10.1002/chem.201705208
Chemistry - A European Journal
COMMUNICATION
Under standard reaction conditions, phenylacetaldehyde was converted to 13C-NMR, by performing the reaction in a sealed J. Young NMR-tube,
product 2a in only 25% isolated yield. Under identical conditions except for though neither CO2 nor CO was observed.
the exclusion of base, the non-decarbonylated coupled aldol condensation
Concerning the active catalyst speciation, we followed the reaction by
product was obtained in 56% yield. This led us to believe that the base has GC analysis of aliquots taken during the course of the reaction and plotted a
an important role in the net decarbonylation step. When performing the same a kinetic trace of yield versus time. An induction period is observed, figure
reaction exclusively with base and no Ru-precatalyst, we indeed observed S2, indicating formation of
a new catalytically active species likely
both aldol coupling, condensation and decarbonylation albeit in only 22% involving formation of nanoparticles. Following this trace, the reaction
yield. We then considered the possibility that after performing mixture was commonly observed by visual inspection to gradually darken
dehydrogenation of the alcohol, the Ru-catalyst could be facilitating the during this time and turn black, though we still cannot fully exclude the
aldol chemistry through its Lewis acidic properties. A Lewis acid pathway possibility of formation of another active homogeneous catalyst species.
was tested through performing the reaction in the presence of Cu(OTf)2 with Although the mercury test is not necessarily applicable to ruthenium
and without base, respectively. In the presence of base, the same product was catalysis, in a separate experiment we added a drop of mercury to the
obtained in a similar yield as without the Lewis acid, and without base 2- reaction mixture without any observable decrease in activity. While the
phenylnaphthalene was obtained.[21] No reaction occurred with 2- kinetic trace is highly supportive of that RuCl2(PPh3)3 is only a pre-catalyst,
phenylethanol with only 5 mol% Cu(OTf)2. From these reactions, we cannot the negative mercury test should most likely only be considered as an
completely exclude the possibility of ruthenium promoting the coupling indication of its non-applicability in this catalytic system.[25] In the kinetic
through Lewis acid-type participation, though it is clear that its presence is experiments only the starting material and product were observed, no
required for the overall reaction to occur. This is further indicated by the intermediates could be detected.
significant reduction in aldol-coupling yield in the absence of ruthenium pre-
In summary, the fully deoxygenative coupling of arylethanols is herein
catalyst. On this basis, we consider three mechanistic possibilities for the reported to produce 1,3-diarylpropenes in good yields. The reaction is
coupling and net decarbonylation reaction, scheme 6.
performed by the use of versatile and commercially available RuCl2(PPh3)3.
The observed dark reaction mixture combined with that the kinetic trace,
which shows a sigmoidal curvature is indicative of that the Ru-complex is
only a pre-catalyst. Control experiments further indicate that the base has an
important role in the second deoxygenation or net decarbonylation to
produce the hydrocarbon product. To the best of our knowledge, this is the
first report of a Ru-catalyzed fully deoxygenative coupling of alcohols. We
believe that these results could enable further development in the field of
alcohol alkylation reactions and other deoxygenation reactions as for
example relevant to biomass upgrading. Further and more detailed
mechanistic studies are currently in progress in our laboratories.
OH
Ph
"Ru"
O
Ph
cat. aldol
-O
OH
Ph
O
path A
Ph
Ph
- OH-
- CO2
- H2
Ph
Ph
Ph
OH
OH
tBuO- + H2
O
OH- + tBuOH
O
H2
O
Acknowledgements
path B
Ph
Ph
- OH-
- H2O
Ph
Ph
Ph
Ph
- CO
Financial support by the Olle Engkvist Byggmästare foundation, the Royal
Swedish Academy of Forestry and Agriculture, the Magnus Bergvall
Foundation and the Royal Physiographic Society in Lund and the Science
and Engineering Research Board, India, SB/OS/PDF-004/2015-16 (SM) is
gratefully acknowledged.
"Ru"
O
Ru
1. Decarbonylation
2. Red. elimination
H
path C
Ph
Ph
Ph
Ph
Scheme 6. Possible mechanistic pathways for the net decarbonylation.
The first step involves dehydrogenation by ruthenium, followed by a
catalytic aldol coupling to generate the β-hydroxyaldehyde. Typically in
hydrogen-borrowing chemistry, the base is generally thought to be
responsible for catalyzing the aldol reaction and condensation, though it is
clear in this work that the aldol reaction from the aldehyde intermediate can
also occur in the absence of base as shown in scheme 5. The synthesis of
1,3-diphenylpropene has on several occasions been reported to be achieved
quantitatively through self-condensation of phenylacetaldehyde in ethanol
solvent in the presence of KOH.[22] The β-hydroxyaldehyde could as in path
A proceed through nucleophilic attack of hydroxide at the carbonyl, to
eliminate formate in the form of CO2 and H2, in addition to regenerating the
base. This reaction was, in a basic ethanol solution without transition metal
catalysis, proposed by Stoermer.[22c] Decarbonylation involving a carbanion
intermediate as in path B has been proposed with 1,3-dihalosubstituted
benzaldehydes[23], and compared to aryl anions, benzylic anions should be
significantly more accessible from an energetic perspective. Although
transition metal-catalyzed decarbonylation is considerably more common
with rhodium complexes, ruthenium complexes are also known to catalyze
this transformation as in path C.[24] In an early attempt to get further insight
into the mechanism, we attempted to characterize the gaseous byproduct by
Keywords: dehydrogenation; hydrogen-borrowing catalysis; ruthenium
complexes; metal hydrides; decarbonylation
[1]
[2]
[3]
M. H. S. A. Hamid, P. A. Slatford, J. M. J. Williams, Adv. Synth. Catal.
2007, 349, 1555-1575.
a) F. Huang, Z. Q. Liu, Z. K. Yu, Angew. Chem. Int. Ed. 2016, 55, 862-
875; b) Y. Obora, ACS Catal. 2014, 4, 3972-3981.
a) R. Yamaguchi, K. Fujita, M. W. Zhu, Heterocycles 2010, 81, 1093-
1140; b) Q. Yang, Q. F. Wang, Z. K. Yu, Chem. Soc. Rev. 2015, 44,
2305-2329.
[4]
[5]
D. Wang, D. Astruc, Chem. Rev. 2015, 115, 6621-6686.
C. S. Cho, B. T. Kim, H. S. Kim, T. J. Kim, S. C. Shim, Organometallics
2003, 22, 3608-3610.
[6]
[7]
R. Martinez, D. J. Ramon, M. Yus, Tetrahedron 2006, 62, 8982-8987.
A. Prades, M. Viciano, M. Sanau, E. Peris, Organometallics 2008, 27,
4254-4259.
[8]
[9]
H. W. Cheung, T. Y. Lee, H. Y. Lui, C. H. Yeung, C. P. Lau, Adv. Synth.
Catal. 2008, 350, 2975-2983.
a) D. Gnanamgari, C. H. Leung, N. D. Schley, S. T. Hilton, R. H.
Crabtree, Org. Biomol. Chem. 2008, 6, 4442-4445; b) D. Gnanamgari,
This article is protected by copyright. All rights reserved.