Highly Active, Selective, and Recyclable Water-Soluble Glutathione-Stabilized Pd and Pd-Alloy Nanoparticle Catalysts in Biphasic Solvent

Article abstract of DOI:10.1002/cctc.201901968

Glutathione-protected Pd, PdAu and PdPt nanoparticles (NPs) show greatly increased reactivity and stability towards the hydrogenation/isomerization of allyl alcohol in a biphasic organic/aqueous solvent mixture compared to single-phase alkanethiol-protected Pd-based NPs. NPs synthesized under aerobic conditions have higher activity compared to those synthesized under nitrogen, with the highest TOF of 676 mole product/mole metal/hour for 75 : 25 PdAu NPs at 20 mL/min H₂ flow. 75 : 25 PdAu NPs prepared under nitrogen have a maximum TOF of 507 at 20 mL/min H₂ flow, but can be recycled 4 times (9 times at 8 mL/min H₂ flow). Ethyl acetate is a better organic phase solvent compared to dichloromethane and chlorobenzene in terms of TOF and recyclability. All of the NPs studied show >80 % selectivity for the isomer product (propanal). The branched, open structure of the glutathione and use of the biphasic solvent provides high catalytic activity, high selectivity, and easy removal of the products in the organic phase for good recyclability of the water-soluble catalysts.

Full text of DOI:10.1002/cctc.201901968

DOI: 10.1002/cctc.201901968
Full Papers
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
Highly Active, Selective, and Recyclable Water-Soluble
Glutathione-Stabilized Pd and Pd-Alloy Nanoparticle
Catalysts in Biphasic Solvent
[a]
[a]
[b]
[a]
Shekhar Bhama, Tirtha R. Sibakoti, Jacek B. Jasinski, and Francis P. Zamborini*
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
Glutathione-protected Pd, PdAu and PdPt nanoparticles (NPs)
show greatly increased reactivity and stability towards the
hydrogenation/isomerization of allyl alcohol in a biphasic
organic/aqueous solvent mixture compared to single-phase
alkanethiol-protected Pd-based NPs. NPs synthesized under
aerobic conditions have higher activity compared to those
synthesized under nitrogen, with the highest TOF of 676 mole
product/mole metal/hour for 75:25 PdAu NPs at 20 mL/min H2
flow. 75:25 PdAu NPs prepared under nitrogen have a
maximum TOF of 507 at 20 mL/min H flow, but can be recycled
2
4 times (9 times at 8 mL/min H flow). Ethyl acetate is a better
2
organic phase solvent compared to dichloromethane and
chlorobenzene in terms of TOF and recyclability. All of the NPs
studied show >80% selectivity for the isomer product
(propanal). The branched, open structure of the glutathione
and use of the biphasic solvent provides high catalytic activity,
high selectivity, and easy removal of the products in the organic
phase for good recyclability of the water-soluble catalysts.
Introduction
from the reaction products, heterogeneous systems often suffer
from low turnover frequency (TOF), limited lifetime and poor
selectivity, primarily due to the principle of diffusion and the
Recent developments in nanotechnology have led to the
synthesis of advanced materials with desirable reactivity as well
as optical and electronic properties that have use in many
[
11]
involvement of two different phases in the catalytic system.
Also, their complex method of preparation and poor particle
[1,12]
applications, such as H storage, drug delivery, sensing, and
morphological control
has redirected studies on homoge-
2
[1–3]
catalysis.
The unique electronic structure and properties of
neous systems. Homogeneous systems have recently drawn
great interest due to their potential for higher activity and
nanoparticles (NPs) are largely due to the high surface area-to-
volume ratio along with low coordination number of the
surface atoms of NPs. This leads to size-dependent properties
very different from the bulk materials. These properties brand
NPs as a suitable candidate for the catalysis of a variety of
organic reactions. The key issue is to obtain maximum catalytic
efficiency by controlling the NP size, shape, composition and
functionality, which can be accomplished by using various
[13,14]
selectivity.
Some of the current problems include leaching
of delicate core metals and imperfect recycling and recovery.
Homogeneous NPs with weak stabilizers often have high
activity but become unstable in solution, leading to changes in
size and morphology, dissolution, aggregation, or precipitation.
These stability issues can be addressed by synthesizing
[
15,16]
metal NPs with dendrimers, polymers or surfactants.
[4]
[5]
[6]
organic stabilizers, such as polymers, dendrimers, peptides,
Another option is the use of organic ligand stabilizers, such as
[7]
[7]
[17]
[18]
[19]
and ligands. The big challenge is to find a protecting stabilizer
that prevents irreversible aggregation under the reaction
conditions, but also allows high catalytic activity and controlled
selectivity.
thiols, phosphines,
amines
or ammonium salts.
The
ligands usually form a monolayer coating on the NP surface and
[
20]
provide control over the core size of the NPs
and the
[21]
solubility and catalytic activity. The NPs are usually highly
stable and can be stored as solid powders, but often suffer
significantly from lower catalytic activity due to the blocking of
Transition metal NPs in solution have been frequently used
over a solid support for catalyzing a variety of organic and
inorganic reactions.
[8–10]
[22]
While it is easy to separate the catalyst
reactive sites. Crooks and co-workers reported the synthesis
and substrate size-selective catalytic activity of various Pd NPs
encapsulated within poly(amidoamine) dendrimers of different
generations and different end groups. Steric crowding on the
dendrimer periphery led to lower TOF for larger substrates.
Also, PdPt and PdAu dendrimer-encapsulated alloy NPs had
[a] Dr. S. Bhama, T. R. Sibakoti, Prof. F. P. Zamborini
Department of Chemistry
[23]
University of Louisville
Louisville, KY-40292 (USA)
E-mail: f.zamborini@louisville.edu
b] Dr. J. B. Jasinski
higher catalytic activity (TOF=200–1500 mole H /mole metal/
2
[
Conn Center for Renewable Energy Research
University of Louisville
Louisville, KY-40292 (USA)
hour) than dendrimer-encapsulated Pd, Au, or Pt single metal
NPs (TOF=0–800 mole H /mole metal/hour) for the hydro-
2
genation of allyl alcohol. The work of Sadeghmoghaddam et al.
represents an example of thiol ligand-stabilized Pd NPs for
catalysis, where they focused on understanding the mechanism
and regioselectivity of hexanethiolate- and dodecanethiolate-
Supporting information for this article is available on the WWW under
https://doi.org/10.1002/cctc.201901968
This manuscript is part of the Special Issue on the Sustainable and Afford-
able Chemistry to Meet Future Challenges in the Pharmaceutical Industry.
ChemCatChem 2020, 12, 1–10
1
© 2020 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
��
These are not the final page numbers!

Full Papers
capped Pd NPs in different environments. They found that both
hydrogen and the type of solvent are essential for the catalytic
workers used Glu-stabilized PdAu NPs supported on carbon
sheets as an oxygen reduction reaction (ORR) electrocatalyst
after calcination while Niu et al. combined Glu Pd NPs with
carbon microspheres for methanol oxidation electrocatalysis.
Guan et al. combined Glu Pd NPs with hydrogels to improve the
activity of a heterogeneous porous catalyst. In our work, we use
a single-phase synthesis of Glu Pd, PdAu, and PdPt NPs under
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
[24]
isomerization reaction.
Shon and co-workers reported the
synthesis of dodecanethiolate-capped Pd NPs, by employing a
[20]
Bunte Salt precursor, for the hydrogenation of allyl alcohol.
They showed that alkanethiolate-coated Pd and PdAu NPs
selectively catalyze the isomerization of allyl alcohol over
hydrogenation, where the reaction was 80% complete after
[31]
both inert and oxygen conditions.
Control was further
[25,26]
one hour.
Our group showed the same result of high
explored by varying the composition of different metals to
synthesize bimetallic alloy NPs. Furthermore, the catalytic
properties of Pd and Pd-alloy NPs were studied in different
biphasic solvent systems to obtain further insights into the role
of solvent and ligand structure on the stability, activity and
selectivity of these NPs. We also studied the recyclability of Glu-
coated NPs synthesized under inert and oxygen conditions. This
research provides insight into factors that control the optimum
catalytic performance, in terms of activity, selectivity, and
recyclability, of Pd NPs stabilized with open, branched thiols
operating under biphasic solvent conditions.
isomerization selectivity but low reactivity for hexanethiolate-
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
[18]
coated Pd NPs.
In comparison, long chain alkylamine
stabilizers were much more active, but less stable and more
[18]
difficult to separate from the products. More recently, Shon
and co-workers investigated the water-soluble ω-carboxyl-1-
hexanethiolate-capped (PdNP-1) and ω-carboxyl-1-octanethio-
late-capped (PdNP-2) Pd NPs as micelle biphasic catalysts for
hydrogenation and isomerization of 1-octen-3-ol, which showed
À 1 À 1
[27]
a TOF of 1.43 site
h
based on the active site of PdNP-1.
Recyclability studies for PdNP-1 showed similar activity and 63–
3% hydrogenation selectivity for up to 3 cycles. These
examples show that strong stabilizing ligands on the Pd NPs
7
surface leads to fairly stable NPs, but often give modest TOFs Results and Discussion
due to their metal-ligand interactions.
Here we describe the synthesis of Pd and Pd-alloy (PdAu
and PdPt) NPs coated with glutathione (Glu) ligands for use in
catalyzing the hydrogenation/isomerization of allyl alcohol as a
model system. We hypothesized that the more open and
branched Glu structure would improve the accessibility to the
Pd surface, improving the catalytic activity, while the strong Pd-
thiol interaction would provide stability and isomerization
selectivity as observed for other non-polar alkanethiol-coated
Pd NPs. While the Glu ligand has not been as widely studied as
Pd Nanoparticle Synthesis and Characterization. To design a
catalyst with high stability, efficiency and recyclability, we
synthesized Pd and Pd alloy NPs by co-reduction of Pd with Au
or Pt salts in different metal compositions [Pd:(Au or Pt) in
90:10, 75:25 and 50:50 ratios] in the presence of glutathione
(Glu) under both inert and oxygen atmosphere as shown in
[31]
Scheme 1. Sharma et al.
showed that Pd NPs synthesized
under oxygen causes expansion in their lattice parameter from
3.88 Å to 4.12 Å due to the incorporation of oxygen. We
hypothesized that the presence of oxygen in the lattice might
result in different catalytic behavior due to different substrate
or hydrogen binding properties caused by different metal-metal
distance and potential electronic effects. All NPs were charac-
[28]
[7]
other thiols, there are examples of Glu-stabilized Au and Pt
[7]
[28]
NPs for homogeneous and heterogeneous catalysis. In the
case of Pt, use of the Glu stabilizer did not lead to a higher TOF
[7]
compared to alkanethiols. Glu-stabilized Pd NPs have been
specifically studied previously for use as a homogeneous
catalyst for the hydrogenation of nitroarenes in the aqueous
1
terized using H NMR and FTIR spectroscopy to show the
attachment of ligands to the metal core of the NPs. TGA
analysis of Pd-Glu (Inert) NPs, for example, gave a 25.6%
organic composition and 74.4% Pd metal. The diameter of Pd-
[29]
phase and for the synthesis of heterogeneous NPs anchored
[30]
onto metal oxides, such as ZnO and TiO2.
Chen and co-
Scheme 1. Single-phase synthesis of water-soluble glutathione-coated Pd and Pd-alloy NPs.
ChemCatChem 2020, 12, 1–10
www.chemcatchem.org
2
© 2020 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
��
These are not the final page numbers!

Full Papers
Glu NPs was 1.8�0.3 nm and 2.6�0.5 nm for inert and oxy
NPs, respectively, based on TEM analysis while the (50:50)
Pd:Au-Glu NPs were 2.3�0.5 nm and 2.1�0.6 nm for inert and
oxy NPs, respectively (Figure S1 and S2, supporting informa-
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
[31]
tion), consistent with the literature. The exact composition of
the alloy NPs was not determined. We assumed that bimetallic
alloy NPs formed with a composition similar to the synthesis
ratio based on the different catalytic activity for the pure Pd
and Pd alloy NPs and based on previous results of Pd alloy NPs
[18,32–34]
synthesized by co-reduction.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
Hydrogen stability of glutathione-coated Pd NPs. Since
hydrogen is one of the reactants in the hydrogenation/isomer-
ization reaction of allyl alcohol, it was important for the NPs to
be stable under a hydrogen atmosphere. Molecular hydrogen
gets reversibly adsorbed on the surface of Pd where it rapidly
dissociates into atomic hydrogen and then diffuses into the
[35]
metal lattice and occupies octahedral interstitial sites. This
increases the lattice constant. At low hydrogen partial pressure,
α-phase PdH forms, which undergoes a phase transition into β-
x
[36,37]
phase PdH at higher hydrogen partial pressure.
x
We determined the stability of the aqueous solutions of Pd-
Glu (Inert) and Pd-Glu (Oxy) NPs by monitoring the UV-Vis
absorbance spectra of their solutions from 200–900 nm as a
function of exposure time to pure hydrogen bubbling through
the solution at a flow rate of 19.9�0.7 mL/min as shown in
Figure 1. Due to the strong binding of Glu ligands with Pd NPs,
the NPs showed a very similar spectrum for at least 2 hours,
confirming that they were stable against hydrogen-induced
aggregation. Interestingly, the same trend was observed for Pd-
alloy NPs with Au and Pt with different ratios confirming that
Glu ligands provide stability against hydrogen-induced aggre-
gation (data not shown). The UV-Vis spectra of these small Glu-
coated Pd NPs shows an exponential decay in absorbance with
an increase in wavelength and lack any surface plasmon
resonance bands.
Figure 1. UV-Vis plots of absorbance vs. wavelength for exposure time 0–
1
20 min to 100% hydrogen bubbling through solutions of Pd-Glu (Inert) (A)
and Pd-Glu (Oxy) NPs (B). Insets shows pictures of solutions without
hydrogen at 0 minute and with hydrogen after 120 minutes.
We also studied the stability of Pd-Glu (Inert) NPs and Pd-
Glu (Oxy) NPs against hydrogen-induced aggregation under the
biphasic condition of chlorobenzene and nanopure water in the
(Scheme 2) with various Glu-coated Pd- and Pd-alloy NPs and
compared them to previous work by others and our
[18,20,25–27,32]
presence of allyl alcohol at 19.9�0.7 mL/min H flow rate for 0,
lab.
The set-up is shown in Figure S5 of supporting
2
5
, 15, 30, 60, 120, 240 minutes using UV-Vis spectroscopy as
information. Figure 2 shows several GCs of the aqueous and
organic phase at different times up to 25 min during the
shown in Figure S3 in the Supporting Information. UV-Vis
spectra show Pd-Glu (Inert) NPs to be comparatively more
stable than Pd-Glu (Oxy) NPs based on the lower overall
absorbance at 250 nm and appearance of a fairly well-defined
peak in this region for the latter, which is due to chlorobenzene,
propanal, Pd(II)-thiolate species, or some combination (Fig-
ure S4 of supporting information shows their UV-vis spectra).
The peak is better defined in this area due to greater instability
of the Pd-Glu (Oxy) NPs by aggregation, precipitation, or Pd(II)
formation, all of which would decrease the absorbance from
the original NPs in the region. In both Pd-Glu NP samples, the
absorbance at high wavelengths drops to some degree, but the
overall stability is fairly good, although better for Pd-Glu (inert)
NPs.
reaction of allyl alcohol with H (19.9�0.7 mL/min) catalyzed by
2
Pd-Glu (Oxy) NPs. In the aqueous phase, the peak correspond-
ing to allyl alcohol with a retention time (t ) of 4.4 minutes
r
decreased with time, eventually disappearing by 25 minutes,
whereas the peaks corresponding to propanal and 1-propanol
at t of 1.5 and 3.6 minutes, respectively, increased with time. In
r
the organic phase, the peak heights for both products also
increased with time as the allyl alcohol peak height decreased
and eventually vanished completely by 25 minutes.
Qualitatively, the data clearly show that the catalyst is highly
selective to the isomerization product, propanal, based on the
much higher intensity for that peak in the GC compared to the
peak for 1-propanol. A comparison of the peak intensities of the
aqueous versus the organic phases shows that the products
favor the organic phase more, which makes it easy to separate
the products from the aqueous-soluble catalyst. This allows
Biphasic catalysis with glutathione-coated NPs. We moni-
tored the homogeneous biphasic liquid-liquid catalysis and
selectivity for the hydrogenation/isomerization of allyl alcohol
ChemCatChem 2020, 12, 1–10
www.chemcatchem.org
3
© 2020 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
��
These are not the final page numbers!

Full Papers
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
Scheme 2. Illustration shows the biphasic catalytic reaction of allyl alcohol with hydrogen using various glutathione-coated nanoparticles.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
than Pd NPs at all compositions, generally increasing TOF with
an increasing amount of Au in the NP. For Pd:Pt NPs, the
activity is smaller than Pd for 10% Pt, about the same for 25%
Pt, and higher for 50% Pt. Also, the alloy Pd:Au and Pd:Pt NPs
synthesized under oxygen atmosphere showed higher catalytic
activity than the same NPs synthesized under inert conditions.
In the case of pure Pd NPs, the activity was about the same for
those synthesized under oxygen or inert conditions. The
presence of other metal helps to increase the rate of the
reaction, especially for oxygen synthesized Pd-alloy NPs. Pd:Au-
Glu (Inert/Oxygen) NPs in all ratios showed higher TOFs than
Pd:Pt-Glu (Inert/Oxygen) NPs with the same ratio. We believe
the higher electronegativity of Au (2.54) compared to Pt (2.28)
is the main reason for the bigger improvement in the overall
catalytic activity compared to Pd only (2.20). Increased electro-
negativity of the second metal makes the Pd more electro-
positive and can improve substrate binding through the olefin
[34]
group. Also, the addition of the second metal may alter the
geometry of the crystal lattice, allowing for more active sites to
be available for catalysis that provides multiple binding/
interacting points with the substrate and/or hydrogen, increas-
[33,34]
ing the rate of product formation.
All NPs showed a high
selectivity of >80% for the isomerized product. (75:25) Pd:Au-
Glu (Oxy) NPs showed the highest TOF of 590 moles of
product/mole metal/hour. It is interesting that the water-soluble
Glu ligand leads to high selectivity for the isomerization product
as previously shown for organic-soluble alkanethiols but the
TOF is 2–6 times higher, likely due to the open structure of the
Glu ligands compared to alkanethiols.
[18]
Role of different hydrogen flow rates on TOF and
selectivity of the catalytic reaction. It is important to study the
effect of flow rates of hydrogen on the TOF and the selectivity
of the overall reaction in order to better understand the
reaction and optimize the turnover. We used 3 different flow
rates of 100% hydrogen (8.0�0.2, 19.9�0.7 and 39.9�0.3 mL/
min) and chlorobenzene/aqueous biphasic solvent system for
Pd-Glu (Inert/Oxygen) NPs. It was observed that the TOF
increased from about 200 to 300 as the hydrogen flow rate
increased from 8.0�0.2 to 39.9�0.3 mL/min as shown in
Table 2, whereas the selectivity didn’t change.
Figure 2. Example of biphasic catalysis using Pd-Glu (Oxy) NPs with 2 mL
flow
rate for 25 minutes. GC-FID chromatograms of the aqueous phase (A) and
nanopure water and 2 mL chlorobenzene using 19.9�0.7 mL/min H
2
organic phase (B) after 0, 5, 10, 15, 20 and 25 minutes as indicated.
easy recovery and reuse of the catalyst for several cycles.
Control experiments using UV-vis and GC show that we can
extract all 1-propanol and propanal from the aqueous phase
into the organic phase for easy product separation (Figure S5
and S6).
Table 1 shows the turnover frequencies (TOF) for all
We also studied the kinetics of the catalytic reaction for Pd-
Glu (Inert) NPs by observing the total percentage completion of
the reaction with time (minutes) at different hydrogen flow
rates (8.0�0.2, 19.9�0.7 and 39.9�0.3 mL/min) using chlor-
synthesized Pd and Pd-alloy NPs at 19.9�0.7 mL/min H flow
2
rate using chlorobenzene as the organic solvent along with the
aqueous phase. The catalytic activities of Pd:Au NPs are higher
ChemCatChem 2020, 12, 1–10
www.chemcatchem.org
4
© 2020 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
��
These are not the final page numbers!

Full Papers
Table 1. Table of turnover frequencies (TOF) in mole product/mole total metal/hour and selectivity towards hydrogenated and isomerized products for all
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
synthesized Pd and Pd-alloy NPs under inert/oxygen conditions at 19.9�0.7 mL/min H
2
flow rate using chlorobenzene organic solvent with aqueous phase.
5
0:50
50:50
Pd:Pt-
Glu
75:25
Pd:Pt-
Glu
75:25
Pd:Pt-
Glu
90:10
Pd:Pt-
Glu
90:10
Pd:Pt-
Glu
Pd-
Glu
Pd-
Glu
90:10
Pd:Au-
90:10
Pd:Au-
Glu
75:25
Pd:Au-
Glu
75:25
Pd:Au-
Glu
50:50
Pd:Au-
Glu
50:50
Pd:Au-
Glu
Pd:Pt-
Glu
(Inert) (Oxy) Glu
(Oxy)
(Inert)
(Oxy)
(Inert)
(Oxy)
(Inert)
(Inert)
(Oxy)
(Inert)
(Oxy)
(Inert)
(Oxy)
Hyd
TOF
Iso
TOF
Overall 252
36
20
26
31
30
7
14
11
20
56
22
66
29
39
216
215
235
186
212
150
181
142
172
95
102
191
205
189
200
398
418
435
491
305
327
524
590
380
409
510
549
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
TOF
Hyd %
Iso %
17
83
10
90
13
87
22
78
16
84
11
89
10
90
7
93
9
91
15
85
8
92
16
84
9
91
10
90
Table 2. Effect of flow rates of 100% hydrogen (8.0�0.2, 19.9�0.7 and 39.9�0.3 mL/min) on turnover frequencies (TOF) in mole product/mole Pd/hour
and selectivity of the overall catalytic reaction using Pd-Glu (Inert/Oxygen) NPs.
Hydrogen flow-rates [mL/min]
8.0�0.2
19.9�0.7
39.9�0.3
Nanoparticles
Pd-Glu (Inert)
5
152
157
5
Pd-Glu (Oxy)
9
183
192
8
Pd-Glu (Inert)
14
192
206
10
Pd-Glu (Oxy)
11
120
201
7
Pd-Glu (Inert)
13
291
304
6
Pd-Glu (Oxy)
20
293
313
9
Hydrogenation TOF
Isomerization TOF
Overall conversion TOF
Hydrogenation [%]
Isomerization [%]
95
92
90
93
94
91
obenzene/aqueous biphasic solvent system. It clearly showed
that at 8.0�0.2 mL/min flow rate, it took 5 minutes for the NPs
to become catalytically active as compare to higher flow rates
which showed activity from the beginning of the reaction as
shown in Figure S7 in the Supporting Information. While
ethyl acetate solvent [TOF=507/676 for (75:25) Pd:Au-Glu
(Inert/Oxygen) NPs] is the best organic solvent in terms of
activity compared to dichloromethane (TOF=388/451) and
chlorobenzene (TOF=327/590). All solvents favored >80%
isomerized product (propanal). The TOF values were higher
with Pd-Glu (Oxy) NPs compared to Pd-Glu (Inert) NPs for all
three organic solvents in general, but the difference was much
less for dichloromethane. We believe that the activity could
depend on the aromaticity and polarity index of the organic
solvent. As the solvent polarity decreases, the stability of the
increased H flow rate increased the TOF for the pure Pd NPs, it
2
became clear that the stability is compromised for the higher
flow rates, reducing the potential for recyclability. For this
reason, we did not test all of the alloy NPs at the higher flow
rates.
Solvent effects in the biphasic mixture. We studied
dichloromethane (DCM), chlorobenzene (Clben) and ethyl
acetate (EtOAc) as organic solvents for the biphasic solvent
mixture to optimize the reaction using the (75:25) Pd:Au-Glu
NPs increases likely due to greater H solubility in the organic
2
phase, which prevents NP instability caused by too much H2
adsorption on the Pd NPs. The non-aromatic solvents may be
better as they do not directly interact with the Pd surface as
well as aromatics. These trends are pure speculation, however,
as we didn’t explore it further. More work will be needed to
better understand these trends.
Recyclability of glutathione-coated Pd and Pd-alloy NPs.
Due to their good stability, we compared the catalytic
recyclability of (75:25) Pd:Au-Glu (Inert) NPs for hydrogena-
(
0
Inert/Oxy) NPs as our model catalysts at a H flow rate of 19.9�
2
.7 mL/min as shown in Table 3. Preliminary experiments
indicated that the (75:25) Pd:Au-Glu NPs were highly active
and among the most stable and most recyclable NPs at low to
medium flow rates, which is the rationale for using these NPs
for optimization of the solvent. TOF values showed that the
Table 3. Effects of organic solvent (dichloromethane, chlorobenzene and ethyl acetate) on turnover frequencies (TOF) in mole product/mole total metal/
hour and selectivity of the overall catalytic reaction at 19.9�0.7 H flow rate using (75:25) Pd:Au-Glu (Inert/Oxygen) NPs.
2
Organic Solvent
used
Dichloromethane
Chlorobenzene
Ethyl acetate
Nanoparticles
Inert)
Hydrogenation TOF 47
(75:25) Pd:Au-Glu (75:25) Pd:Au-Glu
(75:25) Pd:Au-Glu (75:25) Pd:Au-Glu
(75:25) Pd:Au-Glu (75:25) Pd:Au-Glu
(
(Oxygen)
66
(Inert)
22
(Oxygen)
66
(Inert)
73
(Oxygen)
102
Isomerization TOF
Overall conversion
TOF
Hydrogenation [%]
Isomerization [%]
341
388
385
451
305
327
524
590
434
507
574
676
12
88
16
84
8
92
16
84
16
84
15
85
ChemCatChem 2020, 12, 1–10
www.chemcatchem.org
5
© 2020 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
��
These are not the final page numbers!

Full Papers
tion/isomerization of allyl alcohol in various organic (chloroben-
zene, ethyl acetate and dichloromethane)/aqueous biphasic
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
solvent systems for 7–9 cycles at 19.9�0.7 mL/min H flow rate
2
using UV-Vis spectroscopy as shown in Figure S8 in the
Supporting Information. UV-Vis spectra for NPs in the presence
of allyl alcohol and H showed stability for 7 cycles with ethyl
2
acetate, 5 cycles with dichloromethane, and 3 cycles with
chlorobenzene, where each cycle reaction went until comple-
tion. We only counted cycles where full conversion of the
reactants occurred. The Pd:Au-Glu NPs with low recyclability
showed low absorbance and a more defined peak near 250 nm
in the UV-vis spectrum. We also compared the recyclability of
single metal Pd-Glu (Inert) and Pd-Glu (Oxy) NPs for hydro-
genation/isomerization of allyl alcohol for at least 6 cycles in
chlorobenzene/aqueous biphasic solvent system by UV-Vis
spectrometry as shown in Figure S9 in the Supporting Informa-
tion. The data clearly show that Pd-Glu (Inert) NPs were stable
for more cycles than Pd-Glu (Oxy) NPs, again based on the
lower absorbance around 250 nm for the latter. Note the high
recyclability is not possible with the same clusters used in a
single phase aqueous reaction. As shown in Figure S10, the UV-
vis spectrum of Pd-Glu NPs in single phase versus biphasic
solvent on the second cycle shows a large degree of
aggregation on the second cycle for the former. The gas
chromatogram in Figure S10 shows mostly non-transformed
reactant (allyl alcohol) in the GC.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
Our data reveal that the ethyl acetate solvent leads to the
highest TOF and highest recyclability. In addition, NPs synthe-
sized under oxygen exhibit higher TOF but lower recyclability,
whereas inert synthesized NPs exhibit lower TOF but higher
recyclability. These trends are clear in the TOF and recyclability
data for (75:25) Pd:Au-Glu (Oxy) NPs compared to (75:25)
Pd:Au-Glu (Inert) NPs as shown in Table 4, all of which were
obtained at a H flow rate of 19.9 mL/min.
2
Maximizing the recycling capacity of the catalysts decreases
the overall production cost of the chemical products signifi-
cantly at the industrial scale. Using a biphasic solvent system
with nanopure water and an immiscible organic solvent makes
it very easy to extract the products into the organic phase and
separate them from the aqueous-phase catalyst, which can
then be reused in another reaction. Figure 3 shows GC-FID
results demonstrating the catalytic recyclability of (75:25)
Pd:Au-Glu (Inert) NPs using dichloromethane as the organic
Figure 3. Example of recyclability using (75:25) Pd:Au-Glu (Inert) NPs
showing 100% completion of the catalytic reaction for at least 9 times
(checked 11 times) with dichloromethane at 8.0�0.2 mL/min H flow rate.
2
GC-FID chromatograms of the organic phase (A) and aqueous phase (B) after
each cycle as indicated.
minor decrease in the TOF with each cycle. We focused on the
lowest flow rate for this study since that leads to the highest NP
stability.
Table S1 of supporting information shows the recyclability
of various other Pd and Pd-alloy NPs using ethyl acetate,
solvent and a H flow rate of 8.0�0.2 mL/min. The chromato-
2
grams of the organic and aqueous phases show only the
presence of the propanal and 1-propanol products for the first
9
cycles. Each cycle shows high selectivity for propanal with a
Table 4. Table showing turnover frequencies (TOF) in mole product/mole total metal/hour and recyclability correlation for (75:25) Pd:Au-Glu (Inert/Oxygen)
NPs in ethyl acetate, chlorobenzene and dichloromethane organic solvents.
Reaction Conditions
Nanoparticles
Turnover frequencies [TOF]
Recyclability
Hydrogen at 19.9�0.7 mL/min using organic solvent
Ethyl acetate
(75:25) Pd:Au-Glu (Oxy)
676
507
590
327
451
388
2
4
1
3
1
3
(
75:25) Pd:Au-Glu (Inert)
75:25) Pd:Au-Glu (Oxy)
75:25) Pd:Au-Glu (Inert)
75:25) Pd:Au-Glu (Oxy)
75:25) Pd:Au-Glu (Inert)
Chlorobenzene
Dichloromethane
(
(
ChemCatChem 2020, 12, 1–10
www.chemcatchem.org
6
© 2020 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
��
These are not the final page numbers!

Full Papers
chlorobenzene or dichloromethane organic solvents at different Conclusions
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
H2 flow rates. Ethyl acetate is the best solvent in terms of
activity and recyclability. High recyclability is also generally
associated with low flow rate and NPs synthesized under inert
conditions.
In summary, we have described the successful synthesis of
various water-soluble Pd and Pd-alloy NPs stabilized with Glu,
which shows highly active catalytic performance for selective
isomerization of allyl alcohol with good recyclability. We
observed that oxygen synthesized NPs are catalytically more
active but have lower recyclability than the inert ones. The
catalytic activity increased in the order of PdAu>PdPt>Pd,
with increasing activity as the amount of Au or Pt increased,
except when a small amount of Pt was used. All Glu-coated Pd
and Pd-alloy NPs showed >80% selectivity towards the
Stability comparison of glutathione-coated Pd NPs with
previously reported alkanethiol- and alkylamine-coated Pd
NPs. To compare the stability of Pd-Glu (Inert) NPs in biphasic
solvent with C6SÀ Pd NPs synthesized previously in our lab and
[18]
used in a monophasic organic solvent (dichloromethane), we
studied the hydrogenation and isomerization turnover rates of
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
allyl alcohol using Pd-Glu (Inert) NPs at 3 different H flow rates
2
8
.0�0.2, 19.9�0.7 and 39.9�0.3 mL/min as shown in Fig-
ure S11 in the Supporting Information. Although both Pd-Glu
Inert) NPs and C6SÀ Pd NPs are stabilized by thiol-coated
isomerized product for H flow rates from 8.0 to 39.9�0.2 mL/
2
min and after multiple reaction cycles (at low flow rate). The
ethyl acetate solvent showed higher TOFs than chlorobenzene
and dichloromethane, but the reasons are not well understood.
The best recyclability performance was that of the (75:25)
Pd:Au-Glu (Inert) NPs, which showed 100% completion of the
catalytic reaction for at least 9 cycles using water/dichloro-
(
ligands, Pd coated with Glu in the biphasic solvent system
showed greater stability even at 39.9�0.3 mL/min flow rates.
The C6SÀ Pd NPs became unstable and precipitated. Also, Pd-
Glu NPs showed higher TOF of up to ~300 moles product/
moles Pd/hour as compared to a max TOF of 93 for C6SÀ Pd
NPs. The selectivity for both NPs was the same. This confirms
that Glu-coated Pd NPs in biphasic solvent are more stable and
suitable for some catalytic applications. We have also synthe-
sized Pd NPs stabilized with alkylamines of various chain
lengths (C8NH , C12NH and C16NH ) previously in our lab and
methane and a slow 8.0�0.2 mL/min H flow rate. These
2
studies help us to better understand the effect of monolayer
ligand chemistry, metal composition, and solvent on metal NP
catalysis in terms of activity, selectivity, and stability/recycla-
bility.
2
2
2
investigated the catalytic activity and selectivity for the hydro-
[18]
genation/isomerization reaction of allyl alcohol. These studies Experimental Section
showed a TOF of 120 mole product/mole Pd/hour for C8NH Pd
2
Chemicals. Dichloromethane (99.5%), ethanol (200 proof), meth-
NPs, 194 for C12NH Pd NPs and 719 for C16NH Pd NPs with
2
2
anol (99%), acetone (99%) and 2-propanol (99.9%) were purchased
from VWR Scientific Products and used as received. Sodium
borohydride (98%), potassium tetrachloroplatinate(II), reduced-L-
glutathione, ethyl acetate (99%), allyl alcohol (99%), propyl alcohol
the selectivity of only 1:1 or 3:2 ratios towards the hydro-
genation:isomerization products. The stability of the shorter
chain alkylamine-coated NPs was poor due to weak interaction
of the ligands with the Pd core and all alkylamines showed
modest recyclability due to difficulty in separation of products.
Clearly, with Glu as a stabilizing ligand and the biphasic solvent
system as the reaction mixture, we have successfully prepared a
catalytic system that is stable, highly active, and easily
separated from the products for high recyclability.
(
99.7%) and propionaldehyde (99%) were purchased from Aldrich
Chemical Co. and used as received. Potassium tetrachloropalladate
II) (99%) and chlorobenzene (98%) were purchased from Alfa
(
Aesar Co. and used as received. Hydrogen tetrachloroaurate(III) was
synthesized in our lab. Deuterium oxide and chloroform-d were
purchased from Cambridge Isotope Laboratories. Water was
purified using Barnstead nanopure water (18.3 MΩcm) and used
for all aqueous solutions.
Glu-coated Pd-containing NPs showed >80% selectivity
towards the isomerized product, which is similar to other thiol-
Synthesis of glutathione-coated Pd NPs under Inert condition.
We followed the synthesis reported by Sharma et al.
[31]
[18,25,26]
In the
modified Pd, and Pd-alloy NPs described in the literature.
synthesis of “Inert” glutathione-coated Pd NPs, 0.50 g (1.531 mmol)
of K PdCl was dissolved in 30 mL of nanopure water and nitrogen
The isomerization reaction is an important one-pot trans-
formation to carbonyl compounds that proceeds with high
efficiency, selectivity and atom economy while avoiding the
two-step sequential oxidation and reduction reactions. Selective
formation of the isomerized product is therefore a desirable
attribute. Sensitive substrates do not survive the conditions
involving the use of toxic and/or expensive oxidation and
reduction reagents, making catalytic redox isomerization under
mild reaction conditions especially useful as shown here. Also,
in terms of atom economy, no byproducts are formed. Only the
saturated aldehyde or ketone are formed, which are valuable
intermediates for various pharmaceuticals, agrochemicals and
2
4
gas was purged for at least 1 hour. Separately, 0.165 g (0.536 mmol)
of reduced-L-glutathione was dissolved in 5 mL of nanopure water
and purged in nitrogen for 1 hour. The two solutions were then
2À
combined and stirred under nitrogen atmosphere until PdCl
4
formed a complex with glutathione as indicated by the change in
color from yellow to wine red. The reaction mixture was further
stirred at 360 rpm for another 1 hour, which was then cooled down
using an ice-bath. Subsequently, a fresh aqueous solution of NaBH4
was prepared by dissolving 0.58 g (15.31 mmol) of NaBH in 5 mL of
4
nanopure water that was already purged in nitrogen for 1 hour.
This solution was immediately added to the reaction mixture under
stirring, giving an immediate change in color from wine red to
black, indicating the formation of Pd nanoparticles. The solution
was further stirred for 4 hours from which Pd NPs were separated
by adding an equal volume of methanol (40 mL), stirring for 15
minutes, and centrifuging at 4000 rpm for 10 minutes to obtain
precipitates of the Pd NPs. The mother liquor was removed and the
[38]
fine chemicals. For all of these syntheses, Pd NPs stabilized
with Glu ligands showed high catalytic activity, selectivity and
recyclability.
ChemCatChem 2020, 12, 1–10
www.chemcatchem.org
7
© 2020 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
��
These are not the final page numbers!

Full Papers
precipitated Pd NPs were dissolved in nanopure water and trans-
ferred into a round bottom flask to remove the water using a rotary
evaporator heated at 45 C. The black NPs were suspended in
100 mL of methanol and collected by vacuum filtration on a glass-
fritted Büchner funnel. The NPs were then washed thoroughly with
methanol, ethanol and acetone successively two times before
thoroughly drying and collecting. These NPs are referred to as Pd-
Glu (Inert) NPs.
TGA analysis was conducted using a 2950 TGA HR V5.4 A instru-
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
ment with a flow rate of 100 mL/min of N over a temperature
2
°
range of 25–800 C at a heating rate of 20 C/min using sample sizes
from 5.0–11.4 mg. TGA data were used to observe the change in
weight of NPs to calculate the organic percentage present on the
metal core.
°
°
Transmission electron microscopy (TEM) images were obtained
using a FEI Tecnai F20 operated in TEM mode with an acceleration
voltage of 200 kV. TEM analysis was carried out using a digital
micrograph software (version 3.11.0) and ImageJ software for
particles size analysis. TEM samples of water-soluble Pd-Glu and
(50:50) Pd:Au-Glu NPs were prepared by dissolving in water and
then drop casting the aqueous solution onto a 400 mesh Formvar/
carbon-coated copper grid, letting it air-dry for at least 4 hours.
Synthesis of glutathione-coated Pd NPs under oxygen condition.
Pd NPs coated with glutathione ligands under “Oxygen” conditions
were prepared similarly by performing the above experiment under
air atmosphere without purging nitrogen gas and keeping the
molar concentration of K PdCl , reduced-L-glutathione and NaBH
4
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
2
4
the same. The only visible difference was the formation of an
orange colored solution during the addition of glutathione solution
to the Pd salt solution.
Catalytic Reactions. The catalytic reaction of allyl alcohol to either
hydrogenated and/or isomerized product was performed by
dissolving 6.0 mg of Pd or Pd alloy NPs in 2 mL of nanopure water
in a 10 mL glass vial with an outlet for spent hydrogen gas passed
through a glass pipette. 2 mL of organic solvent [Chlorobenzene
(Clben), Dichloromethane (DCM) or Ethylacetate (EtOAc)] and
200 μL of allyl alcohol were added, forming a biphasic reaction
mixture as shown in Figure S12 in the Supporting Information. The
reaction mixture was then stirred at 500–600 rpm at room temper-
ature and atmospheric pressure. Hydrogen gas was purged at a
given flow rate (8.0�0.2, 19.9�0.7 or 39.9�0.3 mL/min) through a
glass pipette and 20 μL aliquots were removed from both aqueous
and organic phases at particular intervals of time. NPs in the
aqueous phase aliquot were precipitated by adding THF solvent in
2-fold excess (40 μL) and centrifuging at 4000 rpm for 15 min.
Synthesis of glutathione-coated Pd-alloy NPs under inert and
oxygen conditions. Pd-alloy NPs with different ratios of Pt or Au
were synthesized ranging from 90:10, 75:25 and 50:50 under both
inert and oxygen conditions by co-reducing them using NaBH4.
Here, both K PdCl and K PtCl or HAuCl .3H O metal salts in their
2
4
2
4
4
2
respective molar ratio were dissolved together in nanopure water
and used for further synthesis as described above. All of the
synthesized NPs were relatively stable as confirmed by comparing
the catalytic activity using 2 years old synthesized (75:25) Pd:Au-
Glu (Inert) NPs as catalyst for hydrogenation/isomerization of allyl
alcohol at 19.9�0.7 mL/min H2 flow rate. The reactant was
converted into products within 10 minutes as observed before. The
only difference observed was that % hydrogenation (1-propanol
formation) increased from 16% to 33% whereas % isomerization
(propanal formation) decreased from 84% to 67%. The reason for
this is unknown, but could be due to the loss of some ligands from
the Pd surface with time.
The progress of the catalysis reaction was followed by gas
chromatography (GC) of samples before exposure to hydrogen and
after addition of catalyst and exposure to hydrogen at different
intervals of time. The GC data was recorded on a Buck Scientific
1
th
Characterization. H NMR spectra of metal NPs were recorded on a
model 910 GC equipped with a 1/8 inch packed column (10%
400 MHz INOVA spectrometer with 64 scans from À 2 to 18 ppm at
Carbowax 20 M on silica 80/100 mesh, 6 foot) using a flame
ionization detector (FID) and helium as the carrier gas. The method
developed for GC includes injecting 1 μL of the reaction aliquot
and varying temperatures from (a) 80°C to 80°C, holding for 2 min
(b) 80°C to 135°C, ramp at 25°C/min (c) 135°C to 135°C, holding
for 1 min. The pressure varied from (a) 12 psi helium for 3 min (b)
14 psi helium for 2 min. The turnover frequency (TOF) of the
products were calculated using the slope of the linear portion of
>60% of the conversion of reactant to isomer and hydrogenated
product in the plot of % hydrogenation and % isomerization versus
1
room temperature and pressure. H NMR spectroscopy gives
information about the successful attachment of ligands onto the
metal core of the synthesized NPs and confirms the removal of
non-bound ligands. A residual solvent peak at δ 4.79 ppm of D O
2
1
was used as an internal reference. H NMR spectrum of pure
glutathione showed proton resonances at δ 2.21 (À CH
CHCOOH),
2
2
.58 (À CH CONH), 2.95 (À CH SH), 3.84 (À CHNH ), 4.00 (À CH COOH)
2
2
2
2
and 4.59 (À CHCH SH) ppm. Protons from carboxylic acid, amide and
2
amine were exchanged with D O and therefore, couldn’t be
2
1
[18]
observed in the spectrum. H NMR of Pd-Glutathione NPs differed
time as described previously.
from that of pure glutathione in that the peaks for the thiol-coated
Pd NPs were all broadened relative to those of pure glutathione.
The peak for the two protons α to the thiol group at 2.95 ppm
wasn’t observed in the glutathione-coated Pd NPs, due to the
presence of Pd core that creates large inhomogeneity in the
magnetic field around the local chemical environment. Also, a lack
of any sharp peaks in the spectra confirmed the absence of any
Catalytic Recyclability Studies. For recycling studies, the first cycle
was the same as described for the catalytic reaction. The organic
layer with most of the product was separated from the aqueous
layer containing the catalyst. 2 mL of additional organic solvent was
added again to extract the remaining product from the aqueous
layer. The two organic fractions were added together and 1 μL was
anlyzed by GC-FID. A 20 μL aliquot of the aqueous phase was
removed and the NPs were precipitated by adding 40 μL of THF
and centrifuging for 15 minutes prior to analyzing 1 μL of the
supernatant by GC-FID. The NPs in the aqueous phase were used
again as catalysts by adding organic solvent, adding allyl alcohol,
and purging with hydrogen gas for another reaction cycle. The
entire process was repeated until the reaction stopped going to
completion.
1
other impurities or free ligand. H NMR spectroscopy showed
similar results for glutathione-coated Pd-alloy NPs synthesized
under inert and oxygen condition.
UV-Vis data were obtained using a Varian Cary 50 Bio UV-Visible
spectrophotometer ranging from 200–900 nm. All aqueous solu-
tions were placed in a quartz cuvette with an optical path length of
1
cm and each spectrum was corrected using the solvent spectrum.
UV-Vis data were used to assess the NP stability in the presence of
H and throughout the conditions of the reaction. We also obtained
2
the spectra of allyl alcohol, 1-propanol, propanal, the various
solvents tested, and Pd(II) salts or thiolate complexes to correlate
with the spectra obtained under the reaction conditions.
ChemCatChem 2020, 12, 1–10
www.chemcatchem.org
8
© 2020 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
��
These are not the final page numbers!

Full Papers
Acknowledgments
[15] A. Roucoux, J. Schulz, H. Patin, Chem. Rev. 2002, 102, 3757–3778.
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
[
16] J. Durand, E. Teuma, M. Gómez, Eur. J. Inorg. Chem. 2008, 2008, 3577–
3586.
We gratefully acknowledge the National Science Foundation
[
17] K. Esumi, S. Sarashina, T. Yoshimura, Langmuir 2004, 20, 5189–5191.
(Grant CHE-1611170) for partial support of this research.
[18] M. Moreno, L. N. Kissell, J. B. Jasinski, F. P. Zamborini, ACS Catal. 2012, 2,
602–2613.
[19] S. R. Isaacs, E. C. Cutler, J.-S. Park, T. R. Lee, Y.-S. Shon, Langmuir 2005,
1, 5689–5692.
[20] D. J. Gavia, Y.-S. Shon, Langmuir 2012, 28, 14502–14508.
2
2
Conflict of Interest
[
[
[
21] F. Porta, Ž. Krpetić, L. Prati, A. Gaiassi, G. Scarì, Langmuir 2008, 24, 7061–
7064.
22] P. B. Murphy, F. Liu, S. C. Cook, N. Jahan, D. G. Marangoni, T. B. Grindley,
P. Zhang, Can. J. Chem. 2009, 87, 1641–1649.
The authors declare no conflict of interest.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
23] S.-K. Oh, Y. Niu, R. M. Crooks, Langmuir 2005, 21, 10209–10213.
Keywords: Biphasic · Glutathione · Selectivity · Homogeneous ·
Catalysis · Recyclability · Nanoparticles · Allyl Alcohol ·
Hydrogenation · Isomerization
[24] E. Sadeghmoghaddam, H. Gu, Y.-S. Shon, ACS Catal. 2012, 2, 1838–1845.
[
[
[
[
25] E. Sadeghmoghaddam, C. Lam, D. Choi, Y.-S. Shon, J. Mater. Chem.
011, 21, 307–312.
26] E. Sadeghmoghaddam, K. Gaïeb, Y.-S. Shon, Appl. Catal. A 2011, 405,
2
137–141.
27] M. S. Maung, T. Dinh, C. Salazar, Y.-S. Shon, Colloids Surf. A 2017, 513,
367–372.
[
1] S. Mandal, D. Roy, R. V. Chaudhari, M. Sastry, Chem. Mater. 2004, 16,
714–3724.
[2] R. M. Crooks, M. Zhao, L. Sun, V. Chechik, L. K. Yeung, Acc. Chem. Res.
001, 34, 181–190.
3] R. Sardar, A. M. Funston, P. Mulvaney, R. W. Murray, Langmuir 2009, 25,
3840–13851.
[4] Y. Xia, Y. Xiong, B. Lim, S. E. Skrabalak, Angew. Chem. Int. Ed. 2009, 48,
0–103.
5] R. W. J. Scott, O. M. Wilson, R. M. Crooks, J. Phys. Chem. B 2005, 109,
92–704.
[6] D. B. Pacardo, M. Sethi, S. E. Jones, R. R. Naik, M. R. Knecht, ACS Nano
009, 3, 1288–1296.
28] Z. Song, W. Li, F. Niu, Y. Xu, L. Niu, W. Yang, Y. Wang, J. Liu, J. Mater.
3
Chem. A 2017, 5, 230–239.
[
[
29] S. Sharma, J. Colloid Interface Sci. 2015, 441, 25–29.
30] J. Lee, H. S. Shim, M. Lee, J. K. Song, D. Lee, J. Phys. Chem. Lett. 2011, 2,
2
[
2840–2845.
1
[31] S. Sharma, B. Kim, D. Lee, Langmuir 2012, 28, 15958–15965.
[
[
32] F. J. Ibañez, F. P. Zamborini, J. Am. Chem. Soc. 2008, 130, 622–633.
33] R. W. J. Scott, A. K. Datye, R. M. Crooks, J. Am. Chem. Soc. 2003, 125,
6
[
3708–3709.
6
[34] R. W. J. Scott, O. M. Wilson, S.-K. Oh, E. A. Kenik, R. M. Crooks, J. Am.
Chem. Soc. 2004, 126, 15583–15591.
35] J. RaviPrakash, A. H. McDaniel, M. Horn, L. Pilione, P. Sunal, R. Messier,
2
[
[
7] S. E. Eklund, D. E. Cliffel, Langmuir 2004, 20, 6012–6018.
8] D. M. Dotzauer, S. Bhattacharjee, Y. Wen, M. L. Bruening, Langmuir 2009,
25, 1865–1871.
R. T. McGrath, F. K. Schweighardt, Sens. Actuators B 2007, 120, 439–446.
[36] M. Hakamada, H. Nakano, T. Furukawa, M. Takahashi, M. Mabuchi, J.
Phys. Chem. C 2010, 114, 868–873.
[
[9] Y. Lin, K. A. Watson, M. J. Fallbach, S. Ghose, J. G. Smith, D. M. Delozier,
[
[
37] G. Sandrock, J. Alloys Compd. 1999, 293–295, 877–888.
38] R. Uma, C. Crévisy, R. Grée, Chem. Rev. 2003, 103, 27–52.
W. Cao, R. E. Crooks, J. W. Connell, ACS Nano 2009, 3, 871–884.
10] P. Wagener, A. Schwenke, S. Barcikowski, Langmuir 2012, 28, 6132–
[
[
[
6140.
11] S. Panigrahi, S. Basu, S. Praharaj, S. Pande, S. Jana, A. Pal, S. K. Ghosh, T.
Pal, J. Phys. Chem. C 2007, 111, 4596–4605.
12] R. Narayanan, M. El-Sayed, Top. Catal. 2008, 47, 15–21.
[13] D. Astruc, F. Lu, J. R. Aranzaes, Angew. Chem. Int. Ed. 2005, 44, 7852–
872.
14] N. Semagina, L. Kiwi-Minsker, Catal. Rev. 2009, 51, 147–217.
Manuscript received: October 17, 2019
Revised manuscript received: January 7, 2020
Version of record online: ■■■, ■■■■
7
[
ChemCatChem 2020, 12, 1–10
www.chemcatchem.org
9
© 2020 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
��
These are not the final page numbers!

Full Papers
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
FULL PAPERS
Protected nanoparticles: Gluta-
thione-protected Pd, PdAu and PdPt
nanoparticles (NPs) show greatly
increased reactivity and stability
towards the hydrogenation/isomer-
ization of allyl alcohol in a biphasic
organic/aqueous solvent mixture
compared to single-phase alkane-
thiol-protected Pd-based NPs. The
branched, open structure of the glu-
tathione and use of the biphasic
solvent provides high catalytic
activity, high selectivity, and easy
removal of the products in the
organic phase for good recyclability
of the water-soluble catalysts.
Dr. S. Bhama, T. R. Sibakoti, Dr. J. B.
Jasinski, Prof. F. P. Zamborini*
1 – 10
Highly Active, Selective, and Recy-
clable Water-Soluble Glutathione-
Stabilized Pd and Pd-Alloy Nano-
particle Catalysts in Biphasic
Solvent
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5