ◦C for three days. The small
crystals produced data of marginal quality, but was successfully
refined to obtain the solid state structure of (PNN)Ir(CH3)I(THF)
which is presented in Fig. 2.
(2)
A strong interaction between the (PNN)Ir fragment and cy-
clooctene is evident in the significant thermostability of 1-COE
which tolerates heating at 150 ◦C in sealed NMR tubes with
benzene-d6 solvent for hours with no observable decomposition.
In fact, even substitution by a strong donor ligand, such as
triphenylphosphine, required thermolysis at 80 ◦C for five days
to achieve complete conversion of 1-COE to (PNN)IrPPh3 (1-
PPh3) (eqn (3)).14 However, oxidative substitution of 1-COE
proved considerably more facile. Addition of methyl iodide to
1-COE at ambient temperature resulted in immediate loss of the
cyclooctene ligand and oxidation of the iridium center to yield the
deep green complex, (PNN)Ir(CH3)I (1-MeI). The five-coordinate
iridium(III) methyl iodide product was isolated by filtration from
pentane and characterized by multinuclear NMR spectroscopy
The data reveal a generalized octahedral geometry about
the metal center, with the THF ligand bound trans to the
˚
iridium–methyl bond. The Ir(1)–O(2) bond length of 2.36(1) A
is comparable (within the low resolution of the structure) to
previously reported values for both cationic and neutral iridum–
tetrahydrofuran adducts.20c,21b Interestingly, the normally strong
trans influencing methyl ligand does not drastically elongate the
iridium–oxygen bond, although it may contribute to weakening
the ether adduct (vide infra).
1
and combustion analysis. The H NMR spectrum of 1-MeI in
This journal is
The Royal Society of Chemistry 2011
Dalton Trans., 2011, 40, 4300–4306 | 4303
©

View Article Online
The equilibrium constant of THF binding was determined from
additions of excess cyclic ether to benzene-d6 solutions of 1-MeI.
The concentrations of THF were determined by 1H NMR integra-
tion of the signals for THF versus an internal ferrocene standard.18
The concentrations of (PNN)Ir(CH3)I and (PNN)Ir(CH3)I(THF)
1
were obtained from the 31P { H} NMR chemical shift (with respect
to the limiting values of 8.54 and -0.71 ppm) and the total
concentration of iridium introduced to the sample.22 While the
observation of only one time averaged 31P NMR signal suggests
the reaction approaches equilibrium rapidly, this was confirmed
by monitoring the component concentrations over two days which
afforded no detectable change. These experiments yielded an
equilibrium constant of 1.66(9) M for the reversible binding of
THF to 1-MeI at 23 ◦C, indicating only a weak coordination
of THF. The implication that the PNN ligand imparts a fairly
electron rich iridium center (vide supra) suggests the ability of 1-
MeI to bind THF is may have a substantial steric component, with
the low metal electrophilicity being responsible for the weakness
of coordination.
The reversible coordination of THF to 1-MeI did not prevent
further alkylation of the metal center. Addition of a small excess
of dimethylmagnesium to 1-MeI in THF afforded (PNN)Ir(CH3)2
(1-Me2) upon recrystallization from pentane or diethyl ether. The
iridium(III) dimethyl complex could also be formed using other
methylating agents, such as methyl lithium or methyl magnesium
bromide, but these produced lower yields and significant quantities
of uncharacterized by-products. The 1-Me2 complex was charac-
terized by single crystal X-ray diffraction, elemental analysis, and
Fig. 2 Molecular structure of (PNN)Ir(CH3)I(THF) at 30% probability
ellipsoids. All hydrogen atoms are omitted for clarity. Select bond distances
◦
˚
(A) and angles ( ): Ir(1)–C(27) 2.06(2), Ir(1)–O(2) 2.36(1), Ir(1)–P(1)
2.219(5), Ir(1)–N(1) 2.22(2), Ir(1)–N(2) 2.00(1), Ir(1)–I(1) 2.690(2);
C(27)–Ir(1)–O(2) 177.8(6), P(1)–Ir(1)–N(1) 164.4(5), P(1)–Ir(1)–N(2)
84.4(5), N(2)–Ir(1)–N(1) 80.2(7).
1
The strength of the iridium–tetrahydrofuran interaction was
examined by a set of titration experiments. A sample of 1-
MeI in benzene-d6 was treated progressively with 5, 10, 100,
200, and 600 equivalents of THF. Each successive addition of
ether diminished the green hue of the sample and altered the
NMR spectroscopy. The H NMR spectrum of 1-Me2 indicated
formation of a Cs symmetric molecule with doublets centered at
2.08 and 1.15 ppm assigned to the N–CH3 and Ir–CH3 protons,
respectively. The resonance at 1.15 ppm was correlated to a peak at
-17.22 ppm in the 13C NMR spectrum by means of 1H–13C HSQC
NMR spectroscopy. In addition, a single resonance was observed
1
1
observed chemical shifts in the H and 31P{ H} NMR spectra.
Changes in the 1H NMR spectra were relatively minor with
chemical shift changes of less than 0.6 ppm observed under
conditions where the signals were not obscured by added THF. The
1
in the 31P { H} NMR spectrum at 12.01 ppm.
1-Me2 exhibited limited solubility in pentane, which enabled
formation of X-ray quality crystals from concentrated solutions
of the hydrocarbon solvent at -35 ◦C. The molecular structure
obtained from these dark purple blocks is depicted in Fig. 4.
The iridium(III) dimethyl complex shows a generalized trigonal
bipyramid geometry with the diphenylphosphine and dimethy-
lamino groups of the PNN ligand occupying the axial sites
and the two iridium–methyls and anilido nitrogen arrayed in
equatorial positions. The iridium–carbon bond lengths of 2.074(6)
1
influence of added THF in the 31P{ H} NMR spectra was more
dramatic with a single resonance shifting gradually upfield from
8.53 ppm in the absence of THF down to -0.71 ppm with
600 equivalents of ether. Further additions of THF effected
no detectable change in the NMR spectra. Significantly, the
observation of a single sharp 31P resonance in the presence of
ether is consistent with a rapid and reversible binding of THF to
1-MeI at ambient temperature, where the chemical shift reflects the
relative proportion of (PNN)Ir(CH3)I and (PNN)Ir(CH3)I(THF)
in a given concentration of THF (Fig. 3).9
˚
and 2.094(6) A are similar to the metal–carbon bond distance
observed in the THF adduct of 1-MeI. Notably, the Ir(1)–
˚
N(1) distance of 2.003(4) A is substantially shorter than the
˚
2.190(5) A bond length found for Ir(1)–N(2). A similar trend
can be observed in (PNN)Ir(CH3)I(THF). The short Ir(1)–N(1)
bond length may be due to the localized anionic charge on the
central nitrogen, but may also indicate a significant degree of p-
donation from the anilido nitrogen to the metal center. In fact,
this bond is slightly shorter than the analogous Ir–N bond length
i
15
˚
reported for [N(SiMe2CH2P Pr2)2]Ir(CH3)I (2.079(7) A). Thus
the crystallographic data offers some support for p-donation from
the PNN ligand as an influencing factor in the relatively red shifted
stretching frequency of 1-CO (vide supra).
Complex 1-Me2 has proven to be considerably less stable than
1-MeI or 1-COE. While 1-Me2 will persist in rigorously dry
Fig. 3 Reversible coordination of tetrahydrofuran to 1-MeI.
4304 | Dalton Trans., 2011, 40, 4300–4306
This journal is
The Royal Society of Chemistry 2011
©

View Article Online
The iridium(III) dihydride complex was characterized primarily
by NMR spectroscopy, as the species degraded over approximately
6 h with or without the presence of excess H2. 1-H2 could also be
formed by addition of dihydrogen to 1-COE, although this trans-
formation was slow and substantial degradation of the iridium
dihydride complex occurred prior to complete consumption of
the cyclooctene iridium(I) complex. Interestingly, hydrogenation
of the cyclooctene was not detected upon treatment of 1-COE
with H2, instead simple substitution of the olefin was observed by
NMR spectroscopy. The 1H NMR spectrum of 1-H2 in benzene-d6
indicates a high symmetry structure with an upfield shifted metal–
1
hydride resonance at -22.76 ppm (2JP–H = 16 Hz). The 31P{ H}
NMR spectrum of 1-H2 displayed a singlet at 21.35 ppm. The
modest lifetime of 1-H2 also permitted characterization by both
1
one-dimensional 13C{ H} and two-dimensional 1H–13C HSQC
NMR spectroscopy.18
The identification of 1-H2 as a classical iridium(III) dihydride
complex is supported by the synthesis of the (PNN)IrHD (1-
HD) isotopologue which was prepared in a manner analogous to
the perproteo complex using HD gas. The NMR spectroscopy
for 1-HD indicated a small isotopic perturbation of the 31P
chemical shift downfield to 21.47 ppm and an upfield shift of the
iridium–hydride resonance to -22.80 ppm. Significantly, the new
metal–hydride signal appeared as a broad doublet with a peak
width at half height of only 6.7 Hz. This indicates a maximum
possible JH–D coupling constant of ~3.4 Hz, which is roughly an
order of magnitude lower than that typically observed in non-
classical dihydrogen complexes.26 This observation, along with the
substantial cis 2JP–H = 16 Hz coupling constant, are consistent with
the assignment of 1-H2 as an iridium(III) species.27
The limited stability of 1-H2 has thus far hampered efforts to
assay the reactivity of this sterically accessible hydride species. The
exact pathways for the degradation of the iridium(III)) dihydride
species are unknown to date, but the observation of dark insoluble
particles and small amounts of free PNN ligand in the NMR
spectra upon loss of 1-H2 suggest at least some reduction of
the metal center. The presence of a basic anilido nitrogen site
in the PNN ligand may facilitate chelate loss, as intramolecular
iridium–hydride deprotonation has been observed in similar
complexes.19a,24b,28 Additionally, the relatively small substituents
on the PNN ligand may permit formation of iridium–hydride
aggregate species.29 These possibilities are the subject of continued
investigation.
Fig. 4 Molecular structure of 1-Me2 at 30% probability ellipsoids. All
hydrogen atoms are omitted for clarity. Select bond distances (A) and
˚
angles (◦): Ir(1)–C(23) 2.074(6), Ir(1)–C(24) 2.094(6), Ir(1)–N(1) 2.003(4),
Ir(1)–N(2) 2.190(5), Ir(1)–P(1) 2.202(1); C(23)–Ir(1)–C(24) 77.6(2),
P(1)–Ir(1)–N(1) 84.4(1), P(1)–Ir(1)–N(2) 163.4(1), N(2)–Ir(1)–N(1)
79.4(2).
hydrocarbon solutions for approximately two days at ambient
temperature, even mild thermolysis at 55 ◦C induced complete
sample degradation over the course of hours. Recent studies have
indicated that five-coordinate iridium(III) dialkyl complexes such
as 1-Me2 may be expected to undergo facile C–C bond reductive
elimination upon heating.19c Such a process would then permit
access to a potentially reactive 14-electron, (PNN)Ir fragment.
Unfortunately, no ethane was detected by 1H NMR spectroscopy
in the mixture of thermolysis products23 and considerable quanti-
ties of metallic particles were observed precipitating from solution.
Conducting thermolysis experiments with the exclusion of light
also provided similar results. The limited thermal stability of 1-Me2
then directed investigations toward more mild routes to activate
the dimethyl iridium complex.
Iridium alkyl complexes have previously been employed as pre-
cursors for hydrogenolysis reactions to prepare reactive iridium–
hydride species24 which are known to mediate C–H bond ac-
tivation, olefin hydrogenation and hydrosilylation among other
reactions.25 An analogous synthesis was attempted with 1-Me2 by
addition of one atmosphere of dihydrogen at ambient temperature.
Hydrogenolysis of 1-Me2 occurred immediately upon mixing of
the gas into solution, bleaching the dark purple color to afford
methane and (PNN)IrH2 (1-H2) (Fig. 5).
Concluding remarks
Organoiridium complexes of the anionic, tridentate N-
(dimethylaminoethyl)-2-diphenylphosphinoaniline ligand have
Fig. 5 Synthetic routes for the formation of 1-H2.
The Royal Society of Chemistry 2011
This journal is
Dalton Trans., 2011, 40, 4300–4306 | 4305
©

View Article Online
been synthesized, representing the first transition metal complexes
of this sterically open ligand platform. The combination of
hard and soft binding sites supported the formation of a highly
stable iridium(I) cyclooctene complex. The chelate ligand imparts
an unexpectedly electron rich metal center, as judged by the
preparation of the corresponding monocarbonyl complex. The
iridium(I) cyclooctene adduct serves as a convenient precursor
for preparation of an iridium(III) methyl iodide species, which
undergoes reversible coordination of tetrahydrofuran. Binding
of the cyclic ether was confirmed by X-ray diffraction, and the
equilibrium constant for THF dissociation was determined by
NMR spectroscopy. A modestly stable iridum(III) dimethyl com-
plex was prepared by alkylation of (PNN)Ir(CH3)I and the dark
purple compound crystallographically characterized, revealing a
trigonal bipyramidal geometry. Hydrogenolysis of (PNN)Ir(CH3)2
afforded a classical iridium(III) dihydride complex, which was
characterized by NMR spectroscopy despite its modest stability.
These studies suggest that the open coordination sphere of the N-
(dimethylaminoethyl)-2-diphenylphosphinoaniline ligand can be
used to easily prepare organoiridium complexes as precursors to
potentially reactive iridium–hydride species. However, the limited
stability of the (PNN)IrH2 has hindered efforts to further probe
its activity. Ongoing investigations are pursing the fate of the
(PNN)IrH2 complex as well as small steric alteration to the PNN
ligand to effect enhanced lifetimes of the iridium–hydride species
while still maintaining a largely open coordination sphere.
Budzelaar, N. N. P. Moonen, R. de Gelder, J. M. M. Smits and
A. W. Gal, Eur. J. Inorg. Chem., 2000, 753; (c) W. H. Bernskoetter,
E. Lobkovsky and P. J. Chirik, Organometallics, 2005, 24, 6250;
(d) A. Friedrich, R. Ghosh, R. Kolb, E. Herdtweck and S. Schneider,
Organometallics, 2009, 28, 708.
12 (a) C. A. Tolman, Chem. Rev., 1977, 77, 313; (b) C. E. Zachmanoglou,
A. Docrat, B. M. Bridgewater, G. E. Parkin, C. G. Brandow, J. E.
Bercaw, C. N. Jardine, M. Lyall, J. C. Green and J. B. Kiester, J. Am.
Chem. Soc., 2002, 124, 9525.
13 A. S. Goldman and K. Krogh-Jespersen, J. Am. Chem. Soc., 1996, 118,
12159.
14 Z. Keming, P. D. Achord, X. Zhang, K. Krogh-Jespersen and A. S.
Goldman, J. Am. Chem. Soc., 2004, 126, 13044.
15 M. D. Fryzuk, P. A. MacNeil and S. J. Rettig, Organometallics, 1986,
5, 2469.
16 M. T. Whited and R. H. Grubbs, J. Am. Chem. Soc., 2008, 130, 5874.
17 D. M. Tellers, S. J. Skoog, R. G. Bergman, T. B. Gunnoe and W. D
Harman, Organometallics, 2000, 19, 2428.
18 See Experimental section for full details.
19 (a) M. D Fryzuk, P. A. MacNeil and S. J. Rettig, Organometallics,
1985, 4, 1145; (b) M. D. Fryzuk, K. Joshi, R. K. Chadha and S. J.
Rettig, J. Am. Chem. Soc., 1991, 113, 8724; (c) R. Ghosh, T. J. Emge,
K. Krogh-Jespersen and A. S. Goldman, J. Am. Chem. Soc., 2008, 130,
11317.
20 (a) R. H. Crabtree, P. C. Demou, D. Eden, J. M. Mihelcic, C. A. Parnell,
J. M. Quirk and G. E. Morris, J. Am. Chem. Soc., 1982, 104, 6994; (b) B.
Olgemoeller, H. Bauer, H. Loebermann, U. Nagel and W. Beck, Chem.
Berichte, 1982, 115, 2271; (c) X. Luo, G. K. Schulte and R. H. Crabtree,
Inorg. Chem., 1990, 29, 682; (d) C. Chardon, O. Eisenstein, T. Johnson
and K. G. Caulton, New J. Chem., 1992, 16, 781; (e) C. Bianchini,
K. G. Caulton, T. J. Johnson, A. Meli, M. Peruzzini and F. Vizza,
Organometallics, 1995, 14, 933; (f) M. J. Alcon, M. Iglesias, F. Sanchez
and I. Viani, J. Organomet. Chem., 2000, 601, 284; (g) J. T. Golden,
R. A. Andersen and R. G. Bergman, J. Am. Chem. Soc., 2001, 123,
5837; (h) P. Sangtrirutnugul and T. D. Tilley, Organometallics, 2007,
26, 5557; (i) C. Y. Tang, A. L. Thompson and A. Simon, J. Am. Chem.
Soc., 2010, 132, 10578.
Acknowledgements
21 (a) J. M. O’Connor, A. Closson and P. Gantzel, J. Am. Chem. Soc.,
2002, 124, 2434; (b) X. Sung and K. S. Chan, Organometallics, 2007,
26, 965.
22 Small changes in the chemical shift value due to attenuations in solvent
polarity are neglected in this estimate.
We gratefully acknowledge Brown University for financial sup-
port. N.G.L also thanks Brown University for a Summer Under-
graduate Teaching and Research Award.
23 Some quantities of methane, methane-d1, and free ligand were observed
in the 1H NMR spectrum from thermolysis of 1-Me2 in benzene-d6
solution.
24 For representative examples of iridium-hydride syntheses via alkyl
hydrogenolysis see: (a) W. M. Rees, M. R. Churchill, Y. Li and J. D.
Atwood, Organometallics, 1985, 4, 1162; (b) M. D. Fryzuk, P. A.
MacNeil and S. J. Rettig, J. Am. Chem. Soc., 1987, 109, 2803; (c) P. P.
Deutsch and R. Eisenberg, J. Am. Chem. Soc., 1990, 112, 714.
25 L. A. Oro, Iridium Complexes in Organic Synthesis, Wiley-VCH,
Weinheim, 2009.
References
1 D. Morales-Morales and C. M. Jensen, The Chemistry of Pincer
Compounds, Elsevier B.V., Amsterdam, 2010.
2 A. Martin and G. van Koten, Angew. Chem. Int Ed., 2001, 40, 3750.
3 D. Milstein, Top. Catal., 2010, 53, 915, and references therein.
4 (a) A. M. Winter, K. Eichele, H. G. Mack, S. Potuznik, H. A. Mayer
and W. C. Kaska, J. Organomet. Chem., 2003, 683, 149; (b) L. C. Liang,
J. M. Lin and C. H. Hung, Organometallics, 2003, 22, 3007.
5 L. Fan, B. M. Foxman and O. V. Ozerov, Organometallics, 2004, 23,
326.
26 P. G. Jessop and R. H. Morris, Coord. Chem. Rev., 1992, 121, 954.
27 G. J. Kubas, Acc. Chem. Res., 1988, 21, 120.
6 W. Lee and L. Liang, Dalton Trans., 2005, 1952.
28 I. C. M. Wehman-Ooyevaar, I. F. Luitwieler, K. Vatter, D. M. Grove,
W. J. J. Smeets, E. Horn, A. L. Spek and G. van Koten, Inorg. Chim.
Acta., 1996, 252, 55.
29 (a) H. H. Wang and L. H. Pignolet, Inorg. Chem., 1980, 19, 1470;
(b) H. H. Wang, A. L. Casalnuovo, B. J. Johnson, A. M. Mueting
and L. H. Pignolet, Inorg. Chem., 1988, 27, 325; (c) L. Garlaschelli, F.
Greco, G. Peli, M. Manassero, M. Sansoni, R. Gobetto, L. Salassa and
R. Della Pergola, Eur. J. Inorg. Chem., 2003, 11, 2108; (d) A. Rifat, G.
Kociok-Kohn, J. W. Steed and A. S. Weller, Organometallics, 23, 428.
7 J. L. Herde, L. C. Lambert and C. V. Senoff, Inorg. Syn., 1974, 15, 18.
8 A. B. Pangborn, M. A. Giardello, R. H. Grubbs, R. K. Rosen and F. J.
Timmers, Organometallics, 1996, 15, 1518.
9 J. Sandstro¨m, Dynamic NMR SpectroscopyAcademic Press: New York,
1982.
10 J. M. O’Connor, Science of Synthesis, 2002, 617.
11 For selected examples of activated Ir-COE complexes see: (a) R. S.
Tanke and R. H. Crabtree, Inorg. Chem., 1989, 28, 3444; (b) P. H. M.
4306 | Dalton Trans., 2011, 40, 4300–4306
This journal is
The Royal Society of Chemistry 2011
©