[Cp*Ru(CO)2]2-catalyzed hydrodeoxygenation and hydrocracking of diols and epoxides

Article abstract of DOI:10.1021/om200447z

The hydrogenolysis of glycols to hydrocarbons is a reaction of potential value for the production of chemicals and fuels from biomass-derived polyols. We find that [Cp*Ru(CO)_2]2 (Cp* = 1,2,3,4,5- pentamethylcyclopentadienyl) catalyzes the hydrogenolysis of glycols to alkanes and of epoxides to alkanes/alkenes by molecular hydrogen at 170200 °C and 427 atm; the product yields range from moderate to excellent. Moreover, we also established a novel process that produces alkanes with a lower number of carbons from glycols: e.g., toluene from 1-phenyl-1,2-ethanediol.

Full text of DOI:10.1021/om200447z

Communication
pubs.acs.org/Organometallics
[Cp*Ru(CO)2]2-Catalyzed Hydrodeoxygenation and Hydrocracking of
Diols and Epoxides
Sylvain Stanowski,† Kenneth M. Nicholas,‡ and Radhey S. Srivastava*,†
†Department of Chemistry, University of Louisiana at Lafayette, Lafayette, Louisiana 70504, United States
‡Department of Chemistry & Biochemistry, University of Oklahoma, Norman, Oklahoma 73019, United States
S
* Supporting Information
ABSTRACT: The hydrogenolysis of glycols to hydrocarbons is a reaction of
potential value for the production of chemicals and fuels from biomass-derived
polyols. We find that [Cp*Ru(CO)2]2 (Cp* = 1,2,3,4,5-pentamethylcyclopenta-
dienyl) catalyzes the hydrogenolysis of glycols to alkanes and of epoxides to
alkanes/alkenes by molecular hydrogen at 170−200 °C and 4−27 atm; the
product yields range from moderate to excellent. Moreover, we also established a
novel process that produces alkanes with a lower number of carbons from glycols: e.g., toluene from 1-phenyl-1,2-ethanediol.
he development of new, effective processes for the
preparation of reduced-oxygen-content products from
Gable and co-workers provided mechanistic insights into the
catalytic process involving LReO3 complexes, especially the
olefin-producing metallo-glycolate fragmentation step.10 Re-
cently, Abu-Omar and co-workers achieved the deoxygenation
of epoxides and deoxydehydration of glycols to alkenes and
alkanes with MeReO3 as catalyst and the more practical
reductant H2.11 An Au−C catalyst has also been found to be
effective for the epoxide to alkene hydrogenolysis under mild
conditions.12 Contemporaneously, the Nicholas group reported
sulfite-driven glycol DODH to olefins catalyzed by MeReO3
and Z+ReO4− and examined its scope and catalytic pathway
(Scheme 1).13 Finally, Bergman et al. disclosed that Re2(CO)10
(in air) catalyzes glycol DODH with secondary alcohols as
reductants.14
In view of the efficacy of ruthenium catalysts in hydro-
genation and hydrogenolysis reactions,15 we initiated a project
seeking the development of new DODH reaction systems that
would utilize readily available catalysts and the practical
reductant H2. Hydrogen is a valuable reductant due to its low
cost and low toxicity, its strong reducing potential, and its
production of water as a benign byproduct. In this report we
reveal the first examples of glycol hydro-didehydroxylation and
epoxide deoxygenation catalyzed by a low-valent ruthenium
complex. Also uncovered is a novel C−C hydrogenolysis
(cracking) process that produces reduced-carbon-number
alkanes.
T
polyols and carbohydrates is an important step toward the
practical utilization of these renewable resources to produce
fuels and value-added chemicals.1,2 Metal-catalyzed deoxygena-
tion of glycols and epoxides is much less developed than the
reverse oxygenation reactions, epoxidation and dihydroxylation
of alkenes, which have been extensively applied in organic
synthesis and industry.3 Recently hydrodehydroxylation
(hydrogenolysis) of polyols has been reported employing
various heterogeneous catalyst systems, usually resulting in
removal of one hydroxyl group, in some cases with significant
regioselectivity.4 Homogeneous catalytic monodehydroxylation
of glycols has also been recently achieved by Bullock and Fagan,
who found that {[Cp*Ru(CO)2]2(μ-H)}+OTf− is a precatalyst
for the selective hydrogenolysis of 1,2-propanediol to 1-
propanol (Scheme 1).5 Schlaf and co-workers reported the
Scheme 1
total hydrodeoxygenation of glycerol and 1,2-hexanediol to
saturated hydrocarbons by [LRu(H2O)x(diimine)]2+ and
[Cp*Ru(OH2)(N−N)]+.6
We began our investigation with 1-phenyl-1,2-ethanediol as a
representative glycol substrate (Scheme 2; R = Ph) and readily
16
available [Cp*Ru(CO)2]2
(1; Cp* = η5-C5Me5) as
Deoxydehydration (DODH) reactions, which remove two
adjacent hydroxyl groups, are attractive for producing
unsaturated hydrocarbons (Scheme 1). There are several
multistep stoichiometric methods for glycol DODH,7 but
single-step8 and catalytic methods are few, and the latter are
almost entirely based on oxo−rhenium complexes. Metal-
catalyzed DODH reactions of glycols were first discovered by
Andrews and Cook9 using PPh3 as the reductant and Cp*ReO3
as the catalyst. Experimental and computational studies by
precatalyst. The reaction between the glycol and H2 (4 atm)
was first conducted in benzene with 10 mol % of [Cp*Ru-
(CO)2]2 in a steel autoclave at 170 °C. GC and GC-MS
analysis of the reaction mixture over 30 h indicated the gradual
and nearly simultaneous formation of ethylbenzene and,
Received: May 25, 2011
Published: January 9, 2012
© 2012 American Chemical Society
515
dx.doi.org/10.1021/om200447z | Organometallics 2012, 31, 515−518

Organometallics
Communication
48 h (entry 3). At 200 °C with a hydrogen pressure of 6 atm,
10% octane and 16% of the cracking product heptane were
produced (entry 4); at 20 atm H2 and 170 °C, the octane and
heptane yields increased substantially to 46% and 49%,
respectively (entry 5). The long-chain 1,2-tetradecanediol was
even less reactive, with no hydrocarbons being detected at 200
°C and 6 atm H2 in 48 h. However, when the reaction was
conducted at 200 °C with 26 atm of H2, tetradecane (3%) and
truncated linear alkanes were produced in low yield: i.e.,
nonane (4%), decane (4%), undecane (5%), and tridecane
(4%) after 47 h (Table 1, entry 6). In contrast, the cyclic glycol
cis-1,2-cyclohexanediol was converted to cyclohexene (56%),
with no detectable alkane or cracking products (entry 7). We
also investigated the reaction of 1,4-anhydroerythritol (a
biomass-derived glycol). When 1,4-anhydroerythritol was
hydrogenated by 1 under our reaction conditions, it afforded
THF in modest yield (12%) (entry 11).
Scheme 2
remarkably, toluene (ca. 2:1), with one less carbon than the
starting glycol (entry 1, Table 1);17 neither styrene nor
monoalcohols were detected.
Table 1. Glycol and Epoxide Deoxygenation by H2/
[Cp*Ru(CO)2]2
a
The hydrodeoxygenation of epoxides is also catalyzed by 1.
Thus, styrene oxide was converted to ethylbenzene (68%) and
toluene (25%) at 170 °C/4 atm of H2 (Table 1, entry 8), the
same products as formed from the corresponding glycol.
Similarly, cyclohexene oxide (entry 9), like the corresponding
diol (entry 7), afforded the alkene cyclohexene in 44% yield
after 8 h (entry 9), again with no alkane detected. 1,2-Hexene
oxide (like the corresponding glycol) was converted to n-
hexane in good yield (70%, entry 10).
The complete hydrodeoxygenation and hydrocracking
reactions found in the glycol and epoxide reactions catalyzed
by 1 have little precedent in homogeneous Ru catalysis. Aside
from the glycol to alkane hydrogenation catalyzed by
[LRu(H2O)x(diimine)]2+ complexes,5 the hydrocracking of
aldoses catalyzed by H2Ru(PMe3)4 under basic conditions has
been reported,19 but this reaction likely involves base-promoted
retroaldol C−C cleavage of the β-hydroxy aldehyde, a pathway
not available to saturated polyols. In fact, we can find no reports
of complex 1 serving as a hydrogenation precatalyst for any
class of substrates.
Although the catalytic pathways for the hydrodehydrox-
ylation, deoxygenation, and hydrocracking reactions promoted
by 1 are presently unclear, a few observations and conclusions
can be noted at this time. The similar product profiles in both
the glycol and epoxide hydrogenations are suggestive of the
involvement of common intermediates. Regarding possible
organic intermediates in the catalytic process leading to alkanes
and cracking products, such as monoalcohols or olefins, we
have found that both styrene and the mono-ol PhCH2CH2OH
are completely converted to ethylbenzene (80%) and a small
amount (4%) of toluene within 6 h under the standard catalytic
reaction conditions. Hence, while these may be hydrogenation
intermediates from the glycol, they are unlikely to be primary
precursors to the hydrocracking products (e.g., via olefin
metathesis/hydrogenation). With respect to intervening Ru
complexes, the preponderance of alkane products (and absence
of monoalcohols) together with the observed reduced-carbon
alkanes in the reactions promoted by 1 appear to exclude the
a
Reaction conditions: 0.14 mmol of glycol or epoxide in benzene (6−
10 mL), 0.014 mmol of [Cp*Ru(CO)2]2, 170 °C, 4 atm or as
indicated in the table. Product yields were determined by GC and GC-
MS with n-dodecane as internal standard.
+
The hydrogenolysis reactions promoted by 1 of representa-
tive aliphatic glycols, better models for biomass-derived polyols,
were also examined to determine the products, selectivity, and
efficiency (Table 1). When it was subjected to the standard
reaction conditions (170 °C, 4 atm of H2, 10 mol % 1), 1,2-
hexanediol produced hexane in 70% yield (entry 2) as the only
detected product.18 The reaction of 1,2-octanediol under the
same conditions was sluggish and produced only 4% octane in
active involvement of [Cp*Ru(CO)2(μ-H)]2 or [Cp*Ru-
(CO)2(H2)]+, implicated in the Bullock−Fagan studies of
glycol monohydrogenolysis.6 Under neutral conditions more
likely ruthenium hydride intermediates are Cp*Ru(CO)2H20 or
[Cp*Ru(μ-H4)RuCp*].21 Since a sparingly soluble black
material was present at the conclusion of the reactions, we
sought to determine its role in the catalytic process. The
material was largely soluble in DMSO, but its IR and 1H NMR
516
dx.doi.org/10.1021/om200447z | Organometallics 2012, 31, 515−518

Organometallics
Communication
spectra were inconclusive.22 Importantly, the material was itself
catalytically inactive, as was ruthenium powder, indicating the
former to be a catalytic dead end and suggesting that the active
catalyst is homogeneous.
The production of both alkanes and truncated alkanes in the
present system is most simply explained by the intermediacy of
a metalloglycolate species (Scheme 3), which could undergo
Organic Transformations: A Guide to Functional Group Preparations;
VCH: New York, 1993; pp 155−156.
(4) (a) Deng, W.; Tan, X.; Fang, W.; Zhang, Q.; Wang, Y. Catal. Lett.
2009, 133, 167. (b) Yan, N.; Zhao, C.; Luo, C.; Dyson, P. J.; Liu, H.;
Kou, Y. J. Am. Chem. Soc. 2006, 128, 8714−8715.
(5) (a) Taher, D.; Thibault, M. E.; Di Mondo, D.; Jennings, M.;
Schlaf, M. Chem. Eur. J. 2009, 15, 10132−10143. (b) Thibault, M. E.;
DiMondo, D.; Jennings, M.; Abdelnur, P. V.; Eberlin, M. N.; Schlaf, M.
Green Chem. 2011, 13, 357−366.
(6) (a) Ghosh, P.; Fagan, P. J.; Marshall, W. J.; Hauptman, E.;
Bullock, R. M. Inorg. Chem. 2009, 48, 6490. (b) Schlaf, M.; Ghosh, P.;
Fagan, P. J.; Hauptman, E.; Bullock, R. M. Adv. Synth. Catal. 2009,
351, 789. (c) Schlaf, M.; Ghosh, P.; Fagan, P. J.; Hauptmann, E.;
Bullock, R. M. Angew. Chem., Int. Ed. 2001, 40, 3887.
(7) (a) Block, E. In Organic Reactions; Dauben, W. G., Ed.; Wiley:
New York, 1984; Vol. 30. (b) Jang, D. O.; Joo, Y. H. Synth. Commun.
1998, 28, 871.
Scheme 3. Possible Catalytic Pathway for Hydrogenolysis
and Hydrocracking
(8) Arceo, E.; Marsden, P.; Bergman, R. G.; Ellman, J. A. Chem.
Commun. 2009, 23, 3357−3359.
(9) Cook, G. K.; Andrews, M. A. J. Am. Chem. Soc. 1996, 118, 9448.
(10) (a) Gable, K. P.; Phan, T. N. J. Am. Chem. Soc. 1994, 116, 833.
(b) Gable, K. P. Organometallics 1994, 13, 2486. (c) Gable, K. P.;
Juliette, J. J. J. Am. Chem. Soc. 1996, 118, 2625. (d) Gable, K. P.;
Zhuravlev, F. A. J. Am. Chem. Soc. 2002, 124, 3970. (e) Gable, K. P.;
AbuBaker, A.; Zientara, K.; Wainwright, A. M. Organometallics 1999,
18, 173. (f) Gable, K. P.; Ross, B. ACS Symp. Ser. 2006, 921
(Feedstocks for the Future), 143.
(11) Ziegler, J. E.; Zdilla, M. J.; Evans, A. J.; Abu-Omar, M. M. Inorg.
Chem. 2009, 48, 9998.
(12) Yamane, Y.; Liu, X.; Hamasaki, A.; Ishida, T.; Haruta, M.;
Tokunaga, M. Org. Lett. 2009, 5162.
(13) (a) Vkuturi, S.; Chapman, G.; Ahmad, I.; Nicholas, K. M. Inorg.
Chem. 2010, 49, 4744. (b) Ahmad, I.; Chapman, G.; Nicholas, K. M.
Organometallics 2011, 30, 2810−2818.
(14) Arceo, E.; Ellman, J. A.; Bergman, R. G. J. Am. Chem. Soc. 2010,
132, 11408.
alternative redox fragmentations either to olefin10 (which is
subsequently hydrogenated) or to reduced-carbon aldehydes23
(which also are then reduced24).25 Both mono- and bimetallic
(bridged) glycolate species (A and B) can be envisioned, the
latter being precedented by [Cp*Ru(L)(μ-OR)2]2.26
In conclusion, we have disclosed here a new system for glycol
and epoxide conversion to alkanes which employs H2 as
reductant and [Cp*Ru(CO)2]2 as a precatalyst. A rare
hydrocracking reaction is also promoted by 1. Future
investigations will focus on elucidating the important catalytic
species and mechanisms of these reactions and their application
to biomass-derived polyols.
(15) Clapham, S. E.; Hadzovic, A.; Morris, R. H. Coord. Chem. Rev.
2004, 248, 2201−2237.
(16) King, R. B.; Iqbal, M. Z.; King, A. D. J. Organomet. Chem. 1979,
171, 53.
(17) GC analyses of aliquots provide yield of ethyl benzene/toluene,
time: 20/10%, 8 h; 26/16%, 23 h; 40/31%, 30 h.
ASSOCIATED CONTENT
■
(18) The potential cracking product, pentane, would not be
detectable because of its short GC retention time.
(19) Andrews, M. A.; Klaeren, S. A. J. Am. Chem. Soc. 1989, 111,
S
* Supporting Information
Text and figures giving experimental procedures and
representative gas chromatograms. This material is available
free of charge via the Internet at http://pubs.acs.org.
4131−4133.
(20) Cheng, T. -Y.; Bullock, R. M. Organometallics 2002, 21, 2325.
(21) (b) Marsh, R. E. Organometallics 1989, 8, 1584. (a) Suzuki, H.;
Omori, H.; Lee, D. H.; Yoshida, Y.; Moro-oka, Y. O. Organometallics
1988, 7, 2243−2245.
AUTHOR INFORMATION
■
Corresponding Author
(22) The black precipitate did not melt below 120 °C and almost
entirely dissolved in DMSO (10 mg/mL). The decomposition
temperature of [Cp*Ru(CO)2]2 is 297 °C;16 therefore, the black
material does not likely result from simple thermal decomposition of 1.
Its IR spectrum (KBr) showed no significant peaks in the 1800−2100
*Tel: 337 482 5677. Fax: +1 337 482 5676. E-mail: rss1805@
louisiana.edu.
ACKNOWLEDGMENTS
■
Financial support from the Louisiana Board of Regents (R.S.S.)
and from the Oklahoma Bioenergy Center (K.M.N.) is greatly
appreciated.
1
cm−1 M−CO region; its H NMR spectrum showed broad peaks at
1.091, 4.96, 2.5, 2.96, 5.9, and 7.4 ppm. When styrenediol (20 mg) and
15 mg of the black material were subjected to the usual reaction
conditions (24 h, 4 atm of H2), only the unreacted diol was detected.
The same result was obtained when the diol (20 mg) and Ru powder
(1.4 mg) were heated together under hydrogen.
REFERENCES
■
(1) Litchenthaler, F. W.; Mondal, S. Pure Appl. Chem. 1997, 69, 1853.
(2) (a) Chheda, J. N.; Huber, G. W.; Dumesic, J. A. Angew. Chem.,
Int. Ed. 2007, 46, 7164−7183. (b) Huber, G. W.; Iborra, S.; Corma, A.
Chem. Rev. 2006, 106, 4044. (c) Corma, A.; Iborra, S.; Velty, A. Chem.
Rev. 2007, 107, 2411.
(23) Pt−glycolateoxidative fragmentation: Andrews, M. A.; Gould,
G. L. Organometallics 1991, 10, 387−389.
(24) (a) Strohmeier, W.; Weigelt, L. J. Organomet. Chem. 1978, 145,
189−94. (b) Sanchez-Delgado, R. A.; De Ochoa, O. L. J. Mol. Catal.
1979, 6, 303−5. (c) Grey, R. A.; Pez, G. P.; Wallo, A. J. Am. Chem. Soc.
1981, 103, 7536−42. (d) Clarke, M. L.; Diaz-Valenzuela, M. B.;
Slawin, A. M. Z. Organometallics 2007, 26, 16−19.
(3) (a) Sharpless, K. B.; Umbreit, M. A.; Nieh, M. T.; Flood, T. C. J.
Am. Chem. Soc. 1972, 94, 6538. (b) Corey, E. J.; Hopkins, P. B.
Tetrahedron Lett. 1982, 23, 1979. (c) Larock, R. C. Comprehensive
517
dx.doi.org/10.1021/om200447z | Organometallics 2012, 31, 515−518

Organometallics
Communication
(25) Although the pathway by which C9−C13 hydrocarbons are
formed from tetradecanediol is unclear, we speculate that these could
be derived from isomerization of the terminal glycol to internal glycols
(via reversible dehydration/hydration), which in turn undergo
hydrocracking.
(26) Loren, S. D.; Campion, B. K.; Heyn, R. H.; Tilley, T. D.;
Bursten, B. E.; Luth, K. W. J. Am. Chem. Soc. 1989, 111, 4712−4718.
518
dx.doi.org/10.1021/om200447z | Organometallics 2012, 31, 515−518