Light Mediated Preparation of Palladium Nanoparticles as Catalysts for Alkyne cis-Semihydrogenation

Article abstract of DOI:10.1021/acs.orglett.7b00999

A bisacylphosphine oxide photoinitiator was used for the light mediated preparation of palladium nanoparticles (PdNPs) with a small diameter of 2.8 nm. All starting materials are commercially available, and PdNP synthesis is experimentally very easy to conduct. The PdNP-hybrid material was applied as catalyst for the semihydrogenation of various internal alkynes to provide the corresponding alkenes in excellent yields (up to 99%) and Z-selectivities (Z/E ratios up to 99/1).

Full text of DOI:10.1021/acs.orglett.7b00999

Letter
pubs.acs.org/OrgLett
Light Mediated Preparation of Palladium Nanoparticles as Catalysts
for Alkyne cis-Semihydrogenation
Florian Masing,† Harald Nusse,‡ Jurgen Klingauf,‡ and Armido Studer*,†
̈
̈
̈
†Institute of Organic Chemistry, University of Munster, Corrensstrasse 40, 48149 Munster, Germany
̈
̈
‡Institute of Medical Physics and Biophysics, University of Munster, Robert-Koch-Strasse 31, 48149 Munster, Germany
̈
̈
S
* Supporting Information
ABSTRACT: A bisacylphosphine oxide photoinitiator was
used for the light mediated preparation of palladium
nanoparticles (PdNPs) with a small diameter of 2.8 nm. All
starting materials are commercially available, and PdNP
synthesis is experimentally very easy to conduct. The PdNP-
hybrid material was applied as catalyst for the semi-
hydrogenation of various internal alkynes to provide the
corresponding alkenes in excellent yields (up to 99%) and Z-
selectivities (Z/E ratios up to 99/1).
homolysis to form a phosphinoyl/benzoyl (3/4) radical pair.2,3
The phosphinoyl radical 3 then acts as a one-electron reductant
for the Ag/Au-metal salts to generate metal nanoparticles
(Scheme 1).2,3 Second, the photochemically formed BAPO-
fragments 3 and 4 can initiate the polymerization of monomers
that are present in the reaction mixture as cosolvents. These in
situ formed polymers then prevent the aggregation of the
generated Ag/Au-nanoparticles.2,3 However, to our knowledge
commercial P-based radical photoinitiators have not been used
for Pd-nanoparticle preparation to date.
Metal nanomaterials are well-known for their unique
reactivity. For instance, Pd, Fe, Ni, or Au based nanoparticles
have been applied as catalysts for alkyne semihydrogenation to
form Z-alkenes.4 These new nanomaterials may replace in
future Lindlar’s catalyst, the current standard semihydrogena-
tion catalyst in synthesis. It is well-known that the Lindlar
catalyst suffers from several drawbacks: toxic Pb(OAc)2 and
large amounts of quinoline are required to suppress over-
hydrogenation of the target alkenes to alkanes.
henylbis(2,4,6-trimethylbenzoyl)phosphine oxide (1,
PBAPO) and diphenyl(2,4,6-trimethylbenzoyl)phosphine
oxide (2, TMDPO) are well established and widely used
photoinitiators for radical polymerization in industry that have
found applications in various processes such as UV-curing of
coatings, adhesives, inks, or dental fillings (Scheme 1).1
Moreover, the application of BAPO 1 for the photochemical
preparation of AgNP- or AuNP-polymer hybrid materials was
reported.2 In these processes, BAPO 1 plays a dual role: first,
BAPO 1 engages in a photomediated Norrish type 1 P−C bond
Scheme 1. (a) Photoinitiators Investigated for the Light
Mediated Preparation of PdNPs; (b) Norrish Type 1
Photolysis of Photoinitiators 1 and 2; (c) Phosphinoyl
Radical As Reductant for Metal Ions
Furthermore, low Z-selectivity due to Z/E-isomerization,
irreproducibility, and a limited substrate scope are challenging
factors.4i−k,5 Considering metal nanocatalysts for alkyne
semihydrogenation as alternatives, they generally have to be
prepared by a complex multistep procedure including the use of
noncommercial chemicals. Furthermore, many of them operate
under harsh reaction conditions limiting their applicability.6
Recently, our group introduced a light mediated process for
the preparation of polymer coated Au and Pd nanoparticles by
using well-defined but noncommercial homo- and copolymers
bearing photoactive α-hydroxyalkyl ketone (HAK) substi-
tuents.4l,7 During irradiation with UV-light the HAK moieties
undergo Norrish type 1 cleavage to generate ketyl radicals able
Received: April 3, 2017
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.7b00999
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
to reduce metal ions to form metal nanoparticles. The
photochemically modified polymers formed as byproducts
then stabilize the metal nanoparticles.4l,7 Inspired by these
results, we decided to investigate PdNP preparation using
acylphosphine oxide photoinitiators 1 and 2 as Norrish type 1
active species. In contrast to the photoactive polymers used
before,4l,7 initiators 1 and 2 are commercially available and
cheap, offering an easy and valuable entry to palladium
nanomaterials. Herein, we present first results along those lines.
For nanoparticle preparation, phenylbis(2,4,6-trimethyl-
benzoyl)phosphine oxide (1) or diphenyl(2,4,6-trimethyl-
benzoyl)phosphine oxide (2) was diluted in DMF and
Pd(OAc)2 was added. The ratio of photoactive moieties to
Pd(OAc)2 was set as 20:1. Irradiation under an argon
atmosphere in a photoreactor (λ = 254 nm) for 10 min using
bisacylphosphine oxide 1 provided a deep brown solution
indicating Pd nanoparticle formation. Indeed, TEM inves-
tigations revealed successful preparation of Pd nanoparticles
Scheme 2. Proposed Mechanism of the Light Mediated
PdNP Formation
(Pd@1*) with a diameter of 2.8
1.0 nm (Figure 1a). By
and unreacted Norrish initiators (1 and 2) stabilize the thus
generated PdNPs. Isolation of the PdNP-hybrid material as a
black solid material was achieved by evaporation of DMF, and
this hybrid was then used directly in catalysis as is.
As a first reaction the hydrogenation of ethyl 3-phenyl-
propiolate (9a) in DMF at 40 °C with H2 (1 atm, balloon) as a
reductant was investigated. Pleasingly, with nanoparticles
prepared from BAPO 1 (Pd@1*, 1 mol %) the semi-
hydrogenation product 10a was obtained in near-quantitative
yield (99%) and excellent Z/E selectivity (99/1) after 3 h
(Table 1, entry 1). The corresponding overhydrogenation
Figure 1. TEM images of the prepared palladium nanoparticles: (a)
Pd@1* and (b) Pd@2*.
Table 1. Semihydrogenation of Ethyl 3-Phenylpropiolate
(9a) Using Various Pd Catalysts
switching to acylphosphine oxide 2 under otherwise identical
conditions, polydisperse Pd nanoparticles (Pd@2*) between
40 and 900 nm (mean diameter: 177 126 nm) were obtained,
documenting the large effect of the initiator structure on
nanoparticle formation (Figure 1b). The Pd oxidation state in
the hybrid material was investigated by powder X-ray
diffraction (see the Supporting Information). Only the broad
diffraction signals characteristic for Pd0 nanoparticles were
detected proving full photochemical reduction of the Pd-salt
within the 10 min of irradiation.8
By mass spectrometric analysis of the irradiated PdNP
solutions, phosphinic acid amide 6 and amide 8 were identified
along with unreacted 1 or 2 revealing that the photochemical
decomposition of the initiators was not complete within 10 min
of irradiation (Scheme 2). Considering the experimentally
identified fragmentation products, the following mechanism for
PdNP formation is suggested. Phosphinoyl/benzoyl (3/4)
radical pairs are generated by Norrish Type 1 homolysis.9 The
P-radical 3 then acts as a one-electron donor to reduce Pd-ions
thereby being oxidized to an oxophosphonium cation 5, which
is eventually trapped by dimethylamine derived from DMF to
amide 6.3 The concomitantly generated benzoyl radical 4 is
trapped with dimethylamine to give after deprotonation a ketyl
radical anion 7, which is able to act as a one-electron donor for
Pd salt reduction explaining the formation of amide 8 as a
second byproduct (Scheme 2).4l,10 We do not assume that
DMF and the trace amounts of dimethylamine are the
reductants for Pd-ions in this system, since the formation of
PdNPs was significantly slower in a control experiment in the
absence of photoinitiators 1 and 2.11 It is likely that amides 6, 8,
conv 9a
a
yield 10a (%)
yield 11a
a
a
entry Pd source t (h)
(%)
(Z/E)
(%)
1
2
3
4
5
6
7
8
Pd@1*
Pd@1*
Pd@1*
Pd@2*
Pd/C
3
>99
>99
>99
1
99 (99/1)
99 (99/1)
93 (97/3)
1 (n.d.)
33 (91/9)
0
<1
<1
6
8
b
2.5
8
0
1.5
2
98
65
99
74
99
Pd/C
>99
>99
>99
Pd(OAc)2
Pd(OAc)2
3
25 (91/9)
0
4
a
Conversion of alkyne 9a, yield and Z/E ratio of alkene 10a, and yield
of alkane 11a determined by GC-FID using an internal standard. GC
peaks assigned by GC-MS. Structure of 10a confirmed by NMR
spectroscopy. Catalyst recycling, reaction performed at room
temperature. n.d. = not determined.
b
product 11a was identified in traces only (<1%) by gas
chromatography. Extending the reaction time to 8 h did not
change the outcome, clearly showing that Z/E isomerization
and overhydrogenation are kinetically disfavored processes
under the applied conditions (Table 1, entry 2).
The Pd@1* catalyst was recycled via removal of the solvent
and all reaction products by distillation. Pd@1* remaining in
the reaction flask was applied to the next cycle by recharging
with DMF, alkyne 9a, and H2. Interestingly, the recycled
catalyst was found to be more active and the reaction was
B
DOI: 10.1021/acs.orglett.7b00999
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
completed within 2.5 h at room temperature. The desired
alkene 10a was obtained in a good yield (93%) (Table 1, entry
3). However, selectivity slightly dropped (97/3) and over-
hydrogenated alkane 11a was formed in 6% yield, revealing that
the original catalyst performs better.
Scheme 3. Scope
PdNPs prepared with TMDPO 2 (Pd@2*) are not active
semihydrogenation catalysts, and only 1% conversion was
achieved after 8 h under otherwise identical conditions (Table
1, entry 4). This change in reactivity may be explained
considering the large diameter differences of the hybrids (see
Figure 1) or by ligand effects.12 To document the unique
reactivity of Pd@1* exerted by the ligands 1, 6, and 8,
semihydrogenation of alkyne 9a was repeated with Pd/C and
Pd(OAc)2 under otherwise identical conditions. Pd/C was
found to be more active than Pd@1*, and complete conversion
was obtained within 1.5 h (Table 1, entry 5). However, the
yield of alkene 10a was low (33%) and the Z/E-selectivity was
only 91/9. In this case, alkane 11a was formed as the major
product in 65% yield. Extending the reaction time to 2 h
afforded alkane 11a in a quantitative yield (Table 1, entry 6). A
similar result was achieved with Pd(OAc)2 as a precatalyst
(Table 1, entries 7, 8).13
We next tested the substrate scope using alkynes 9b−s as
substrates (Scheme 3). Reactions were followed by GC
analysis, and the reaction time was adapted individually
according to the alkyne reactivity. Conversion, yields, and Z/
E ratios of alkenes 10b−s and yields of alkanes 11b−s were
determined by GC-FID using an internal standard. GC peaks
were assigned by GC-MS, and structures of Z-alkenes 10b−s
were confirmed by NMR spectroscopy. Isolated yields for Z-
alkenes 10a−s that are provided in the Supporting Information
are slightly lower. For methyl 3-phenylpropiolate (9b) full
conversion was achieved at room temperature within 4.5 h to
give 10b in quantitative yield (99%) and excellent Z/E ratio
(99/1). Various arylpropyl alkynes 9c−h were investigated, and
electronic effects exerted by the aryl substituent studied. The 3-
pyridyl derivate 9c was fully converted at 40 °C for 3.5 h to the
alkene 10c (99%, Z/E = 99/1). Arylpropyl alkynes 9d−h
bearing formyl-, keto-, nitro-, chloro-, or trifluormethoxy-
substituents were smoothly converted to the corresponding
alkenes 10d−h in good to excellent yields (92−97%) with
excellent Z/E ratios (97/3−99/1) in 3−8 h. Importantly,
functional groups such as formyl, keto, or nitro, which are
readily reduced using classical Pd-catalysts, are tolerated.14 In
these reactions, up to 3% of the undesired alkanes 11d−h were
identified. We then tested more sterically shielded alkynes and
found the aryl isopropyl congener 9i to be converted in 3.5 h at
40 °C to 10i in excellent yield (97%) and Z/E selectivity (98/
2). However, the catalyst loading had to be increased to 2 mol
%. The cyclopentyl derivate 9j was readily transferred to alkene
10j (97% yield, Z/E ratio 99/1) in 4 h at 1 mol % catalyst
loading. Arylcyclohexyl alkynes also turned out to be good
substrates. Various substituents such as bromo, chloro,
methoxy, or ethoxycarbonyl (9k−n) are tolerated, and the
corresponding alkenes 10k−n were obtained in 2.5−9 h in
good yields (90−96%) and excellent Z/E ratios (98/2 or 99/
1). For these substrates, 2−6% of the overhydrogenated alkanes
11k−n were formed as byproducts. The 3-pyridylcyclohexyl
congener 9o was a very unreactive substrate and an 85%
conversion at 2% catalyst loading was noted after 24 h. The
alkene 10o was formed in 84% yield with excellent Z/E
selectivity (98/2). Under the same conditions, diphenylacety-
lene (9p) was reduced to stilbene 10p (99%, Z/E ratio = 97/3).
a
Reaction performed with 2 mol % of the catalyst.
Arylalkyl alkynes bearing a methylester or nitrile functionality at
the alkyl terminus were successfully semihydrogenated to give
10q and 10r in good yields (96%, 97%) and excellent Z/E
ratios (99/1). We studied the semihydrogenation of bisalkyl
alkyne 9s and obtained alkene 10s within 2.5 h in a quantitative
yield (99%) and high selectivity (Z/E = 96/4). Finally, we
performed the semihydrogenation of alkyne 9a at 1 mmol scale
and the desired Z-alkene (Z/E = 99/1) was obtained after 1 h
in 90% isolated yield.
In summary, a new method for the photochemical
preparation of Pd nanoparticles using cheap and commercially
available photoinitiators 1 and 2 was developed. Simple mixing
of the photoinitiator and Pd(OAc)2 in DMF and subsequent
irradiation with UV-light leads to the formation of PdNPs that
are stabilized by unreacted photoinitiator and amide byproducts
6 and 8 derived therefrom. PdNP-hybrids obtained are
C
DOI: 10.1021/acs.orglett.7b00999
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Catal. 2010, 272, 246−252. (i) Mitsudome, T.; Takahashi, Y.;
Ichikawa, S.; Mizugaki, T.; Jitsukawa, K.; Kaneda, K. Angew. Chem.,
Int. Ed. 2013, 52, 1481−1485. (j) Slack, E. D.; Gabriel, C. M.;
Lipshutz, B. H. Angew. Chem., Int. Ed. 2014, 53, 14051−14054.
(k) Verho, O.; Zheng, H.; Gustafson, K. P. J.; Nagendiran, A.; Zou, X.;
characterized by TEM, XRD, and mass spectrometry.
Importantly, PdNP preparation is very easy to conduct and
reliably provides small PdNPs within a few minutes. Whereas
PdNPs derived from 2 turned out to be not active in the
reduction of alkynes, Pd@1* is a highly efficient catalyst for the
highly stereoselective semihydrogenation of various internal
alkynes to provide Z-alkenes in excellent yields (up to 99%)
and Z/E ratios (up to 99/1). The semihydrogenation is very
robust, scalable and exhibits a broad substrate scope.
Backvall, J.-E. ChemCatChem 2016, 8, 773−778. (l) Masing, F.; Wang,
̈
̈
X.; Nusse, H.; Klingauf, J.; Studer, A. Chem. - Eur. J. 2017, 23, 6014−
̈
6018.
(5) (a) Lindlar, H. Helv. Chim. Acta 1952, 35, 446−450. (b) Oger,
C.; Balas, L.; Durand, T.; Galano, J.-M. Chem. Rev. 2013, 113, 1313−
́
́ ́ ́
1350. (c) Molnar, A.; Sarkany, A.; Varga, M. J. Mol. Catal. A: Chem.
ASSOCIATED CONTENT
* Supporting Information
2001, 173, 185−221. (d) Hauwert, P.; Maestri, G.; Sprengers, J. W.;
Catellani, M.; Elsevier, C. J. Angew. Chem., Int. Ed. 2008, 47, 3223−
3226.
■
S
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.or-
glett.7b00999.
(6) Witte, P. T.; Boland, S.; Kirby, F.; van Maanen, R.; Bleeker, B. F.;
de Winter, D. A. M.; Post, J. A.; Geus, J. W.; Berben, P. H.
ChemCatChem 2013, 5, 582−587.
(7) Masing, F.; Mardyukov, A.; Doerenkamp, C.; Eckert, H.; Malkus,
̈
Detailed experimental procedures, spectroscopic data for
all products (PDF)
U.; Nusse, H.; Klingauf, J.; Studer, A. Angew. Chem., Int. Ed. 2015, 54,
̈
12612−12617.
(8) (a) Luo, C.; Zhang, Y.; Wang, Y. J. Mol. Catal. A: Chem. 2005,
229, 7−12. (b) Lee, D.-W.; Jin, M.-H.; Lee, Y.-J.; Park, J.-H.; Lee, C.-
B.; Park, J.-S. Sci. Rep. 2016, 6, 26474.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: studer@uni-muenster.de.
ORCID
(9) Jockusch, S.; Koptyug, I. V.; McGarry, P. F.; Sluggett, G. W.;
Turro, N. J.; Watkins, D. M. J. Am. Chem. Soc. 1997, 119, 11495−
11501.
(10) Kawamoto, T.; Sato, A.; Ryu, I. Chem. - Eur. J. 2015, 21, 14764−
Armido Studer: 0000-0002-1706-513X
Notes
14767.
(11) (a) Storr, T. E.; Baumann, C. G.; Thatcher, R. J.; De Ornellas,
S.; Whitwood, A. C.; Fairlamb, I. J. S. J. Org. Chem. 2009, 74, 5810−
5821. (b) Reay, A. J.; Fairlamb, I. J. S. Chem. Commun. 2015, 51,
16289−16307.
(12) (a) Wilson, O. M.; Knecht, M. R.; Garcia-Martinez, J. C.;
Crooks, R. M. J. Am. Chem. Soc. 2006, 128, 4510−4511. (b) Chen, J.;
Zhang, Q.; Wang, Y.; Wan, H. Adv. Synth. Catal. 2008, 350, 453−464.
(13) Henglein, A. J. Phys. Chem. B 2000, 104, 6683−6685.
(14) Mori, A.; Miyakawa, Y.; Ohashi, E.; Haga, T.; Maegawa, T.;
Sajiki, H. Org. Lett. 2006, 8, 3279−3281.
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
Dr. Xi Wang (Institute of Organic Chemistry, University of
Munster) is acknowledged for preparing some of the alkynes
used. Dr. Uta Rodehorst (Munster Electrochemical Energy
■
̈
̈
Technology Institute, University of Munster) is acknowledged
̈
for performing the XRD measurements. We thank the
Deutsche Forschungsgemeinschaft (SFB 858) for funding this
work.
REFERENCES
■
(1) (a) Muller, G.; Zalibera, M.; Gescheidt, G.; Rosenthal, A.;
̈
Santiso-Quinones, G.; Dietliker, K.; Grutzmacher, H. Macromol. Rapid
̈
Commun. 2015, 36, 553−557. (b) Chibac, A. L.; Melinte, V.; Buruiana,
T.; Buruiana, E. C. J. Polym. Sci., Part A: Polym. Chem. 2014, 52, 728−
738. (c) Dietliker, K.; Braig, A.; Ricci, A. Photochemistry 2010, 38,
344−368.
(2) (a) Cook, W. D.; Nghiem, Q. D.; Chen, Q.; Chen, F.;
Sangermano, M. Macromolecules 2011, 44, 4065−4071. (b) Balan, L.;
Melinte, V.; Buruiana, T.; Schneider, R.; Vidal, L. Nanotechnology
2012, 23, 415705. (c) Buruiana, E. C.; Chibac, A. L.; Buruiana, T.;
Melinte, V.; Balan, L. J. Nanopart. Res. 2013, 15, 1335.
(3) (a) Yagci, Y.; Tasdelen, M. A.; Jockusch, S. Polymer 2014, 55,
3468−3474. (b) Jockusch, S.; Turro, N. J. J. Am. Chem. Soc. 1998, 120,
11773−11777. (c) El-Zaatari, B. M.; Shete, A. U.; Adzima, B. J.;
Kloxin, C. J. Phys. Chem. Chem. Phys. 2016, 18, 25504−25511.
(4) For selected publications, see: (a) Zhao, M. Chem. - Asian J. 2016,
11, 461−464. (b) Gieshoff, T. N.; Welther, A.; Kessler, M. T.; Prechtl,
M. H. G.; von Wangelin, A. J. Chem. Commun. 2014, 50, 2261−2264.
(c) Mitsudome, T.; Yamamoto, M.; Maeno, Z.; Mizugaki, T.;
Jitsukawa, K.; Kaneda, K. J. Am. Chem. Soc. 2015, 137, 13452−
13455. (d) Alonso, F.; Osante, I.; Yus, M. Adv. Synth. Catal. 2006, 348,
305−308. (e) Liang, S.; Hammond, G. B.; Xu, B. Chem. Commun.
2016, 52, 6013−6016. (f) Vasilikogiannaki, E.; Titilas, I.;
Vassilikogiannakis, G.; Stratakis, M. Chem. Commun. 2015, 51,
2384−2387. (g) Yang, S.; Cao, C.; Peng, L.; Zhang, J.; Han, B.;
Song, W. Chem. Commun. 2016, 52, 3627−3630. (h) Evangelisti, C.;
Panziera, N.; D’Alessio, A.; Bertinetti, L.; Botavina, M.; Vitulli, G. J.
D
DOI: 10.1021/acs.orglett.7b00999
Org. Lett. XXXX, XXX, XXX−XXX