Acceptorless dehydrogenation of C-C single bonds adjacent to functional groups by metal-ligand cooperation

Article abstract of DOI:10.1021/ja409672w

Unprecedented direct acceptorless dehydrogenation of C-C single bonds adjacent to functional groups to form α,β-unsaturated compounds has been accomplished by using a new class of group 9 metal complexes. Metal-ligand cooperation operated by the hydroxycyclopentadienyl ligand was proposed to play a major role in the catalytic transformation.

Full text of DOI:10.1021/ja409672w

Communication
pubs.acs.org/JACS
Acceptorless Dehydrogenation of C−C Single Bonds Adjacent to
Functional Groups by Metal−Ligand Cooperation
Shuhei Kusumoto, Midori Akiyama, and Kyoko Nozaki*
Department of Chemistry and Biotechnology, Graduate School of Engineering, The University of Tokyo, 7-3-1 Hongo, Bunkyo-Ku,
Tokyo 113-8656, Japan
S
* Supporting Information
Scheme 1
ABSTRACT: Unprecedented direct acceptorless dehy-
drogenation of C−C single bonds adjacent to functional
groups to form α,β-unsaturated compounds has been
accomplished by using a new class of group 9 metal
complexes. Metal−ligand cooperation operated by the
hydroxycyclopentadienyl ligand was proposed to play a
major role in the catalytic transformation.
ince the first report by Bergman, Graham, and Jones in the
Searly 1980s, cyclopentadienyl rhodium or iridium phosphine
complexes have been known to undergo oxidative addition of
unactivated sp3 C−H bonds to form alkylmetal hydride species.1
Because the resulting alkylmetal hydride is a coordinatively
saturated 18-electron complex (complex B in Scheme 1a),
further chemical transformation of the alkyl group was rather
limited, and examples have been confined to a few stoichiometric
reactions.1a Hence, catalytic functionalization of a sp3 C−H bond
had required photoirradiation to make vacant site(s) except for a
few thermal dehydrogenations.2,3 In 1996, Jensen reported a
PCP-type pincer iridium dihydride 16-electron complex, which
shows quite high activity in thermal alkane dehydrogenation
reaction.4 Intensive efforts have been devoted to further
development of pincer-type complexes by Jensen, Goldman
and Brookhart in alkane and amine dehydrogenation.5−7 As
another type of sp3 C−H functionalization, rhodium-catalyzed
alkane borylation via σ-bond metathesis established by Hartwig
may be also referred.8
In the field of dehydrogenation of alcohols or hydrogenation
18-electron alkyl hydride complex F having a hydroxyl group on
the Cp ring would release dihydrogen via metal−ligand
cooperative heterolytic bond formation to form a 16-electron
alkyl complex G (Scheme 1c). Coordinatively unsaturated G is
expected to be active for further functionalizations, e.g., β-
hydride elimination to afford an alkene and hydride complex H,
which would exist under equilibrium with E.13 By developing
these hydroxycyclopentadienyl group 9 metal complexes, we
have accomplished the first direct acceptorless dehydrogenation
of C−C single bonds adjacent to functional groups with metal−
ligand cooperation. In the precedents for α,β-dehydrogenation of
of carbonyl groups, metal−ligand cooperation has attracted
much attention.9 A pioneering work by Shvo with cyclo-
pentadienone−hydroxycyclopentadienyl based ruthenium com-
plex established a heterolytic formation of dihydrogen from
proton and hydride (from complex D to C in Scheme 1b) and
heterolytic cleavage of dihydrogen into proton and hydride
(from C to D).10 As another example of efficient metal−ligand
cooperation, highly active acceptorless dehydrogenation of
alcohols was reported by Fujita with pyridone−hydroxypyridine
based iridium catalyst, which can release hydride on the metal as
dihydrogen by the assistance of proton on the hydroxypyridine
ligand.11,12
Here in this work, we designed hydroxycyclopentadienyl
group 9 metal complexes E in the aim of metal−ligand
cooperation in C−H functionalizations (Scheme 1c). While
the 18-electron Cp* complex B does not have a coordination site
for further functionalizations (Scheme 1a), we hypothesized the
carbonyl compounds, stoichiometric oxidants such as DDQ,14
16,17
IBX15 and O2
for palladium catalysts or alkenes for
ruthenium catalysts18 were indispensable.19
Received: September 18, 2013
Published: December 3, 2013
© 2013 American Chemical Society
18726
dx.doi.org/10.1021/ja409672w | J. Am. Chem. Soc. 2013, 135, 18726−18729

Journal of the American Chemical Society
Communication
Scheme 2
Desired hydroxycyclopentadienyl iridium and rhodium
complexes 4a1, 4a2 and 4b were synthesized in three steps
from cyclopentadienone metal(I) chloride dimer 1a or 1b
(Scheme 2). Reaction of iridium and rhodium chloride dimer 1a
and 1b with di-tert-butylmethylphosphine resulted in the
formation of tetracoordinated complexes 2a1 and 2b.20 These
tetracoordinated complexes were air and moisture stable and
could be purified by silica-gel column chromatography. Complex
2a1 showed a distorted tetrahedral geometry by X-ray
crystallography.21 Treatment of chloride complexes 2a1 and 2b
with 2 equiv of lithium triethylborohydride led to clean
formation of dihydride complexes 3a1 and 3b, confirmed by
31P and 1H NMR spectroscopy. Iridium complex 3a1 was isolated
in a nearly quantitative yield as a rare anionic Ir(I) dihydride
complex22 (Figure 1a), while rhodium complex 3b was too
Table 1. Physical Parameters of Complexes 3a1, 4a1, and 6
3a1
4a1
6
a
a
IR ν (Ir−H, sym) (cm−1
1H NMR (Ir-H) (ppm)
bond length (Ir−C1)
(C1−O1) (Å)
)
2181
2210
2162
−18.54
−
b
b
c
−19.21
2.479(9)
−18.50
2.330(6)
1.382(7)
1.261(10)
−
a
b
c
Measured by KBr method. In THF-d8. In C6D6.
Å) and shorter Ir−Cl distance (2.330(6) Å) in 4a1 correspond to
a structure where cyclopentadienone ring is reduced and
coordinates to Ir(III) as an η5-cyclopentadienyl anion. 1H
NMR signal of hydrides in complex 3a1 was observed at −19.21
ppm, in a higher field than in 4a1 (−18.50 ppm), implying
stronger anionic feature of the hydrogens in 3a when compared
to 4a1. IR absorbance of Ir−H symmetric stretching is slightly
weaker in anionic Ir complex 3a1 (2181 cm−1) than in complex
4a1 (2210 cm−1), showing weaker Ir−H bond in anionic Ir(I)
complex 3a1 than neutral Ir(III) complex 4a1. The Ir−H bond in
4a1 is stronger in comparison to Cp* complex 6 (2162 cm−1),
probably because of the weaker donation from tetraphenylhy-
droxycyclopentadienyl ring than Cp*.24
Direct acceptorless dehydrogenation of C−C single bonds
attached to functional groups were accomplished by using
hydroxycyclopentadienyl iridium complex 4a1 (Table 2).25 First,
reactivities of the Ir complexes were compared in the
dehydrogenation of a simple cycloalkane, cyclooctane in the
presence of a hydrogen acceptor (tbutylethylene, hereafter
abbreviated as TBE). The hydroxycyclopentadienyl complex 4a1
showed higher activity (TON 24, entry 1) when compared to
silyl-protected complexes 5a1me and 5a1ipr (TON 5 for each,
entries 2 and 3) or Cp* complex 6 (TON 4, entry 4), although
the values were lower than the highest record with PCP-type
pincer Ir complexes.3,5 Dehydrogenation with 4a1 also
proceeded in the absence of TBE (entry 6). Next, we applied
the Ir complexes to other substrates. It should be noted that the
hydroxycyclopentadienyl complexes 4a1 and 4a2 were uniquely
active in dehydrogenation of α-tetralone to 1-naphthol in the
absence of dihydrogen acceptor. Complexes 4a1 and 4a2 showed
19 and 4 turnovers in the dehydrogenation of α-tetralone to 1-
naphthol in 30 min (entries 7 and 8), while rhodium complex 4b,
Cp* iridium complex 6 or even PCP-type pincer iridium
complex IrH2(C6H3-2,6-(CH2PtBu2)2) (7) showed no activity
(entries 9, 11 and 12). The reaction is strongly suppressed when
the hydroxyl group on the Cp ring was protected by a silyl group
(5a1ipr), indicating that the free hydroxyl group plays an
important role in this metal−ligand cooperation (compare
entries 10 with 7). It is worth noting that no oxidant was
necessary and evolution of 116 μmol of dihydrogen (almost
Figure 1. X-ray structures of complex (a) 3a1 and (b) 4a1 (thermal
ellipsoids are drawn with 10% probability for 3a1, 50% for 4a1.
Hydrogens on carbon atoms and carbon atoms of THFs are omitted for
clarity.).
unstable to be fully characterized. This anionic iridium(I)
complex 3a1 readily reacted with a proton source such as
moisture, alcohol and triethylaminehydrochloride to afford the
desired hydroxyl-substituted cyclopentadienyl iridium(III) dihy-
dride complex 4a1.23 Anionic Iridium(I) complex 3a1 can be also
trapped by trialkylchlorosilane to afford trimethylsilyloxyl-
(5a1me) or triisopropylsilyloxyl- (5a1ipr) substituted complexes.
Although anionic complex 3b could not be isolated because of its
instability, the corresponding hydroxycyclopentadienyl rhodium
complex 4b was isolated by flash silica-gel column chromatog-
raphy. Triphenylphosphine ligated iridium complex 4a2 was also
synthesized by the same procedure as for 4a1.
X-ray structures of 3a1 and 4a1 are shown in Figure 1. Physical
parameters (1H NMR, IR and bond lengths observed in X-ray
analysis) of 3a1, 4a1 and Cp*IrH2PMetBu2 (6) are summarized in
Table 1. The short C1−O1 distance in 3a1 (1.261(10) Å) and the
long Ir−C1 length (2.474(9) Å) suggest that 3a1 is an anionic
Ir(I) complex with the cyclopentadienone coordination in an η4
manner. In sharp contrast, longer C1−O1 bond length (1.382(7)
18727
dx.doi.org/10.1021/ja409672w | J. Am. Chem. Soc. 2013, 135, 18726−18729

Journal of the American Chemical Society
Communication
and naphthopiranone, respectively (entries 16 and 17). Cyclic
amide, dihydroquinolinone, was also dehydrogenated to
quinolinol with 37 turnovers (entry 18). The applicable substrate
was not limited to the ones to form α,β-unsaturated carbonyl
compounds: Isochromanone was also dehydrogenated to the
corresponding alkenyl ester with 4 turnovers (entry 19). To the
best of our knowledge, no direct dehydrogenation of alkyl ester
to alkenyl ester has been reported even in the presence of
stoichiometric oxidant. Dihydrobenzofuran could be applied and
showed relatively high activity (TON 66, entry 20). On the other
hand, however, the reaction was not applicable to the following
substrates: Unsubstituted cyclohexanone resulted in formation
of a complex mixture including cyclohexenone, phenol and
higher molecular weight compounds originated from the
competing aldol type condensation (see Supporting Informa-
tion). Structurally related cyclooctanone or acyclic esters did not
show any activity for the dehydrogenation to result in recovery of
the starting materials.26
Table 2. Dehydrogenation of C−C Single Bonds Catalyzed by
a b
,
Group 9 Metal Complexes
In the dehydrogenation reaction of cycloketones, β-hydride
elimination was suggested to be the turnover limiting step by H/
D exchange experiment and DFT calculation. To clarify the
reason for the different reactivity of cyclohexanone and
cyclooctanone in this reaction, we first examined H/D exchange
experiment in the presence of 2-naphthol-d1 (Scheme 3). After
Scheme 3
the reaction with complex 4a1 in 30 min, α-position of
cyclooctanone was deuterated by 2%, while only a trace amount
of H/D exchange was observed with 2,6-dimethylcyclohexanone.
Because this H/D exchange is thought to undergo via oxidative
addition to Ir(I) followed by replacement of the Ir(III) by D+, the
higher activity observed for cyclooctanone than 2,6-dimethylcy-
clohexanone suggests that the C−H bond cleavage did take place
with cyclooctanone. DFT calculation (B3LYP/LanL2DZ, 6-
31G*) was further performed to compare the following β-
hydride elimination step for cyclooctanone and cyclohexanone
(Figure 2).26 As a result, β-hydride elimination barrier relative to
the corresponding α-metallocarbonyl complex G is 2.4 kcal/mol
a
Reaction condition for entries 1−4 and 15: mixture of complex (5
μmol), substrate (4.0 mL) abd tbutylethylene (200 μL) were stirred in
a stainless autoclave at 200 °C for 29 h. Reaction condition for other
entries: mixture of substrates (200 μL for liquid or 100 mg for solid)
and complex (5 μmol) were stirred in a glas tube connected to an Ar
line at 200 °C for 19 h. Determined by GV analysis with dodecane as
b
c
d
an internal standard. 116 μmol of dihydrogen was detected by GC.
e
f
Isolated yield. Tetrahydronaphthalene (107 μmol) and naphthalene
g
(171 μmol) were observed as side products. Quinoline (22.7) μmol
was observed as a sided product.
‡
equimolar to 1-naphthol) was confirmed in entry 7. In longer
reaction time, 29 h, turnover number increased up to 97 in entry
13. However, reduction of the carbonyl moiety by evolved
dihydrogen took place, and significant amounts of naphthalene
and tetrahydronaphthalene were detected as side products. This
undesired reduction of the carbonyl moiety could be suppressed
by removing the evolved dihydrogen by Ar bubbling.26 Aliphatic
ketone, 2,6-dimethylcyclohexanone, can be also dehydrogenated
to afford 2,6-dimethylcyclohexenone and 2,6-dimethylphenol as
two major products with total TON of 12 (entry 14). The
addition of hydrogen acceptor TBE improved the TON by about
double (entry 15). The reaction was applicable to cyclic esters,
dihydrocoumarin and splitomicin showing 5 and 6 turnovers to
afford the corresponding α,β-unsaturated lactone, coumarine
higher for cyclooctanone (ΔG8 ) than for cyclohexanone
‡
(ΔG6 ). When these two transition states are compared to the
common intermediate E, the transition state with cyclooctanone
(TS8) was located 9.8 kcal/mol higher than that with
‡
‡
cyclohexanone (TS6). The difference of ΔG8 and ΔG6 could
be attributed to the larger ring strain of cyclooctanone than
cyclohexanone, which caused the larger dihedral angle (Ir−Cα−
Cβ−H) in the transition state (22.1° in TS8 and 13.6° in TS6).
Also, dehydrogenation to unsaturated ketone is thermodynami-
cally more favored for cyclohexanone than for cylooctanone by 3
kcal/mol.26
In conclusion, we synthesized novel hydroxycyclopentadienyl
group 9 metal complexes. Hydroxycyclopentadienyl iridium
complex 4a1 was found to be active in direct acceptorless
18728
dx.doi.org/10.1021/ja409672w | J. Am. Chem. Soc. 2013, 135, 18726−18729

Journal of the American Chemical Society
Communication
Goldman, A. S. Chem. Commun. 2003, 2060−2061. (c) Bernskoetter, W.
H.; Brookhart, M. Organometallics 2008, 27, 2036−2045.
(6) Choi, J.; MacArthur, A. H. R.; Brookhart, M.; Goldman, A. S. Chem.
Rev. 2011, 111, 1761−1779.
(7) An recent review for dehydrogenation reaction: Dobereiner, G. E.;
Crabtree, R. H. Chem. Rev. 2010, 110, 681−703.
(8) Chen, H.; Schlecht, S.; Semple, T. C.; Hartwig, J. F. Science 2000,
287, 1995.
(9) Recent review articles for metal-based bifunctional catalysts:
(a) Ikariya, T.; Murata, K.; Noyori, R. Org. Biomol. Chem. 2006, 4, 393−
406. (b) Muniz, K. Angew. Chem., Int. Ed. 2005, 44, 6622−6627.
̃
(c) Shibasaki, M.; Matsunaga, S. Chem. Soc. Rev. 2006, 35, 269−279.
(d) Ma, J.-A.; Cahard, D. Angew. Chem., Int. Ed. 2004, 43, 4566−4583.
(e) Groger, H. Chem.Eur. J. 2001, 7, 5246−5251. (f) Gunanathan, C.;
̈
Milstein, D. Acc. Chem. Res. 2011, 44, 588−602.
(10) (a) Conley, B. L.; Pennington-Boggio, M. K.; Boz, E.; Williams, T.
J. Chem. Rev. 2010, 110, 2294. (b) Karvembu, R.; Prabhakaran, R.;
Natarajan, K. Coord. Chem. Rev. 2005, 249, 911. (c) Shvo, Y.; Czarkie, D.
J. Organomet. Chem. 1989, 368, 357−365.
(11) Fujita, K.; Tanino, N.; Yamaguchi, R. Org. Lett. 2007, 9, 109−111.
(12) High yielding dehydrogenation of tetrahydroquinolines was also
reported by Yamaguchi, R.; Iketa, C.; Takahashi, Y.; Fujita, K. J. Am.
Chem. Soc. 2009, 131, 8410−8412.
(13) For related Rhenium complex: Landwehr, A.; Dudle, B.; Fox, T.;
Blacque, O.; Berke, H. Chem.Eur. J. 2012, 18, 5701−5714.
(14) (a) Walker, D.; Hiebert, J. D. Chem. Rev. 1967, 67, 153.
(b) Buckle, D. R. 2,3-Dichloro-5,6-dicyano-1,4-benzoquinone. In
Encyclopedia of Reagents for Organic Synthesis; John Wiley & Sons,
Inc.: New York, 2010.
(15) (a) Nicolaou, K. C.; Zhong, Y.-L.; Baran, P. S. J. Am. Chem. Soc.
2000, 122, 7596. (b) Nicolaou, K. C.; Montagnon, T.; Baran, P. S.;
Zhong, Y.-L. J. Am. Chem. Soc. 2002, 124, 2245. (c) Nicolaou, K. C.;
Montagnon, T.; Baran, P. S. Angew. Chem., Int. Ed. 2002, 41, 993.
(d) Nicolaou, K. C.; Gray, D. L. F.; Montagnon, T.; Harrison, S. T.
Angew. Chem., Int. Ed. 2002, 41, 996.
(16) (a) Theisen, R. J. J. Org. Chem. 1971, 36, 752−757. (b) Muzart, J.
Eur. J. Org. Chem. 2010, 20, 3779. (c) Diao, T.; Wadzinski, T. D.; Stahl,
S. S. Chem. Sci. 2012, 3, 887−891. (d) Keßler, M. T.; Prechtl, M. H. G.
ChemCatChem 2012, 4, 326.
(17) (a) Diao, T.; Stahl, S. S. J. Am. Chem. Soc. 2011, 133, 14566.
(b) Izawa, Y.; Pun, D.; Stahl, S. S. Science 2011, 333, 209.
(18) Yi, C. S.; Lee, D. W. Organometallics 2009, 28, 947−949.
Dehydrogenation of several amines including tetrahydroquinoline to
quinoline were also reported in the paper.
(19) Acceptorless dehydrogenation of quinolines are already known
with cyclometalated iridium catalysts. Wu, J.; Talwar, D.; Johnston, S.;
Yan, M.; Xiao, J. Angew. Chem. Int. Ed. 2013, 52, 6983−6987.
(20) Related synthesis of rhodium complex: McVey, S.; Maitlis, P. M.
Can. J. Chem. 1966, 44, 2429.
(21) In solid state, complex 2a1 has a tortion angle of 61.6° in average
between two coordination planes (P−Ir−Cl and cyclopentadienone−
Ir).
(22) Simpson, R. D.; Marshall, W. J.; Farischon, A. A.; Roe, D. C.;
Grushin, V. V. Inorg. Chem. 1999, 38, 4171−4173.
Figure 2. DFT calculation for β-hydride elimination (kcal/mol).
dehydrogenation of C−C single bonds adjacent to functional
groups by metal−ligand cooperation. This report provides a new
strategy for C−H bond functionalization and synthesis of α,β-
unsaturated compounds.
ASSOCIATED CONTENT
* Supporting Information
■
S
Experimental details for the synthesis of complexes and
dehydrogenation reactions, crystallographic data (CIF) for
complexes 3a1, 4a1, 5a1me, and 6. Optimized geometries for
calculated transition states and intermediates. This material is
available free of charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
nozaki@chembio.t.u-tokyo.ac.jp
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported by JSPS Grant-in-Aid for JSPS Fellows.
A part of this work was conducted in Research Hub for Advanced
Nano Characterization, The University of Tokyo, supported by
MEXT, Japan. The computations were performed using
Research Center for Computational Science, Okazaki, Japan.
REFERENCES
■
(1) (a) Janowicz, A. H.; Bergman, R. G. J. Am. Chem. Soc. 1983, 105,
3929. (b) Janowicz, A. H.; Bergman, R. G. J. Am. Chem. Soc. 1982, 104,
352. (c) Hoyano, J. K.; McMaster, A. D.; Graham, W. A. G. J. Am. Chem.
Soc. 1983, 105, 7190. (d) Jones, W. D.; Feher, F. J. J. Am. Chem. Soc.
1984, 106, 1650. (e) Arndtsen, B. A.; Bergman, R. G.; Mobley, T. A.;
Peterson, T. H. Acc. Chem. Res. 1995, 28, 154−162. (f) Jones, W. D.;
Feher, F. J. Acc. Chem. Res. 1989, 22, 91−100.
(23) Complexes 3a1 and 4a1 were confirmed as dihydride complexes
not dihydrogen complexes by measuring the T1 of hydrogen on the
iridium with inversion recovery method (404 ms for 3a1, 415 ms for
4a1).
(24) Same tendency is known for Shvo’s ruthenium complex and Cp*
hydridoruthenium dicarbonyl complex (a) Shvo, Y.; Czarkie, D. J.
Organomet. Chem. 1989, 368, 357−365. (b) Chen, T.-Y.; Bullock, M.
Organomettalics 2011, 21, 2325−2331.
(2) (a) Crabtree, R. H.; Mihelcic, J. M.; Quirk, J. M. J. Am. Chem. Soc.
1979, 101, 7738−7740. (b) Baudry, D.; Ephritikhine, M.; Felkin, H.;
Holmes-Smith, R. J. Chem. Soc., Chem. Commun. 1983, 788−789.
(3) (a) Sakakura, T.; Sodeyama, T.; Sasaki, K.; Wada, K.; Tanaka, M. J.
Am. Chem. Soc. 1990, 112, 7221−7229. (b) Maguire, J. A.; Boese, W. T.;
Goldman, A. S. J. Am. Chem. Soc. 1989, 111, 7088−7093.
(4) Gupta, M.; Hagen, C.; Flesher, R. J.; Kaska, W. C.; Jensen, C. M.
Chem. Commun. 1996, 2083.
(5) (a) Gu, X.-Q.; Chen, W.; Morales-Morales, D.; Jensen, C. M. J. Mol.
(25) Recent reports for the reactivity of group 9 metal complexes
toward carbonyl compounds: (a) Masuda, Y.; Hasegawa, M.; Yamashita,
M.; Nozaki, K.; Ishida, N.; Murakami, M. J. Am. Chem. Soc. 2013, 135,
7142−7145. (b) Zhu, Y.; Smith, D. A.; Herbert, D. D.; Gatard, S.;
Ozerov, O. V. Chem. Commun. 2012, 48, 218−220.
(26) See Supporting Information for further experimental details and
theoretical calculation results.
Catal. 2002, 189, 119−124. (b) Zhang, X.; Fried, A.; Knapp, S.;
18729
dx.doi.org/10.1021/ja409672w | J. Am. Chem. Soc. 2013, 135, 18726−18729