Improved stability and catalytic activity of palladium nanoparticle catalysts using phosphine-functionalized imidazolium ionic liquids

Article abstract of DOI:10.1002/adsc.201100551

Palladium nanoparticles (Pd NPs) stabilized by 6 different phosphine-functionalized ionic liquids (PFILs) were synthesized in imidazolium-based ionic liquids (ILs) using H_2(g) (4 bar) as a reductant. Characterization showed well-dispersed parti

Full text of DOI:10.1002/adsc.201100551

FULL PAPERS
DOI: 10.1002/adsc.201100551
Improved Stability and Catalytic Activity of Palladium
Nanoparticle Catalysts using Phosphine-Functionalized
Imidazolium Ionic Liquids
Kylie L. Luskaa and Audrey Mooresa,*
a
Department of Chemistry, McGill University, Otto Maass Chemistry Building, 801 Sherbrooke Street West, Montreal,
Quebec, Canada H3A 2K6
Fax. (+1)-514-398-3797; phone: (+1)-514-398-4654
E-mail: audrey.moores@mcgill.ca
Received: July 12, 2011; Revised: September 10, 2011; Published online: November 16, 2011
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/adcs.201100551.
Abstract: Palladium nanoparticles (Pd NPs) stabi- sions of the Pd NPs were employed as biphasic hy-
lized by 6 different phosphine-functionalized ionic drogenation catalysts for the reduction of the olefinic
liquids (PFILs) were synthesized in imidazolium- bond in styrene under mild conditions (508C, 4 bar
based ionic liquids (ILs) using H2(g) (4 bar) as a re-
H2(g), 1.5 h). The PFIL-stabilized Pd NPs were effec-
ductant. Characterization showed well-dispersed par- tive hydrogenation catalysts and showed superior ac-
ticles of ~3 nm (TEM) and confirmed the PFIL sta- tivity and recyclability over NPs synthesized in the
bilization of the NPs (XPS). The PFILs were com- absence of PFILs. Poisoning tests of the Pd NP cata-
posed of an imidazolium functionality separated lysts and characterization of the electronic properties
from the phosphine group by a propyl or undecyl of the phosphine were also performed.
chain. The counter anions for both FILs and IL sol-
vents were chosen from N-bis(trifluoromethanesulfo- Keywords: biphasic catalysis; functionalized ionic liq-
nyl)imide (Tf2NÀ), trifluoromethanesulfonate (TfOÀ) uids; hydrogenation; ionic liquids; nanoparticles;
À
or hexafluorophosphate (PF6 ). Colloidal suspen- palladium
Introduction
The most common use of metal NPs embedded in
ILs has been in the field of catalysis as the IL allows
Ionic liquids (ILs) have been employed as useful for a facile separation of the product, through decant-
media for the generation of transition metal nano- ation or extraction, and reuse of the IL phase contain-
structures of various sizes and shapes as a result of ing the NP catalyst. These NPs possess high catalytic
their ability to act as both a solvent and stabilizing activities and have been used most effectively in the
ligand.[1] 1,3-Dialkylimidazolium ILs have been of hydrogenation of alkenes[5] and arenes.[4c] The catalyt-
particular interest since their pre-organized structures ic activity of these NPs is attributed to the ability of
provide both a hydrophilic and hydrophobic domain the loosely bound IL species to be displaced, which
to control the growth of nanoparticles (NPs).[2] The allows the substrate access to catalytically active
resulting NPs can be isolated from the IL or em- sites.[2] Although the ILs provide short-term stabiliza-
ployed as colloidal suspensions since electrostatic sta- tion, their displacement can also lead to aggregation
bilization of the NPs is provided by the IL through of the metal NPs under catalytic conditions.[5b,6] Fur-
the formation of supramolecular species of the form ther stabilization of metal NPs can be provided by
n+
[(DAI)x(X)xÀn
]
and [(DAI)xÀn(X)x]nÀ (where DAI= classic transition metal ligands (i.e., bipyridine,[1e,7]
1,3-dialkylimidazolium cation and XÀ =counter phenanthroline),[8] polymers (i.e., poly-vinylpyrroli-
anion).[2] Imidazolium ILs are interesting species for done,[9] IL co-polymers)[10] or various supports
the synthesis of NPs since changing the nature of the (SiO2,[11] carbon nanotubes)[12] and provide systems
IL (i.e., N-alkyl chain length,[3] counter anion volu- possessing increased stabilities, while maintaining
me)[1c,4] can influence the properties of the resulting high catalytic activities.
NPs.
Adv. Synth. Catal. 2011, 353, 3167 – 3177
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3167

Kylie L. Luska and Audrey Moores
FULL PAPERS
Functionalized ILs (FILs) containing such metal- sequent use as efficient and recyclable biphasic hydro-
binding moieties as amines,[12b,13] bipyridine,[6,7c,d,14] car- genation catalysts. Halide-free NPs were synthesized
boxylates,[12b] nitriles[15] or thiolates[16] have been pre- from Pd
ACTHNUTRGNEU(GN acac)2 in [BDMI]X under an atmosphere of
viously employed in the synthesis of transition metal H2(g) in the presence of PFILs (Scheme 2). Well-dis-
NPs. FILs are effective stabilizing ligands since they persed NPs were characterized by transmission elec-
provide both a covalent and an electrostatic stabiliza- tron microscopic (TEM) and X-ray photoelectron
tion, while non-functionalized ILs can only provide spectroscopic (XPS) analysis, which provided evi-
an electrostatic stabilization to the NPs. Covalent dence for the PFIL stabilization of the NPs. Pd NPs
binding of FILs to the NP surface would also influ- were employed as olefin hydrogenation catalysts for
ence the electronic properties of the metal as a result the conversion of styrene to ethylbenzene. Compari-
of the interaction with the binding moiety and the sons between PFIL-stabilized NPs and those obtained
presence of the highly polar IL layer surrounding the without PFIL stabilization demonstrate the superior
NP. In the context of catalysis, this effect could lead activity and stability of these systems. Furthermore,
to a change of the electron density at the surface the nature of the FIL employed, in terms of the N-
metal atoms and/or control of the solubility of sub- alkyl chain length (n) and the anionic species (XÀ),
strates at the metal surface, providing both highly was shown to influence the catalytic activity of the
active and selective catalytic systems.[7c] In addition to NPs.
improving NP stability, FIL-stabilized NPs can pro-
vide highly recyclable biphasic catalytic systems alle-
viating difficulties with aggregation of NPs and leach- Results and Discussion
ing of metal or organic species into the organic
phase.[7c] Several strategies have employed FILs in the Synthesis of Phosphine-Functionalized Ionic Liquids
synthesis of NPs including: the use of FILs as both a
reaction solvent and NP stabilizer,[15] the incorpora- The synthesis of PFILs 1–6 followed a similar proce-
tion of FIL-stabilized NPs onto carbon nanotubes or dure to that reported by Dupont and co-workers in
fibers[12b,c,14,16a] and the use of FILs as an IL-soluble li- which a radical chain addition of diphenylphosphine
gand.[6,7c,d] Recently, several groups have employed bi- to alkene FILs was catalyzed by 1,1’-azobis(cyclohex-
pyridine FILs in the synthesis of efficient biphasic hy- anecarbonitrile) (ABCN) to provide the desired phos-
drogenation NP catalysts.[6,7c,d]
phines in almost quantitative yields (Scheme 1).[17]
Herein, we report the employment of phosphine- Two pathways were employed to provide a series of
functionalized ionic liquids (PFILs) 1–6 (Scheme 1) as alkene FILs in which the length of the N-alkyl chain
stabilizing ligands in the generation of colloidal sus- (n), for n=1, 9, and the counter anion (XÀ), for XÀ =
pensions of Pd NPs in imidazolium ILs and their sub- bis(trifluoromethanesulfonyl)imide (Tf2NÀ), trifluoro-
Scheme 1. Synthesis of phosphine-functionalized ionic liquids (PFILs) 1–6.
Scheme 2. Formation of PFIL-stabilized palladium nanoparticles in [BDMI]X.
3168
asc.wiley-vch.de
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2011, 353, 3167 – 3177

Improved Stability and Catalytic Activity of Palladium Nanoparticle Catalysts
methanesulfonate
(TfOÀ),
hexafluorophosphate serve as stabilizing ligands and/or NP catalyst poi-
À
(PF6 ), were varied. In pathway 1, alkenylimidazoli- sons.[18]
um salts were synthesized through the quaternization
An X-ray photoelectron spectroscopic (XPS) analy-
of 1,2-dimethylimidazole with alkenyl bromides and a sis of the Pd NPs was employed to elucidate the
salt exchange with either LiNTf2, LiOTf or nature of the stabilizing layer of the particles. XPS
NH4PF6.[17] Pathway 2 was employed to synthesize the analysis of Pd NPs stabilized by PFIL 3 showed the
undecenylimidazolium TfOÀ salt via a halide-free presence of palladium, phosphorus, fluorine, nitrogen,
methodology involving the quarternization of 1,2-di- sulfur, oxygen and carbon, which signified the pres-
methylimidazole with undec-10-enyl trifluorometh- ence of the PFIL in the ligand sphere of the NPs. The
A
P 2p(3/2) binding energy (BE) for PFIL 3-stabilized
NPs (130.9 eV) was comparable to that of PFIL 3
(130.7 eV) (see Supporting Information, Figure S6
and Figure S7). A shift in the XPS spectrum for the
Pd P signal was not expected since the electronega-
tivities of Pd and P are similar. The BEs for Pd,
Synthesis and Characterization of Palladium
Nanoparticles
À
The synthesis of Pd NPs was achieved through the re- 3d(5/2) =335.7 eV and 3d(3/2) =341.0 eV, indicated the
duction of Pd(acac)2 (acac=acetylacetonate) in NPs were composed of Pd(0) (see Supporting Infor-
ACHTUNGTRENNUNG
[BDMI]X under an atmosphere of H2(g) (4 bar) in the mation, Figure S5) and were in agreement with other
presence of 1.0 equivalent of PFILs 1–6 (Scheme 2). phosphine-stabilized Pd NPs.[19] The Pd 3d spectrum
À
After reaction for 4 h at 808C, the IL transforms from also indicated the presence of Pd O bonds, which
a bright yellow solution to a brown suspension and could result from either the oxidation of Pd due to air
TEM analysis of the IL phase shows the presence of exposure or due to the interaction of the TfOÀ coun-
small, well-dispersed NPs having diameters of ~3 nm ter anion with the Pd surface (see Supporting Infor-
(Table 1). A correlation between the size of the Pd mation, Figure S8 and Figure S9).[20] The XPS data
NPs and the parameters of the phosphine ligand, in suggest that the primary stabilization of the Pd NPs
terms of n and XÀ, was not observed. The NP:IL col- was provided by the phosphine moiety of the PFIL;
loidal suspension was used directly for catalytic stud- although, the possible stabilization by the IL anion
ies or the NPs were isolated from the IL by repeated and/or a metal oxide layer cannot be excluded. To
washing using an acetone:pentane (2:1) solution and this end, the TfOÀ anion could have been provided by
centrifugation. The [PFIL]/[Pd] molar ratio of 1.0 was either PFIL 3 or the IL solvent and as such the pres-
selected as a good compromise between stability and ence of [BDMI]OTf in the ligand layer of the NPs
activity, whereas a ratio <1.0 provided Pd aggregates could not be ruled out as contributing to the stabiliza-
(Figure 1) and a ratio of 2.0 provided mono-dispersed tion of the NPs. XPS analysis of [BDMI]OTf-stabi-
NPs with dramatically reduced activity. It should be lized Pd NPs was not undertaken since only Pd aggre-
noted that the 1-alkyl-2,3-dimethylimidazolium salts gates were produced and did not provide a compara-
of PFILs 1–6 and [BDMI]X ILs were chosen to pre- ble sample to our mono-dispersed PFIL-stabilized
vent the formation of carbene species during NP syn- NPs. However, Dupont and co-workers have reported
thesis and allow for the synthesis of well-defined NPs XPS evidence for IL anion stabilization of NPs syn-
free of carbene contamination. The formation of 1,3- thesized in IL solvents.[5b,20] Small NPs stabilized by
dialkylimidazolium carbenes has been proposed to
Table 1. Particle size and distribution of Pd NPs stabilized by PFILs 1–6 determined by TEM before and after catalysis.[a]
Solvent
Stabilizing Ligands
Particle Size [nm] (standard deviation s)
PFIL
n
X
Before catalysis
After three cycles
A
none[b]
1
aggregates
3.0 (Æ0.6)
3.2 (Æ0.8)
aggregates
3.5 (Æ1.0)
3.0 (Æ0.6)
aggregates
2.8 (Æ0.7)
2.7 (Æ0.5)
aggregates
3.1 (Æ0.7)
3.5 (Æ0.7)
aggregates
3.7 (Æ0.8)
2.8 (Æ0.7)
aggregates
2.7 (Æ0.5)
2.5 (Æ0.6)
1
9
Tf2N
Tf2N
2
A
none[b]
3
4
1
9
TfO
TfO
A
none[b]
5
6
1
9
PF6
PF6
[a]
[b]
Reaction conditions: PdACTHUNTGRNEGN(U acac)2 (0.1 mmol), PFIL 1–6 (0.1 mmol), [BDMI]X (2.0 mL), H2(g) (4 bar), 808C, 4 h, stirred at
1000 rpm.
Identical reaction conditions without the addition a PFIL stabilizer.
Adv. Synth. Catal. 2011, 353, 3167 – 3177
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
3169

Kylie L. Luska and Audrey Moores
FULL PAPERS
Figure 1. TEM micrographs of Pd NPs synthesized in [BDMI]NTf2 employing 0.0 equiv. of PFIL (left) and 1.0 equiv. of PFIL
1 (right).
À
À
IL solvents containing either BF4 or PF6 anions ure S23), caused the formation of a broad signal at
have shown a metal-fluorine interaction in XPS.
À17.8 ppm corresponding to free phosphine. Heating
31P NMR studies of PFIL 1 stabilized Pd NPs in these samples to 50 or 758C caused the broadening or
[BDMI]NTf2 were conducted. The NPs were synthe- complete disappearance of the free phosphine signal.
sized as outlined above employing 1.0 equivalent of These two points indicate that an exchange process
PFIL 1 and care was taken to keep the sample under between free PFIL and the PFIL-stabilized NPs is oc-
an inert atmosphere. Analysis of the NPs showed only curring and demonstrates the lability of the ligand. In
a weak signal at 29.5 ppm (see Supporting Informa- the context of catalysis, this lability would allow the
tion, Figure S10), which does not correspond to PFIL Pd NPs to be stable against aggregation while creating
1 (À17.6 ppm) (see Supporting Information, Figure active sites during catalysis to provide effective cata-
S11a) or the PFIL 1 oxide (31.8 ppm) (see Supporting lyst systems.
Information, Figure S11b). This signal is consistent
with complex
a
Pd(II)
of
a
diphenyl-
A
A
was taken after the first cycle of styrene hydrogena- Nanoparticles
tion (see Supporting Information, Figure S12) and no
signal was detected. The disappearance of the species The catalytic properties of Pd NPs stabilized by
observed at 29.5 ppm correlates well with the reduc- PFILs 1–6 were tested by use in the biphasic olefin
tion of the intermediate Pd complex with PFIL 1 into hydrogenation of styrene (Scheme 3). The PFIL-stabi-
Pd NPs during the first hydrogenation cycle. In liquid lized Pd NPs embedded in [BDMI]X were efficient
NMR, probed nuclei attached to a surface do not ex- catalysts for the reduction of styrene to ethylbenzene
hibit any signal. All these observations thus indicate under mild conditions (508C, 4 bar H2(g), 1.5 h). Aro-
that the majority of the phosphine was attached to matic hydrogenation of styrene to ethylcyclohexane
the NPs surface.
was not observed under these reaction conditions.
The NP:IL mixtures were reused for consecutive
Lastly, the dynamics of the PFIL 1-stabilized NPs
were investigated using 31P NMR. Heating a sample hydrogenation cycles after pentane extraction and
of PFIL 1-stabilized NPs to either 50 or 758C (see
Supporting Information, Figure S13 and Figure S14)
did not change the spectrum (i.e., liberate free PFIL
1 from the NPs). Addition of greater equivalents of
PFIL 1, 1.0, 2.0 or 4.0 equivalents with respect to Pd
(see Supporting Information, Figure S15 to Fig- rene employing Pd NPs stabilized by PFILs 1–6.
Scheme 3. Reaction conditions for the hydrogenation of sty-
3170
asc.wiley-vch.de
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2011, 353, 3167 – 3177

Improved Stability and Catalytic Activity of Palladium Nanoparticle Catalysts
Table 2. Catalytic activity of palladium nanoparticles stabilized by PFILs 1–6 for the hydrogenation of styrene.[a]
Solvent
Stabilizing Ligands
Styrene Conversion [%][b]
Cycle 2
PFIL
n
X
Cycle 1
Cycle 3
A
blank[c]
1
23
85
73
18
48
46
30
28
52
31
100
83
21
79
61
31
52
73
37
100
85
26
78
70
28
65
70
1
9
Tf2N
Tf2N
2
A
blank[c]
3
4
1
9
TfO
TfO
A
blank[c]
5
6
1
9
PF6
PF6
[a]
Reaction conditions: PdACTHUNTGRNEGN(U acac)2 (0.1 mmol), PFIL 1–6 (0.1 mmol), [BDMI]X (2.0 mL), styrene (25.0 mmol), H2(g) (4 bar),
508C, 1.5 h, stirred at 1000 rpm.
Determined by GC analysis employing dodecane as internal standard.
Identical reaction conditions without the addition a PFIL stabilizer.
[b]
[c]
drying under vacuum for at least 1 h. As outlined in vided well-dispersed NPs, while the synthesis without
Table 2, each NP catalyst was employed for three con- PFIL yielded a mixture of small NPs and large aggre-
secutive hydrogenation cycles of styrene; the catalytic gates. The increased activities for PFIL-stabilized NPs
activity of the Pd NPs was observed to increase upon were attributed to an increase in the number of cata-
recycling for both PFIL and [BDMI]X-stabilized cata- lytically active sites for PFIL-stabilized NPs resulting
lysts. This induction period could result from: (i) an from a larger surface-to-volume ratio of the smaller
incomplete reduction of the Pd(II) precursor, which Pd NPs.
allowed the continued formation of Pd NPs during
Both the N-alkyl chain length and the counter
the initial hydrogenation cycles; or (ii) a restructuring anion of the PFIL ligands were shown to influence
of the NPs under catalytic conditions to provide more the catalytic activity of the resulting Pd NPs. The
active catalytic sites. The latter phenomenon has been nature of the PFIL counter anion (XÀ) affected the
reported by Janiak and co-workers for [1-butyl-3- activity of the NPs greatly as the Tf2NÀ PFILs 1 and 2
methylimidazolium]X-stabilized NPs, which exhibited possessed a higher catalytic activity then the TfOÀ or
À
an increase in the initial rate of cyclohexene hydroge- PF6 PFILs 3–6. The most active catalyst was PFIL 1-
nation with recycling.[5a,22] The increased activity was stabilized Pd NPs as complete conversion of styrene
attributed to the restructuring of the metal surface was achieved during the second and third hydrogena-
and/or the formation of N-heterocyclic carbene tion cycles (Table 2; entry 2). Increased activity for
(NHC) surface species.[5a,22] In regard to our Pd NPs, the Tf2NÀ-based catalyst could result from the in-
the formation of Pd-NHC species is less likely to creased hydrophobicity Àof PFIL 1 and [BDMI]NTf2,
occur since the 1-alkyl-2,3-dimethylimidazolium salts as both TfOÀ- and PF6 -based ILs are known to be
of PFILs and [BDMI]X were employed. Furthermore, more hydrophilic.[23] It was also observed that the
TEM analyses of the NP:IL phase before catalysis length of the N-alkyl chain (n) influenced the activity
and after the third hydrogenation cycle were per- of the NPs; although, to a lesser extent than that of
formed for each catalyst (Table 1) and showed consis- the counter anion. In the case of the Tf2NÀ-based
tent NP sizes before and after three cycles of catalysis. PFILs 1 (n=1) and 2 (n=9), the longer chain PFIL 2
Therefore, the induction period most likely results possessed a lower catalytic activity than the shorter
from the incomplete decomposition of the Pd(II) pre- chain PFIL 1. The lower catalytic activity could result
cursor under the reaction conditions employed for NP from an increase in the hydrophobicity of the ligand
synthesis.
sphere, which would increase the solubility of styrene
Hydrogenation of styrene employing Pd NPs stabi- or ethylbenzene at the Pd surface and thus hinder
lized by PFILs 1–6 were compared against Pd NPs access to catalytically active sites.
synthesized in [BDMI]X in the absence of a phos-
The long-term stability of the NP catalysts was also
phine stabilizer to investigate the influence of the evaluated over ten consecutive hydrogenation cycles
PFIL on catalytic activity. NP:IL mixtures synthesized for the hydrogenation of styrene employing the most
solely in [BDMI]X were shown to have some catalytic active catalysts based on the short N-alkyl chain phos-
activity (Table 2; entries 1, 3, 5); however, PFIL-stabi- phines, PFIL 1, 3 and 5 (see Supporting Information,
lized Pd NPs showed a significant increase in catalytic Figure S1). PFIL 1 stabilized NPs were again the
activity (Table 2; entries 2–4, 6/7, 8/9). As mentioned most active catalyst over the series of ten hydrogena-
above, NP synthesis in the presence of PFILs 1–6 pro- tion cycles. In comparison, PFIL 3 provided slightly
Adv. Synth. Catal. 2011, 353, 3167 – 3177
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
3171

Kylie L. Luska and Audrey Moores
FULL PAPERS
Table 3. 31P NMR data for the phosphine selenides of PFILs.
lower styrene conversions, while the activity of PFIL
5 stabilized NPs was significantly lower. The average
styrene conversions over the ten hydrogenation cycles
were: 85, 61 and 38% for PFIL 1, 3, and 5 respective-
ly. TEM was employed to analyzes the NP:IL phase
Stabilizing Ligands
d [ppm]
1JP,Se [Hz]
PFIL
n
X
1
3
5
6
1
1
1
9
Tf2N
TfO
PF6
36.2
36.2
36.2
36.6
726
726
726
722
after ten runs and showed Pd NPs sizes of 3.3
3.8
(Æ0.9),
E
N
PF6
spectively. These diameters were similar to those ob-
served before and after the third cycle of catalysis
(Table 1).
In view of the 31P NMR studies outlined above, a tronic parameters of the phosphine. The synthesis in-
catalyst poison test employing CS2 was performed to volved stirring the PFIL and selenium powder in
determine whether the active catalyst species was het- DMSO-d6 overnight at room temperature. The
erogeneous. The addition of 0.5 equivalents (w.r.t. Pd) 31P NMR of the reaction mixtures showed quantita-
of CS2 to the NP:IL mixture completely deactivated tive conversion of PFILs into the corresponding phos-
the catalyst such that no ethylbenzene was formed phine selenide (Table 3, see Supporting Information,
under typical hydrogenation conditions for Pd NPs Figure S24 to Figure S28).
stabilized by PFIL 1. Therefore, this indicated that
According to the data in Table 3, it was determined
the residual Pd complex that was present after NP that the nature of the counter anion does not influ-
synthesis was not the active catalyst species.
ence the electronic properties of the PFILs as 1, 3,
and 5 had identical JP,Se values. The length of the N-
1
alkyl chain has a small influence on the electronics of
1
Characterization of the Electronic Properties of
Phosphine-Functionalized Ionic Liquids
the phopshine moiety as PFIL 6 had a smaller JP,Se
than PFIL 5 and is thus a slightly better s-donor. This
comparison also shows that a propyl alkyl chain is of
Functionalization of phosphine ligands with ionic sufficient length to isolate the phosphine moiety from
moieties has been a popular strategy in the develop- the electron-withdrawing ability of the imidazolium
ment of biphasic catalysts since the use of tris(3-sulfo- group. Furthermore, the selenide of PPh3 was also
phenyl)phosphine trisodium salt (TPPTS) in the synthesized in DMSO-d6 (1JP,Se =736 Hz) and as ex-
Ruhrchemie/Rhꢁne–Poulenc process.[24] Important to pected our PFILs are stronger s-donors than PPh3
this field in the context of catalysis is the influence, due the presence of the alkyl substituent.
both steric and electronic, that the ionic group has on
the coordination chemistry of the phosphine moiety.
The steric effect of a phosphine is often evaluated by Synthesis of Rhenium Carbonyl Complexes
the Tolman cone angle, while the electronic influence
can be estimated from the synthesis the phosphine se- PFILs 1, 3, 5 and 6 were also used in the synthesis of
lenides or the metal carbonyl complexes.[24b] To this Re carbonyl complexes to evaluate the electronic
end, the influence of the PFIL parameters (i.e., N- properties of these phosphines through the CO
alkyl chain length and counter anion) on the electron- stretching frequencies of these complexes (Table 4).
ic properties of the phosphine moiety were deter- ReBr(CO)3ACTHNUTRGNEU(GN PFIL)2 complexes were synthesized in a
mined through the synthesis of the corresponding similar procedure as that reported by Storhoff and
phosphine selenides and rhenium carbonyl complexes. Lewis.[26] In short, ReBr(CO)5 was refluxed in THF to
produce [ReBr(CO)3ACTHNUTRGNE(NUG THF)]2, an intermediate THF
complex, which was refluxed in CH2Cl2 with
Synthesis of Phosphine Selenides
Table 4. Carbonyl stretching frequencies of ReBr(CO)3
AHCTUNGTRENNUNG
The electronic properties of phosphines can be evalu-
1
ated from the JP,Se coupling constant in the 31P NMR
of the corresponding phosphine selenide. Allen and
co-workers have shown that the JP,Se coupling con-
stant can be related to the degree of s-character in
CO Stretching Frequency [cmÀ1][a]
1
PFIL
n
X
1
3
5
6
1
1
1
9
Tf2N
TfO
PF6
2028, 1946, 1901
2028, 1944, 1900
2022, 1935, 1912
2023, 1938, 1897
the lone pair of the phosphorus atom and as such
1
stronger s-donors have smaller JP,Se coupling con-
stants.[25] The series of phosphine selenides based on
PFILs, 1, 3, 5 and 6 was synthesized to determine
whether the N-alkyl chain length (n=1, 9) or counter
PF6
[a]
All signals strong and samples recorded as an ATR anal-
ysis.
anion (XÀ =Tf2NÀ, TfOÀ, PF6 ) influenced the elec-
À
3172
asc.wiley-vch.de
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2011, 353, 3167 – 3177

Improved Stability and Catalytic Activity of Palladium Nanoparticle Catalysts
thesis and application of metal NPs in non-functional-
ized ILs. Furthermore, the N-alkyl chain length and
the counter anion of the PFILs have been shown to
influence the catalytic activity of the resulting NPs.
Scheme 4. Synthesis of model rhenium complexes.
Experimental Section
2.0 equivalents of PFIL to yield the desired fac-
ReBr(CO)
3ACHTUNGTRENNUNG(PFIL)2 complex (Scheme 4). ReBr(CO)3 General
ACHTUNGTRENNUNG
All syntheses were carried out under an argon atmosphere
propyl (n=1) chain and allowed for the evaluation of
the counter anion. The CO stretching frequencies for
the Tf2NÀ (PFIL 1) and TfOÀ complexes (PFIL 3)
were similar, while those of the PF6 complex (PFIL
5) were shifted to lower wavenumbers. This shift indi-
cates that the PF6 counter anion interacts with the
Re metal center, causing a shift in the CO frequen-
cies. Furthermore, ReBr(CO)3ACHTNUTRGNEUNG(PFIL)2 complexes syn-
thesized using PFIL 5 and 6 allowed for the compari-
son between the short chain (n=1) and long chain
(n=9) PFILs, respectively. CO stretching frequencies
for ReBr(CO)
numbers compared to ReBr(CO)
ReBr(CO)3A(PFIL 3)2; however, the shift was not as
employing Schlenk techniques. [1-Allyl-2,3-dimethylimida-
[17]
zolium]Br,[17] [1-allyl-2,3-dimethyl-imidazolium]NTf2
and
PFIL 1[17] were prepared following known literature proce-
dures. Acetonitrile (dried over 4 ꢂ molecular sieves), allyl
bromide (filtration through basic alumina), dichloromethane
(distillation over calcium hydride), diethyl ether (distillation
over sodium and benzophenone), pyridine (distillation over
calcium hydride), and tetrahydrofuran (Grubbs apparatus)
were purified prior to use. All other chemicals and solvents
were purchased from commercial sources and used without
further purification. Melting points (mp) were determined
on a Gallenkamp melting point apparatus and are uncor-
rected. FT-infrared spectra were measured on a Perkin–
Elmer Spectrum 400 Spectrometer equipped with a Pike
GladiATR. Nuclear magnetic resonance (NMR) spectra
were recorded on a 300 MHz Varian Mercury spectrometer.
1H and 13C NMR spectra were calibrated to TMS using the
residual solvent signal and 31P NMR spectra were calibrated
using 85% H3PO4. Mass spectra (MS) were recorded in posi-
tive electrospray mode with an IonSpec 7.0 tesla FTMS
(Lake Forest, CA). Transmission electron microscopy
(TEM) was performed on a Philips CM 200 microscope. X-
ray photoelectron spectrometry (XPS) was performed on a
VG ESCALAB 3 MKII spectrometer (VG, Thermo Elec-
tron Corporation, UK) equipped with an MgKa source.
High-pressure experiments were performed employing a
Parr Instruments 5000 Series Multiple Reactor System
equipped with 100 mL reaction vessels.
À
À
(PFIL 6)2 were shifted to lower wave-
CTHUNGTRENNUNG
CHTUNGTRENNUNG
great as that observed for ReBr(CO)
(PFIL 5)2.
À
Therefore, the PF6 anion interacts with the Re
center regardless of the length of the alkyl chain; al-
though, this interaction decreases as the length of the
alkyl chain of the PFIL is increased. Taking this find-
ing and that of the phosphine selenides into account,
it can be concluded that the electronic properties of
the phosphine moiety are not significantly influenced
by changes to either n or XÀ of the PFILs.
The model Re complexes provide insight into the
À
catalytic activity of the PF6 PFILs 5 and 6 (Table 2).
These complexes evidenced a possible interaction be-
tween the PF6 anion and the Re metal center, which
À
can also be envisaged to occur within the ligand
sphere of the Pd NPs. As discussed above, Dupont
and co-woÀrkers have evidenced an interaction be-
Synthesis of Phosphine Functionalized Ionic Liquids
1-Allyl-2,3-dimethylimidazolium trifluoromethanesulfonate:
À
1-Allyl-2,3-dimethyl-imidazolium
bromide
(5.78 g,
tween BF4 and PF6 IL anions with the surface of
26.6 mmol) was dissolved in 25 mL H2O and LiOTf (4.35 g,
27.9 mmol) was added in small portions to the solution. The
mixture was stirred at room temperature for 18 h. The aque-
ous solvent was removed, the crude product was dissolved
in MeCN, filtered through celite, dried over MgSO4 and sol-
vent removed to afford a white solid; yield: 4.49 g (59%);
metal NPs synthesized in IL solvents.[5b,20] This inter-
action could explain the reduced catalytic activity of
À
NPs stabilized by PFILs 5 and 6 as the PF6 anion
could occupy catalytically active sites on the Pd sur-
face.
1
mp 44–458C. H NMR (300 MHz, CD3OD): d=7.50 (d, J=
2.0 Hz, 1H), 7.48 (d, J=2.0 Hz, 1H), 6.08–5.95 (m, 1H),
5.30 (dd, J=46.5, 13.5 Hz, 2H), 4.84–4.78 (m, 2H), 3.83 (s,
3H), 2.60 (s, 3H); 13C NMR (75.5 MHz, CD3OD): d=146.3
(s), 132.0 (s), 123.8 (s), 122.4 (s), 121.9 (q, JC,F =318.6 Hz),
120.2 (s), 51.6 (s), 35.6 (s), 9.6 (s); ESI/MS(+): m/z=
137.1073, calcd. for [C8H13N2]+: 137.1075.
1-Allyl-2,3-dimethylimidazolium hexafluorophosphate: 1-
Allyl-2,3-dimethylimidazolium bromide (5.37 g, 24.7 mmol)
was dissolved in 25 mL H2O and NH4PF6 (4.43 g,
27.9 mmol) was added in small portions to the solution. The
Conclusions
PFIL-stabilized Pd NP suspensions were used as
active and reusable biphasic hydrogenation catalysts
for the reduction of olefins. Development of PFILs as
effective NP stabilizing ligands has extended the
classes of FILs that have been explored and provides
an alternative class of ligand that can be used to alle-
viate particle aggregation issues observed in the syn- mixture was stirred at room temperature for 18 h and ex-
Adv. Synth. Catal. 2011, 353, 3167 – 3177
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
3173

Kylie L. Luska and Audrey Moores
FULL PAPERS
tracted with CH2Cl2 (50 mL). The organic layer was separat-
ed, washed with H2O (3ꢃ50 mL) and dried over MgSO4.
Upon solvent removal, the product was dried under vacuum
for 18 h at 508C to afford an off-white solid; yield: 4.59 g
29.0 (s), 26.6 (s), 11.3 (s); ESI/MS(+): m/z=249.2330, calcd.
for [C16H29N2]+: 249.2325.
1-(Undec-10-enyl)-2,3-dimethylimidazolium bis(trifluoro-
methanesulfonyl)amide: 1-(Undec-10-enyl)-2,3-dimethylimi-
dazolium bromide (4.77 g, 14.5 mmol) was dissolved in H2O
(50 mL). Lithium bistrifluoromethanesulfonimidate (4.37 g,
15.2 mmol) was added in portions and the mixture was
stirred vigorously for 18 h. The mixture was extracted with
CH2Cl2 (50 mL), washed with H2O (3ꢃ50 mL) and dried
over MgSO4. Upon solvent removal, the product was dried
under vacuum to afford a yellow liquid; 6.43 g (84%).
1H NMR (300 MHz, CDCl3): d=7.16 (d, J=2.1 Hz, 1H),
7.12 (d, J=2.2 Hz, 1H), 5.76 (ddt, J=17.1, 10.2, 6.8, 1H),
5.01–4.80 (m, 2H), 3.98 (t, J=7.5 Hz, 2H), 3.74 (s, 3H), 2.54
(s, 3H), 1.99 (q, J=7.0 Hz, 2H), 1.73 (pentet, J=7.0 Hz,
2H), 1.48–1.09 (m, 12H); 13C NMR (75.5 MHz, CDCl3): d=
1
(66%); mp 39–408C. H NMR (300 MHz, CDCl3): d=7.14
(d, J=1.0 Hz, 1H), 7.12 (d, J=2.1 Hz, 1H), 5.95–5.70 (m,
1H), 5.16 (dd, J=37.0, 13.7 Hz, 2H), 4.57 (d, J=5.6 Hz,
2H), 3.65 (s, 3H), 2.43 (s, 3H); 13C NMR (75.5 MHz,
CDCl3): d=144.6 (s), 130.3 (s), 122.5 (s), 121.0 (s), 120.1 (s),
50.4 (s), 34.9 (s), 8.9 (s); ESI/MS(+); m/z=137.1074, calcd.
for [C8H13N2]+: 137.1073.
Undec-10-enyl trifluoromethanesulfonate: 10-Undecen-1-
ol (9.0 mL, 44.9 mmol) and pyridine (4.3 mL, 53.9 mmol)
were dissolved in 30 mL of dry CH2Cl2 and stirred for
15 min. Trifluoromethanesulfonic anhydride (8.3 mL,
49.4 mmol) was dissolved in 40 mL CH2Cl2 and cooled to
08C. The alcohol solution was added dropwise to the cooled
solution over 30 min. After stirring for 30 min at room tem-
perature, the mixture was washed with 1M HCl, water and
dried over MgSO4. The solvent was removed and the crude
product was purified on silica gel (5% EtOAc:hexanes) to
afford a pale yellow liquid; yield: 10.5 g (77%). 1H NMR
(300 MHz, CD3OD): d=5.79 (m, 1H), 4.93 (m, 2H), 4.51 (t,
J=6.5 Hz, 2H), 2.02 (q, J=7.1 Hz, 2H), 1.81 (quintet, J=
7.0 Hz, 2H), 1.33 (m, 12H); 13C NMR (75.5 MHz, CD3OD):
d=139.3 (s), 118.9 (q, JC,F =319.6 Hz), 114.4 (s), 78.0 (s),
34.0 (s), 29.5 (m, 3C), 29.2 (s), 29.1 (s), 29.0 (s), 25.3 (s).
1-(Undec-10-enyl)-2,3-dimethylimidazolium trifluorome-
thanesulfonate: 1,2-Dimethylimidazole (3.28 g, 34.1 mmol)
was dissolved in 30 mL of dry MeCN and cooled to 08C.
143.6 (s), 139.1 (s), 122.5 (s), 120.7 (s), 119.7 (q, JC,F
=
321.6 Hz), 114.1 (s), 48.8 (s), 35.2 (s), 33.7 (s), 29.5 (s), 29.2
(s), 29.2 (s), 29.0 (s), 28.9 (s), 28.8 (s), 26.2 (s), 9.5 (s); ESI/
MS(+): m/z=249.2329, calcd. for [C16H29N2]+: 249.2325.
1-(Undec-10-enyl)-2,3-dimethylimidazolium hexafluoro-
phosphate: 1-(Undec-10-enyl)-2,3-dimethylimidazolium hex-
afluorophosphate was prepared as outlined for 1-(undec-10-
enyl)-2,3-dimethylimidazolium bis(trifluoromethylsulfonyl)-
AHCTUNGTERNaGNUN mide employing NH4PF6 in place of LiNTF2 to afford an
off-white solid; yield: 3.22 g (86%). 1H NMR (300 MHz,
CDCl3): d=7.48 (d, J=2.1 Hz, 1H), 7.43 (d, J=2.1 Hz, 1H),
5.80 (ddt, J=17.0, 10.2, 6.7, 1H), 5.03–4.88 (m, 2H), 4.12 (t,
J=7.5 Hz, 2H), 3.80 (s, 3H), 2.60 (s, 3H), 2.04 (q, J=
6.9 Hz, 2H), 1.81 (pentet, J=6.0 Hz, 2H), 1.44–1.28 (m,
12H); 13C NMR (75.5 MHz, CDCl3): d=145.9 (s), 140.3 (s),
123.7 (s), 122.2 (s), 114.9 (s), 49.6 (s), 35.4 (s), 35.0 (s), 30.9
(s), 30.6 (s), 30.6 (s), 30.3 (s), 30.3 (s), 30.2 (s), 27.4 (s), 9.5
(s); ESI/MS(+): m/z=249.2324, calcd. for [C16H29N2]+:
249.2325.
Undec-10-enyl
trifluoromethanesulfonate
(10.3 g,
34.1 mmol) was dissolved in 20 mL of dry MeCN. This solu-
tion was added dropwise to the cooled imidazole solution
over 5 min. The mixture was allowed to stir at 08C for 1 h,
then warmed to room temperature and stirred for 18 h. The
solvent was removed and the product was dried under
vacuum to afford a pale yellow liquid; yield: 12.9 g (95%).
1H NMR (300 MHz, CD3OD): d=7.51 (d, J=2.1 Hz, 1H),
7.46 (d, J=2.1 Hz, 1H), 5.80 (m, 1H), 4.96 (m, 2H), 4.14 (t,
J=7.5 Hz, 2H), 3.81 (s, 3H), 2.62 (s, 3H), 2.04 (q, J=
6.6 Hz, 2H), 1.81 (quintet, J=7.1 Hz, 2H), 1.34 (m, 12H);
13C NMR (75.5 MHz, CD3OD): d=145.9 (s), 140.2 (s), 123.7
(s), 122.3 (s), 121.9 (q, JC,F =318.4 Hz), 114.9 (s), 49.6 (s),
35.5 (s), 30.9 (s), 30.6 (m, 2C), 30.3 (m, 2C), 30.2 (s), 27.5
(s), 9.6 (s); ESI/MS(+): m/z=249.2325, calcd. for
[C16H29N2]+: 249.2324.
1-(Undec-10-enyl)-2,3-dimethylimidazolium bromide: 1,2-
Dimethylimidazole (1.14 g, 11.9 mmol) was dissolved in de-
gassed CH3CN (12 mL) and 11-bromo-1-undecene (2.60 mL,
11.9 mmol) was added dropwise. The reaction mixture was
stirred at 708C for 18 h. The solvent was removed to yield
an off-white solid, which was suspended in Et2O (3ꢃ50 mL)
to wash the crude product. The product was collected and
dried under vacuum; yield: 3.18 g (81%); mp 74–758C.
1H NMR (300 MHz, CDCl3): d=7.70 (d, J=2.1 Hz, 1H),
7.41 (d, J=2.1 Hz, 1H), 5.77 (ddt, J=16.9, 10.1, 6.7, 1H),
5.00–4.86 (m, 2H), 4.16 (t, J=7.5 Hz, 2H), 4.01 (s, 3H), 2.79
(s, 3H), 2.00 (q, J=7.0 Hz, 2H), 1.79 (pentet, J=7.0 Hz,
2H), 1.38–1.19 (m, 12H); 13C NMR (75.5 MHz, CDCl3): d=
144.0 (s), 139.3 (s), 123.3 (s), 121.1 (s), 114.4 (s), 49.3 (s),
36.5 (s), 34.0 (s), 30.0 (s), 29.5 (s), 29.5 (s), 29.2 (s), 29.2 (s),
1-[11-(Diphenylphosphino)undecyl]-2,3-dimethylimidazo-
lium bis(trifluoromethanesulfonyl)amide (PFIL 2): 1-
(Undec-10-enyl)-2,3-dimethylimidazolium bis(trifluorome-
thylsulfonyl)amide (2.06 g, 3.88 mmol), diphenylphosphine
(1.45 g, 7.76 mmol) and 1,1’-azobis(cyclohexanecarbonitrile)
(ABCN) (20 mg, 0.08 mmol) were combined in a Schlenk
flask. The mixture was stirred at 808C for 72 h, during
which time ABCN (100 mg, 0.40 mmol) was recharged
twice. The volatiles were removed under vacuum and the re-
sulting oil was washed with Et2O (3ꢃ20 mL). The product
was dried under vacuum to afford a pale yellow oil; yield:
2.14 g (77%). 1H NMR (300 MHz, DMSO-d6): d=7.64 (d,
J=2.1 Hz, 1H), 7.61 (d, J=2.1 Hz, 1H), 7.44–7.30 (m,
10H), 4.08 (t, J=7.3 Hz, 2H), 3.74 (s, 3H), 2.57 (s, 3H),
2.12–1.97 (m, 2H), 1.81–1.57 (m, 2H), 1.47–1.09 (m, 16H);
13C NMR (75.5 MHz, DMSO-d6): d=144.2 (s), 138.7 (d,
J
C,P =13.9 Hz), 132.3 (d, JC,P =18.5 Hz), 128.5–128.4 (m, 2C),
122.3 (s), 120.8 (s), 119.5 (q, JC,F =322.1 Hz), 47.5 (s), 34.6
(s), 30.4 (d, JC,P =12.7 Hz), 29.2 (s), 28.9 (m, 3C), 28.7 (s),
28.5 (s), 26.7 (d, JC,P =11.1 Hz), 25.6 (s), 25.5 (d, JC,P
=
15.9 Hz), 9.1 (s); 31P NMR (81 MHz, DMSO-d6): d=À16.4
(s); ESI/MS(+): m/z=435.2930, calcd. for [C28H40N2P]+:
435.2924.
1-[3-(Diphenylphosphino)propyl]-2,3-dimethylimidazoli-
um trifluoromethanesulfonate (PFIL 3): PFIL 3 was synthe-
sized as outline for PFIL 2 to obtain a white powder; yield:
1
1.92 g (94%); mp 76–788C. H NMR (300 MHz, DMSO-d6):
3174
asc.wiley-vch.de
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2011, 353, 3167 – 3177

Improved Stability and Catalytic Activity of Palladium Nanoparticle Catalysts
d=7.64 (d, J=2.0 Hz, 1H), 7.60 (d, J=2.0 Hz, 1H), 7.39 (m,
10H), 4.21 (t, J=7.0 Hz, 2H), 3.72 (s, 3H), 2.54 (s, 3H),
2.10 (m, 2H), 1.76 (m, 2H). 13C NMR (75.5 MHz, DMSO-
d6): d=144.4 (s, 1C), 137.8 (d, JC,P =13.2 Hz, 1C), 132.4 (d,
Synthesis of Rhenium Complexes
ReBr(CO)3ACHTNUGRTENUNG(PFIL)2 complexes were synthesized by employ-
ing an analogous procedure to that reported by Storhoff and
Lewis.[26] This method involves the production of
J
C,P =19.8 Hz, 1C), 128.8 (m, 2C), 122.4 (s, 1C), 120.8 (s,
[ReBr(CO)3ACTHNUTRGNE(UNG THF)]2 by refluxing ReBr(CO)5 in THF, which
is subsequently reacted with 2.0 equiv. of the desired PFIL
in CH2Cl2.
1C), 48.1 (d, JC,P =15.5 Hz, 1C), 34.7 (s, 1C), 26.1 (d, JC,P
=
18.3 Hz, 1C), 22.9 (d, JC,P =12.4 Hz, 1C), 9.2 (s, 1C);
31P NMR (81 MHz, DMSO-d6): d=À16.4 (s); ESI/MS(+):
m/z=323.1672, calcd. for [C20H24N2P]+: 323.1666.
ReBr(CO)3ACHTNUTRGNEUNG(PFIL 1)2: ReBr(CO)5 (113 mg, 0.28 mmol)
was dissolved in THF (10 mL) and allowed to reflux for
18 h. After the reaction mixture had cooled to room temper-
ature, the solvent was removed under vacuum and the resi-
due dried for 1 h. 1-[3-(Diphenylphosphino)propyl]-2,3-di-
1-[11-(Diphenylphosphino)undecyl]-2,3-dimethylimidazo-
lium trifluoromethanesulfonate (PFIL 4): PFIL 4 was syn-
thesized as outlined for PFIL 2 to obtain a colourless oil;
1
yield: 92%. H NMR (300 MHz, DMSO-d6): d=7.63 (d, J=
methylimidazolium
bis(trifluoromethanesulfonyl)amide
2.1 Hz, 2H), 7.60 (d, J=2.1 Hz, 2H), 7.37 (m, 10H), 4.08 (t,
J=7.3 Hz, 2H), 3.73 (s, 3H), 2.56 (s, 3H), 2.04 (t, J=6.6 Hz,
2H), 1.69 (quintet, J=6.6 Hz, 2H), 1.47–1.03 (m, 16H);
13C NMR (75.5 MHz, DMSO-d6): d=144.9 (s, 1C), 139.3 (d,
(344 mg, 0.57 mmol) was dissolved in degassed CH2Cl2
(10 mL) and added to the residual solids. The solution was
refluxed for 4 h. After the reaction mixture had cooled to
room temperature, pentane was added until the product pre-
cipitated from solution. The white powder was collected,
washed with pentane (3ꢃ5 mL) and dried under vacuum;
J
C,P =14.6 Hz, 1C), 133.0 (d, JC,P =18.6 Hz, 1C), 129.2 (m,
2C), 123.0 (s, 1C), 121.5 (s, 1C), 121.4 (q, JC,F =321.9 Hz,
1C), 48.2 (s, 1C), 35.3 (s, 1C), 31.1 (d, JC,P =12.5 Hz, 1C),
29.9 (s, 1C), 29.8–29.4 (m, 3C), 29.3 (s, 1C), 29.2 (s, 1C),
27.5 (d, JC,P =11.4 Hz, 1C), 26.5–26.0 (m, 2C), 9.8 (s, 1C);
31P NMR (81 MHz, DMSO-d6): d=À16.4 (s); ESI/MS(+):
m/z=435.2924, calcd. for [C28H40N2P]+: 435.2917.
1
yield: 245 mg (56%); mp 84–858C. H NMR (300 MHz, ace-
tone-d6): d=7.65–7.27 (m, 24H), 4.28 (t, J=7.3 Hz, 4H),
3.91 (s, 6H), 2.65 (dm, JH,P =102 Hz, 4H), 2.65 (s, 6H), 1.68
(dm, JH,P =66 Hz, 4H); 13C NMR (75.5 MHz, acetone-d6):
d=145.7 (s, 1C), 133.6 (t, JC,P =4.5 Hz, 1C), 133.4 (t, JC,P
=
1-[3-(Diphenylphosphino)propyl]-2,3-dimethylimidazoli-
um hexafluorophosphate (PFIL 5): PFIL 5 was synthesized
as outlined for PFIL 2 to obtain a white solid; yield: 100%;
4.5 Hz, 1C), 131.5 (d, J=12.1 Hz, 1C), 129.7 (m, 1C), 123.5
(s, 1C), 122.1 (s, 1C), 49.2 (t, JC,P =7.2 Hz, 1C), 35.6 (s, 1C),
25.6 (s, 1C), 24.2 (t, JC,P =14.3 Hz, 1C), 9.9 (s, 1C);
31P NMR: (81 MHz, DMSO-d6): d=À8.5 (s); IR: (ATR):
1
mp 112–1138C. H NMR (300 MHz, DMSO-d6): d=7.61 (d,
J=2.0 Hz, 1H), 7.56 (d, J=2.0 Hz, 1H), 7.43–7.31 (m,
10H), 4.18 (t, J=7.0 Hz, 2H), 3.70 (s, 3H), 2.50 (s, 3H),
2.14–2.04 (m, 2H), 1.84–1.66 (m, 2H); 13C NMR (75.5 MHz,
DMSO-d6): d=144.4 (s, 1C), 137.8 (d, JC,P =13.3 Hz, 1C),
132.4 (d, JC,P =18.7 Hz, 1C), 129.1–128.5 (m, 2C), 122.4 (s,
1C), 120.8 (s, 1C), 48.2 (d, JC,P =15.0 Hz, 1C), 34.6 (s, 1C),
26.1 (d, JC,P =18.2 Hz, 1C), 23.0 (d, JCP =11.6 Hz, 1C), 9.2 (s,
1C); 31P NMR (81 MHz, DMSO-d6): d=À16.4 (s); ESI/
MS(+): m/z=323.1675, calcd. for [C20H24N2P]+: 323.1672.
1-[11-(Diphenylphosphino)undecyl]-2,3-dimethylimidazo-
lium hexafluorophosphate (PFIL 6): PFIL 5 was synthesized
as outlined for PFIL 2 to obtain a pale yellow oil; yield:
100%. 1H NMR (300 MHz, DMSO-d6): d=7.64 (d, J=
2.1 Hz, 1H), 7.61 (d, J=2.1 Hz, 1H), 7.45–7.31 (m, 10H),
4.08 (t, J=7.3 Hz, 2H), 3.74 (s, 3H), 2.56 (s, 3H), 2.05 (t,
J=7.5 Hz, 2H), 1.75–1.60 (m, 2H), 1.48–1.14 (m, 16H);
13C NMR (75.5 MHz, DMSO-d6): d=144.2 (s, 1C), 138.6 (d,
vco =2029, 1946, 1901 cmÀ1
; ESI/MS(+): m/z=497.0958,
calcd. for [C43H48O3N4BrP2Re]2+: 497.0952.
ReBr(CO)3A(PFIL 3)2: ReBr(CO)3A(PFIL 3)2 was synthe-
sized as outlined for ReBr(CO)3A(PFIL 1)2 to obtain an off-
G
CHTUNGTRENNUNG
CTHUNGTRENNUNG
1
white solid; yield: 65%; mp 134–1358C. H NMR (300 MHz,
acetone-d6): d=7.75–7.26 (m, 24H), 4.27 (t, J=7.3 Hz, 4H),
3.89 (s, 6H), 2.68 (dm, JH,P =84 Hz, 4H), 2.60 (s, 6H), 1.63
(m, 4H); 13C NMR (75.5 MHz, acetone-d6): d=145.6 (s,
1C), 133.8 (t, JC,P =4.5 Hz, 1C), 133.3 (t, JC,P =4.5 Hz, 1C),
131.4 (d, J=14.3 Hz, 1C), 129.6 (dt, JC,P =12.8, 4.5 Hz, 1C),
123.4 (s, 1C), 122.1 (s, 1C), 48.9 (t, JC,P =7.2 Hz, 1C), 35.5
(s, 1C), 25.5 (s, 1C), 23.5 (t, JC,P =13.6 Hz, 1C), 10.0 (s, 1C);
31P NMR: (81 MHz, DMSO-d6): d=À7.7 (s); IR: (ATR)
vco =2028, 1945, 1900 cmÀ1
;
ESI/MS(+): m/z=497.0957,
calcd. for [C43H48O3N4BrP2Re]2+: 497.0952.
ReBr(CO)3A(PFIL 5)2: ReBr(CO)3A(PFIL 5)2 was synthe-
sized as outlined for ReBr(CO)3A(PFIL 1)2 to obtain a white
G
CHTUNGTRENNUNG
CTHUNGTRENNUNG
J
C,P =13.9 Hz, 1C), 132.4 (d, JC,P =18.4 Hz, 1C), 128.7–128.3
1
solid; yield: 48%; mp 186–1878C. H NMR (300 MHz, ace-
tone-d6): d=7.62–7.26 (m, 24H), 4.26 (t, J=7.3 Hz, 4H),
3.90 (s, 6H), 2.65 (dm, JH,P =105 Hz, 4H), 2.63 (s, 6H), 1.67
(dm, JH,P =57 Hz, 4H); 13C NMR (75.5 MHz, acetone-d6):
(m, 2C), 122.3 (s, 1C), 120.8 (s, 1C), 47.5 (s, 1C), 34.7 (s,
1C), 30.4 (d, JC,P =12.8 Hz, 1C), 29.2 (s, 1C), 28.9 (m, 3C),
28.6 (s, 1C), 28.5 (s, 1C), 26.7 (d, JC,P =11.1 Hz, 1C), 25.6–
25.3 (m, 2C), 9.1 (s, 1C); 31P NMR (81 MHz, DMSO-d6):
d=À16.4 (s); ESI/MS(+): m/z=435.2921, calcd. for
[C28H40N2P]+: 435.2924.
d=144.8 (s, 1C), 132.8 (t, JC,P =4.5 Hz, 1C), 132.4 (t, JC,P
=
4.5 Hz, 1C), 130.5 (d, J=14.3 Hz, 1C), 128.7 (m, 1C), 122.6
(s, 1C), 121.1 (s, 1C), 48.1 (t, JC,P =7.6 Hz, 1C), 34.6 (s, 1C),
24.6 (s, 1C), 22.9 (t, JC,P =13.6 Hz, 1C), 8.9 (s, 1C);
31P NMR (81 MHz, acetone-d6): d=À8.2 (s); IR (ATR)
Synthesis of Phosphine Selenides
vco =2022, 1935, 1912 cmÀ1
; ESI/MS(+): m/z=497.0955,
PFIL (0.12 mmol) and excess selenium powder (40 mg)
were dissolved in DMSO-d6 (1.0 g) and stirred overnight
under an inert environment. The mixture was filtered
through celite and analyzed by 31P NMR.
calcd. for [C43H48O3N4BrP2Re]2+: 497.0952.
ReBr(CO)3A(PFIL 6)2: ReBr(CO)3A(PFIL 6)2 was synthe-
3A
G
CHTUNGTRENNUNG
sized as outlined for ReBr(CO) CHTUNGTRENN(UNG PFIL 1)2 to obtain a white
solid; yield: 38%; mp 124–1258C. 1H NMR (300 MHz,
CD3CN): d=7.54–7.31 (m, 20H), 7.27–7.19 (m, 4H), 4.00 (t,
J=7.5 Hz, 4H), 3.68 (s, 6H), 2.48 (s, 6H), 2.25 (dm, JH,P
=
186 Hz, 4H), 1.73 (m, 4H), 1.42–0.87 (m, 32H); 13C NMR
Adv. Synth. Catal. 2011, 353, 3167 – 3177
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
3175

Kylie L. Luska and Audrey Moores
FULL PAPERS
(75.5 MHz, CD3CN): d=145.4 (s, 1C), 134.1 (t, JC,P =4.5 Hz,
1C), 133.8 (t, JC,P =4.5 Hz, 1C), 131.3 (d, J=7.6 Hz, 1C),
129.4 (dt, JC,P =12.8, 4.5 Hz, 1C), 123.2 (s, 1C), 121.8 (s,
1C), 49.2 (s, 1C), 35.8 (s, 1C), 31.3 (t, JC,P =6.0 Hz, 1C),
30.3 (s, 1C), 30.1 (s, 2C), 29.9 (s, 1C), 29.7 (s, 1C), 29.6 (s,
1C), 27.2 (t, JC,P =13.6 Hz, 1C), 26.8 (s, 1C), 24.7 (s, 1C),
10.1 (s, 1C); 31P NMR (81 MHz, DMSO-d6): d=À8.5 (s); IR
Supporting Information
The Supporting Information contains additional catalytic hy-
drogenation, TEM, XPS and NMR data for PFIL-stabilized
Pd NPs, NMR data for PFIL phosphine selenides, NMR and
HR-MS data for the synthesis of PFILs.
(ATR): vco =2023, 1938, 1897 cmÀ1
; ESI/MS(+): m/z=
609.2208, calcd. for [C59H80O3N4BrP2Re]2+: 609.2204.
Acknowledgements
Synthesis of Palladium Nanoparticles
We thank the Natural Science and Engineering Research
Council of Canada (NSERC), the Canada Foundation for In-
novation (CFI), the Canada Research Chairs (CRC), the
Fonds de Recherche sur la Nature et les Technologies
(FQRNT), the Center for Green Chemistry and Catalysis
(CGCC) and McGill University for their financial support.
We also thank Dr. Alain LeSimple and Nadim Saade
(McGill University) for HR-MS analysis and Dr. Suzie
Poulin (ꢀcole Polytechnique de Montrꢁal) for XPS measure-
ments.
In a typical experiment, PdACHTNUTRGENNG(U acac)2 (30 mg, 0.10 mmol), PFIL
1–6 (0.10 mmol) and [BDMI]X (2 mL) (BDMI=1-butyl-2,3-
À
dimethylimidazolium; XÀ =Tf2NÀ, TfOÀ, PF6 ) were com-
bined in a glovebox and placed in a high-pressure reactor.
After stirring the mixture at 808C under an atmosphere of
argon for 45 min, a constant pressure of H2(g) (4 bar) was ad-
mitted to the system and and the content was stirred for 4 h
at 808C. The Pd NPs embedded in ionic liquid were em-
ployed for hydrogenation studies and TEM analysis. Isola-
tion of the Pd NPs for XPS analysis was achieved by dissolv-
ing the mixture in acetone (10 mL), centrifuging (7500 rpm
for 10 min), washing with acetone:pentane (2:1) (2ꢃ10 mL)
and drying under vacuum.
References
[1] a) M. Scariot, D. Silva, J. Scholten, G. Machado, S.
Teixeira, M. Novak, G. Ebeling, J. Dupont, Angew.
Chem. 2008, 120, 9215–9218; Angew. Chem. Int. Ed.
2008, 47, 9075–9078; b) T. Kim, W. Kim, S. Hong, J.
Kim, K. Suh, Angew. Chem. 2009, 121, 3864–3867;
Angew. Chem. Int. Ed. 2009, 48, 3806–3809; c) E.
Redel, R. Thomann, C. Janiak, Inorg. Chem. 2008, 47,
14–16; d) J. Dupont, G. S. Fonseca, A. P. Umpierre,
P. F. P. Fichtner, S. R. Teixeira, J. Am. Chem. Soc. 2002,
124, 4228–4229; e) B. Lꢄger, A. Denicourt-Nowicki, H.
Olivier-Bourbigou, A. Roucoux, Inorg. Chem. 2008, 47,
9090–9096.
Hydrogenation Studies
In a typical experiment, the synthesized Pd NPs embedded
in [BDMI]X were placed in a high-pressure reactor with sty-
rene (2.9 mL, 25 mmol). After evacuating and backfilling
the vessel, the mixture was stirred at 508C for 1.5 h under a
constant pressure of H2(g) (4 bar). The mixture was cooled to
room temperature and extracted with pentane (3ꢃ8 mL).
The organic phase was analyzed by gas chromatography em-
ploying dodecane as an internal standard and the IL phase
was dried under vacuum for at least 1 h. Recycling experi-
ments were performed with the dried IL phase under the re-
action conditions outline above. The IL phase was stored
under vacuum between catalytic cycles.
[2] J. Dupont, J. D. Scholten, Chem. Soc. Rev. 2010, 39,
1780–1804.
[3] a) T. Gutel, C. C. Santini, K. Philippot, A. Padua, K.
Pelzer, B. Chaudret, Y. Chauvin, J. M. Basset, J. Mater.
Chem. 2009, 19, 3624–3631; b) P. Migowski, G. Macha-
do, S. Texeira, M. Alves, J. Morais, A. Traverse, J.
Dupont, Phys. Chem. Chem. Phys. 2007, 9, 4814–4821.
[4] a) D. O. Silva, J. D. Scholten, M. A. Gelesky, S. R. Teix-
eira, A. C. B. Dos Santos, E. F. Souza-Aguiar, J.
Dupont, ChemSusChem 2008, 1, 291–294; b) E. Redel,
R. Thomann, C. Janiak, Chem. Commun. 2008, 1789–
1791; c) M. H. G. Prechtl, M. Scariot, J. D. Scholten, G.
Machado, S. R. Teixeira, J. Dupont, Inorg. Chem. 2008,
47, 8995–9001.
[5] a) C. Vollmer, E. Redel, K. Abu-Shandi, R. Thomann,
H. Manyar, C. Hardacre, C. Janiak, Chem. Eur. J. 2010,
16, 3849–3858; b) A. Umpierre, G. Machado, G.
Fecher, J. Morais, J. Dupont, Adv. Synth. Catal. 2005,
347, 1404–1412.
Carbon Disulfide Poisoning Tests
Catalyst poisoning tests employing CS2 were performed as a
typical hydrogenation experiment with the addition of CS2
(0.05 mmol, 0.5 equiv.) to the IL:styrene mixture prior to
start of the experiment.
Transmission Electron Microscopy
TEM images of Pd NPs were obtained on a Philips CM 200
microscope operating at an accelerating voltage of 200 kV
with a point resolution of 0.24 nm. Sample preparation in-
cluded dilution of 3 drops of the Pd NP solution in ~1.5 mL
of MeCN, followed by deposition on a carbon-coated
copper grid (400 mesh) at room temperature. The NP size
distribution was determined from the measurement of >100
spherical particles chosen in arbitrary areas of enlarged mi-
crographs.
[6] B. Lꢄger, A. Denicourt-Nowicki, H. Olivier-Bourbigou,
A. Roucoux, Tetrahedron Lett. 2009, 50, 6531–6533.
[7] a) B. Lꢄger, A. Denicourt-Nowicki, A. Roucoux, H.
Olivier-Bourbigou, Adv. Synth. Catal. 2008, 350, 153–
159; b) B. Lꢄger, A. Denicourt-Nowicki, H. Olivier-
Bourbigou, A. Roucoux, ChemSusChem 2008, 1, 984–
3176
asc.wiley-vch.de
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2011, 353, 3167 – 3177

Improved Stability and Catalytic Activity of Palladium Nanoparticle Catalysts
987; c) R. R. Dykeman, N. Yan, R. Scopelliti, P. J.
Dyson, Inorg. Chem. 2011, 50, 717–719; d) Y. Hu, Y. Y.
Yu, Z. S. Hou, H. Li, X. G. Zhao, B. Feng, Adv. Synth.
Catal. 2008, 350, 2077–2085.
H. Lee, Langmuir 2004, 20, 556–560; c) H. Itoh, K.
Naka, Y. Chujo, J. Am. Chem. Soc. 2004, 126, 3026–
3027; d) K. L. Luska, A. Moores, Can. J. Chem. 2011,
DOI: 10.1039/v11-146.
[8] J. Huang, T. Jiang, B. X. Han, H. X. Gao, Y. H. Chang,
G. Y. Zhao, W. Z. Wu, Chem. Commun. 2003, 1654–
1655.
[9] Y. A. Kou, X. D. Mu, D. G. Evans, Catal. Lett. 2004, 97,
151–154.
[10] X. D. Mu, J. Q. Meng, Z. C. Li, Y. Kou, J. Am. Chem.
Soc. 2005, 127, 9694–9695.
[17] C. S. Consorti, G. L. P. Aydos, G. Ebeling, J. Dupont,
Org. Lett. 2008, 10, 237–240.
[18] a) L. Ott, R. Finke, Coord. Chem. Rev. 2007, 251, 1075–
1100; b) J. D. Scholten, G. Ebeling, J. Dupont, Dalton
Trans. 2007, 5554–5560; c) L. S. Ott, S. Campbell, K. R.
Seddon, R. G. Finke, Inorg. Chem. 2007, 46, 10335–
10344.
[11] M. A. Gelesky, S. S. X. Chiaro, F. A. Pavan, J. H. Z. dos
Santos, J. Dupont, Dalton Trans. 2007, 5549–5553.
[12] a) F. Yan, L. H. Qiu, B. Q. Liu, Y. J. Peng, Chem.
Commun. 2011, 47, 2934–2936; b) H. Zhang, H. Cui,
Langmuir 2009, 25, 2604–2612; c) Y. S. Chun, J. Y. Shin,
C. E. Song, S. G. Lee, Chem. Commun. 2008, 942–944.
[13] Z. Wang, Q. Zhang, D. Kuehner, A. Ivaska, L. Niu,
Green Chem. 2008, 10, 907–909.
[14] M. Crespo-Quesada, R. R. Dykeman, G. Laurenczy,
P. J. Dyson, L. Kiwi-Minsker, J. Catal. 2011, 279, 66–74.
[15] a) D. B. Zhao, Z. F. Fei, T. J. Geldbach, R. Scopelliti,
P. J. Dyson, J. Am. Chem. Soc. 2004, 126, 15876–15882;
b) Z. F. Fei, D. B. Zhao, D. Pieraccini, W. H. Ang, T. J.
Geldbach, R. Scopelliti, C. Chiappe, P. J. Dyson, Orga-
nometallics 2007, 26, 1588–1598; c) M. H. G. Prechtl,
J. D. Scholten, J. Dupont, J. Mol. Catal. A: Chem. 2009,
313, 74–78.
[19] K. Sawai, R. Tatumi, T. Nakahodo, H. Fujihara, Angew.
Chem. 2008, 120, 7023–7025; Angew. Chem. Int. Ed.
2008, 47, 6917–6919.
[20] G. Fonseca, G. Machado, S. Teixeira, G. H. Fecher, J.
Morais, M. Alves, J. Dupont, J. Colloid Interface Sci.
2006, 301, 193–201.
[21] R. A. Komoroski, A. J. Magistro, P. P. Nicholas, Inorg.
Chem. 1986, 25, 3917–3925.
[22] E. Redel, J. Kramer, R. Thomann, C. Janiak, J. Orga-
nomet. Chem. 2009, 694, 1069–1075.
[23] P. Wasserscheid, T. Welton, Ionic Liquids in Synthesis,
2nd edn., Wiley-VCH, Weinheim, 2008.
[24] a) K. H. Shaughnessy, Chem. Rev. 2009, 109, 643–710;
b) D. J. Snelders, G. van Koten, R. J. Klein Gebbink,
Chem. Eur. J. 2011, 17, 42–57.
[25] a) D. W. Allen, I. W. Nowell, B. F. Taylor, J. Chem. Soc.
Dalton Trans. 1985, 2505–2508; b) D. W. Allen, B. F.
Taylor, J. Chem. Soc. Dalton Trans. 1982, 51–54.
[26] B. N. Storhoff, H. C. Lewis, Synth. React. Inorg. Met.-
Org. Chem. 1974, 4, 467–475.
[16] a) N. Kocharova, J. Leiro, J. Lukkari, M. Heinonen, T.
Skala, F. Sutara, M. Skoda, M. Vondracek, Langmuir
2008, 24, 3235–3242; b) K. Kim, D. Demberelnyamba,
Adv. Synth. Catal. 2011, 353, 3167 – 3177
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
3177