Acidolyis and oxygen atom transfer reactivity of a diiridium hydroperoxo complex

Article abstract of DOI:10.1039/c2dt32887a

Oxygenation of Ir₂^II,II(tfepma)₂(CN ^tBu)₂Cl₃H (1, tfepma = CH ₃N[P(OCH₂CF₃)₂]₂) furnishes Ir₂^II,II(tfepma)₂(CN ^tBu)₂Cl₃(OOH) (2), the first isolable iridium hydroperoxo complex. This complex transfers an oxygen atom to triphenylphosphine, producing triphenylphosphine oxide and Ir₂ ^II,II(tfepma)₂(CN^tBu)₂Cl ₃(OH) (3) in high yield. Reaction of 2 with acid induces cleavage of the O-O bond, initially forming [Ir₂^II,II(tfepma) ₂(CN^tBu)₂Cl₃(OH₂)]Cl (4), which was identified by NMR spectroscopy. With HCl as the acid source, the chloride anion displaces water to form the pseudo-C_2h isomer of Ir₂^II,II(tfepma)₂(CN^tBu) ₂Cl₄ (5). With 2,6-lutidinium chloride as the acid, complex 4 forms initially and is deprotonated by the conjugate base to afford 3 as the major product.

Full text of DOI:10.1039/c2dt32887a

Dalton
Transactions
View Article Online
View Journal | View Issue
PAPER
Acidolyis and oxygen atom transfer reactivity of a
diiridium hydroperoxo complex†
Cite this: Dalton Trans., 2013, 42, 3521
Thomas S. Teets and Daniel G. Nocera*
Oxygenation of Ir2II,II(tfepma)2(CNtBu)2Cl3H (1, tfepma = CH3N[P(OCH2CF3)2]2) furnishes Ir2II,II(tfep-
t
2
ma) (CN Bu)2Cl3(OOH) (2), the first isolable iridium hydroperoxo complex. This complex transfers an
oxygen atom to triphenylphosphine, producing triphenylphosphine oxide and Ir2II,II(tfepma)2(CNtBu)2-
Cl3(OH) (3) in high yield. Reaction of 2 with acid induces cleavage of the O–O bond, initially forming
[Ir2II,II(tfepma)2(CNtBu)2Cl3(OH2)]Cl (4), which was identified by NMR spectroscopy. With HCl as the
acid source, the chloride anion displaces water to form the pseudo-C2h isomer of Ir2II,II(tfepma)2(CNtBu)2Cl4
(5). With 2,6-lutidinium chloride as the acid, complex 4 forms initially and is deprotonated by the conju-
gate base to afford 3 as the major product.
Received 3rd December 2012,
Accepted 3rd December 2012
DOI: 10.1039/c2dt32887a
www.rsc.org/dalton
Introduction
The synthesis and reactivity of O2 complexes of transition
metals have garnered interest owing to the prominence of
such complexes in oxidative biochemistry,1–3 O2 reduction cata-
lysis,4 and aerobic oxidation catalysis.5 One class of reduced
oxygen complexes is metal hydroperoxos, which are commonly
prepared by aerobic oxidation of metal hydrides,6–23 or by
Scheme 1 Synthesis of hydroperoxo complex 2.
addition of hydrogen peroxide to
a
terminal metal
hydroxide.24–26 Although interest in the mechanism of the for-
mation of metal hydroperoxo complexes has been preponder-
ant, reactivity of these species has been interrogated for select
systems. Oxygen-atom-transfer reactions are among the most
prevalent.24–30 Oxygenation of phosphines to phosphine
oxides, sulphides to sulphoxides, olefins to epoxides or carbo-
nyl compounds, and benzaldehyde to benzoic acid has been
promoted by late metal hydroperoxo and alkylperoxo com-
plexes. Although acidolysis reactions of coordinated hydro-
peroxos are less studied, there are several examples of hydrogen
or alkyl peroxide liberation upon treatment of a hydroperoxo
or alkylperoxo complex with acid.25,29,39
complex has been proposed as a key intermediate in the
reduction of oxygen to water mediated by a dirhodium hydride
akin to 1,32 though the rhodium analogue of 2 has not been
observed spectroscopically. Though several examples of
iridium alkylperoxo complexes are known,33–35 complex 2 rep-
resents, to the best of our knowledge, the first example of an
isolable and structurally characterized iridium hydroperoxo
complex. With 2 in hand, we became interested in interrogat-
ing its reactivity on two fronts: (i) its oxygen atom transfer to
various substrates, and (ii) its reactions with acid sources,
leading to an understanding of the O–O bond cleavage step in
the O2 reduction chemistry of the analogous dirhodium system.
The initial results of our studies on both of these fronts are dis-
closed here. We show that diiridium hydroperoxo complex 2
transfers an oxygen atom to phosphines, generating a novel di-
iridium hydroxo complex, and that treatment of 2 with acids
induces rapid O–O bond cleavage and water formation.
We recently reported that oxygenation of a diiridium
hydride species, Ir2II,II(tfepma)2(CNtBu)2Cl3H (1, tfepma =
CH3N[P(OCH2CF3)2]2), produces the diiridium terminal hydro-
peroxo complex, Ir2II,II(tfepma)2(CNtBu)2Cl3(OOH) (2),31
depicted in Scheme 1. An analogous dirhodium hydroperoxo
Department of Chemistry, 6-335, Massachusetts Institute of Technology, 77
Massachusetts Avenue, Cambridge, MA 02139-4307. E-mail: nocera@mit.edu
†Electronic supplementary information (ESI) available: Full experimental
Experimental
General considerations
details, crystallographic details, and select NMR spectra. CCDC 914838 and
914839. For ESI and crystallographic data in CIF or other electronic format see
DOI: 10.1039/c2dt32887a
All reactions involving air-sensitive materials were executed in
a nitrogen-filled glovebox using solvents previously dried by
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 3521–3527 | 3521

View Article Online
Paper
Dalton Transactions
passage through an alumina column under argon. HCl (4 M in and several side products. The solution was concentrated
dioxane) was obtained from Sigma-Aldrich, whereas PPh3 and in vacuo to leave a yellow residue. The residue was taken up in
PEt3 were obtained from Strem. Anhydrous 2,6-lutidinium 4 mL of 1 : 1 Et2O–hexane, and concentrated in vacuo to leave a
chloride was prepared by adding anhydrous 2,6-lutidine to 1 M yellow solid. The solid was recrystallized from 3 mL of 1 : 2
HCl–Et2O under nitrogen, as described previously.37 The CH2Cl2–hexane at −35 °C. The supernatant was decanted, and
complex Ir2II,II(tfepma)2(CNtBu)2Cl3(OOH) (2; tfepma = CH3N- the yellow microcrystals were washed with 2 mL of hexane and
[P(OCH2CF3)2]2) was prepared by oxygenation of Ir2II,II(tfep- dried in vacuo. Yield: 22 mg (59%). 1H NMR (500 MHz,
t
31
2
ma) (CN Bu)2Cl3H (1), as described previously.
CD2Cl2) δ/ppm: 6.88 (br, s, 1H), 5.26 (m, 2H), 5.08 (m, 2H),
4.76–4.91 (m, 4H), 4.68 (m, 4H), 4.52 (m, 2H), 4.35 (m, 2H),
2.86 (pseudo-quintet, 6H), 1.46 (s, 9H), 1.23 (s, 9H). 31P{1H}
Physical methods
NMR spectra were recorded at the MIT Department of Chem- NMR (121.5 MHz, THF-d8) δ/ppm: 85.7–87.9 (m, 2P), 77.0–79.2
istry Instrumentation Facility on a Varian Mercury 300 NMR (m, 2P) (AA′BB′, 20 lines resolved, δavg = 82.5 ppm). Anal. Calcd
Spectrometer or
a Varian Inova-500 NMR Spectrometer. for C28H41Cl4F24Ir2N4O9P4: C, 20.40; H, 2.51; N, 3.40. Found:
31P{1H} NMR spectra were referenced to an external standard C, 20.74; H, 2.55; N, 3.22.
1
of 85% D3PO4 and H spectra were referenced to the residual
proteo solvent resonances. IR spectra were recorded on a Perkin-
Reaction of Ir2II,II(tfepma)2(CNtBu)2Cl3(OOH) (2) with HCl
Elmer Spectrum 400 FT-IR/FT-FIR Spectrometer outfitted Ir2II,II(tfepma)2(CNtBu)2Cl3(OOH) (2) (15 mg, 0.0090 mmol)
with a Pike Technologies GladiATR attenuated total reflectance was dissolved in 0.68 mL of THF-d8. A solution of HCl (0.48 M,
accessory with
pressure clamp.
a
monolithic diamond crystal stage and 20 μL, 0.0097 mmol, 1.1 eq.) in THF-d8 (prepared by diluting a
4.2 M HCl–dioxane solution) was introduced via autopipet.
The solution lightened in colour slightly, and was immediately
Preparation of Ir2II,II(tfepma)2(CNtBu)2Cl3(OH) (3) by reaction
of Ir2II,II(tfepma)2(CNtBu)2Cl3(OOH) (2) with PPh3
transferred to a J. Young NMR tube. Initial NMR spectra indi-
cate complete consumption of 2, with [Ir2II,II(tfepma)2(CNtBu)2-
A sample of Ir2II,II(tfepma)2(CNtBu)2Cl3(OOH) (2) (25 mg, Cl3(OH2)]Cl (4) as the major species. After 48 h, the NMR
0.015 mmol) was dissolved in 0.7 mL of THF-d8. This solution spectra show complete conversion to the pseuso-C2h isomer of
was used to dissolve solid PPh3 (8 mg, 0.03 mmol, 2 eq.). The Ir2II,II(tfepma)2(CNtBu)2Cl4 (5), with several side products
solution was transferred to a J. Young NMR tube and held at evident in the 1H NMR spectrum. Crystals of 5 suitable for
room temperature for 12 h. At this time, 31P{1H} NMR indi- X-ray diffraction were obtained from this mixture by removing
cates the formation of 3 and OvPPh3 as major products, the solvent in vacuo, redissolving in 2 mL of ca. 1 : 3 CH2Cl2–
along with unreacted PPh3 and some minor side products. The hexane and chilling to −35 °C overnight. Spectroscopic data
solution was concentrated in vacuo to yield a yellow residue, for 4: 1H NMR (500 MHz, THF-d8) δ/ppm: 7.83 (br, s, 2H),
which was redissolved in ∼2 mL of 1 : 1 Et2O–hexane. This solu- 4.45–5.45 (m, 16H), 3.14 (pseudoquintet, 6H), 1.60 (s, 9H),
tion was concentrated in vacuo to leave a yellow solid. The 1.47 (s, 9H). 31P{1H} NMR (121.5 MHz, THF-d8) δ/ppm:
solid was washed with hexane and dried in vacuo. We were 79.9–81.7 (br, m, 2P), 69.8–71.6 (br, m, 2P) (AA′BB′, 8 lines
unable to satisfactorily separate 3 from the OvPPh3 and resolved, δavg = 75.8 ppm). IR (solid): ˜νCuN = 2193 cm−1. Spectro-
unreacted PPh3, precluding a satisfactory elemental analysis. scopic data for 5: 1H NMR (500 MHz, THF-d8) δ/ppm: 5.18
Crystals of 3 suitable for X-ray diffraction were grown from (m, 4H), 5.04 (m, 8H), 4.62 (m, 4H), 3.03 (pseudoquintet, 6H),
CH2Cl2–hexane at −20 °C and manually separated from the 1.50 (s, 18H). 31P{1H} NMR (121.5 MHz, THF-d8) δ/ppm: 76.8
1
OvPPh3/PPh3 also present. Spectroscopic data for 3: H NMR (s, 4P). IR (Nujol): ˜νCuN = 2188 cm−1
.
(500 MHz, THF-d8) δ/ppm: 6.96 (br, s, 1H), 5.48 (m, 2H), 5.26
Preparation and isolation of Ir2II,II(tfepma)2(CNtBu)2Cl4 (5)
(pseudo-C2h isomer)
(m, 2H), 4.98 (m, 4H), 4.84 (m, 2H), 4.76 (m, 2H), 4.66 (m, 2H),
4.47 (m, 2H), 2.93 (pseudoquintet, 6H), 1.49 (s, 9H), 1.29 (s,
9H). 31P{1H} NMR (121.5 MHz, THF-d8) δ/ppm: 86.0–88.2 (m, A sample of 2 (45 mg, 0.027 mmol) was dissolved in 2 mL of
2P), 76.9–79.1 (m, 2P) (AA′BB′, 21 lines resolved, δavg
82.6 ppm). IR (Nujol): ˜νO–H = 3336 cm−1, ˜νCuN = 2187 cm−1
=
THF and sealed in a septum-capped vial. An aliquot of HCl in
dioxane (4.2 M, 13 μL, 0.054 mmol, 2.0 eq.) was introduced via
syringe. The solution lightened in colour, and after ∼15 min,
an aliquot of the headspace gas was withdrawn for GC analy-
sis, which showed no gaseous reaction products. The solution
.
Preparation and isolation of Ir2II,II(tfepma)2(CNtBu)2Cl3(OH)
(3)
In the glovebox, Ir2II,II(tfepma)2(CNtBu)2Cl3(OOH) (2) (37 mg, was concentrated in vacuo to leave a yellow residue, which was
0.022 mmol) was dissolved in 1 mL of THF in a scintillation redissolved in Et2O–hexane. Removal of the solvents in vacuo
vial. In a separate vial, a stock solution of PEt3 (20 μL, left a pale yellow solid, which was redissolved in THF-d8. At
0.14 mmol) in 0.20 mL of THF was prepared. An aliquot of this stage, 1H NMR showed a majority of aqua product 4, as
1
this stock solution (66 μL), corresponding to 2 eq. of PEt3, was described above. After 48 h, the crude H NMR spectrum indi-
added to the stirring solution of 2. The yellow solution was cated complete consumption of 4, with the desired product 5
stirred for 2 h at room temperature, at which time the crude as the major species (ca. 60% crude yield). The solution was
31P{1H} NMR spectrum shows a mixture of 3, OvPEt3, PEt3, concentrated in vacuo, and the resulting residue suspended in
3522 | Dalton Trans., 2013, 42, 3521–3527
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
Et2O–hexane. Removal of the solvents in vacuo left a yellow dichloromethane solvent molecule was disordered about a
solid, which was redissolved in 1.5 mL of CH2Cl2/3 mL of special position. In the structure of 5, one of the –CF3 groups
hexane. The yellow solution was chilled at −35 °C for 24 h, pro- was modelled as a two-part disorder. The (1,2) and (1,3) dis-
ducing pure product 5 as yellow microcrystals. Yield: 12 mg tances of all disordered parts were restrained to be similar
(27%). Spectral data are identical to those given above. Anal. using the SADI command; the rigid-bond restraints SIMU and
Calcd for C28H40Cl4F24Ir2N4O8P4: C, 20.18; H, 2.42; N, 3.36. DELU were also used on disordered parts. Unit cell para-
Found: C, 20.41; H, 2.44; N, 3.22.
meters, morphology, and solution statistics for the structures
of 3 and 5 are summarized in Table S1.† All thermal ellipsoid
plots are drawn at the 50% probability level, with carbon-
bound hydrogen atoms, solvent molecules and –CH2CF3
Reaction of Ir2II,II(tfepma)2(CNtBu)2Cl3(OOH) (2) with
2,6-lutidinium chloride
A solution of Ir2II,II(tfepma)2(CNtBu)2Cl3(OOH) (2) (10 mg, groups omitted for clarity.
0.0060 mmol) in 0.7 mL of THF was prepared in a J. Young
NMR tube. In a scintillation vial, a stock solution of 2,6-lutidi-
nium chloride (10.5 mg in 0.50 mL of CD3CN) was prepared. A
41 μL aliquot of this stock solution, corresponding to 1 eq. of
2,6-lutidinium chloride, was added to the NMR tube. Initially
Results and discussion
Oxygen-atom transfer
Hydroperoxo complex 2 persists for appreciable time in THF
some of the 2,6-lutidinium hydrochloride precipitated, but
with occasional shaking all solids dissolved after ∼2 h. After
40 h, the 1H NMR spectrum indicated the formation of 3
as the major product, with some unreacted 2 also present
(ca. 15% relative to 3). In addition to several metal-containing
side products, the NMR spectrum also indicated a mixture of
2,6-lutidinium chloride and free 2,6-lutidine in the product
mixture. The reaction can also be executed in the same
manner with a molar excess (2 eq.) of 2,6-lutidinium chloride,
and in this case all of 2 is consumed within 15 h, with a
similar distribution of side products.
solution, both in the presence or absence of additional O2.
Under an inert atmosphere, 2 decomposes to a complex
mixture of products with a half-life of approximately 2 weeks
at room temperature. Complex 2 is unreactive towards the sub-
strates cyclohexene, 2-cyclohexen-1-one, benzaldehyde, and
1,4-cyclohexadiene; treatment of 2 with 5–10 eq. of these sub-
strates at room temperature gives no evidence for product for-
mation after prolonged time periods. Attempts to heat reaction
mixtures led to the decomposition of hydroperoxo complex 2
to a multitude of intractable products, with no evidence of
substrate oxygenation. The inability of 2 to oxidize unsaturated
organic substrates may be in part due to its coordinatively
saturated octahedral geometry; previous examples of olefin
oxygenation involved square planar hydroperoxo com-
plexes,24,26 and in a related study of alkylperoxo complexes it
was explicitly shown that coordination of the alkene precedes
oxygen atom transfer.36 Complex 2 undergoes nonspecific
decomposition in the presence of dimethylsulphide at room
temperature, again with no evidence for oxygen atom transfer.
In contrast to the foregoing substrates, treatment of 2 with
PPh3 leads to near quantitative oxygen atom transfer, produ-
cing OvPPh3 and Ir2II,II(tfepma)2(CNtBu)2Cl3(OH) (3) as
depicted in Scheme 2. This transformation proceeds cleanly
with ca. 2 equivalents of PPh3; stoichiometric reaction con-
ditions result in significant decomposition in addition to the
X-Ray crystallographic details
Single crystals of 3 were obtained from CH2Cl2–hexane at
−20 °C, whereas crystals of 5 were grown from the same
solvent system at −35 °C. The crystals were mounted on a
Bruker three circle goniometer platform equipped with an
APEX detector. A graphite monochromator was employed for
wavelength selection of the Mo Kα radiation (λ = 0.71073 Å).
The data were processed and refined using the program SAINT
supplied by Siemens Industrial Automation. Structures were
solved by Patterson methods in SHELXS and refined by stan-
dard difference Fourier techniques in the SHELXTL program
suite (6.10 v., Sheldrick G. M., and Siemens Industrial Auto-
mation, 2000). Hydrogen atoms bonded to carbon were placed
in calculated positions using the standard riding model and
refined isotropically; all other atoms were refined anisotropi-
cally. The oxygen-bound hydrogen atom in 3 was tentatively
located in the difference map, but owing to the substantial dis-
order in the crystal was constrained using a standard riding
model for O–H protons. The crystal of 3 was found to be a non-
merohedral twin; two domains of the unit cell were located,
the absorption correction was applied using the program
TWINABS, and for the final refinement cycles the model was
refined against both twin domains. In addition, the structure
of 3 was substantially disordered. By virtue of a crystallographi-
cally-imposed inversion centre at the molecule’s centroid, two
orientations of the six monodentate ligands were present. In
addition, three of the four crystallographically independent
desired products. The initial 31P{1H} and H NMR spectra for
1
the reaction mixture are consistent with a weak interaction of
2 with PPh3, as evidenced by the substantial broadening of the
free PPh3 resonances, with no apparent change in the spectral
features attributed to 2. The reaction is complete within 12 h,
–CF3 groups were disordered about two positions, and the Scheme 2 Oxygen-atom transfer by 2.
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 3521–3527 | 3523

View Article Online
Paper
Dalton Transactions
and furnishes the 31P{1H} NMR spectrum depicted in Fig. S3,†
1
with the corresponding H NMR spectrum shown in Fig. S4.†
The 31P{1H} NMR spectrum of the hydroxo product 3 is con-
sistent with an AA′BB′ spin system, with chemical shifts and
II,II
splitting patterns reminiscent of other Ir2
complexes that
we have recently described.31,37 Two distinct CuNC(CH3)3 reso-
nances in the H NMR spectrum provide further evidence for
1
an asymmetric complex. A slightly broadened singlet at
6.96 ppm in the 1H NMR spectrum (THF-d8), integrating to
one proton, is attributed to the O–H group of 3. The NMR
spectra also clearly show the presence of OvPPh3 and
unreacted PPh3, with the latter resonances remaining broad-
ened in the final product spectra. The IR spectrum of the iso-
lated product contains a weak resonance at 3336 cm−1
attributed to the O–H stretch, which occurs at a slightly lower
energy than that of hydroperoxo complex 2 (3355 cm−1).
Complex 3 can be prepared in pure form by using PEt3 instead
of PPh3 as the O-atom acceptor. In this case, recrystallization
removes all other products and 3 can be obtained in 59% iso-
lated yield, with purity verified by NMR (Fig. S5 and S6†) and
elemental analysis.
Fig. 1 shows the single crystal X-ray diffraction structure of
complex 3‡. A crystallographically imposed inversion centre
defines two disordered orientations of the monodentate
ligands that surround the metal–metal axis. The relative con-
figuration of the ligands is identical to that of hydroperoxo
complex 2, i.e. the hydroxide in 3 occupies the former site of
the hydroperoxide, with no structural rearrangement upon
Fig. 1 X-ray crystal structure of 3, shown at the 50% probability level. In (a) all
non-carbon atoms are labelled, whereas in (b) a view down the metal–metal
axis with an analogous colour scheme is shown. Solvent molecules, carbon-
bound hydrogen atoms, and –CH2CF3 groups are omitted for clarity. Only one of
the two disordered orientations is shown.
oxygen-atom transfer. The Ir(1)–Ir(1A) distance of 2.6584(5) Å
II,II
is ca. 0.1 Å shorter than our other recent Ir2
complexes, but
nonetheless is suggestive of a metal–metal single bond.31,37 As
shown in Fig. 1b, the two equatorial coordination spheres
about the iridium atoms are very nearly eclipsed, as judged by
the two independent P–Ir–Ir–P torsion angles, one of which is
fixed at 180° by the inversion centre, whereas the other is quite
small at 4.40(7)°.
Acidolysis
During the reduction of O2 to water mediated by dirhodium
hydride complexes,32 the putative hydroperoxo intermediate,
proposed to form after the rate-determining insertion of O2,
reacts with acid to induce cleavage of the O–O bond to release
water.32 Since this O–O bond cleavage occurs beyond the rate-
determining formation of the hydroperoxo, information about
the mechanism was not afforded by the subsequent kinetic
analysis.31 Rather, insight into the O–O bond cleavage is
gleaned by examination of the reaction of the isolable iridium
hydroperoxo complex 2 with acidic reagents. To this end, the
reactivity of 2 was examined with the two acids, HCl and 2,6-
lutidinium chloride.
Scheme 3 Reactivity of 2 with acids.
As shown in Scheme 3, the reaction of hydroperoxo 2 with
HCl proceeds in two distinct steps. Immediately upon addition
of HCl, oxygen–oxygen bond cleavage is induced, and an
unstable species appears that is assigned to the cationic
complex [Ir2II,II(tfepma)2(CNtBu)2Cl3(OH2)]Cl (4), on the basis
of NMR spectroscopy. The 31P{1H} NMR spectrum (Fig. S7†)
‡Crystallographic data for 3·CH2Cl2: C29H43Cl5F24Ir2N4O9P4, M = 1733.20, Tricli-
shows a slightly broadened AA′BB′ splitting pattern, suggestive
ˉ
nic, P1, a = 10.4031(11), b = 12.5695(13), c = 12.6361(13), α = 65.033(2)°, β =
II,II
of an Ir2
complex but distinct from the spectrum of 2. In
70.612(2)°, γ = 84.392(2)°, V = 1411.2(3), Z = 1, μ = 5.187 mm−1, T = 200(2) K, R1
=
the 1H NMR spectrum (Fig. S8†), the expected resonances
0.0638, wR2
= 0.1362 (based on all reflections), GooF = 1.098, reflections
assigned to tfepma and CNtBu are present. In addition, a
measured = 8499, unique reflections = 8499, Rint = 0.0557.
3524 | Dalton Trans., 2013, 42, 3521–3527
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
broad resonance at ∼7.8 ppm, integrating to two protons, is
present and is assigned to a bound water. Notably, the NMR
spectral data suggest that only protons originating from the
hydroperoxo and the added acid are present in the
bound water; no deuterium incorporation from solvent is
observed. This aqua resonance persists in the isolated
product, and diminishes at a rate commensurate with the dis-
appearance of the other resonances of 4, verifying that this
broad resonance is indeed attributed to bound O–H protons
(i.e. and not to H2O2 or some other free species in solution).
The formation of 4 is analogous to the reactivity observed
when an isolable monorhodium peroxo complex is treated
with HCl, as we have recently reported.38 Over time, the NMR
resonances attributed to 4 give way to a symmetric product,
with a singlet in the 31P{1H} NMR spectrum (76.8 ppm) and a
1
single CNtBu resonance in the H spectrum (Fig. S7 and S9†).
The disappearance of 4 is independent of the presence of
added acid, and the final 1H NMR spectrum also shows a
broad singlet at ∼2.6 ppm, which overlaps with some of the
peaks for side products but is tentatively assigned to free H2O.
The final diiridium product is assigned to be the centrosym-
metric pseudo-C2h isomer of Ir2II,II(tfepma)2(CNtBu)2Cl4 (5), Fig. 2 X-ray crystal structure of 5, shown at the 50% probability level. In (a) all
non-carbon atoms are labelled, whereas in (b) a view down the metal–metal
distinct from the asymmetric isomer that is prepared by
halogen oxidation.37 Complex 5 was isolated in pure form in
axis with an analogous colour scheme is shown. Carbon-bound hydrogen atoms
and –CH2CF3 groups are omitted for clarity.
low yields (27%) by recrystallization (see Fig. S10† for the 1H
NMR spectrum of pure 5).
This reactivity with acids, in which the oxygen–oxygen bond
is cleaved in both cases, stands in contrast to the reactivity of
previously reported hydroperoxo complexes, which are
observed to release H2O2 upon protonolysis.24,28,39 We have
considered the possibility here that H2O2 is initially formed,
followed by rapid disproportionation to H2O and O2. However,
after treating 2 with acid, immediate addition of N,N-diethyl-p-
phenylenediamine (DPD), which is oxidized by H2O2 and other
oxygen-containing oxidants,40 gave no evidence for the for-
mation of the stable, bright pink N,N-diethyl-p-phenylenedi-
amine radical (DPD˙). This experiment suggests that if
hydrogen peroxide is initially formed, it does not persist in solu-
tion and decomposes instantaneously. In addition, analysis
of the headspace gases following acidolysis of 2 (conducted in
an inert atmosphere) showed no evidence for the evolution of
O2, which would arise if H2O2 were disproportionated.
Acknowledging that these two negative results do not conclus-
ively rule out a mechanism involving initial H2O2 release, they
do raise the possibility of other modes of O–O bond cleavage,
which may involve the formation of reactive high-valent
iridium species. Higher-oxidation-state iridium complexes are
not observed in the final product mixture, though it is worth
emphasizing that the yields of 4 formed during the acidolysis
reactions are moderate at best (≤60%), so it is likely that the
short-lived intermediates auto-decompose or react with solvent
after forming. Given the apparent high reactivity of these inter-
mediates, and the rapidity at which the overall O–O bond clea-
vage occurs, we are currently looking into trapping strategies
as well as computational investigations to provide insight into
the structures and reaction sequences of key intermediates.
The structure of 5§ was validated by X-ray crystallography,
as shown in Fig. 2. The symmetric nature of the complex is
evident, and the Ir(1)–Ir(2) bond distance of 2.7586(3) Å is
consistent with a metal–metal bond. As shown in Fig. 2b, the
equatorial coordination planes are twisted relative to one
another, with four independent P–Ir–Ir–P dihedral angles of
25.66(4), 22.61(4), 156.13(4) and 155.60(4)°. These values
indicate
a
substantial deviation from the eclipsed
configuration that is observed in the structure of 3 as dis-
cussed earlier. Complex 5 is crystallographically isostructural
with our previously reported dirhodium analogue of 5, with
nearly identical unit cell dimensions and the same space
group (P21/c).
The outcome of the reaction of hydroperoxo complex 2 with
acid is different if the acid is the weaker 2,6-lutidinium chlor-
ide. At early time points, a small amount of aqua complex 4 is
evident (see Fig. S11†). But over the course of 15 h, when 2 eq.
of acid are used, hydroxo complex 3 appears as the major
product, as is evident in the 1H NMR spectrum of Fig. S12.†
This observation suggests that 2,6-lutidinium is a strong
enough acid to protonate the hydroperoxo ligand, but that the
ensuing aqua complex 4 is deprotonated by free lutidine, at a
rate exceeding the substitution of Cl− for the bound water.
§Crystallographic data for 5: C28H40Cl4F24Ir2N4O10P4, M = 1666.72, Monoclinic,
P21/c, a = 12.7559(9), b = 19.5938(14), c = 21.0181(15), β = 97.9880(10)°, V =
5202.2(6), Z = 4, μ = 5.572 mm−1, T = 100(2) K, R1 = 0.0483, wR2 = 0.0811 (based
on all reflections), GooF = 1.040, reflections measured = 117 116, unique reflec-
tions = 15 213, Rint = 0.0527.
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 3521–3527 | 3525

View Article Online
Paper
Dalton Transactions
9 J. F. Endicott, C.-L. Wong, T. Inoue and P. Natarajan, Inorg.
Chem., 1979, 18, 450.
Conclusions
Two key classes of reactions have been demonstrated with a 10 M. T. Atlay, M. Preece, G. Strukul and B. R. James, J. Chem.
late metal hydroperoxo. Though unreactive towards unsatu- Soc., Chem. Commun., 1982, 406.
rated organic substrates, diiridium hydroperoxo complex 2 11 M. T. Atlay, M. Preece, G. Strukul and B. R. James,
undergoes oxygen atom transfer with phosphines to generate a Can. J. Chem., 1983, 61, 1332.
novel diiridium hydroxo product 3. The structural and spectro- 12 A. Bakac, J. Am. Chem. Soc., 1997, 119, 10726.
scopic characterization of 3 adds to a growing library of metal– 13 S. Thyagarajan, C. D. Incarvito, A. L. Rheingold and
metal bonded M2II,II(M = Rh, Ir) complexes supported by two
K. H. Theopold, Chem. Commun., 2001, 2198.
bridging tfepma ligands. Acids induce cleavage of the O–O 14 W. Cui and B. B. Wayland, J. Am. Chem. Soc., 2006, 128,
bond in hydroperoxo complex 2 to furnish an equivalent of 10350.
water, which is initially bound to the iridium in aqua complex 15 D. D. Wick and K. I. Goldberg, J. Am. Chem. Soc., 1999, 121,
4, and gradually is displaced by chloride to form the sym- 11900.
metric Ir2II,IICl4 product 5. With a weaker acid source, 2,6-luti- 16 M. M. Konnick, B. A. Gandhi, I. A. Guzei and S. S. Stahl,
dinium, intermediate 4 is intercepted by the conjugate base Angew. Chem., Int. Ed., 2006, 45, 2904.
and instead forms hydroxo complex 3 as the major product. 17 M. C. Denney, N. A. Smythe, K. L. Cetto, R. A. Kemp and
These reactions with acids provide insight into the O–O bond K. I. Goldberg, J. Am. Chem. Soc., 2006, 128, 2508.
cleavage step involved in the reduction of oxygen to water 18 E. Szajna-Fuller and A. Bakac, Inorg. Chem., 2010, 49,
mediated by isostructural dirhodium complexes, and begin to 781.
define plausible elementary steps that occur beyond the rate- 19 J. M. Keith, R. J. Nielsen, J. Oxgaard and W. A. Goddard,
determining oxygen activation chemistry. Some details related J. Am. Chem. Soc., 2005, 127, 13172.
to this reactivity of the hydroperxo remain outstanding. We are 20 M. M. Konnick and S. S. Stahl, J. Am. Chem. Soc., 2008, 130,
particularly interested in defining short-lived intermediates 5753.
that are involved in the oxygen–oxygen bond cleavage reaction. 21 M. M. Konnick, N. Decharin, B. V. Popp and S. S. Stahl,
1
2
In addition, though we can rule out formation of eq. of O2,
Chem. Sci., 2011, 2, 326.
the fate of the second oxygen atom in the oxygen–oxygen bond 22 N. Decharin and S. S. Stahl, J. Am. Chem. Soc., 2011, 133,
cleavage reaction is unresolved. As we continue our studies of 5732.
oxygen activation and reduction chemistry, subsequent work 23 N. Decharin, B. V. Popp and S. S. Stahl, J. Am. Chem. Soc.,
will aim to more precisely define the pathways by which acido- 2011, 133, 13268.
lysis induces liberation of water from rhodium and iridium 24 G. Strukul, R. Ros and R. A. Michelin, Inorg. Chem., 1982,
hydroperoxo complexes.
21, 495.
25 K. D. Karlin, P. Ghosh, R. W. Cruse, A. Farooq, Y. Gultneh,
R. R. Jacobson, N. J. Blackburn, R. W. Strange and
J. Zubieta, J. Am. Chem. Soc., 1988, 110, 6769.
26 T. Miyaji, M. Kujime, S. Hikichi, Y. Moro-oka and M. Akita,
Inorg. Chem., 2002, 41, 5286.
27 W.-D. Wang, A. Bakac and J. H. Espenson, Inorg. Chem.,
1995, 34, 4049.
28 S. A. Mirza, B. Bocquet, C. Robyr, S. Thomi and
A. F. Williams, Inorg. Chem., 1996, 35, 1332.
29 O. Pestovsky and A. Bakac, J. Am. Chem. Soc., 2003, 125,
14714.
Acknowledgements
This work was supported by NSF Grant CHE-1112154. Grants
from the NSF (CHE-9808061 and DBI-9729592) support the
Department of Chemistry Instrumentation Facility. T.S.T.
acknowledges the Fannie and John Hertz Foundation for a
graduate research fellowship.
30 M. L. Man, J. Zhu, S. M. Ng, Z. Zhou, C. Yin, Z. Lin and
C. P. Lau, Organometallics, 2004, 23, 6214.
Notes and references
1 M. Costas, M. P. Mehn, M. P. Jensen and L. Que, Chem. 31 T. S. Teets and D. G. Nocera, J. Am. Chem. Soc., 2011, 133,
Rev., 2004, 104, 939. 17796.
2 L. M. Mirica, X. Ottenwaelder and T. D. P. Stack, Chem. 32 T. S. Teets, T. R. Cook, B. D. McCarthy and D. G. Nocera,
Rev., 2004, 104, 1013.
J. Am. Chem. Soc., 2011, 133, 8114.
3 E. A. Lewis and W. B. Tolman, Chem. Rev., 2004, 104, 1047.
4 J. Rosenthal and D. G. Nocera, Acc. Chem. Res., 2007, 40,
543.
5 S. S. Stahl, Angew. Chem., Int. Ed., 2004, 43, 3400.
6 H. L. Roberts and W. R. Symes, J. Chem. Soc. A, 1968, 1450.
7 L. E. Johnston and J. A. Page, Can. J. Chem., 1969, 47, 4241.
33 B. L. Booth, R. N. Haszeldine and G. R. H. Neuss, J. Chem.
Soc., Chem. Commun., 1972, 1074.
34 B. L. Booth, R. N. Haszeldine and G. R. H. Neuss, J. Chem.
Soc., Dalton Trans., 1982, 37.
35 B. L. Booth, R. N. Haszeldine and G. R. H. Neuss, J. Chem.
Soc., Dalton Trans., 1982, 511.
8 R. D. Gillard, B. T. Heaton and D. H. Vaughan, J. Chem. 36 H. Mimoun, R. Charpentier, A. Mitschler, J. Fischer and
Soc. A, 1970, 3126.
R. Weiss, J. Am. Chem. Soc., 1980, 102, 1047.
3526 | Dalton Trans., 2013, 42, 3521–3527
This journal is © The Royal Society of Chemistry 2013

View Article Online
Dalton Transactions
Paper
37 T. S. Teets, T. R. Cook, B. D. McCarthy and D. G. Nocera, 39 M. M. Konnick, I. A. Guzei and S. S. Stahl, J. Am. Chem.
Inorg. Chem., 2011, 50, 5223. Soc., 2004, 126, 10212.
38 T. S. Teets and D. G. Nocera, Inorg. Chem., 2012, 51, 40 H. Bader, V. Sturzenegger and J. Hoigné, Water Res., 1988,
7192.
22, 1109.
This journal is © The Royal Society of Chemistry 2013
Dalton Trans., 2013, 42, 3521–3527 | 3527