Tandem semi-hydrogenation/isomerization of propargyl alcohols to saturated carbonyl analogues by dodecanethiolate-capped palladium nanoparticle catalysts

Article abstract of DOI:10.1039/c3ra42119h

The efficient one-pot conversion of propargyl alcohols to their saturated carbonyl analogues is carried out for the first time using metal nanoparticle catalysts, dodecanethiolate-capped Pd nanoparticles. Kinetic studies reveal that the reaction progresses through a semi-hydrogenation intermediate (allyl alcohols) followed by isomerization to carbonyls.

Full text of DOI:10.1039/c3ra42119h

RSC Advances
View Article Online
View Journal | View Issue
COMMUNICATION
Tandem semi-hydrogenation/isomerization of
propargyl alcohols to saturated carbonyl analogues
by dodecanethiolate-capped palladium nanoparticle
catalysts3
Cite this: RSC Advances, 2013, 3, 13642
Received 29th April 2013,
Accepted 7th June 2013
DOI: 10.1039/c3ra42119h
Diego J. Gavia,{ Jordan Koeppen,{ Elham Sadeghmoghaddam and Young-
www.rsc.org/advances
Seok Shon*
The efficient one-pot conversion of propargyl alcohols to their
saturated carbonyl analogues is carried out for the first time using
metal nanoparticle catalysts, dodecanethiolate-capped Pd nano-
particles. Kinetic studies reveal that the reaction progresses
carbonyls has been reported. Until now, therefore, two-step
synthetic methods such as the full hydrogenation followed by
alcohol oxidation or the isomerization followed by selective
reduction of CLC bond are considered to be the most effective
ways for this particular conversion.7,8
through
a semi-hydrogenation intermediate (allyl alcohols)
followed by isomerization to carbonyls.
In our previous studies, the synthesis of dodecanethiolate-
protected palladium nanoparticles generated from sodium
S-dodecylthiosulfate precursors yielded nanoparticles with a lower
ligand surface coverage than those synthesized with dodecanethiol
ligands.9 Used as catalysts, these palladium nanoparticles were
capable of isomerizing 2-propen-1-ol (allyl alcohol) to propanal
The ease of preparation of propargyl alcohols1,2 and the broad
utility of their reaction products have resulted in a vast library of
effective reactions involving these compounds.3–5 For example,
propargyl alcohols can undergo full hydrogenation to saturated
alcohols in the presence of an activated metal such as Pt or Pd in a
pressurized, hydrogen rich environment.3 Moreover, the semi-
hydrogenation of propargyl alcohols, in which a triple bond is
reduced to a double bond without any further hydrogenation, can
be achieved by employing mildly poisoned catalysts, such as
Lindlar’s catalyst.4 Even the isomerization of propargyl alcohols to
a,b-unsaturated carbonyls have been attained utilizing ruthenium
catalysts as reported by Trost et al.5
However, the catalytic reactions of propargyl alcohols are still
somewhat deficient due to the lack of an effective one-pot
synthetic method allowing for the conversion of propargyl alcohols
directly to saturated carbonyls, which are important intermediates
for pharmaceutical and industrial applications. To our knowledge,
the one-pot conversion of the propargylic alcohol subgroup to its
saturated carbonyl has only been achieved by Pd(OAc)2-catalysed
hydrogenation of methyl 4-hydroxy-4-phenyl-2-butynoate, using
excess HCO2H/NBu3 as the hydrogen source, to the corresponding
c-ketoester in a moderate yield of 65%.6 No example involving the
direct conversion of simple propargyl alcohols to small saturated
with
a rather high selectivity over the hydrogenation to
1-propanol.9 However, these catalytic materials revealed limita-
tions when employing allyl alcohol substrates with more than one
substituent around the CLC bond.10 Recently, the optimization of
these palladium nanoparticles was achieved by performing
systematic variations of the synthetic parameters, such as reactant
concentration and temperature.11 This allowed the further
manipulation of the surface ligand density and the core size of
the dodecanethiolate-capped palladium nanoparticles. The opti-
mized palladium nanoparticles with the lower surface ligand
coverage could produce isomerization products in high yields even
for the substituted allyl alcohols.11
In this article, for the first time, we report the successful one-
pot conversion of simple propargyl alcohols to their saturated
carbonyl analogues using metal nanoparticle catalysts.12,13 The
effects of hydrogen gas concentration, reaction temperature, and
substrate structures were investigated. Moreover, the kinetic
profile of this reaction was obtained to understand the reaction
mechanism.
Palladium catalysts employed in this study are two different
nanoparticles for which the detailed synthesis and characteriza-
Department of Chemistry and Biochemistry, California State University, Long Beach.
1250 Bellflower Blvd., Long Beach, CA 90840-9507, United States.
tion are documented in ESI. 11 A palladium nanoparticle catalyst,
3
E-mail: ys.shon@csulb.edu; Fax: (562) 985-8557; Tel: (562) 985-4466
yPd1289(SC12H25)164, which has an average core size of y3.4 nm
with a moderately low ligand surface coverage (y0.34 ligands/
surface Pd atoms),11,14 was initially applied to the catalytic reaction
of 2-propyn-1-ol (propargyl alcohol) in the presence of 10 mmol H2
gas at 0 uC (Table 1). The reaction was analysed with 1H NMR after
Electronic supplementary information (ESI) available: The synthetic procedure for
3
Pd nanoparticles, the characterization methods, the procedure for the catalytic
reactions, the mechanism of the tandem reaction, the catalytic reaction results of
3-butyn-2-ol, TEM results before and after the catalytic reaction, and TGA results. See
DOI: 10.1039/c3ra42119h
{ Authors contributed equally to this study.
13642 | RSC Adv., 2013, 3, 13642–13645
This journal is ß The Royal Society of Chemistry 2013

View Article Online
RSC Advances
Communication
Table 1 The conversion of 2-propyn-1-ol to propanal by dodecanethiolate-capped PdNP catalysts at 0 uCa
Catalysis yields (%)c
TGA
% Pd
TEM
Diameter (nm)
Catalystsb
Ligands/surface atomsb
Semi-hydrogenation
Full-hydrogenation
Tandem reaction
TOFd
Pd1289L164
Pd116L59
80.6
51.0
3.38 ¡ 0.95
1.51 ¡ 0.46
0.34
0.75
31
28
32
5
37
0
123
0
a
b
Reaction conditions: 50 mL 2-propyn-1-ol; 5 mol% Pd; y10 mmol H2; 2 mL CDCl3; 6 h. Calculations were based on theoretical models
provided by Hostetler et al. (ref. 14) along with TGA and TEM data. Yields were determined by 1H NMR. Initial TOFs are based on the mole
c
d
isomerized per mol Pd atoms per hour.
6 h, which inferred the formation of a mixture of 2-propen-1-ol
(semi-hydrogenation product), 1-propanol (full-hydrogenation
product), and propanal (tandem semi-hydrogenation and isomer-
ization product; in short tandem product) with fairly similar
yields. Although there was no clear selectivity over any product, the
generation of the tandem product was clearly observed from the
reaction using these thiolate-poisoned palladium nanoparticle
catalysts. The other palladium nanoparticle catalyst utilized in this
reaction, yPd116(SC12H25)59, featured an average core size of y1.5
nm.11 However, due to the higher surface ligand coverage (y0.73
ligands/surface Pd atoms) of these palladium nanoparticle
catalysts, they exhibited poor catalytic activities and produced
mainly semi-hydrogenation products in a low yield of 28% without
any evidence of the tandem product.
A palladium nanoparticle with a lower surface ligand density,
yPd1289(SC12H25)164, was, therefore, chosen as a catalyst for the
optimization of reaction conditions. An increase in reaction
temperature to room temperature (y25 uC) resulted in a higher
reaction conversion indicated by the absence of 2-propen-1-ol, the
semi-hydrogenation product, along with the upsurge in yields of
both propanal and 1-propanol (Table 2, entry 1). However, the
selectivity between 1-propanol (the full hydrogenation product)
and propanal (the tandem product) remained mostly the same
even at the increased temperature (46 : 54 to 42 : 58).
hydride to the p bond of allyl alcohol.15 The conversion to
saturated carbonyls (the isomerization products) requires the b-H
elimination of a Markovnikov intermediate (a branched Pd–alkyl
intermediate). This b-H elimination step can theoretically be more
favoured over the insertion of another hydride when Pd
nanoparticles have hydride undersupplied surface. Therefore,
the effect of hydrogen gas concentration was examined for the
catalytic reaction of 2-propyn-1-ol after decreasing the hydrogen
gas equivalent for a five-fold while maintaining all other reaction
conditions. The results showed that the selectivity towards
propanal increases with the decrease in concentration of hydrogen
gas (Table 2, entry 2). A further decrease in hydrogen gas
concentration, however, resulted in an incomplete reaction.
Similar results were also observed from the catalytic reaction of
3-butyn-2-ol with the variation of hydrogen gas concentration (see
ESI ).
3
The kinetic profile of the conversion of 2-propyn-1-ol is
illustrated in Fig. 1. As the data dictates, 2-propyn-1-ol was first
converted to its semi-hydrogenation product, 2-propen-1-ol, with-
out producing any full hydrogenation or the tandem products in
the early stage of the reaction. It took about 2 h for the
concentration of 2-propyn-1-ol to decrease below the concentration
of the semi-hydrogenation product, 2-propen-1-ol, in the reaction
mixture. After this, the allyl alcohol intermediate onset its
conversion to either propanal or 1-propanol. The stronger and
thermodynamically driven adsorption of alkyne on the palladium
surface compared to alkene has well been documented in other
publications.16,17 The kinetic profile in Fig. 1 was consistent with
the high adsorption constant of the 2-propyn-1-ol, only leading to
the isomerization of 2-propen-1-ol after consuming more than a
Prior studies involving the isomerization of allyl alcohols by
palladium nanoparticle catalysts in the presence of hydrogen gas
proposed that the catalytic reaction transpires through a Pd–alkyl
mechanism (see ESI for the proposed mechanism of the tandem
3
reaction).15 Pd–H species generated upon the adsorption of
hydrogen on the nanoparticle surface can ultimately insert a
Table 2 The optimization of the conversion of propargyl alcohols to carbonyl analogous using Pd1289L164 catalysts at 25 uC: the effects of temperature, H2 molar
equivalents, and the substratesa
Catalysis yields (%)b
Entry
Propargyl alcohol
R1
R2
mmol H2
Semi-hydrogenation
Full-hydrogenation
Tandem reaction
TOFc
1
2
3
4
5
2-propyn-1-ol
2-propyn-1-ol
3-butyn-2-ol
2-butyn-1-ol
3-phenyl-2-propyn-1-ol
H
H
CH3
H
H
H
H
H
CH3
Ph
10
2
2
2
2
0
0
0
43
24
42
30
20
13
1
58
70
76
41
39
290
350
380
205
195
a
b
c
Reaction condition: 50 mL of propargyl alcohol; 5 mol% Pd, 2 mL CDCl3; 4 h. Yields were determined by 1H NMR. Initial TOFs are based
on the mole isomerized per mole Pd atoms per hour.
This journal is ß The Royal Society of Chemistry 2013
RSC Adv., 2013, 3, 13642–13645 | 13643

View Article Online
Communication
RSC Advances
to that of other mono-substituted semi-hydrogenation intermedi-
ates. Nonetheless, the selectivity towards the isomerization
product (butanal) over the full hydrogenation product (1-butanol)
still remained the same (.3 times more favoured than the full
hydrogenation product). Furthermore, the catalytic reaction of
3-phenyl-2-propyn-1-ol (entry 5), with a phenyl group at the R2
position, resulted in almost 35% of unreacted substrate. The
accessibility of 3-phenyl-2-propyn-1-ol to the surface reaction sites
on Pd nanoparticle was likely lower due to the steric interference
of ligand surrounding nanoparticles.18 The more rigid nature of
the substrate should also have contributed to the slower reaction
by the limited access to the surface reaction sites. Although the
catalytic reactivity was quite lower for this substrate, its selectivity
with regards to the tandem reaction (39%) over the full-
hydrogenation (1%) was much higher.
Fig. 1 The kinetic profile of the catalytic conversion of propagyl alcohol to
1-propanal. Kinetic data was obtained by monitoring the reaction progress using
1H NMR.
half of 2-propyn-1-ol. Subsequent isomerization and hydrogena-
tion of 2-propen-1-ol took place in logarithmic reaction gradients
confirming a first order reaction.15
To examine the stability of catalysts, the Pd nanoparticles were
isolated from the reaction mixture by methanol-induced precipita-
tion from the homogeneous solution. The precipitates were re-
Conclusions
It has been shown that palladium nanoparticles synthesized from
sodium S-dodecylthiosulfate precursors were able to convert small
propargyl alcohols to their saturated carbonyl analogues under a
one-pot condition. The catalytic reactions were performed under
fairly mild conditions and without any other reagents besides
palladium catalysts and H2 gas. Overall, the results clearly showed
the significance of highly active homogeneous nanoparticle
catalysts, which are capable of providing a high selectivity towards
atypical products that are not possible through the traditional
groups of catalysts. Future studies will focus on improving
selectivity towards the tandem products by controlling the
structure/density of capping ligands and introducing chemical
additives for secondary surface poisoning.
dissolved in organic solvents for characterization by TEM (see ESI )
3
and UV-Vis spectroscopy. TEM results showed the high population
of small Pd nanoparticles without any notable morphological
change. UV-Vis results also confirmed the absence of the
absorption bands corresponding to both Pd(II) species and
oxidized Pd. The earlier report from our group has also shown
the high recyclability of dodecanethiolate-capped Pd nanoparticles
in a similar reaction condition.15
The yPd1289(SC12H25)164 catalyst was further tested against
three additional commercially available substituted propargyl
alcohols as summarized in Table 2 (entries 3–5) and Scheme 1.
The conversion of 3-butyn-2-ol (entry 3) resulted in yields far
congruent to its unsubstituted equivalent. Alkyl substituents at the
R1 position (Scheme 1) seemed rather beneficial towards
isomerization to saturated carbonyls since their presence essen-
tially increases the thermodynamic stability of the enol inter-
mediate.10 As for 2-butyn-1-ol (entry 4) which contains a methyl
substituent at the R2 position, the overall conversion from the
semi-hydrogenation product (2-buten-1-ol) to butanal and 1-buta-
nol was far more difficult to achieve. This is likely due to the
higher stability of 2-buten-1-ol (a di-substituted alkene) compared
Acknowledgements
The authors gratefully acknowledge the financial support from
ACS-PRF (49407-UR7) and CSULB (RSCA, MGSS, and RSA).
Notes and references
1 C.-J. Li, Acc. Chem. Res., 2010, 4, 581–590.
2 J. Ito, R. Asal and H. Nishiyama, Org. Lett., 2010, 12, 3860–3862.
3 Handbook of Homogenous Hydrogenation, ed. J. G. de Vries and
C. J. Elsevier, Wiley-VCH, Weinheim, 2007, vol. 1, pp. 375–409.
4 (a) H. Lindlar and R. Dubuis, Org. Synth., 1973, 5, 880–883; (b)
´
´
´
B. Bridier, N. Lopez and J. Perez-Ramırez, Dalton Trans., 2010,
39, 8412–8419.
5 (a) B. M. Trost and R. C. Livingston, J. Am. Chem. Soc., 1995,
117, 9586–9587; (b) B. M. Trost and R. C. Livingston, J. Am.
Chem. Soc., 2008, 130, 11970–11978.
6 A. Arcadi, E. Bernocchi, A. Burini, S. Cacchi, F. Marinelli and
B. Pietroni, Tetrahedron, 1988, 44, 481–490.
7 (a) M. S. Kwon, N. Kim, C. M. Park, J. S. Lee, K. Y. Kang and
J. Park, Org. Lett., 2005, 7, 1077–1079; (b) J. Zheng, S. Lin,
X. Zhu, B. Jiang, Z. Yang and Z. Pan, Chem. Commun., 2012, 48,
6235–6237.
8 M. S. Ide, B. Hao, M. Neurock and R. J. Davis, ACS Catal., 2012,
2, 671–683.
9 E. Sadegmoghaddam, C. Lam, D. Choi and Y.-S. Shon, J. Mater.
Chem., 2011, 21, 307–312.
Scheme 1 The conversion of propargyl alcohols (R1 = H or CH3; R2 = H, CH3, or Ph)
to their carbonyl analogues using PdNP catalysts.
13644 | RSC Adv., 2013, 3, 13642–13645
This journal is ß The Royal Society of Chemistry 2013

View Article Online
RSC Advances
Communication
¨
10 E. Sadegmoghaddam, K. Gaıeb and Y.-S. Shon, Appl. Catal., A,
D. Wignall, G. L. Glish, M. D. Porter, N. D. Evans and R.
W. Murray, Langmuir, 1998, 14, 17–30.
2011, 405, 137–141.
11 D. J. Gavia and Y.-S. Shon, Langmuir, 2012, 28, 14502–14508.
15 E. Sadegmoghaddam, H. Gu and Y.-S. Shon, ACS Catal., 2012,
2, 1838–1845.
´
12 (a) M. Crespo-Quesada, F. Cardenas-Lizana, A.-L. Dessimoz and
L. Kiwi-Minsker, ACS Catal., 2012, 2, 1773–1786; (b) M. Crespo-
Quesada, A. Yarulin, M. Jin, Y. Xia and L. Kiwi-Minsker, J. Am.
Chem. Soc., 2011, 133, 12787–12794.
16 A. Molnar, A. Sarkany and M. Varga, J. Mol. Catal. A: Chem.,
2001, 173, 185.
17 M. Crespo-Quesada, R. R. Dykeman, G. Laurenczy, P. J. Dyson
and L. Kiwi-Minsker, J. Catal., 2011, 279, 66.
18 (a) S. Bhattacharjee and M. L. Bruening, Langmuir, 2008, 24,
2916–2920; (b) S. Kidambi, J. Dai, J. Li and M. L. Bruening, J.
Am. Chem. Soc., 2004, 126, 2658–2659.
13 (a) M. W. van Laren and C. J. Elsevier, Angew. Chem., Int. Ed.,
1999, 38, 3715–3717; (b) M. P. R. Spee, J. Boersma, M. D. Meijer,
M. Q. Slagt, G. van Koten and J. W. Geus, J. Org. Chem., 2001,
66, 1647–1656.
14 M. J. Hostetler, J. E. Wingate, C.-J. Zhong, J. E. Harris, R.
W. Vachet, M. R. Clark, J. D. Londono, S. J. Green, J. J. Stokes, G.
This journal is ß The Royal Society of Chemistry 2013
RSC Adv., 2013, 3, 13642–13645 | 13645