Aqueous-phase biphasic dehydroaromatization of bio-derived limonene into p-cymene by soluble Pd nanocluster catalysts

Article abstract of DOI:10.1016/j.jcat.2008.01.003

A biphasic approach to the dehydroaromatization of bioderived limonene into water-insoluble p-cymene using soluble Pd nanoparticle catalysts in an aqueous phase (≥150 °C, 2 bar H₂) was successfully achieved with a conversion of 93% and a selectivity of 82%. The Pd nanoparticles, operating under forcing conditions (180 °C, 2 bar H₂), can be recycled at least four times without noticeable degradation. The effects of temperature, pressure, reaction time, pH, catalyst concentration, metal type, the type and amount of polymer stabilizer, and the preparation method were systematically investigated to optimize the process and provide insight into the mechanisms involved.

Full text of DOI:10.1016/j.jcat.2008.01.003

Journal of Catalysis 254 (2008) 244–250
www.elsevier.com/locate/jcat
Aqueous-phase biphasic dehydroaromatization of bio-derived limonene
into p-cymene by soluble Pd nanocluster catalysts
Chen Zhao a, Weijia Gan a, Xiaobing Fan a, Zhipeng Cai a, Paul J. Dyson b,∗, Yuan Kou a,∗
a
PKU Green Chemistry Center, Beijing National Laboratory for Molecular Sciences, College of Chemistry and Molecular Engineering, Peking University,
Beijing 100871, China
b
Institut des Sciences et Ingénierie Chimiques, Ecole Polytechnique Fédérale de Lausanne (EPFL), CH-1015 Lausanne, Switzerland
Received 19 October 2007; revised 24 December 2007; accepted 2 January 2008
Available online 31 January 2008
Abstract
A biphasic approach to the dehydroaromatization of bioderived limonene into water-insoluble p-cymene using soluble Pd nanoparticle catalysts
◦
in an aqueous phase (ꢀ150 C, 2 bar H ) was successfully achieved with a conversion of 93% and a selectivity of 82%. The Pd nanoparticles,
2
◦
operating under forcing conditions (180 C, 2 bar H ), can be recycled at least four times without noticeable degradation. The effects of tempera-
2
ture, pressure, reaction time, pH, catalyst concentration, metal type, the type and amount of polymer stabilizer, and the preparation method were
systematically investigated to optimize the process and provide insight into the mechanisms involved.
© 2008 Elsevier Inc. All rights reserved.
Keywords: Limonene; p-Cymene; Dehydroaromatization; Palladium; Nanoparticles; Biphasic catalysis; Terpene; Poly(N-vinyl-2-pyrrolidone); Biomass
1. Introduction
Pd/Al2O3 [8], Pt/Al2O3 [9], Pt/C [10], Cu–Ni [11], and het-
eropolyacid H5PMo10V2O40 [12], has been investigated in de-
tail. From these studies, it appears that palladium-based cata-
lysts are the most efficient; for example, selectivities of 70%
over Pd/C [8] and 90% over Pd/SiO2 were obtained using H2
as carrier gas on a fixed-bed reactor [4,5].
The conversion of natural biomass materials (e.g., terpenes)
to value-added chemical and fine chemical products, as recently
described by Corma et al. [1], is highly important, and such
conversion via green catalytic processes remains a considerable
challenge. Limonene is a waste product obtained in vast quanti-
ties from the fruit juice industry. Because limonene is a fragrant
active mono-terpene and renewable feedstock, its conversion to
the valuable aromatic compound p-cymene is of great commer-
cial interest [2,3].
The dehydroaromatization of limonene to p-cymene under a
hydrogen atmosphere (Scheme 1) comprises a relatively com-
plex series of reactions. Hydrogenation, dehydrogenation, and
isomerization reactions occur simultaneously, making it diffi-
cult to obtain a single product with high selectivity. The de-
hydroaromatization of limonene to p-cymene over heteroge-
neous catalysts, such as Pd/SiO2 [4,5], Pd/zeolite [6,7], Pd/C,
Environmentally benign water-based systems using metallic
nanoparticles as catalysts have not yet been reported. Aqueous–
organic biphasic systems offer certain advantages for the sep-
aration of products, such as water-insoluble p-cymene, from
hydrophilic catalysts [13,14]. Immobilization of a nanoparti-
cle catalyst in water rather than on a surface allows all of the
surface atoms of the nanoparticle to may participate in cataly-
sis, thereby permitting a traditional gas–solid phase heteroge-
neous catalytic process at the gas–liquid–solid interface [15].
Aqueous-phase processes also facilitate the controlled sup-
ply or dissipation of heat in reactions. In previous work, we
found that soluble nanoparticle catalysts exhibited high cat-
alytic activity under mild conditions; examples include Ru
nanoparticles catalyzing the selective hydrogenation of cel-
lobiose in water [16], Rh nanoparticles catalyzing the hydro-
genation of benzene [17] and arenes [18] in ionic liquids,
Pt nanoparticles selectively catalyzing the hydrogenation of
*
Corresponding authors. Faxes: +41 21 693 98 85, +86 10 62751708.
E-mail addresses: paul.dyson@epfl.ch (P.J. Dyson), yuankou@pku.edu.cn
(Y. Kou).
0021-9517/$ – see front matter © 2008 Elsevier Inc. All rights reserved.
doi:10.1016/j.jcat.2008.01.003

C. Zhao et al. / Journal of Catalysis 254 (2008) 244–250
245
Scheme 1. Dehydroaromatization of limonene to p-cymene under a hydrogen atmosphere.
o-chloronitrobenzene [19] and cinnamaldehyde [20] in ionic
2.2. Catalyst preparation
liquids, and Pt nanoparticles catalyzing the aerobic oxidation
of alcohols in aqueous–organic biphases [21]. Among these,
aqueous–organic [21] and ionic liquid–organic systems [17–20]
have shown great potential in reaction control and product sep-
aration.
In general, however, soluble nanoparticles are not thermally
stable at higher temperatures [22]. The highest working temper-
ature reported to date is 160 ◦C for Pd-catalyzed Heck coupling
reactions [23]. Herein we report for the first time that soluble
Pd nanoparticles can efficiently catalyze the dehydroaromati-
zation of limonene to p-cymene at 180 ◦C under 2 bar H2
in an aqueous solution. After reaction, the p-cymene and the
nanoparticle-containing aqueous solution are separated into two
phases. The catalyst may be reused several times without de-
activation after reaction at 180 ◦C. We also characterize the
relationship between the nanoparticles’ structure and their ac-
tivity and stability.
2.2.1. Nanoparticle synthesis using ethanol–water
reduction [25]
Aqueous solutions of PVP-metal nanoparticles were pre-
pared from the appropriate metal salt and PVP-K15, -K30,
-K60, or -K90. In a typical reaction, PdCl2 was converted
to H2PdCl4–H2O using hydrochloric acid, after which the
H2PdCl4–H2O solution (CPd = 0.0221 mol/L, 1.4 mL,
0.03 mmol) and PVP (0.062 g, 0.6 mmol) were added to a
mixture of H2O (30 mL) and ethanol (30 mL). The solution
was stirred under reflux at 80 ◦C for 2 h, during which time it
turned black. The solvent was removed in vacuo, and distilled
water (30 mL) was added to form a Pd nanoparticle solution
in water (CPd = 10−3 mol/L). Other metal nanoparticles were
prepared in the same way but using different precursors (i.e.,
RuCl3, H2PtCl6, RhCl3 and HAuCl4).
2.2.2. Nanoparticle synthesis using H2 reduction
Typically, an H2PdCl4–H2O solution (CPd = 0.0221 mol/L,
1.4 mL, 0.03 mmol) and PVP (0.124 g, 1.2 mmol) were mixed
in H2O (30 mL). The solution was reduced in an autoclave un-
der H2 (40 bar) at 100 ◦C for 3 h to form a Pd nanoparticle
solution.
2. Experimental
2.1. General procedures
Limonene (boiling point, 176–178 ◦C; purity, 98% [GC])
was provided by Nanjing Huagong Co, Ltd., China. RuCl3,
H2PtCl6, PdCl2, RhCl3, and HAuCl4 (all AR grade) were ob-
tained from Shenyang Research Institute of Nonferrous Met-
als, China. Poly(N-vinyl-2-pyrrolidone) (PVP) with various
alkyl chain lengths (i.e., K15, K30, K60, K90; average Mw =
10,000, 30,000, 220,000, and 630,000, respectively) were pur-
chased from J & K Chemicals Ltd. N,N-dimethyl-N-cetyl-N-
(2-hydroxyethyl)-ammonium bromide was synthesized as de-
scribed previously [24]. Ethanol (GR grade) was refluxed with
state agent and distilled under nitrogen (99.9995%) before use.
All other chemicals were used as received without further pu-
rification.
Gas chromatography (GC) analysis was performed using an
Agilent 6820 gas chromatograph with a flame ionization detec-
tor equipped with a 30 m (0.25-mm-i.d) HP Innowax column.
GC–mass spectroscopy (GC–MS) analysis was performed us-
ing an Agilent GC 6890 with an Agilent 5973 inevt mass selec-
tive detector.
2.2.3. Nanoparticle synthesis using NaBH4 reduction [24]
PVP (0.670 g, 6.6 mmol) or N,N-dimethyl-N-cetyl-N-
(2-hydroxyethyl)-ammonium bromide (0.260 g, 0.66 mmol)
was dissolved in H2O (45 mL), and then sodium borohydride
(36 mg, 0.95 mmol) was added. This mixture was added, with
vigorous agitation, to an aqueous solution of H2PdCl4 (CPd
0.0221 mol/L, 15 mL) to give Pd nanoparticles.
=
2.3. Limonene conversion using the nanoparticle solutions
In a typical reaction, limonene (1.36 g, 0.01 mol) and the
metal nanoparticles in water (10 mL, 1×10−3 mol) were added
to an autoclave (reactor volume, 100 mL). The reaction was
carried out under 2 bar H2 at 150 ◦C for 3 h at a stirring speed
of 500 rpm. After reaction, the products and the nanoparticle
solution were separated into two phases, and the upper organic
product layer was collected and analyzed by GC and GC–MS.

246
C. Zhao et al. / Journal of Catalysis 254 (2008) 244–250
2.4. Catalyst recycling experiments
a pronounced effect on the reaction. Isomerization and dehy-
drogenation reactions are endothermic, and, as expected, the
selectivity toward p-cymene increased with temperature; for
example, a selectivity of 9.3% was observed at 120 ◦C, increas-
ing to 33% at 150 ◦C over the same Pd nanoparticle catalyst.
The selectivity of p-cymene should be sensitive to hydrogen
pressure, because 1 mol of hydrogen is released during the de-
hydrogenation of limonene. Preliminary results showed that the
selectivity increased from 4.2 to 58% as the hydrogen pres-
sure was reduced from 40 to 0 bar; however, Pd nanoparticles
were not stable in an N2 atmosphere without H2, indicating that
the H2 atmosphere helps stabilize the Pd nanocatalysts, espe-
cially at higher temperatures. In addition, under lower pressure
of H2 (2 bar), the reaction reached equilibrium in 2–3 h, and a
turnover frequency (TOF) > 700 h−1 was maintained.
For the Pd nanoparticle catalysts operating in aqueous solu-
tions, acidic (pH 2) or neutral (pH 7) solutions gave the best
activity. Decreasing the pH to 0 not only eroded the stainless
steel autoclave, but also led to low activity and selectivity. Al-
though basic catalysts, such as CaO, can selectively catalyze
limonene to p-cymene [29], high-pH (i.e., pH 10, 12, and 14)
solutions containing the nanoparticle catalysts produced con-
versions of only ca. 30%.
The Pd nanoparticle solution (60 mL, 5×10−3 mol), synthe-
sized by ethanol–water reduction and stabilized by PVP-K90,
was mixed with limonene (2.04 g, 0.015 mol). This mixture was
added to the autoclave and reacted at 180 ◦C under 2 bar H2 for
3 h. After reaction, the upper organic layer was decanted and
analyzed by GC and GC–MS, and the recovered Pd nanoparti-
cles were reused with no further treatment.
2.5. Determination of Pd leaching by inductively coupled
plasma-atomic emission spectroscopy
The Pd content in the organic phase was determined by in-
ductively coupled plasma-atomic emission spectroscopy (ICP-
AES). The organic phase was removed under vacuum, and
HNO3 (10 mL) and HCl (30 mL) were added to the residue.
The mixture was heated at 100 ◦C for 12 h until the Pd metal
was thoroughly dissolved. The resulting transparent solution
was diluted to 500 mL with deionized water and analyzed
by ICP-AES (Profile Spec, Leeman Labs; detection limit,
1 µg–mg/mL).
2.6. Nanoparticle characterization by transmission electron
microscopy
A Pd concentration of 10−3 mol/L was generally used in
this work because this concentration yielded good selectivity
with a high conversion at 150 ◦C. In principle, more dilute Pd
solutions should facilitate control of the nanoparticle assembly
process, but at the expense of catalytic activity. Three nanopar-
ticle solutions with different Pd concentrations were evaluated,
as shown in Table 1 (entries 1–3). Increasing the concentration
from 0.001 to 0.005 to 0.01 mol/L led to slightly increased con-
version but significantly increased selectivity. The most concen-
trated Pd nanoparticle solution (0.01 mol/L; entry 3) provided a
selectivity of 78% with a near-quantitative conversion, although
Transmission electron microscopy (TEM) was performed
with a Hitachi H-9000 HRTEM at 300 keV. The nanoparticles
were diluted in methanol, and one drop of solution was placed
on a copper grid coated with carbon film. To determine the size
distribution, more than 300 particles were counted from each
sample.
3. Results and discussion
the reaction rate (TOF) was reduced to 96 h−1
.
3.1. Dehydroaromatization of limonene under different
conditions
Table 1
Conversion of limonene using soluble Pd nanoparticles under different condi-
tions
To find a suitable catalyst for the conversion of limonene
to p-cymene, nanoparticle catalysts composed of five different
transition metals (i.e., Pd, Rh, Ru, Pt, and Au) were prepared
and evaluated under the same conditions (150 ◦C, 2 bar H2, 1 h).
The metal nanoparticles were prepared by ethanol–water reduc-
tion of the appropriate metal salts with PVP. Among these, the
Au nanoparticles demonstrated very low activity, possibly due,
at least in part, to their instability under the reaction conditions.
Au deposits were observed along with a colorless solution at
the end of the reaction. Pt, Ru, and Rh all exhibited good ac-
tivity in the reaction but with poor selectivity (<10%). Only
the Pd nanoparticles exhibited high stability under the reaction
conditions and gave relatively high activity and selectivity, that
is, a conversion of 70% with a selectivity of 25% toward the p-
cymene product. It is interesting that Pd nanoparticles prepared
by other methods also have demonstrated very good selectivity
in previous studies [26–28].
a
Entry
T
Time PVP Pd
Conversion p-Cymene TOF
◦
−1 b
)
( C) (h)
type concentration (GC-%)
selectivity (h
(GC-%)
(mol/L)
c
1
180
180
180
150
150
150
150
180
180
180
180
3
3
3
3
3
3
3
1
1
1
1
K30 0.001
K30 0.005
K30 0.01
K15 0.001
K30 0.001
K60 0.001
K90 0.001
K15 0.001
K30 0.001
K60 0.001
K90 0.001
91
99
96
92
94
93
82
97
68
74
74
44
70
78
31
32
25
26
49
32
38
45
677
195
96
c
2
c
3
4
5
6
7
720
750
710
730
970
677
735
741
c
8
c
9
c
10
11
a
−3
General conditions: PVP:Pd (mol:mol) = 20, Pd (1×10 mol/L, 10 mL),
nanoparticles were synthesized by alcohol–water reduction, water as solvent,
pH 2, limonene (1.360 g, 0.01 mol), 2 bar H , stirred at 500 rpm.
2
b
Temperature, hydrogen pressure, and reaction time were
evaluated to optimize the reaction conditions. The tempera-
ture was varied from 100 to 200 ◦C and was found to have
Tested after 1 h, defined as number of moles of consumed H per mole of
2
Pd per hour, determined by GC.
c
Nanoparticle aggregation observed.

C. Zhao et al. / Journal of Catalysis 254 (2008) 244–250
247
Fig. 1. (a) TEM micrograph of Pd nanoparticles protected by PVP-K90 in water before catalysis (scale bar = 20 nm). Nanoparticles were synthesized according
to the method described in Section 2.2.1. (b) Histogram showing the particle size distribution from (a). (c) TEM micrograph of Pd nanoparticles after the first run
of recycling experiment, as described in Section 2.4. Scale bar = 20 nm. (d) Histogram showing the particle size distribution from (c). (e) TEM micrograph of Pd
◦
−3
nanoparticles stabilized by PVP-K15 after reaction under 2 bar H at 180 C for 3 h, stirred at 500 rpm, [Pd] = 5×10 mol/L, limonene/catalyst = 50/1 mol/mol.
2
Scale bar = 20 nm.
Four different PVP polymers with various alkyl chain
lengths (i.e., K15, K30, K60, and K90, with average Mw
Table 1 (entries 4–7) shows that the overall catalytic efficiencies
of the Pd nanoparticles protected by these four types of PVP
were essentially the same. However, under more severe condi-
tions (i.e., 180 ◦C), the lower molecular weight polymers (i.e.,
=
10,000, 30,000, 220,000, and 630,000, respectively) were eval-
uated under relatively mild conditions (150 ◦C, 2 bar H2, 3 h).

248
C. Zhao et al. / Journal of Catalysis 254 (2008) 244–250
Table 2
a
Conversion of limonene using hydrogen acceptors and/or using different Pd nanoparticle preparation methods
Entry
Stabilizer
Preparation method
T
Stabilizer/Pd
(mol/mol)
Substrate/Pd
(mol/mol)
Conversion
(GC-%)
p-Cymene
selectivity (GC-%)
TOF
(h
◦
−1 b
)
( C)
c
1
PVP-K30
PVP-K90
PVP-K30
Surfactant
PVP-K30
PVP-K90
PVP-K15
PVP-K30
PVP-K90
PVP-K90
PVP-K90
PVP-K90
PVP-K90
H
H
reduction
reduction
150
180
180
180
180
180
180
180
180
180
180
180
180
40
20
20
2
100
100
20
20
20
20
20
20
20
1000
1000
200
200
1000
1000
200
200
200
50
89
68
99
97
93
89
98
99
99
93
99
85
90
22
29
66
70
52
34
77
70
70
82
87
81
70
705
620
170
160
910
850
158
180
180
46
2
2
c
2
c
3
NaBH reduction
NaBH reduction
4
4
c
4
c
5
Alcohol water reduction
Alcohol water reduction
Alcohol water reduction
Alcohol water reduction
Alcohol water reduction
Alcohol water reduction
Alcohol water reduction
Alcohol water reduction
Alcohol water reduction
6
c
7
c
8
9
10
11
12
d
d
50
50
50
50
30
40
d
d
13
a
−3
−3
General conditions: 2 bar H , 3 h, stirred at 500 rpm, Pd (1 × 10 mol/L, 10 mL in entries 1–6; 5 × 10 mol/L, 20 mL in entries 7–9), water as solvent,
2
pH 2.
b
c
d
Determined after 1 h, defined as number of moles of consumed H per mole of Pd per hour, determined by GC.
2
Nanoparticle aggregation observed.
−3
Pd (5 × 10 mol/L, 60 mL), limonene (2.040 g, 0.015 mol), recycling data.
PVP-K15, PVP-K30, and PVP-K60) could not protect the par-
ticles against aggregation (see Table 1, entries 8–10). Clearly,
there is a fine balance between stability and reactivity. For ex-
ample, the Pd nanoparticles stabilized by PVP-K15 exhibited
the highest activity with a TOF of 970 h−1 (Table 1, entry 8),
although these nanoparticles agglomerated under the reaction
conditions. The Pd nanoparticles protected by the PVP-K90
but with no aggregation observed after operating at 180 ◦C (Ta-
ble 2, entry 9), demonstrating again that PVP-K90 was the most
effective stabilizer for the nanoparticles operating at higher
temperatures.
3.2. Catalyst recycling
were stable even at 180 ◦C, albeit with a lower TOF of 741 h−1
.
To test recyclability, a Pd catalyst prepared by ethanol–water
reduction and protected by PVP-90 was recycled four times
at 180 ◦C and 2 bar H2 (Table 2, entries 10–13). The bipha-
sic approach made separation of the product very easy, and
no Pd leaching was found within the detection limit (1 µg–
mg/mL; standard ICP-AES analysis). The catalysts recovered
in the aqueous phase were reused with no further treatment. It
can be seen from Table 2 that the first run successfully reached a
conversion of 93% and a selectivity of 82% (entry 10), whereas
the second run successfully reached a conversion of 99% and
a selectivity of 87% (entry 11). Good selectivity (ca. >80%)
and near-quantitative conversion were obtained throughout the
recycling.
TEM micrographs of the Pd nanoparticles with different
PVP stabilizers (Figs. 1 and S1) show that the Pd nanoparticles
protected by PVP-K15 contained both small and large parti-
cles, ranging in size from 2.4 to 3.2 nm. The nonuniformity of
the particles can lead to good initial activity (TOF = 970 h−1
)
but poor stability (Fig. 1e). The Pd nanoparticles stabilized
by PVP-K90 are more uniform than those protected by PVP-
K15. A narrow unimodal size distribution with a diameter of
2.4 0.3 nm was observed for the freshly prepared Pd nanopar-
ticles (Figs. 1a and 1b), and a diameter of 2.8 0.5 nm was
observed after the catalysis (Figs. 1c and 1d). In general, it can
be concluded from Fig. S1 that the smaller the Pd nanoparticle
size, with a narrower size distribution, the higher the catalytic
activity and stability.
3.3. Proposed mechanism for the dehydroaromatization of
limonene over the Pd nanoparticles
Three reducing agents (i.e., H2, NaBH4, and C2H5OH) were
used to prepare the Pd nanoparticles. Table 2 shows that the Pd
nanoparticles reduced by H2 in either pure water or ethanol–
water mixture [30,31] gave the most unstable catalysts under
equivalent reaction conditions, resulting in poor selectivity and
pronounced aggregation. The Pd nanoparticles prepared by re-
duction with NaBH4 in the presence of either the surfactant,
N,N-dimethyl-N-cetyl-N-(2-hydroxyethyl)-ammonium bro-
mide (HEA16 Br) [23], or PVP were more stable than those
prepared by H2 reduction, resulting in a selectivity of 70% and
a conversion of nearly 100%; however, extensive aggregation of
the nanoparticles also was observed after the reaction. The Pd
nanoparticles obtained by reduction with ethanol gave the same
results, a selectivity of 70% and a conversion of nearly 100%,
The conversion of limonene to p-cymene on a Pd/SiO2 het-
erogeneous catalyst has been proposed to occur through the
dehydroaromatization mechanism shown in Scheme 2a [9].
Two reaction routes were determined by changing the catalysts
and/or the carrier gas. When the Pd/SiO2 catalyst and H2 at-
mosphere were used simultaneously, the reaction was believed
to follow route 1 (including full hydrogenation followed by
dehydrogenation), and thus the only byproduct was the fully hy-
drogenated compound p-menthane. In the absence of Pd (i.e.,
simply using SiO2 as catalyst), or in the case where N2 was used
in place of H2 but still with Pd/SiO2, additional isomerization
products were obtained.

C. Zhao et al. / Journal of Catalysis 254 (2008) 244–250
249
Scheme 2. Proposed mechanism of limonene conversion over (a) heterogeneous Pd/SiO and (b) soluble Pd catalysts.
2
Nonetheless, when we used Pd nanoparticles and H2, the re-
action did not appear to involve full hydrogenation followed
by dehydrogenation. This hypothesis is supported by the fol-
lowing observations. First, it can be seen from Fig. 2 that the
increased selectivity of p-cymene was accompanied by a de-
crease in selectivity toward the isomerization products, indicat-
ing that the isomerization products were transformed mainly
to p-cymene. Second, the Pd nanoparticles were unable to de-
hydrogenate p-menthane to p-cymene. When p-menthane was
used as a substrate, even in the presence of freshly prepared
Pd nanoparticles, no dehydrogenation products were observed,
indicating an alternative pathway to route 1. The observation
of terpinolene and α-terpinolene indicates that the correspond-
ing isomerization was the main reaction pathway. The process
of isomerization on the soluble Pd nanoparticles likely pro-
ceeded by a single route (i.e., the production of terpinolene
and then of α-terpinolene), suggesting a highly selective iso-
merization pathway for the nanoparticle catalysis. Thus, the
Pd nanoparticles would appear to facilitate the isomerization
of the double bond in a selective manner (Scheme 2b). If this
were not the case, then many isomerization products other
than terpinolene and α-terpinolene would be produced, as has
been observed over supported Pd catalysts. Based on all of
these findings, we illustrate a plausible reaction pathway in-
volving initial isomerization followed by dehydrogenation in
Scheme 2b.

250
C. Zhao et al. / Journal of Catalysis 254 (2008) 244–250
authors thank Nanjing Huagong Co. Ltd. for the generous gift
of the limonene used in this study.
Supplementary material
The online version of this article contains additional supple-
mentary material.
Please visit DOI: 10.1016/j.jcat.2008.01.003.
References
[1] A. Corma, S. Iborra, A. Velty, Chem. Rev. 107 (2007) 2411.
[2] K.A.D. Swift, Top. Catal. 27 (2004) 143.
[3] J.M. Derfer, M.M. Derfer, in: Kirk-Othmer Encyclopedia Chem. Technol.,
vol. 22, John Wiley & Sons, Inc., New York, 1978, p. 709.
[4] P. Lesage, J.P. Candy, C. Hirigoyen, F. Humblot, J.M. Basset, J. Mol.
Catal. A Chem. 112 (1996) 431.
[5] D. Buhl, D.M. Roberge, W.F. Hölderich, Appl. Catal. A Gen. 188 (1999)
287.
[6] P.A. Weyrich, W.F. Hölderich, M.A. Daelen, A.M. Gorman, Catal. Lett. 52
(1998) 7.
[7] P.A. Weyrich, W.F. Hölderich, Appl. Catal. A Gen. 158 (1997) 145.
[8] R.J. Grau, P.D. Zgolicz, C. Gutierrez, H.A. Taher, J. Mol. Catal. A
Chem. 148 (1999) 203.
[9] H. Pines, R.C. Olberg, V.N. Ipatieff, J. Am. Chem. Soc. 74 (1952) 4872.
[10] R.P. Linstead, K.O.A. Michaelis, S.L.S. Thomas, J. Chem. Soc. (1940)
1139.
[11] R.C. Palmer, Ind. Eng. Chem. 34 (1942) 1028.
[12] R. Neumann, M. Lissel, J. Org. Chem. 54 (1989) 4607.
[13] D.J. Cole-Hamilton, Science 299 (2003) 1702.
[14] R.T. Baker, W. Tumas, Science 284 (1999) 1477.
[15] D.B. Zhao, M. Wu, Y. Kou, E.Z. Min, Catal. Today 74 (2002) 157.
[16] N. Yan, C. Zhao, C. Luo, P.J. Dyson, H. Liu, Y. Kou, J. Am. Chem.
Soc. 128 (2006) 8714.
[17] X.D. Mu, J.Q. Meng, Z.C. Li, Y. Kou, J. Am. Chem. Soc. 127 (2005) 9694.
[18] C. Zhao, H.Z. Wang, N. Yan, C.X. Xiao, X.D. Mu, P.J. Dyson, Y. Kou,
J. Catal. 250 (2007) 33.
[19] C.X. Xiao, H.Z. Wang, X.D. Mu, Y. Kou, J. Catal. 250 (2007) 25.
[20] M. Zou, X.D. Mu, N. Yan, Y. Kou, Chin. J. Catal. 28 (2007) 389.
[21] T. Wang, C.X. Xiao, L. Yan, L. Xu, J. Luo, H. Shou, Y. Kou, H. Liu, Chem.
Commun. (2007) 4375.
[22] J.A. Widegren, R.G. Finke, J. Mol. Catal. A Chem. 191 (2003) 187.
[23] M.T. Reetz, G. Lohmer, Chem. Commun. (1996) 1921.
[24] A. Roucoux, J. Schulz, H. Patin, Adv. Synth. Catal. 345 (2003) 222.
[25] T. Teranishi, M. Miyake, Chem. Mater. 10 (1998) 594.
[26] J.L. Pellegatta, C. Blandy, R. Choukroun, C. Lorber, B. Chaudret,
P. Lecante, E. Snoeck, New J. Chem. 27 (2003) 1528.
[27] V. Kogan, Z. Aizenshtat, R. Neumann, New J. Chem. 26 (2002) 272.
[28] A.P. Umpierre, G. Machado, G.H. Fecher, J. Morais, J. Dupont, Adv.
Synth. Catal. 347 (2005) 1407.
Fig. 2. Product distributions for limonene conversion as a function of tempera-
ture. Reaction conditions: Reactions were carried out in an autoclave in water
containing Pd nanoparticles under 2 bar H for 5 h, stirred at 500 rpm. Further
2
reaction parameters and experimental data are given in Table S1, entries 1–5.
4. Conclusion
Biphasic dehydroaromatization of the renewable biomass
material limonene into p-cymene using soluble Pd nanoparticle
catalysts in aqueous solution in the presence of hydrogen occurs
under severe conditions (ꢀ150 ◦C), achieving a conversion of
93% and a selectivity of 82%. The Pd nanoparticle catalysts
can be recycled at least four times without deactivation even af-
ter reaction at 180 ◦C. To the best of our knowledge, this is the
highest operating temperature that soluble metal nanoparticles
can tolerate in the liquid phase.
The effects of temperature, pressure, reaction time, pH, cat-
alyst concentration, metal type, and stabilizer were systemati-
cally investigated to optimize the reaction and provide insight
into its mechanism. The reaction is believed to follow a dif-
ferent pathway than that of traditional heterogeneous catalysts,
which likely is responsible for the excellent yield and selectivity
obtained with the Pd nanocatalysts. The relationship between
the nanoparticles’ structure and their activity and stability also
was explored; in general, the smaller and more uniform the
nanoparticles size, the higher the activity and stability obtained.
Acknowledgments
[29] W.T. Reichle, J. Catal. 94 (1985) 547.
This work was supported by the National Science Founda-
tion of China (projects 20533010, 20773005, J0630421). The
[30] F. Lu, J. Liu, J. Xu, Adv. Synth. Catal. 348 (2006) 857.
[31] F. Lu, J. Liu, J. Xu, J. Mol. Catal. A Chem. 271 (2007) 6.