Base-free, acceptorless, and chemoselective alcohol dehydrogenation catalyzed by an amide-derived NNN -ruthenium(II) hydride complex

Article abstract of DOI:10.1021/om4000677

The bmpi (1,3-bis(6′-methyl-2′-pyridylimino)isoindoline) pincer Ru(II) hydride complex catalyzes base-free, acceptorless, and chemoselective dehydrogenation of alcohols with liberation of dihydrogen under moderate (<120 C) conditions. Primary alcohols and diols are converted to ester and lactone products with high conversion efficiencies. The catalyst system is remarkably selective for the oxidation of secondary alcohols in the presence of primary alcohols.

Full text of DOI:10.1021/om4000677

Communication
pubs.acs.org/Organometallics
Base-Free, Acceptorless, and Chemoselective Alcohol
Dehydrogenation Catalyzed by an Amide-Derived NNN-
Ruthenium(II) Hydride Complex
Kuei-Nin T. Tseng, Jeff W. Kampf, and Nathaniel K. Szymczak*
Department of Chemistry, University of Michigan, 930 North University Avenue, Ann Arbor, Michigan 48109, United States
S
* Supporting Information
ABSTRACT: The bmpi (1,3-bis(6′-methyl-2′-pyridylimino)-
isoindoline) pincer Ru(II) hydride complex catalyzes base-free,
acceptorless, and chemoselective dehydrogenation of alcohols
with liberation of dihydrogen under moderate (<120 °C)
conditions. Primary alcohols and diols are converted to ester
and lactone products with high conversion efficiencies. The
catalyst system is remarkably selective for the oxidation of
secondary alcohols in the presence of primary alcohols.
n the past decade, catalytic alcohol dehydrogenation
reactions have evolved as an atom-economic methodology
To promote rapid dehydrogenation reactivity, we targeted
amide-derived NNN pincer Ru complexes that incorporate the
bmpi (1,3-bis(6′-methyl-2′-pyridylimino)isoindoline) ligand.6,7
Metalation was achieved by addition of the deprotonated bmpi
ligand to Ru(PPh3)3Cl2 over 21 h in THF solvent, which
afforded Ru(bmpi)(PPh3)Cl (1)8 in 92% yield (Figure 1). The
31P{1H} NMR spectrum of 1 exhibits a singlet at 43.50 ppm,
I
to generate H2 from biologically relevant alcohols1 and/or to
reveal a reactive organic fragment that can undergo follow up
reactivity to form higher order liquid products.2 In the absence
of an added H2 acceptor, primary alcohols can couple to form
esters, with concomitant release of H2, which is a variant of
acceptorless dehydrogenative coupling (ADC) reactivity.
Recently, this reaction class has seen substantial growth,
typified by catalytic systems incorporating bifunctional
metal−ligand scaffolds capable of promoting ADC.3 For high
atom economy, reactions that function in the absence of
exogenous additives are particularly desirable, when this trait is
coupled with mild reaction conditions. However, many alcohol
dehydrogenation catalysts require the addition of super-
stoichiometric quantities (with respect to Ru catalyst) of a
basic reagent for efficient catalysis.4 The bases (e.g., KtBuO) are
highly caustic, require specialized equipment, and also
contribute to unwanted waste products; thus, systems that
operate under base-free conditions are highly desirable.
1
while the H NMR spectrum features a single set of ligand-
based resonances, including a singlet at 1.71 ppm, consistent
Most known catalysts that effect alcohol dehydrogenation
and/or ADC reactivity incorporate pincer-type ligand frame-
works that can operate by a ligand−metal cooperative
mechanism. Often containing a central pyridine ring, the
pincer framework can exhibit multifunctionality by dearomati-
zation/aromatization of the pyridinyl group, which ensues from
the deprotonation/protonation of a methylene spacer.3,5 We
targeted an alternative anionic NNN pincer framework that
substitutes the methylene spacer groups with an imine linker
whose orbitals preclude concerted hydrogen transfer to an
adjacent acceptor, in order to probe whether a cooperative
mechanism is required for efficient dehydrogenation.6 Herein,
we report the synthesis and characterization of new pincer-
ligated ruthenium complexes, which efficiently catalyze the
dehydrogenation of primary and secondary alcohols without
the requirement of added hydrogen acceptor or base.
Figure 1. Synthesis of HRu(bmpi)(PPh3)2 (2) and ORTEP diagram
of complex 2 with thermal ellipsoids at the 50% probability level. PPh3
phenyl groups and hydrogen atoms, except for the hydride, are
omitted for clarity.
Received: January 28, 2013
© XXXX American Chemical Society
A
dx.doi.org/10.1021/om4000677 | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Communication
with symmetric binding of the bmpi ligand. This molecule
represents the first reported ruthenium complex with a bpi
(1,3-bis(2′-pyridylimino)isoindoline) ligand having substituents
ortho to the pyridyl nitrogens.7b,9
Ruthenium hydride complexes are implicated as catalytic
intermediates in alcohol dehydrogenation with pincer-type
ligands;3b−d,5,10 thus, we targeted hydride variants of Ru−bmpi
complexes. The complex HRu(bmpi)(PPh3)2 (2) was isolated
in 89% yield by allowing 1 to react with 1.05 equiv of NaHEt3B
and 1.2 equiv of PPh3 in THF solution at room temperature for
2 h (Figure 1). The 31P{1H} NMR spectrum of 2 exhibits a
1
singlet at 50.97 ppm, and the H NMR spectrum revealed a
solution structure consistent with C2 symmetry. A single peak
for the ortho-substituted CH3 units was visualized at 3.12 ppm,
in addition to a high-field triplet resonance at −9.58 ppm (JPH
=
20.0 Hz); the latter resonance is consistent with a hydride
ligand trans to an amido nitrogen.11 Crystals suitable for single-
crystal X-ray diffraction were obtained from vapor diffusion of
pentane into a THF solution of 2, and the solid-state structure
of 2 (Figure 1) confirmed the proposed geometry, exposing a
distorted-octahedral geometry around the Ru(II) center with a
hydride ligand (located from the difference map) trans to the
pyrrolidine nitrogen atom.
Figure 2. Reaction profiles of 1-phenylethanol dehydrogenation with
and without added catalyst poisons. Conditions for each curve were
identical ([2] = 7.44 × 10−4 M; [1-phenylethanol] = 7.53 M; [Ph-
■
TMS] = 0.53 M; 120 °C) with the following modifications: (black
)
●
▲
no modifications; (red ) Hg(0) added after 2 h; (blue ) 1.0 equiv
▼
of 1,10-phenanthroline added after 2 h; (pink ) 0.25 equiv of 1,10-
phenanthroline added after 2 h.
The ability of 1 and 2 to effect catalytic transfer
i
hydrogenation with PrOH was evaluated using acetophenone
volatility, is amenable to reaction sampling by GCMS and/or
1H NMR spectroscopy. When the dehydrogenation reaction
was monitored over 24 h in an open system inside an inert-
atmosphere glovebox, the reaction profile displayed two
regions: a linear region (where the maximum rate was
measured) and culmination. The reaction profile was linear
after 4 min15 and reached culmination after approximately 12 h
(Figure 2, Standard).
as the substrate. Heating a 0.1 M acetophenone iPrOH solution
containing 0.5 mol % of 1 and 1.0 mol % of KOtBu to 80 °C for
1 h resulted in quantitative conversion (99%) of acetophenone
to 1-phenylethanol. No reaction took place in the absence of a
base. Complex 2 effected similar hydrogenative transformations
with fewer required reagents. For example, without added base,
under the reaction conditions listed above, complex 2
catalytically reduced acetophenone to 1-phenylethanol in 99%
conversion within 1 h.
The catalyst identity of Ru-catalyzed transfer hydrogenation
reactions with alcohols was recently reported to be distinct
from that of a presumed homogeneous precursor, which
formed catalytically active heterogeneous Ru(0) nanosized
clusters.16 Accordingly, we undertook initial experiments to
probe whether complex 2 participates as a homogeneous
catalyst or as a precursor to a heterogeneous species. Among
the myriad techniques that collectively can be used to establish
the catalyst identity, reproducible kinetic data and poisoning
experiments are regarded as highly supportive evidence for
homogeneous catalysis.17 Consistent with an operative
homogeneous system, the catalytic activity was unaffected by
Hg(0) (∼800 equiv) addition (Figure 2, Hg), when added
during catalysis. Additionally, a substoichiometric ligand poison-
ing experiment was conducted.18 Complete poisoning with
1,10-phenanthroline required 1 equiv (∼ 25% decrease in the
rate was observed with 0.25 equiv), inconsistent with a
heterogeneous system, where low surface area aggregates are
typically poisoned by ≪1 equiv of added ligand poison.18
Finally, in the absence of any poisoning reagent, highly
reproducible (nonsigmoidal) reaction kinetic profiles were
observed (Figure S9, Supporting Information). The combined
results of these tests suggest that the active catalytic species is
indeed a homogeneous ruthenium complex.
In addition to transfer hydrogenation reactivity by 2, we
hypothesized that dehydrogenation of PrOH should also be
i
possible without an added hydrogen acceptor. Indeed, in the
absence of an exogenous ketone, such as acetophenone, iPrOH
was consumed within 2 h at 80 °C using 10 mol % of 2 in
C6D6. H2 and acetone were confirmed as the sole reaction
products by in situ examination of the reaction mixture in a
sealed NMR tube.12 1H NMR spectroscopic analyses revealed
the formation of acetone and dihydrogen in equimolar
quantities after 1 h at 80 °C, observed as singlets at 1.55 and
4.47 ppm, respectively.13 This catalyst system was found to be
remarkably active, and when using 20 ppm of 2, a turnover
number (TON) of 314 and turnover frequency (TOF) of 51
h−1 were obtained, on heating to reflux for 6 h.14 In situ analysis
of the dehydrogenation reaction revealed the release of free
PPh3 during catalysis, as visualized by 31P NMR spectroscopy.
Phosphine dissociation from 2 under catalytic conditions is
consistent with an inner-sphere type pathway, which requires
an open coordination site for substrate binding.
In order to investigate contributors to the overall reaction
efficiency, 1-phenylethanol was chosen as a model 2° alcohol
substrate (eq 1 and Figure 2). This substrate exhibits a
The rates of alcohol dehydrogenation by structurally related
Ru−PNP complexes are often dramatically increased by base
additives,4 and thus we examined whether complex 2 exhibited
similar reactivity toward bases. Addition of 1−1000 equiv of
KOtBu to the dehydrogenation reaction noted above (Table S1,
Supporting Information) modified the reaction rate in a
manner dependent on the quantity of base additive. Increasing
dehydrogenation reaction profile (Figure S7, Supporting
Information) similar to that of PrOH and, due to its low
i
B
dx.doi.org/10.1021/om4000677 | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Communication
the amount of base up to 100 equiv nearly doubled the reaction
rate; however, at 1000 equiv, the reaction rate was suppressed.
We hypothesized that, in the presence of excess base, KOtBu or
tBuOH could compete with 1-phenylethanol coordination, thus
impeding catalytic activity by diverting a competitive β-hydride
elimination pathway. In support, a control experiment using
for the dehydrogenation of secondary alcohols, primary
alcohols were similarly dehydrogenated to afford esters with
no required additives under moderate reaction conditions. For
instance, when n-butanol was used as a substrate, butyl butyrate
was generated in 99% yield after 7 h at 111 °C (Table 1, entry
t
1000 equiv of BuOH also exhibited a decreased reaction rate
Table 1. Dehydrogenation of Primary Alcohols to Esters and
a
(Table S1, Supporting Information), consistent with rate
of Diols to Lactones Catalyzed by 2
t
suppression by BuOH.
The release of H2 provides a large entropic contribution to
the overall thermodynamic profile of acceptorless alcohol
dehydrogenation at a given temperature.19 Thus, we probed H2
pressure effects on 1-phenylethanol dehydrogenations (Figure
3) to examine how the overall conversion efficiency would be
amt of 2 time conversn yield of ester or
b
b
entry alcohol or diol
(mol %)
(h)
(%)
lactone (%)
c
c
1
2
3
4
5
ethanol
1
1
5
1
5
24
7
9
99
58
88
99
9
n-butanol
99
50
88
99
benzyl alcohol
1,4-butanediol
1,5-pentanediol
24
24
24
a
Conditions: alcohol (0.5 mmol), 2, and 2 mL of toluene were
b
refluxed in an open system under N2. Determined by GC-MS.
c
1
Determined by H NMR spectroscopy.
2). High yields of ester products were not general, and when
benzyl alcohol was used as a substrate, 50% conversion to
benzyl benzoate and 8% conversion to the aldehyde were
noted. Furthermore, the conversion of ethanol, a biorelevant
alcohol, to ethyl acetate proceeded with low conversion (Table
1, entry 1).21
Figure 3. Dependence of 1-phenylethanol dehydrogenation catalyzed
by 2 on the initial H2 pressure. Conditions: 2 (4.1 × 10−3 mmol), Ph-
TMS (2.9 mmol), and 5 mL of 1-phenylethanol (41 mmol) were
charged with H2 in a Fisher-Porter tube and heated to 120 °C for 24 h.
Intramolecular reactivity was explored with aliphatic diol
substrates, which were converted in high yields to the
corresponding lactones. For instance, yields of 88% and 99%
were obtained for γ-butyrolactone and δ-valerolactone (Table
1, entries 4 and 5), respectively, from 1,4-butanediol and 1,5-
pentanediol. Several groups have previously reported homoge-
neous base-free catalytic oxidation of diols to lactones using a
hydrogen acceptor (i.e., acetone) as the solvent.22 However, to
the best of our knowledge, only three accounts3c,10a,23 of
acceptorless and base-free catalytic oxidation of diols to
lactones have been reported, and two3c,10a of these reports
require significantly high temperatures (>200 °C). Thus,
complex 2 is only the second reported example of catalytic
ADC reactivity of diols to lactones that does not require an
added hydrogen acceptor and base and occurs under moderate
(<120 °C) conditions.
Guided by our results that demonstrated distinct conditions
were required for the oxidation of primary and secondary
alcohols, we reasoned that 2 should be capable of chemo-
selective oxidations of secondary alcohols in the presence of
primary alcohols. Indeed, using temperatures lower than those
required for ADC reactivity (90 °C), 2 chemoselectively
oxidized the secondary alcohol moiety in 1-phenyl-1,2-
ethanediol (3) in the absence of exogenous base or hydrogen
acceptor additives and 4 was obtained as the only product,
produced in 70% yield, which demonstrates the high chemo-
selectivity of 2 for alcohol dehydrogenation (eq 2). Because the
methine C−H bond of 3 is weakened with respect to the
primary alcohol due to the adjacent phenyl group, we
investigated whether alkanediols of less disparate bond
strengths can be similarly oxidized chemoselectively. Gratify-
ingly, when 1,3-butanediol (5) was subjected to modified
attenuated by elevated pressure. As expected, the reaction was
found to be highly sensitive to pressure, and the dehydrogen-
ation reaction was significantly inhibited (66% decrease in
TON) when it was performed in a sealed vessel (80 mL
headspace). Under as little as 2.5 psig of H2 initial pressure, we
observed a further, sharp decrease in TON (81%), which
continued to decrease with increased pressure up to 80 psig
(0% conversion). Note that complete product inhibition was
not observed at moderate pressures (40 psig of H2; 10
turnovers for dehydrogenation in 24 h). To determine whether
the H2 pressure effect was the result of rapid equilibration or a
kinetic effect (inhibiting H2 release), a dehydrogenation
experiment was performed in a sealed NMR tube using 0.15
1
mol % of 2 in 2-propanol-d8. HD was not observed in the H
NMR spectrum after the reaction mixture was heated to 90 °C
for 24 h, suggesting that equilibration of H2/D2 is not an
operative pathway under the operating conditions and instead
inhibition is likely due to a kinetic effect that inhibits H2 release
with increasing pressure. Furthermore, although dehydrogen-
ations are inhibited by product release, pressures of at least 8
psig could be obtained from vessels (5 mL of alcohol; 80 mL of
headspace) sealed at ambient pressure. This observation is
noteworthy if considering dehydrogenation of biorenewable
alcohols as a means to release H2 for use as an energy carrier,
where elevated pressures are required for H2 transport.20
We next sought to explore the ADC reactivity of primary
alcohols and diols, whose dehydrogenation products contain
highly reactive aldehyde fragments. Consistent with our results
C
dx.doi.org/10.1021/om4000677 | Organometallics XXXX, XXX, XXX−XXX

Organometallics
Communication
(4) (a) Zhang, J.; Gandelman, M.; Shimon, L. J. W.; Rozenberg, H.;
Milstein, D. Organometallics 2004, 23, 4026. (b) Junge, H.; Beller, M.
Tetrahedron Lett. 2005, 46, 1031.
(5) Spasyuk, D.; Smith, S.; Gusev, D. G. Angew. Chem., Int. Ed. 2012,
51, 2772.
(6) Siggelkow, B.; Meder, M. B.; Galka, C. H.; Gade, L. H. Eur. J.
Inorg. Chem. 2004, 3424.
(7) (a) The methyl groups were also incorporated to prevent the
formation of homoleptic Ru complexes. (b) Camerano, J. A.;
Rodrigues, A.-S.; Rominger, F.; Wadepohl, H.; Gade, L. H. J.
Organomet. Chem. 2011, 696, 1425.
(8) The connectivity (Figure S3, Supporting Information) of 1 was
confirmed by single-crystal X-ray diffraction analysis.
(9) (a) Gagne, R. R.; Marritt, W. A.; Marks, D. N.; Siegl, W. O. Inorg.
Chem. 1981, 20, 3260. (b) Marks, D. N.; Siegl, W. O.; Gagne, R. R.
Inorg. Chem. 1982, 21, 3140. (c) Gagne, R. R.; Marks, D. N. Inorg.
Chem. 1984, 23, 65.
(10) (a) Zhao, J.; Hartwig, J. F. Organometallics 2005, 24, 2441.
(b) Montag, M.; Zhang, J.; Milstein, D. J. Am. Chem. Soc. 2012, 134,
10325.
(11) (a) Jazzar, R. F. R.; Varrone, M.; Burrows, A. D.; Macgregor, S.
A.; Mahon, M. F.; Whittlesey, M. K. Inorg. Chim. Acta 2006, 359, 815.
(b) Nag, S.; Butcher, R. J.; Bhattacharya, S. Eur. J. Inorg. Chem. 2007,
1251.
(12) Different conditions were used: 15 mol % of 2 in 0.7 mL of
iPrOH using phenyltrimethylsilane as an internal standard.
(13) The yield of hydrogen gas was determined using Ph-TMS as an
internal standard and corrected for dissolved and free H2, using
Henry’s Law, with a constant of 0.01907.
dehydrogenation conditions (15 mol % of 2, 100 °C) selective
oxidation of the secondary alcohol was achieved (eq 2) as the
sole reaction product in 52% yield. To the best of our
knowledge, such chemoselective oxidation of secondary
alcohols in the presence of primary alcohols by homogeneous
catalysis are exceptionally rare.24
In conclusion, we have developed an amide-derived NNN-
Ru(II) hydride complex capable of catalyzing acceptorless
dehydrogenation and dehydrogenative couplings of secondary
and primary alcohols/diols, respectively, without requirements
of added exogenous base or acceptor additives. Although prior
reports demonstrated ADC reactivity of primary alcohols to
esters and H2, few catalysts accomplish this without base or
acceptor additives.3a,d,5,22a,25 Thus, when base-free, acceptorless
alcohol oxidation catalysts under moderate (<120 °C)
conditions are compared, the activity of 2 ranks among the
best known ADC catalysts. In addition, 2 is particularly
noteworthy because it mediates the chemoselective oxidation of
secondary alcohols in the presence of primary alcohols without
exogenous base or hydrogen acceptor additives, a difficult
selective transformation.24b Work is ongoing to elucidate the
mechanism of alcohol dehydrogenation, as well as the coupling
reactivity of primary alcohols. Furthermore, we are targeting
catalytic dehydrogenations of alternate polar substrates in order
to examine the scope of ADC reactions with alcohols and/or
amines.
(14) When it is dissolved in solution, complex 2 is air-sensitive; thus,
catalytic dehydrogenation is not tolerated in air.
(15) Monitoring the internal reaction temperature as a function of
time required 3.85 min for the reaction solution to reach 120 °C
(Figure S8, Supporting Information), consistent with temperature
equilibration as the primary contributor to the induction period.
(16) Toubiana, J.; Sasson, Y. Catal. Sci. Technol. 2012, 2, 1644.
(17) (a) Widegren, J. A.; Finke, R. G. J. Mol. Catal. A: Chem. 2003,
198, 317. (b) Crabtree, R. H. Chem. Rev. 2012, 112, 1536. (c) This
test works best with Pt, Pd, and Ni metals, but there has been evidence
of Hg(0) affecting Ru(0) heterogeneous catalysts.16
(18) Bayram, E.; Finke, R. G. ACS Catal. 2012, 2, 1967−1975.
(19) (a) For example, the dehydrogenation of iPrOH to acetone and
H2 is endothermic by 16.5 kcal/mol, and the entropic contribution for
a reaction with H2 release at room temperature is 8 kcal/mol.
(b) Wiberg, K. B.; Crocker, L. S.; Morgan, K. M. J. Am. Chem. Soc.
1991, 113, 3447. (c) Watson, L. A.; Eisenstein, O. J. Chem. Educ. 2002,
79, 1269.
(20) Hydrogen as a Future Energy Carrier; Zuttel, A., Borgschulte, A.,
Schlapbach, L., Eds.; Wiley-VCH: Weinheim, Germany, 2008.
(21) Thermodynamically, the process of H2 formation from alcohols
is generally unfavorable. For example, the ΔG° value for formation of
H2 plus acetal from ethanol is 9.8 kcal/mol.
(22) (a) Murahashi, S.; Naota, T.; Ito, K.; Maeda, Y.; Taki, H. J. Org.
Chem. 1987, 52, 4319. (b) Lin, Y.; Zhu, X.; Zhou, Y. J. Organomet.
Chem. 1992, 429, 269. (c) Endo, Y.; Baeckvall, J.-E. Chem. Eur. J. 2011,
17, 12596. (d) Buntara, T.; Noel, S.; Phua, P. H.; Melian-Cabrera, I.;
de Vries, J. G.; Heeres, H. J. Angew. Chem., Int. Ed. 2011, 50, 7083.
(23) Muniz, K. Angew. Chem., Int. Ed. 2005, 44, 6622.
(24) (a) Manzini, S.; Urbina-Blanco, C. A.; Nolan, S. P.
Organometallics 2013, 32, 660. (b) de, G. C. F.; Peters, J. A.; van, B.
H.; Huskens, J. Synthesis 1994, 1007.
ASSOCIATED CONTENT
* Supporting Information
■
S
CIF files, text, figures, and tables giving crystallographic data for
1 and 2, experimental procedures, and characterizations of new
compounds. This material is available free of charge via the
Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
*E-mail for N.K.S.: nszym@umich.edu.
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank the University of Michigan Department of Chemistry
for generous support of this research and the NSF (CHE-
0840456) for X-ray instrumentation. Andrew Rizzi is gratefully
acknowledged for preliminary experiments on the dehydrogen-
ation of diols with 2.
REFERENCES
■
(1) Johnson, T. C.; Morris, D. J.; Wills, M. Chem. Soc. Rev. 2010, 39,
81.
(25) Blum, Y.; Shvo, Y. J. Organomet. Chem. 1985, 282, C7.
(2) Dobereiner, G. E.; Crabtree, R. H. Chem. Rev. 2010, 110, 681.
(3) (a) Zhang, J.; Leitus, G.; Ben-David, Y.; Milstein, D. J. Am. Chem.
Soc. 2005, 127, 10840. (b) Nielsen, M.; Kammer, A.; Cozzula, D.;
Junge, H.; Gladiali, S.; Beller, M. Angew. Chem., Int. Ed. 2011, 50, 9593.
(c) Zhang, J.; Balaraman, E.; Leitus, G.; Milstein, D. Organometallics
2011, 30, 5716. (d) He, L.-P.; Chen, T.; Gong, D.; Lai, Z.; Huang, K.-
W. Organometallics 2012, 31, 5208.
D
dx.doi.org/10.1021/om4000677 | Organometallics XXXX, XXX, XXX−XXX