Homogeneous catalytic reduction of dioxygen using transfer hydrogenation catalysts

Article abstract of DOI:10.1021/ja073774p

Solutions of Cp*IrH(rac-TsDPEN) (TsDPEN = H ₂NCHPhCHPhN(SO₂C₆H₄CH ₃)^-) (1H(H)) with O₂ generate Cp*Ir(TsDPEN-H) (1) and 1 equiv of H₂O. Kinetic analysis indicates a third-order rate law (second order in [1H(H)] and first order in [O₂]), resulting in an overall rate constant of 0.024 ± 0.013 M^-2 s^-1. Isotopic labeling revealed that the rate of the reaction of 1H(H) + O₂ was strongly affected by deuteration at the hydride position (k_HH2/k_DH2 = 6.0 ± 1.3) but insensitive to deuteration of the amine (k_HH2/k_HD2 = 1.2 ± 0.2); these values are more disparate than for conventional transfer hydrogenation (Casey, C. P.; Johnson, J. B. J. Org. Chem. 2003, 68, 1998-2001). The temperature dependence of the reaction rate indicated Δ ^? = 82.2 kJ/mol, ΔS^? = 13.2 J/mol·K, and a reaction barrier of 85.0 kJ/mol. A CH₂Cl ₂ solution under 0.30 atm of H₂ and 0.13 atm of O ₂ converted to H₂O in the presence of 1 and 10 mol % of H(OEt₂)₂BAr_F⁴ (BAr^F ₄^- = B(C₆H₃-3,5-(CF ₃)₂)₄^-). The formation of water from H₂ was verified by ²H NMR for the reaction of D ₂ + O₂. Solutions of 1 slowly catalyze the oxidation of amyl alcohol to pentanal; using 1,4-benzoquinone as a cocatalyst, the conversion was faster. Complex 1 also catalyzes the reaction of O₂ with RNH₂BH₃ (R = H, t-Bu), resulting in the formation of water and H₂. The deactivation of the catalyst 1 in its reactions with O₂ was traced to degradation of the Cp* ligand to a fulvene derivative. This pathway is not observed in the presence of amine-boranes, which were shown to reduce fulvenes back to Cp*. This work suggests the potential of transfer hydrogenation catalysts in reactions involving O ₂.

Full text of DOI:10.1021/ja073774p

Published on Web 10/25/2007
Homogeneous Catalytic Reduction of Dioxygen Using
Transfer Hydrogenation Catalysts
Zachariah M. Heiden and Thomas B. Rauchfuss*
Contribution from the School of Chemical Sciences, UniVersity of Illinois,
Urbana, Illinois 61801
Received May 25, 2007; E-mail: rauchfuz@uiuc.edu
Abstract: Solutions of Cp*IrH(rac-TsDPEN) (TsDPEN ) H2NCHPhCHPhN(SO2C6H4CH3)-) (1H(H)) with
O2 generate Cp*Ir(TsDPEN-H) (1) and 1 equiv of H2O. Kinetic analysis indicates a third-order rate law
(second order in [1H(H)] and first order in [O2]), resulting in an overall rate constant of 0.024 ( 0.013 M-2
s-1. Isotopic labeling revealed that the rate of the reaction of 1H(H) + O2 was strongly affected by deuteration
at the hydride position (kHH /kDH ) 6.0 ( 1.3) but insensitive to deuteration of the amine (kHH /kHD ) 1.2
2
2
2
2
( 0.2); these values are more disparate than for conventional transfer hydrogenation (Casey, C. P.; Johnson,
J. B. J. Org. Chem. 2003, 68, 1998-2001). The temperature dependence of the reaction rate indicated
∆Hq ) 82.2 kJ/mol, ∆Sq ) 13.2 J/mol‚K, and a reaction barrier of 85.0 kJ/mol. A CH2Cl2 solution under
0.30 atm of H2 and 0.13 atm of O2 converted to H2O in the presence of 1 and 10 mol % of H(OEt2)2BArF
4
(BArF ) B(C6H3-3,5-(CF3)2)4-). The formation of water from H2 was verified by H NMR for the reaction
-
2
4
of D2 + O2. Solutions of 1 slowly catalyze the oxidation of amyl alcohol to pentanal; using 1,4-benzoquinone
as a cocatalyst, the conversion was faster. Complex 1 also catalyzes the reaction of O2 with RNH2BH3 (R
) H, t-Bu), resulting in the formation of water and H2. The deactivation of the catalyst 1 in its reactions with
O2 was traced to degradation of the Cp* ligand to a fulvene derivative. This pathway is not observed in the
presence of amine-boranes, which were shown to reduce fulvenes back to Cp*. This work suggests the
potential of transfer hydrogenation catalysts in reactions involving O2.
Introduction
reactive toward O2 are, after oxidation, generally inert toward
H2.7 Promising is the fact that metal hydrides do react with O2
The Knall gas reaction, 2H2 + O2 f 2H2O, supplies energy
in fuel cells and some bacteria.2 Although platinum metal has
been known since Dobereiner’s time to catalyze this reaction,3
homogeneous catalysts are rare. The realization that the “difficult
side” of the H2-O2 fuel cell is the oxygen reduction reaction
(ORR),4,5 not hydrogen oxidation, motivates the development
of alternative approaches to oxygen reduction. We propose that
the development of new homogeneous catalysts for the hydro-
genation of oxygen could lead to new mechanistic insights
relevant to the design of new heterogeneous catalysts.
to produce hydroxides and hydroperoxides.8-10 For example,
Goldberg and Stahl have recently described the reaction of 16e
palladium hydrides with O2 to give hydroperoxides.9,11-13 The
resulting M-O2H and M-OH complexes are not readily
catalytic since they resist hydrogenolysis, which is required to
regenerate the starting hydride (Scheme 1, reaction A).10,11,14
A possible exception to this pattern is the hydroperoxide derived
from [HCo(CN)5]3-, which hydrolyzes to an aqua complex in
the presence of protons.15 This unusual case highlights the
The hydrogenation of dioxygen by conventional homo-
geneous catalysts is challenging. Many complexes that are
reactive toward H2 are rapidly and irreversibly oxidized upon
treatment with O2.6 Complementarily, complexes that are
(7) Katsuki, T. In ComprehensiVe Coordination Chemistry II; McCleverty, J.
A., Meyer, T. J., Eds.; Elsevier Pergammon: Amsterdam, 2004; Vol. 9,
pp 207-264.
(8) (a) Cui, W.; Wayland, B. B. J. Am. Chem. Soc. 2006, 128, 10350-10351.
(b) Wick, D. D.; Goldberg, K. I. J. Am. Chem. Soc. 1999, 121, 11900-
11901. (c) Thyagarajan, S.; Incarvito, C. D.; Rheingold, A. L.; Theopold,
K. H. Chem. Commun. 2001, 2198-2199.
(1) Casey, C. P.; Johnson, J. B. J. Org. Chem. 2003, 68, 1998-2001.
(2) (a) Cammack, R., Frey, M., Robson, R., Eds. Hydrogen as a Fuel: Learning
from Nature; Taylor & Francis, London, 2001. (b) Buhrke, T.; Lenz, O.;
Krauss, N.; Friedrich, B. J. Biol. Chem. 2005, 280, 23791-23796. (c) Kanai,
R.; Miyachi, S.; Takamiya, A. Nature 1960, 188, 873-875.
(3) (a) Dobereiner, J. J. Chem. (Schweigger) 1823, 38, 321-326. (b) Dobere-
iner, J. Ann. Phys. (Gilbert) 1823, 74, 269-273. (c) Collins, P. M. D.
Platinum Metals ReV. 1986, 30, 141-146.
(4) Basic Research Needs for the Hydrogen Economy; Report of the Basic
Energy Sciences Workshop on Hydrogen Production, Storage and Use;
U.S. Department of Energy, May 13-15, 2003 (http://www.sc.doe.gov/
bes/hydrogen.pdf).
(9) Denney, M. C.; Smythe, N. A.; Cetto, K. L.; Kemp, R. A.; Goldberg, K.
I. J. Am. Chem. Soc. 2006, 128, 2508-2509.
(10) James, B. R.; Morris, R. H.; Kvintorics, P. Can. J. Chem. 1986, 64, 897-
903.
(11) Keith, J. M.; Muller, R. P.; Kemp, R. A.; Goldberg, K. I.; Goddard, W.
A., III; Oxgaard, J. Inorg. Chem. 2006, 45, 9631-9633.
(12) Konnick, M. M.; Gandhi, B. A.; Guzei, I. A.; Stahl, S. S. Angew. Chem.,
Int. Ed. 2006, 45, 2904-2907.
(13) Popp, B. V.; Stahl, S. S. J. Am. Chem. Soc. 2007, 129, 4410-4422.
(14) (a) Shen, J.; Stevens, E. D.; Nolan, S. P. Organometallics 1998, 17, 3875-
3882. (b) Johnston, L. E.; Page, J. A. Can. J. Chem. 1969, 47, 4241-
4246. (c) Jime´nez-Tenorio, M.; Puerta, M. C.; Valerga, P. J. Am. Chem.
Soc. 1993, 115, 9794-9795. (d) Bianchini, C.; Mealli, C.; Meli, A.;
Proserpio, D. M.; Peruzzini, M.; Vizza, F.; Frediani, P. J. Organomet. Chem.
1989, 369, C6-C10. (e) Atlay, M. T.; Preece, M.; Strukul, G.; James, B.
R. Can. J. Chem. 1983, 61, 1332-1338.
(5) (a) Fei, J.; Song, H.-K.; Palmore, G. T. R. Chem. Mater. 2007, 19, 1565-
1570. (b) Boulatov, R. Pure Appl. Chem. 2004, 76, 303-319.
(6) Kubas, G. Metal Dihydrogen and σ-Bond Complexes; Kluwer Academic/
Plenum Publishers: New York, 2001.
9
10.1021/ja073774p CCC: $37.00 © 2007 American Chemical Society
J. AM. CHEM. SOC. 2007, 129, 14303-14310
14303

A R T I C L E S
Heiden and Rauchfuss
separate compartments connected by a load-bearing circuit.4 For
the present report, however, we focus exclusively on the
oxidative reactivity of iridium-based transfer hydrogenation
catalysts in a homogeneous solution.
Scheme 1. Differing Reactivity Patterns Anticipated for Oxygen
toward Classical Metal and Amino-Hydrides
Results
We found that the 18e amino-hydride Cp*IrH(TsDPEN)
reacts with molecular oxygen to give Cp*Ir(TsDPEN-H) (1)
and 1 equiv of water (eq 1). The conversion is signaled by a
color change from pale orange to purple, associated with 1H(H)
and 1, respectively.16 UV-vis experiments indicated that
intermediates do not accumulate (Figure 1).
potential benefits of catalysis in water, which facilitates scission
of M-OxH bonds (where x ) 1 or 2).
A relatively new generation of transfer hydrogenation catalysts,
which operate via heterolytic mechanisms, presents new options
for the regeneration of the metal hydride after formation of
hydroxides or hydroperoxides. In fact, these transfer hydrogena-
tion catalysts operate by the elimination of alcohols (and likely
water) from alkoxy amine intermediates, thus avoiding catalyst
deactivation.16-21,22,23 Given the fact that transfer hydrogenation
catalysts will not deactivate through formation of stable M-OR
species (Scheme 1, reaction B), we investigated them as possible
catalysts for the hydrogenation of O2.
Given that alkoxy (hydroxyl) intermediates will spontaneously
eliminate alcohol (water), one other reaction that is critical to
the development of catalysts for the hydrogenation of oxygen
is the hydrogenation step. The direct addition of H2 to transfer
hydrogenation catalysts has recently been examined.24 The
Mashima-Ikariya system Cp*Ir(TsDPEN-H) (1, TsDPEN )
rac-H2NCHPhCHPhN(SO2C6H4CH3)-) slowly adds H2 to give
the amine hydride Cp*IrH(TsDPEN) (1H(H)).16,18 We recently
reported that this reaction is accelerated by Brønsted acids such
as H(OEt2)2BArF (BArF - ) B(C6H3-3,5-(CF3)2)4-), which
Figure 1. UV-vis spectrum of an unstirred acetonitrile solution of
Cp*IrH(TsDPEN) in air, recorded at 10-min intervals. The absorption bands
arise from formation of Cp*Ir(TsDPEN-H). Inset: the effects of diffusion
of air (O2) into a CD3CN solution of Cp*IrH(TsDPEN). The red color is
due to Cp*Ir(TsDPEN-H).
4
4
converts 1 into the unsaturated iridium(III) amine cation
[Cp*Ir(TsDPEN)]+ ([1H]+, see Scheme 2).23
Scheme 2
The catalytic nature of the H2 + O2 reaction was formally
established stepwise, in a batch process. Thus, after conversion
of 1H(H) into 1 with oxygen, the apparatus was purged with
argon and then treated with 10 mol % of H(OEt2)2BArF to
4
generate some [1H]+. After 1 atm of H2 was introduced, the
solution color changed from purple to light orange over the
course of 15 h, indicative of 1H(H). The apparatus was then
purged with argon, and the addition of O2 induced a quantitative
change to purple 1 over the course of 30 min. Rehydrogenation
of the resulting 1 proceeded without any further addition of acid,
followed by dehydrogenation to 1 to demonstrate 2.5 turnovers.
Assuming the eventual development of new classes of oxygen
reduction catalysts, it will still be necessary in the future to
conduct O2 reduction and H2 (or other substrates) oxidation in
1
Each transformation was verified by H NMR spectroscopy.
(15) (a) Bayston, J. H.; Winfield, M. E. J. Catal. 1964, 3, 123-128. (b) Kwiatek,
J. Catal. ReV. 1967, 1, 37-72.
The progress of the catalytic hydrogenation of O2 is visually
obvious (Figure 2).
(16) Mashima, K.; Abe, T.; Tani, K. Chem. Lett. 1998, 1201-1202.
(17) Mashima, K.; Abe, T.; Tani, K. Chem. Lett. 1998, 1199-1200.
(18) Murata, K.; Ikariya, T.; Noyori, R. J. Org. Chem. 1999, 64, 2186-2187.
(19) (a) Clapham, S. E.; Hadzovic, A.; Morris, R. H. Coord. Chem. ReV. 2004,
248, 2201-2237. (b) Noyori, R.; Sandoval, C. A.; Muniz, K.; Ohkuma, T.
Philos. Trans. R. Soc. London, Ser. A 2005, 363, 901-912. (c) Yamakawa,
M.; Ito, H.; Noyori, R. J. Am. Chem. Soc. 2000, 122, 1466-1478.
(20) Ikariya, T.; Murata, K.; Noyori, R. Org. Biomol. Chem. 2006, 4, 393-
406.
The rate of dehydrogenation of 1H(H) by O2 was found to
be highly solvent-dependent. Under typical conditions using 1
atm of O2, reactions were complete within the time of mixing
in benzene. Reactions required ∼1 h in acetonitrile and several
minutes in dichloromethane. The dehydrogenation of solutions
of 1H(H) was also found to proceed in air but slowly (solid
samples of 1H(H) were dehydrogenated by air over a period of
months). We conducted most experiments in MeCN or CH2Cl2
solutions, wherein reactions proceeded at rates convenient for
mechanistic analysis.
(21) Wu, X.; Xiao, J. Chem. Commun. 2007, 2449-2466.
(22) Many attempts at isolation of metal amine alkoxides have been unsuccess-
ful: Koike, T.; Ikariya, T. Organometallics 2005, 24, 724-730.
(23) Heiden, Z. M.; Rauchfuss, T. B. J. Am. Chem. Soc. 2006, 128, 13048-
13049.
(24) Haack, K.-J.; Hashiguchi, S.; Fujii, A.; Ikariya, T.; Noyori, R. Angew.
Chem., Int. Ed. Engl. 1997, 36, 285-288.
9
14304 J. AM. CHEM. SOC. VOL. 129, NO. 46, 2007

O2 Reduction by Cp*IrH(TsDPEN)
A R T I C L E S
-
4
-
Figure 2. Sequential exposure of a methylene chloride solution of Cp*IrH(TsDPEN) and 10 mol % of H(OEt2)2BArF (BArF ) B(C6H3-3,5-(CF3)2)4
)
4
to O2 and H2. Racemic mixtures of TsDPEN were used; Cp*IrH(S,S-TsDPEN) and Cp*Ir(S,S-TsDPEN-H) are shown only as an example.
Figure 3. Turnover plot for the reaction of Cp*Ir(TsDPEN-H) with 17.5
mol % of [Cp*Ir(TsDPEN)]BArF4, 0.30 atm of H2, and 0.18 atm of O2 in
CD2Cl2. TON is defined as moles of H2O per mole of Cp*Ir(TsDPEN-H).
Figure 5. Plot indicating second-order dependence of the reaction of
Cp*IrH(TsDPEN) with O2 in CD3CN solution. The deviation at early
reaction times is attributed to incomplete temperature equilibration following
rapid thawing of the sample (77-293 K, ∼3 min).
presence of 0.56 atm of D2 and 0.35 atm of O2. The turnover
number for these reactions was limited by mass transfer, i.e.,
mixing, which is important in processes involving gaseous and
dissolved reagents (see below).
Kinetics and Isotope Effects. Studies of the reaction rate
for 1H(H) + O2 were conducted under conditions of excess,
constant PO ([O2]:[1H(H)] > 10). The solvent CD3CN was
2
selected to intentionally slow the reaction to allow for kinetic
analysis (only 8-10 data points could be obtained using
CD2Cl2). An 85-95% conversion of 1H(H) to 1 was observed
(Figure 4). The rate of the dehydrogenation, -d[1H(H)]/dt,
followed a second-order dependence on [1H(H)], as indicated
by the linearity of a plot of 1/[1H(H)] vs time (Figure 5). A
first-order dependence on [O2] was indicated by the observation
of a doubling of the second-order rate constant when [O2] was
doubled. With a first-order dependence on [O2], an overall rate
constant of 0.024 ( 0.013 M-2 s-1 (20.0 °C) was determined
over three half-lives. The presence of 2 equiv of the radical
traps BHT and TEMPO did not affect the yield of 1 by the
reaction of O2 with 1H(H).26,27 The rates of reduction of O2
Figure 4. Time course for the reaction of Cp*IrH(TsDPEN) with O2 in
CD3CN solution. Two minutes were required to set up the data collection.
In the presence of 10 mol % of [Cp*Ir(TsDPEN)]BArF
4
([1H]BArF ), a CD2Cl2 solution of 1 under 0.30 atm of H2 and
4
0.13 atm of O2 produced 4.26 equiv of water over the course of
ca. 300 h (Figure 3).25 The formation of water (D2O) was veri-
fied by 2H NMR spectroscopy for the reaction of 1 with 15 mol %
of [Cp*Ir(TsDPEN)]BArF ([1H]BArF ) in CD2Cl2 in the
4
4
(26) Kirillov, A. M.; Kopylovich, M. N.; Kirillova, M. V.; Karabach, E. Y.;
Haukka, M.; Guedes da Silva, M. F. C.; Pombeiro, A. J. L. AdV. Synth.
Catal. 2006, 348, 159-174.
(27) (a) Burton, G. W.; Ingold, K. U. J. Am. Chem. Soc. 1981, 103, 6472-
6477. (b) Ingold, K. U. Chem. ReV. 1961, 61, 563-589.
(25) Although the dehydrogenation of 1H(H) by O2 in benzene-d6 was rapid,
catalytic hydrogenation was problematic because [1H]BArF is insoluble
in benzene-d6, slowing the hydrogenation of 1 to 40% conve4rsion over 30
vs 16 h, for complete conversion in CD2Cl2.
9
J. AM. CHEM. SOC. VOL. 129, NO. 46, 2007 14305

A R T I C L E S
Heiden and Rauchfuss
were slower only by factors of 2× and 4× for BHT and
TEMPO, respectively, which we attribute to medium effects
since the reduction of O2 is sensitive to solvent polarity. The
rate of the reaction was evaluated over the temperature range
-15 to 20 °C. The temperature dependence of the rate constants
indicated ∆Hq ) 82.8 kJ/mol and ∆Sq ) 13.2 J/mol‚K, for a
net ∆Gq ) 78.9 kJ/mol at 298 K (see Experimental Section).
Isotope effects were examined to probe the relative roles of
N-H and Ir-H in the reactions. In their study on the reduction
of acetophenone by [2,5-Ph2-3,4-Tol2(C4COH)]RuH(CO)2 in
CD2Cl2 solution, Casey and co-workers measured isotope effects
of 1.2 for deuteration at the OH and 2.2 for deuteration at
the hydride.28 Casey and co-workers also estimated the iso-
tope effects for the reduction of acetone using equilibrium
measurements from the dehydrogenation of isopropanol with
(p-cymene)Ru(TsDPEN-H), resulting in isotope effects of 1.6
for deuteration at the amine and 2.4 for deuteration at the
hydride.1 We found a more dramatic isotope effect for the
of 1 in the presence of 11 atm of O2 and t-BuNH2BH3 in CD2Cl2
gave H2 and H2O, which was confirmed by 1H NMR spectros-
copy. No t-BuNH2BH3 was found to remain in the solution after
1
6 h, and the presence of 1 was verified by H NMR spectros-
copy. Quantification of turnover from H2 and H2O was
precluded by the reaction of the cogenerated water with
t-BuNH2-xBH3-x. Hydrolysis of N-alkylborazoles to amines and
boric acid has been reported.35
Alternative Hydrogen Donors. 2. Alcohols. Alcohols have
been recently employed as hydrogen donors in fuel cells36 and
are well known to convert 1 into 1H(H), concomitant with
formation of ketones and aldehydes.16 In conjunction with the
reactivity of 1H(H) toward O2, this dehydrogenation step defines
a catalytic cycle (eqs 3 and 4).
Cp*Ir(TsDPEN-H) + RCH2OH f
Cp*IrH(TsDPEN) + RCHO (3)
Cp*IrH(TsDPEN) + 0.5O2 f
Cp*Ir(TsDPEN-H) + H2O (4)
reduction of O2: deuteration of the hydride gave kHH /kDH
)
2
2
6.0 ( 1.3. An isotope effect of similar magnitude was observed
by Goldberg and coworkers (kH/kD ) 5.8) for the oxygenation
of an organopalladium hydride with dioxygen.9 Deuteration of
the amine of 1H(H) was found to have a relatively minor effect
(kHH /kHD ) 1.2 ( 0.2).
To establish proof of concept, we examined the oxidation of
amyl alcohol in the presence of 5 mol % of 1. Under 3.5 atm
of O2, pentanal was produced, albeit with only 4-5 turnovers.
Oxygen pressures greater than 2.5 atm ([O2]:[1] > 2:1) were
found to have little effect on turnover number (TON) or turnover
frequency. TON was defined as moles of aldehyde per mole of
1. The presence of 25 mol % of triethylamine also had no effect.
Doubling the catalyst loading from 5 to 10 mol % was found
to reduce the TON by a factor of 2.37
We surveyed other hydrogen-transfer catalysts. Ir-based
catalysts were found to be slightly more active in the case of
alcohol oxidation than the Ru derivatives, and the Rh derivative
was the least active. Shvo’s catalyst, {[(η5-Ph4C4CO)]2(µ-H)}-
Ru2(µ-H)(CO)4,38 was also found to catalyze the oxidation of
pentanol and proved more robust than the Mashima-Ikariya
catalyst (Table 1, see also Supporting Information).
The efficiency of the alcohol dehydrogenation reaction was
significantly enhanced by the presence of p-benzoquinone (BQ)
in addition to O2. Thus, using 50 mol % of BQ, 2 atm of O2,
and still 5 mol % of 1, we observed oxidation of 10 equiv of
alcohol over the course of 12 h.39 BQ functions as a hydrogen
acceptor that augments the action of O2. Alcohol oxidation
experiments involving BQ and air were only half as effective
as those with BQ and pure O2. Separate experiments (see below)
demonstrated that BQ rapidly dehydrogenates 1H(H).
2
2
Alternative Hydrogen Donors. 1. Amine-Boranes. We
sought to test the ability of 1 to catalyze the oxidation of amine-
boranes, which have attracted recent attention as hydrogen
sources for possible applications in fuel cells.29 Several metal
complexes have been shown to catalyze the dehydrogenation
of NH3BH3.30 We first investigated the dehydrogenation of
NH3BH3 by 1. We confirmed that NH3BH3 rapidly converted
2 equiV of 1 into 1H(H) in a matter of seconds (eq 2). Excess
NH3BH3 converted 1H(H) into [(Cp*Ir)2(µ-H)3]+, arising from
protonolysis of the TsDPEN ligand. We have previously
reported that this hydrogenolysis is accelerated by protic
reagents.23
2Cp*Ir(TsDPEN-H) + NH3BH3 f
2Cp*IrH(TsDPEN) + 1/n(NHBH)n (2)
The efficient conversion of 1 into 1H(H) by NH3BH3 led us
to couple this dehydrogenation with the reduction of O2. Due
to the low solubility of NH3BH3 in organic solvents,31,32 we
used t-BuNH2BH in O2-coupled reactions.33,34 Using 4 mol %
(28) Casey, C. P.; Johnson, J. B. Can. J. Chem. 2005, 83, 1339-1346.
(29) (a) Zhang, X.-B.; Han, S.; Yan, J.-M.; Chandra, M.; Shioyama, H.; Yasuda,
K.; Kuriyama, N.; Kobayashi, T.; Xu, Q. J. Power Sources 2007, 168,
167-171. (b) Langmi, H. W.; McGrady, G. S. Coord. Chem. ReV. 2007,
251, 925-935.
Dehydrogenation of 1H(H) by Other Oxidants. Experi-
ments were conducted to test the viability of H2O2 as an
(30) (a) Denney, M. C.; Pons, V.; Hebden, T. J.; Heinekey, D. M.; Goldberg,
K. I. J. Am. Chem. Soc. 2006, 128, 12048-12049. (b) Yoon, C. W.;
Sneddon, L. G. J. Am. Chem. Soc. 2006, 128, 13992-13993. (c) Bluhm,
M. E.; Bradley, M. G.; Butterick, R., III; Kusari, U.; Sneddon, L. G. J.
Am. Chem. Soc. 2006, 128, 7748-7749. (d) Keaton, R. J.; Blacquiere, J.
M.; Baker, R. T. J. Am. Chem. Soc. 2007, 129, 1844-1845. (e) Chandra,
M.; Xu, Q. J. Power Sources 2006, 156, 190-194. (f) Clark, T. J.; Russell,
C. A.; Manners, I. J. Am. Chem. Soc. 2006, 128, 9582-9583. (g) Clark, T.
J.; Whittell, G. R.; Manners, I. Inorg. Chem. 2007, 46, 7522-7527.
(31) Meller, A.; Schaschel, E. Inorg. Nucl. Chem. Lett. 1966, 2, 41-43.
(32) (a) Ito, K.; Watanabe, H.; Kubo, M. Bull. Chem. Soc. Jpn. 1960, 33, 1588-
1590. (b) Ohashi, O.; Kurita, Y.; Totani, T.; Watanabe, H.; Nakagawa, T.;
Kubo, M. Bull. Chem. Soc. Jpn. 1962, 35, 1317-1321. (c) Stephens, F.
H.; Pons, V.; Baker, R. T. Dalton Trans. 2007, 2613-2626.
(35) (a) Yoshizaki, T.; Watanabe, H.; Nakagawa, T. Inorg. Chem. 1968, 7, 422-
429. (b) Nagasawa, K. Inorg. Chem. 1966, 5, 442-445. (c) Atkinson, I.
B.; Blundell, D. C.; Clapp, D. B. J. Inorg. Nucl. Chem. 1972, 34, 3037-
3041. (d) Smith, B. C.; Thakur, L.; Wassef, M. A. J. Chem. Soc. A 1967,
1616-1618.
(36) (a) Meng, H.; Wu, M.; Hu, X. X.; Nie, M.; Wei, Z. D.; Shen, P. K. Fuel
Cells 2006, 6, 447-450. (b) Lamy, C.; Lima, A.; LeRhun, V.; Delime, F.;
Coutanceau, C.; Leger, J.-M. J. Power Sources 2002, 105, 283-296.
(37) The reduction of turnover number for a greater catalyst-to-substrate ratio
is indicative of a catalyst deactivation mechanism different than that
observed for the dehydrogenation of ammonia-boranes and 1H(H).
(38) Shvo, Y.; Czarkie, D.; Rahamim, Y.; Chodosh, D. F. J. Am. Chem. Soc.
1986, 108, 7400-7402.
(33) Girolami, G. S.; Rauchfuss, T. B.; Angelici, R. J. Synthesis and Technique
in Inorganic Chemistry; University of Science Books: Sausalito, CA, 1999.
(34) t-BuNH2BH3 was favored over NH3BH3 due to its high solubility in CH2Cl2
and the ability to use 13C NMR as an additional characterization tool for
the ammonia-borane dehydrogenated product.
(39) The ability of BQ to oxidize alcohols prompted an examination of its
hydrogenation by 1H(H). In the presence of 10% H(OEt2)2BArF4, 1H(H)
indeed catalyzes the hydrogenation of BQ to p-C6H4(OH)2 (HQ). At ∼1
atm of H2, in a methylene chloride solution, 5 equiv of BQ was converted
to HQ over the course of 12 h.
9
14306 J. AM. CHEM. SOC. VOL. 129, NO. 46, 2007

O2 Reduction by Cp*IrH(TsDPEN)
A R T I C L E S
Table 1. Effect of Precatalyst on Oxidation of Amyl Alcohol Using
Molecular Oxygen
conversion (Figure 4). This reaction produces a minor product
(2) in 5-15% yield that could be observed by NMR spectros-
copy. Compound 2 was also detected in the reaction of H2O2
and t-BuOOH with both 1 and 1H(H). The yield of 2 was found
to depend on the initial concentration of 1H(H), but was largely
insensitive to [O2]. This finding suggests that the degradation
is competitive with reduction of an intermediate (1H(OOH),
see Conclusions and Mechanistic Considerations) with 1H(H).
1H NMR spectra confirm that 2 contains intact TsDPEN: the
position and multiplicities of the NCH resonances indicate
equatorial phenyl groups in pseudo-octahedral complexes of the
type Cp*IrX(TsDPEN) or [Cp*IrL(TsDPEN)]+.23,44 The Cp*
group has, however, clearly suffered attack, as indicated by an
AB-quartet (2H, δ 2.7-3.4), which is characteristic of a fulvene
derivative.45 Three Cp* methyl groups can be observed (3H
each, see Supporting Information).46 Maitlis, Kirchner, and
Sharp have previously demonstrated the ability of O2 to convert
Cp*M centers into related tetramethylfulvene complexes,42,47,48
which in turn can undergo further activation of the methyl
groups.48 The degradation product 2 is therefore assigned the
formula [η6-(C5Me3(CH2OH)CH2)Ir(TsDPEN)] (Scheme 3).
Compound 2 is proposed to arise via intramolecular H-atom
abstraction from the Cp* methyl by an intermediate hydro-
peroxide. Degradation is competitive with the second-order
reaction of the hydroperoxide with 1H(H). Consistent with this
scenario, the relative yield of 2 was found to increase at high
dilutions (Supporting Information). Further characterization of
2 was precluded by its limited stability. Similar C-H bond
hydroxylation has been observed in studies of copper hydrop-
eroxo complexes.49
O2
pressure
(atm)
precatalyst
substrate
basea
solvent
TONb
Cp*Ir(TsDPEN-H)
Cp*Ir(S,S-TsDPEN-H)
Cp*IrCl(TsDPEN)
Cp*RhCl(TsDPEN)
(p-cymene)RuCl(TsDPEN) amyl alcohol NEt3 CD2Cl2
Shvo’s catalyst amyl alcohol CD2Cl2
amyl alcohol NEt3 CD2Cl2
amyl alcohol NEt3 CD2Cl2
amyl alcohol NEt3 CD2Cl2
amyl alcohol KOHc CD2Cl2
5
5
5
5
5
5
4.26
4.23
3.00
0.54
2.4
5.53d
a 25 mol % (60 µmol) of base relative to alcohol was used; 72 µmol of
base was used in the case of amino-chloride precatalysts. b 5 mol % of
precatalyst (12 µmol) and alcohol (240 µmol) under an O2 atmosphere in
0.8 mL of CD2Cl2 was monitored via 1H NMR for 36 h. c KOH (72 µmol)
was used in the Rh case due to the fact that NEt3 was not significantly
strong enough to dehydrohalogenate the amino-chloride. d 2.5 mol % of
Shvo’s dimer, {[(Ph4(η5-C4CO)]2(µ-H)}Ru2(µ-H)(CO)4, was used; TON is
reported after 29 h, where 53% of Shvo’s dimer remained.38
intermediate in the reduction of O2. No formation of H2O2 was
1
observed via H NMR spectroscopy (δ 8.84 in CD3CN) upon
treating a CD3CN solution of 1H(H) with excess O2 in an NMR
tube. A CD3CN solution of 1H(H) was found to rapidly react
with substoichiometric amounts of H2O2 (0.3 equiv) to give 1
as the main product. Larger amounts (2 equiv) of H2O2 were
found to destroy the organoiridium complexes, as indicated by
the loss of the Cp* and TsDPEN signals in the 1H NMR
spectrum. Treatment of 1H(H) with 1 equiv of t-BuOOH
resulted in the formation of 1, t-BuOH, and H2O (eq 5); further
equivalents of t-BuOOH resulted in destruction of the organo-
iridium complexes, as seen with excess H2O2.
Scheme 3
Cp*IrH(TsDPEN) + ROOH f
Cp*Ir(TsDPEN-H) + H2O + ROH
(R ) t-Bu, H) (5)
In addition to H2O2 and t-BuOOH, nitrosobenzene dehydro-
genated 1H(H) to give N-phenylhydroxylamine.41,42 1H(H) was
found to be dehydrogenated by diethyl azodicarboxylate and
nitric oxide. Azobenzene and nitrous oxide were inert with
respect to 1H(H). Surprisingly, the transfer hydrogenation of
quinones has been lightly studied.39,40 1,4-Benzoquinone rapidly
and cleanly dehydrogenated 1H(H) to give 1 and 1,4-dihy-
droxybenzene (eq 6).
Cp*IrH(TsDPEN) + X f Cp*Ir(TsDPEN-H) + H2X
(X ) p-C6H4O2, PhNO, N2(CO2Et)2) (6)
Catalyst Deactivation. For both the catalytic hydrogenation
of O2 and the oxidations of pentanol, the 1H(H) turned over
ca. 4-10×. We note again that reactions were optimized for
product analysis (e.g., NMR tubes), not turnovers. Nonetheless,
we sought insights into the weaknesses in this catalytic system.
Consistent with the proposed assignment, the degradation
product completely converted into 1H(H) in the presence of
(43) (a) White, C.; Oliver, A. J.; Maitlis, P. M. J. Chem. Soc., Dalton Trans.
1973, 1901-1907. (b) Gill, D. S.; White, C.; Maitlis, P. M. J. Chem. Soc.,
Dalton Trans. 1978, 617-626. (c) Gilbert, T. M.; Bergman, R. G. J. Am.
Chem. Soc. 1985, 107, 3502-3507.
(44) (a) Heiden, Z. M.; Gorecki, B. J.; Rauchfuss, T. B. Manuscript in
preparation. (b) Koike, T.; Ikariya, T. J. Organomet. Chem. 2007, 692,
408-419. (c) Heiden, Z. M.; Letko, C. S.; Rauchfuss, T. B. Manuscript in
preparation.
We confirmed that 1, [1H]BArF , and the hydrogenolysis
4
product of 1H(H) ([Cp*2Ir2H3]+) are stable toward O2.43 Careful
scrutiny showed, however, that the stoichiometric dehydro-
genation of 1H(H) by O2 to give 1 proceeded in only 85-95%
(45) Rais, D.; Bergman, R. G. Chem. Eur. J. 2004, 10, 3970-3978.
(46) 1H NMR (500 MHz, CD3CN): δ 0.75 (s, 3H), 1.48 (s, 3H), 2.02 (s, 3H),
2.36 (s, 3H), 2.77 (d, 1H, J ) 2.7 Hz), 3.40 (d, 1H, J ) 2.7 Hz), 3.87
(ddd, 1H, J ) 3.7, 11.1, 14 Hz, H2NCHPhCHPhNTs), 4.5 (br d, 1H,
HHNCHPhCHPhNTs), 4.80 (d, 1H, J ) 11.1 Hz, HHNCHPhCHPhNTs),
4.91 (br t, 1H, HHNCHPhCHPhNTs), 5.41 (s, 1H), 5.54 (s, 1H), 6.83-
7.38 (m, 14H).
(40) (a) Braude, E. A.; Linstead, R. P.; Mitchell, P. W. D.; Wooldridge, K. R.
H. J. Chem. Soc. 1954, 3595-3598. (b) Nishiguchi, T.; Kurooka, A.;
Fukuzumi, K. J. Org. Chem. 1974, 39, 2403-2405.
(41) Godbout, N.; Sanders, L. K.; Salzmann, R.; Havlin, R. H.; Wojdelski, M.;
Oldfield, E. J. Am. Chem. Soc. 1999, 121, 3829-3844.
(42) Hoard, D. W.; Sharp, P. R. Inorg. Chem. 1993, 32, 612-620.
9
J. AM. CHEM. SOC. VOL. 129, NO. 46, 2007 14307

A R T I C L E S
Heiden and Rauchfuss
Scheme 5. Proposed Catalytic Cycle for Hydrogenation of O2 by
1H(H)
t-BuNH2‚BH3, which explains the high TONs observed with
amine-boranes as substrates (Scheme 3). Maitlis has described
the conversion of an (η6-fulvene)Ru complex to its Cp*
derivatives using NaBH4 (Scheme 4).50
Scheme 4
Conclusions and Mechanistic Considerations
The transfer hydrogenation catalysts developed by Mashima
and Ikariya et al.16-18 exhibit unusual reactivity toward oxygen,
resulting in its catalytic hydrogenation to give water. Very few
complexes are reactive toward both O2 and H2. The required
unusual combination of mutually compatible reductive and
oxidative properties is, however, characteristic of the transfer
hydrogenation catalysts.
from consideration, although it is known to be highly reactive
toward organometallics.53,54 Previous work on the oxygenation
of metal hydrides also assumed O2(3Σg-), not O2(1∆g).11,13,51
3
1
Considerable prior work on 1H(H) and related catalysts has
emphasized reductions, although transfer dehydrogenations have
recently been coupled to oxidations. Ba¨ckvall et al. have
demonstrated aerobic oxidation of amines and alcohols using
Shvo’s catalyst.55,56 The Ba¨ckvall system uses a quinone as a
hydrogen acceptor and a Co-salen complex to aerobically
regenerate the quinone.56 In our case, the iridium complex affects
both the oxidation (dehydrogenation) of the alcohol and the
reduction (hydrogenation) of the oxygen. Recently, Xiao and
co-workers have described Ru and Ir amido-amine catalysts that
are effective transfer hydrogenation catalysts in air.21,57 The
chemistry described here shows that O2 would not be expected
to inhibit the transfer hydrogenation of organic substrates,
instead serving as a mildly competing hydrogen acceptor.
The second-order dependence on [1H(H)] is consistent with
the rate-determining reduction of a hydroperoxo complex
1H(OOH) by 1H(H). The resulting hydroxy amine
Cp*Ir(OH)(TsDPEN) would spontaneously eliminate water,
affording 1, which is poised for reduction by H2 (Scheme 5).
The formation of 1H(OOH) is consistent with the large
difference in the rate of oxidation of Cp*IrD(TsDPEN) vs
Cp*IrH(TsDPEN). The disparate magnitudes of the isotope
effects for deuteration at the amine and the hydride argue against
a concerted mechanism (cyclic transition state), which is
accepted to occur in the transfer hydrogenation of ketones.1,20
Instead, the reactivity is localized at the Ir-H center, analogous
to the reaction of O2 with palladium hydrides.11,13,51 The
efficiency of the O2 + 1H(H) reaction was unaffected by radical
traps,26 which argues against the involvement of homolytic
pathways. We propose that 1H(OOH) and 1H(H) combine to
produce 1, H2O, and 1H(OH). An analogous result was observed
in the treatment of 1H(H) with t-BuOOH, which was found to
produce 1, H2O, and t-BuOH. A key aspect of the mechanism
is the elimination of water from 1H(OH), a reaction akin to the
elimination of alcohol in transfer hydrogenation.
In summary, Ir-based transfer hydrogenation catalysts have
been found to catalyze the hydrogenation of dioxygen using
H2 as well as other hydrogen donors. An intermediate iridium
hydroperoxo complex is indicated. Modifications of the co-
ligands could be expected to confer greater oxidative stability
to these catalysts.58 Furthermore, we have found that counter-
anions (of [1H]+) significantly affect the hydrogenation of O2.59
The significance of this work lies in the use of unconventional
metal hydrides for oxygen reduction. In future work, we will
The solvent effects observed in the reduction of O2 by 1H(H)
are believed to be indicative of ion pairing and possible
aggregation in solution.52 Singlet oxygen (1O2) was excluded
(52) (a) Macchioni, A. Chem. ReV. 2005, 105, 2039-2073. Zuccaccia, D.; Clot,
E.; Macchioni, A. New J. Chem. 2005, 29, 430-433. (b) Ciancaleoni, G.;
Di Maio, I.; Zuccaccia, D.; Macchioni, A. Organometallics 2007, 26, 489-
496.
(53) Selke, M.; Foote, C. S. J. Am. Chem. Soc. 1993, 115, 1166-1167.
(54) Singlet oxygen was also excluded due to its low concentration in the gas
phase ([1O2]:[3O2] ) 2.19 × 10-19:1): (a) Ellis, J. W.; Kneser, H. O. Z.
Physik 1933, 86, 583-591. (b) Long, C.; Kearns, D. R. J. Chem. Phys.
1973, 59, 5729-5736.
(47) Kreindlin, A. Z.; Rybinskaya, M. A. Russ. Chem. ReV. 2004, 73, 417-
432.
(48) (a) Fan, L.; Wei, C.; Aigbirhio, F. I.; Turner, M. L.; Gusev, O. V.;
Morozova, L. N.; Knowles, D. R. T.; Maitlis, P. M. Organometallics 1996,
15, 98-104. (b) O’Leary, S. R.; Adams, H.; Maitlis, P. M. Chem. Commun.
1997, 895-896. (c) Gemel, C.; Mereiter, K.; Schmid, R.; Kirchner, K.
Organometallics 1997, 16, 5601-5603.
(49) (a) Cole, A. P.; Mahadevan, V.; Mirica, L. M.; Ottenwaelder, X.; Stack,
T. D. P. Inorg. Chem. 2005, 44, 7345-7364. (b) Battaini, G.; Monzani,
E.; Perotti, A.; Para, C.; Casella, L.; Santagostini, L.; Gullotti, M.;
Dillinger, R.; Naether, C.; Tuczek, F. J. Am. Chem. Soc. 2003, 125, 4185-
4198.
(55) Prabhakaran, R. Synlett 2004, 2048-2049.
(56) (a) Samec, J. S. M.; Ell, A. H.; Ba¨ckvall, J.-E. Chem. Eur. J. 2005, 11,
2327-2334. (b) Wang, G. Z.; Andreasson, U.; Ba¨ckvall, J. E. J. Chem.
Soc., Chem. Commun. 1994, 1037-1038. (c) Csjernyik, G.; Ell, A. H.;
Fadini, L.; Pugin, B.; Ba¨ckvall, J.-E. J. Org. Chem. 2002, 67, 1657-1662.
(d) Almeida, M. L. S.; Beller, M.; Wang, G.-Z.; Ba¨ckvall, J.-E. Chem.
Eur. J. 1996, 2, 1533-1536.
(57) (a) Wu, X.; Liu, J.; Li, X.; Zanotti-Gerosa, A.; Hancock, F.; Vinci, D.;
Ruan, J.; Xiao, J. Angew. Chem., Int. Ed. 2006, 45, 6718-6722. (b) Li,
X.; Blacker, J.; Houson, I.; Wu, X.; Xiao, J. Synlett 2006, 1155-1160. (c)
Wu, X.; Li, X.; Hems, W.; King, F.; Xiao, J. Org. Biomol. Chem. 2004, 2,
1818-1821.
(50) Fan, L.; Turner, M. L.; Hursthouse, M. B.; Malik, K. M. A.; Gusev, O. V.;
Maitlis, P. M. J. Am. Chem. Soc. 1994, 116, 385-6.
(51) Keith, J. M.; Nielsen, R. J.; Oxgaard, J.; Goddard, W. A., III. J. Am. Chem.
Soc. 2005, 127, 13172-13179.
9
14308 J. AM. CHEM. SOC. VOL. 129, NO. 46, 2007

O2 Reduction by Cp*IrH(TsDPEN)
A R T I C L E S
more explicitly investigate the role of a proton donor (the
coordinated amine) adjacent to the site of oxygen reduction.
Experimental Section
The preparations of 1H(H) and 1 are optimized procedures of
previously described procedures and are presented in the Supporting
Information.16,18,24 Hydrogen (research grade purity, 99.9999%) and
oxygen (99.95%) were used as received from S. J. Smith Co. Seven-
inch J. Young tubes were each fitted with a Wilmad-LabGlass 535-
PP-7 Precision thin-walled NMR tube (o.d. ) 5 mm, i.d. ) 4.24 mm,
and wall thickness ) 0.38 mm) and were used for all NMR-tube-based
reactions. Thin-walled tubes can withstand pressures of ∼20 atm before
fracture. For pressures >15 atm, medium- and thick-walled tubes are
recommended for elevated pressure analysis. All 11B NMR chemical
shifts were referenced to an external BF3‚Et2O standard.
Caution and Warnings: Reactions inVolVing H2 and O2 are danger-
ous! Gas bulbs were filled, with caution, behind blast shields. Pressures
were carefully monitored, especially in experiments inVolVing conden-
sation of O2! No flames were used in sealing NMR tubes. All NMR
tubes were carefully thawed oVer the course of seVeral minutes, from
the top down, before insertion into the spectrometer and transferred
while frozen at liquid nitrogen temperatures.
Figure 6. Temperature dependence of the third-order rate constant for the
reaction Cp*IrH(TsDPEN) + O2 in CD3CN solution.
Dehydrogenation of 1H(H) with t-BuOOH. A 0.75-mL CD3CN
solution of 7.5 mg (10.8 µmol) of 1H(H) and 1.2 mg (7.4 µmol) of
C6Me6 (internal standard) was treated with a 1.4 µL (10.2 µmol) of a
70 wt % solution of t-BuOOH in water, after an initial 1H NMR
spectrum was obtained. A slow color change from pale orange to
reddish-purple was observed over 20 min. Appearance of 1, H2O, and
t-BuOH (δ 1.70) was first observed after 60 s, and continued to grow
over the 20-min time period (referenced to the internal standard).
Additions of further equivalents were found to destroy the organoiridium
complexes and result in a black solution. Compound 2 was found to
form in about 5% yield.
Sequential Addition of H2 and O2. A 25-mL CH2Cl2 solution of
70 mg (101 µmol) of 1H(H) was exposed to an atmosphere of O2 for
30 min. A color change from pale orange to reddish-purple was
observed after 10 min. A 10-mL CH2Cl2 solution of 9.8 mg (9.7 µmol)
of H(OEt2)2BArF was added to the 1 solution. An atmosphere of H2
4
was then introduced for 15 h, resulting in a color change from reddish-
purple to orange-pink. Oxygen was then introduced for 30 min, resulting
in a reddish-purple solution. Sequential exposure to H2 and O2 was
again implemented to demonstrate 2.5 turnovers (Figure 2). The
apparatus was purged with argon for 30 min after each gas addition.
Samples of 0.5 mL volume were removed before addition of Ar and
every 2 h in the presence of H2 for analysis by 1H NMR spectroscopy.
Note: Argon was used to purge the apparatus between additions of H2
or O2 to avoid the presence of explosive concentrations of H2 and O2.
Water Production. A 0.8-mL CD2Cl2 solution of 7.5 mg (10.8
Kinetic Analysis for the Reaction of Cp*IrH(TsDPEN) + O2. A
0.8 mL CD3CN solution of 5 mg (7.2 µmol) of 1H(H), 4 mg (24.7
µmol) of C6Me6 (internal standard), and any other additives (e.g., radical
traps, 14.4 µmol) was prepared under vacuum in a 7-in. J. Young NMR
tube. An initial NMR spectrum was obtained before the addition of O2
(delay time of 15 s) (Supporting Information, Table S1). Oxygen was
transferred and condensed under vacuum using liquid N2 and a bulb
containing about 1 atm of O2. The pressure in the J. Young tube was
monitored using a manometer and calculated using a mole balance
(Supporting Information). An oxygen concentration of 0.10 M was
estimated using the Henry’s law constant for CH3CN (203.8 MPa).60
The affect of solvent deuteration on gas solubility was assumed to be
negligible. The sample was carefully thawed, from the top of the solvent
down, over the course of 2 min before insertion into the spectrometer
from 77 K. Insertion of the sample into the spectrometer was taken as
time ) 0. The sample was a pale orange color upon insertion. Between
120 and 150 s elapsed during the time of insertion and start of data
collection. Data were collected in a pre-acquisition delay array for 60
min, where a spectrum was recorded every 15 s with a pre-acquisition
delay time of 5 s, acquisition time of 5 s, and a post-pulse delay time
of 5 s at a temperature of 293.0 K. The sample was spun at a rate of
20 rpm for the duration of the experiment. A dark purple solution,
characteristic of the presence of 1, was observed upon ejection from
the spectrometer. Rates were determined by following the disappearance
of the Cp* signal in accordance with the signal for C6Me6 (δ 2.19).
The first 110 of the 241 data points were used in the kinetic analysis,
accounting for three half-lives. [O2] dependence was determined to be
first-order from the initial kinetic analysis. Tight seals of J. Young tubes
can be easily analyzed, where leakage of O2 pressure from J. Young
tubes can be verified by NMR spectrometer lock migration and eventual
loss of the lock of the sample and poor data over the course of the
experiment.
µmol) of 1, 3.0 mg (1.9 µmol) of Cp*Ir(TsDPEN)BArF4 ([1H]BArF ),
4
and 3.5 mg (21.6 mmol) of C6Me6 (internal standard) was prepared
under vacuum in a 7-in. J. Young tube, resulting in a reddish-purple
solution. After an initial 1H NMR spectrum was acquired, an atmosphere
of 0.30 atm of H2 and 0.18 atm of O2 was introduced under vacuum
using a premixed bulb containing both H2 and O2 in a 5:2 ratio by
pressure. No immediate changes were observed upon thawing. The
sample was monitored for 310 h by 1H NMR spectroscopy (Figure 3).
Production of D2O. A 0.8-mL CH2Cl2 solution of 11.7 mg (16.9
µmol) of 1, 3.9 mg (2.5 µmol) of Cp*Ir(TsDPEN)BArF4 ([1H]BArF ),
4
and 15 µL (169.3 µmol) of benzene-d6 (internal standard) was prepared
under vacuum in a 7-in. J. Young tube, resulting in a reddish-purple
solution. After an initial 2H NMR spectrum was acquired, an atmosphere
of 0.56 atm of D2 and 0.35 atm of O2 was introduced. No immediate
changes were observed upon thawing. The sample was monitored for
2
310 h by H NMR spectroscopy.
Dehydrogenation of 1H(H) with Dilute H2O2. A 1.0-mL CD3CN
solution of 3.0 mg (4.3 µmol) of 1H(H) and 5.9 mg (36.4 µmol) of
C6Me6 (internal standard) was treated with a 5-µL aliquot of a 0.244
1
M (1.2 µmol) solution of H2O2 in water/CD3CN, after an initial H
NMR spectrum was obtained. An immediate color change from pale
orange to reddish-purple was observed. Addition of 5 µL of the H2O2
solution was repeated five times, resulting in a darker purple color with
each subsequent addition. A 1H NMR spectrum was obtained after each
H2O2 addition, showing the formation of 1 and ∼30% of compound
2.46
Dehydrogenation of Ammonia-Borane. A 0.8-mL CD3CN solu-
tion of 15.5 mg (22.4 µmol) of 1 and 1.0 mg (6.2 µmol) of C6Me6
(58) (a) Matharu, D. S.; Morris, D. J.; Clarkson, G. J.; Wills, M. Chem. Commun.
2006, 3232-3234. (b) Hannedouche, J.; Clarkson, G. J.; Wills, M. J. Am.
Chem. Soc. 2004, 126, 986-987.
(60) Horstmann, S.; Grybat, A.; Kato, R. J. Chem. Thermodyn. 2004, 36, 1015-
(59) Heiden, Z. M.; Rauchfuss, T. B. Manuscript in preparation.
1018.
9
J. AM. CHEM. SOC. VOL. 129, NO. 46, 2007 14309

A R T I C L E S
Heiden and Rauchfuss
(internal standard) was treated with 350 µL of a 0.0324 M (11.3 µmol)
solution of RNH2BH3 (where R ) H, t-Bu). An immediate color change
from reddish-purple to yellow-orange was observed upon mixing. The
presence of 1H(H) and ∼5% [(Cp*Ir)2(µ-H)3]+ was verified by 1H NMR
spectroscopy. Exposure of the solution to air for 24 h resulted in
quantitative formation of 1 from 1H(H).
The values observed do not match with the values reported for tert-
butyl-borazole (1H, δ 1.36; 11B, δ 31.0), indicating the possibility of
hydrolysis (Supporting Information).31,62
Oxidation of Pentanol. A 0.75-mL CD2Cl2 solution of 5 mol % of
catalyst (relative to alcohol) and any other additives (e.g., quinone
cocatalyst (50 mol % relative to alcohol) or base (25 mol % relative to
alcohol)) in the presence of 4 mg (24.7 µmol) of C6Me6 (HMB, internal
standard) was prepared in a 7-in. J. Young tube in air and degassed
prior to oxygen introduction, resulting in a reddish-purple solution. An
initial NMR spectrum was obtained before the addition of O2. Oxygen
was added by condensation at liquid nitrogen temperatures under
vacuum as previously described. Upon addition of O2, reactions were
monitored for 36 h. TONs were determined by comparing the intensity
of the formyl signal at δ 9.74 vs that of the HMB signal (δ 2.22).
Catalytic Dehydrogenation of Ammonia-Borane. A 0.8-mL
CD2Cl2 solution of 8.6 mg (12.4 µmol) of 1, 6.0 mg (24.7 µmol) of
C6Me6 (internal standard), and 27.8 mg (320 µmol) of t-BuNH2BH3
was prepared under vacuum in a 7-in. J. Young NMR tube. Eleven
atmospheres of oxygen was transferred and condensed under vacuum
using liquid N2 as previously described. The pressure in the J. Young
tube was monitored using a manometer. The [O2] was estimated using
an Ostwald coefficient of 0.257 for CH2Cl2 (Supporting Information,
Table S4).61 The sample was carefully thawed over 2 min, from the
top down, resulting in a light yellow solution immediately upon shaking.
After 6 h, the solution was a reddish-purple color, and a white solid
had precipitated. The disappearance of t-BuNH2BH3 and the presence
Acknowledgment. This research was supported by the U.S.
Department of Energy.
Supporting Information Available: Experimental methods,
O2 pressure calculations, kinetic analysis, further characterization
of the t-BuNH2BH3 dehydrogenation product, alcohol oxidation
results, and Arrhenius plot for reduction of O2 by 1H(H). This
material is available free of charge via the Internet at
http://pubs.acs.org.
1
of 1, H2, and H2O were verified by H NMR spectroscopy.
The dehydrogenated “RNBH” product was prepared in a scaled-up
preparation containing 2 mol % of 1 and 0.12 g (1.4 mmol) of t-BuNH2-
BH3 in CH2Cl2, achieving 60.0 mg of white solid in 25 h under 1 atm
of O2. The nitrogen-boron-containing coproduct was poorly soluble
in CD2Cl2 and THF-d8 but soluble in DMSO-d6. 1H NMR (500 MHz,
DMSO-d6): δ 1.19 (9 H), 3.33 (3 H), 6.42 (4 H). 13C NMR (125 MHz,
DMSO-d6): δ 28.41, 50.10. 11B NMR (96 MHz, DMSO-d6): δ 1.76.
Found: C, 47.29; H, 5.35; N, 3.18.
JA073774P
(62) (a) Yamamoto, Y.; Miyamoto, K.; Umeda, J.; Nakatani, Y.; Yamamoto,
T.; Miyaura, N. J. Organomet. Chem. 2006, 691, 4909-4917. (b) Ashby,
E. C.; Kovar, R. A. Inorg. Chem. 1971, 10, 1524-1526. (c) Grace, A.;
Powell, P. J. Chem. Soc. A 1966, 1468-1471. (d) Brown, M. P.; Heseltine,
R. W.; Sutcliffe, L. H. J. Chem. Soc. A 1968, 612-616.
(61) Battino, R., Ed. Oxygen and Ozone; Solubility Data Series 7; Pergamon:
Oxford, 1981.
9
14310 J. AM. CHEM. SOC. VOL. 129, NO. 46, 2007