Mesoporous silica-supported Pd catalysts: An investigation into structure, activity, leaching and heterogeneity

Article abstract of DOI:10.1016/j.jcat.2007.09.007

Mesoporous silica functionalized by mercaptopropyl trimethoxysilane can be loaded with palladium and the resulting material used as a catalyst for a variety of coupling reactions. The ordered silicate reacts with levels of Pd leaching in the low ppm to ppb range. The catalyst can be re-used multiple times with only moderate loss of activity or structure, depending on the method of incorporation of the thiol. A grafting approach gives a significantly more stable material which is likely insulated from attack by the aqueous base, while incorporation of the thiol by co-condensation gives a material with minimal stability under the reaction conditions. Several heterogeneity tests are performed on the catalyst including three-phase tests and poisoning studies. These indicate that the effective heterogeneity depends strongly on the solvent employed and the reaction conditions. Under non-aerobic conditions with a controlled ratio of sulfur ligand to palladium, low levels of reaction due to homogeneous species are observed, but as the organic content of the solvent is increased, more reaction on an immobilized reagent (as judged through the three-phase test) is observed. In addition, these studies highlight the constraints within which the three-phase test can be an accurate assessment of heterogeneity.

Full text of DOI:10.1016/j.jcat.2007.09.007

Journal of Catalysis 252 (2007) 97–109
www.elsevier.com/locate/jcat
Mesoporous silica-supported Pd catalysts: An investigation into structure,
activity, leaching and heterogeneity
Jonathan D. Webb, Stephanie MacQuarrie, Kevin McEleney, Cathleen M. Crudden ∗
Queen’s University, Department of Chemistry, 90 Bader Lane, Kingston, ON, Canada, K7M 3N6
Received 18 July 2007; revised 5 September 2007; accepted 5 September 2007
Abstract
Mesoporous silica functionalized by mercaptopropyl trimethoxysilane can be loaded with palladium and the resulting material used as a catalyst
for a variety of coupling reactions. The ordered silicate reacts with levels of Pd leaching in the low ppm to ppb range. The catalyst can be re-used
multiple times with only moderate loss of activity or structure, depending on the method of incorporation of the thiol. A grafting approach gives
a significantly more stable material which is likely insulated from attack by the aqueous base, while incorporation of the thiol by co-condensation
gives a material with minimal stability under the reaction conditions. Several heterogeneity tests are performed on the catalyst including three-
phase tests and poisoning studies. These indicate that the effective heterogeneity depends strongly on the solvent employed and the reaction
conditions. Under non-aerobic conditions with a controlled ratio of sulfur ligand to palladium, low levels of reaction due to homogeneous species
are observed, but as the organic content of the solvent is increased, more reaction on an immobilized reagent (as judged through the three-
phase test) is observed. In addition, these studies highlight the constraints within which the three-phase test can be an accurate assessment of
heterogeneity.
© 2007 Elsevier Inc. All rights reserved.
Keywords: Palladium; Heterogeneous; Mesoporous; Catalysis; Coupling; Suzuki; Three-phase test
1. Introduction
in the case of Pd on unfunctionalized supports such as Pd/C
and Pd/SiO2, essentially all of the catalysis can be attributed to
Pd into solution [20–29]. Furthermore the severity of the leach-
ing and redeposition leads ultimately to catalyst death due to
losses in solution and also agglomeration of Pd into large, less
active particles on the surface. Similar studies employing Rh
catalysts have also shown that even in the case of homogeneous
precursors, catalysis can often be ascribed to heterogeneous Rh
colloids generated under the reaction conditions [30,31].
A fundamentally different approach to heterogenizing Pd
catalysts is to chemically bind known ligands to the surface of a
support and use these to form strong bonds to Pd. These ligands
include species such as phosphines [32,33] and N-heterocyclic
carbenes [34,35]. In addition to these traditional ligands, pincer
and palladacycle [36–38] ligands which contain a Pd–carbon
bond have also been employed. The high catalytic activity of
these latter ligands generated considerable controversy, since
the complexes are immobilized in the Pd(II) oxidation state,
which raised questions about the mechanism of the coupling re-
The use of silica-supported Pd catalysts in general and meso-
porous silica-supported catalysts in particular has gained in-
creased attention as of late [1–10]. This is presumably driven
by the widespread use and utility of Pd catalysts in indus-
try [11–14] and the resulting difficulties with removing spent
catalysts at the end of the reaction [15–17]. For this reason,
heterogeneous catalysts offer significant advantages over their
often times more reactive homogeneous partners since the lev-
els of Pd remaining in solution are attenuated. However, there
have been numerous cases in which the use of supposedly het-
erogeneous catalysts, such as the ubiquitous Pd on carbon, has
resulted in leaching of Pd at levels which led to questions about
the true heterogeneity of the Pd source [18,19]. Detailed inde-
pendent studies by Kohler, Arai and Conlon have shown that
*
Corresponding author.
E-mail address: cruddenc@chem.queensu.ca (C.M. Crudden).
0021-9517/$ – see front matter © 2007 Elsevier Inc. All rights reserved.
doi:10.1016/j.jcat.2007.09.007

98
J.D. Webb et al. / Journal of Catalysis 252 (2007) 97–109
action and the possibility of a Pd(II)/Pd(IV) catalytic cycle [12,
19,39].
Aldrich. Palladium acetate was obtained from Pressure Chemi-
cal Company.
Detailed studies of the pincer system eventually showed that
with a few possible exceptions, most palladacycle-type cata-
lysts decompose slowly under the reaction conditions, liber-
ating active Pd(0) species that are the true catalysts for the
reaction [29,40–52]. Another class of catalysts that have been
developed utilize less well defined interactions for the immobi-
lization of Pd. Examples of these catalysts include Kobayashi’s
incarcerated catalysts [53,54], ionic gels employed by Kirschn-
ing [55] and Mioskowski [56], Pd-containing perovskites by
Ley [57,58], and the interesting case of physically immobi-
lized catalysts using a zeolite shell by Tang et al. [59]. In-
termediate between these two extremes are systems in which
the support is functionalized, but the interactions between Pd
and the surface are not necessarily well defined. For example,
Ley’s Pd EnCat catalysts contain thiourea ligands for Pd [60–
62], and silica surfaces modified by thiol groups have also been
described [63–65]. There are several fundamental differences
between these catalysts and palladacycle type catalysts, most
notably that these catalysts can act as supports for Pd(0) and/or
Pd(II), and can accommodate an equilibrium between homoge-
neous and heterogeneous Pd, whereas pincer-type catalysts will
not re-adsorb Pd after it has been released as Pd(0), at least not
via the ligand that was initially used to immobilize the Pd. This
is a critical difference since it complicates the interpretation of
various heterogeneity tests.
As part of a study aimed at developing an effective Pd
scavenger based on mesoporous silica modified with mercapto-
propyl trimethoxysilane, we discovered to our surprise that after
scavenging of Pd, the resulting material catalyzed the Suzuki–
Miyaura reaction with low levels of leaching [64]. A similar
study of the activity of Pd-loaded functionalized FSM-16 also
described high catalytic activity with low leaching [63]. In or-
der to further assess the utility, re-usability and heterogeneity of
catalysts such as those prepared from thiol-modified silica, we
have carried out extensive studies of leaching, catalyst integrity
and heterogeneity tests such as three-phase tests and polymeric
poison challenge tests. Interestingly, we find that the catalysts
behave differently in different tests, which highlights the diffi-
culty of interpreting results from only one series of tests [52].
The results of these tests, and tests of the recyclability of the
material are described herein.
2.2. Procedure for catalyst synthesis
In a typical procedure [66], Pluronic 123 (P123) (4.0 g) was
weighed into a 500 mL flask with stir bar. The polymer was
dissolved by adding distilled water (30 mL) and 4 h of vigorous
stirring at 35 ◦C followed by the addition of aqueous 2 M HCl
(120 mL). This solution was stirred for a further 24 h at 35 ◦C
prior to addition of TEOS (9.0 mL) via syringe. A fine white
precipitate formed after about 30 min, this solution was vigor-
ously stirred at 35 ◦C for 24 h after which the suspension was
heated for 48 h at 80 ◦C. The solution was filtered and dried un-
der vacuum for 3 days at 80 ◦C. The polymer was removed via
Soxhlet extraction using refluxing 95% EtOH (600 mL) for 5
days. The recovered silicate was dried at room temperature and
80 ◦C under vacuum both for 18 h. This procedure produced
2.5 g of a fine white powder. SBA-15 (BET SA: 1080 m2/g,
BJH absorption pore diameter: 84 Å, pore volume: 1.13 mL/g).
To prepare SBA-15-SH(g), SBA-15 (900 mg) was sus-
pended in dry toluene (50 mL) in a round bottomed flask
(100 mL) which was flushed with N2. (3-Mercaptopropyl)-
trimethoxysilane (MPTMS) (3.6 mL) was added via syringe
and the mixture heated to 110 ◦C with stirring for 18 h. The
solution was filtered and washed with Et2O (3 × 15 mL). The
recovered powder was extracted via Soxhlet using refluxing
95% EtOH (600 mL) for 24 h. The silicate was dried under
vacuum at room temperature for 18 h. A white powder was re-
covered (625 mg). The final sulfur content varies depending on
the initial batch of SBA-15 (0.6–1.1 mmol/g S) as was deter-
mined by elemental analysis. When comparing two different
materials, equivalent sulfur contents were always employed.
SBA-15-SH(g) (BET SA: 750 m2/g, BJH absorption pore di-
ameter: 72 Å, pore volume: 0.82 mL/g).
Pd(OAc)2, corresponding to 50% of the loading of sulfur
as determined by elemental analysis, was dissolved in THF
(50 mL) and the flask flushed with N2. SBA-15-SH(g) (500 mg)
was then added and the suspension stirred for 2 h. The solution
was filtered and washed with THF (3×50 mL). The red-orange
material was dried under vacuum overnight, yielding 450 mg.
The amount of Pd loaded was determined by an ICP-MS analy-
sis of the supernatant obtained after filtration and washings of
the Pd material prepared as described above. Consistent with
previous work, >99.9% of the initially added Pd was scavenged
from solution [64]. SBA-15-SH(g)·Pd (BET SA: 360 m2/g,
BJH absorption pore diameter: 70 Å, pore volume: 0.44 mL/g).
To prepare SBA-15-SH(cc), 4.0 g of triblock copolymer
Pluronic 123 (P123) was dissolved in 150 mL of water along
with 20 mL of conc. HCl and the resulting mixture stirred
at 40 ◦C for 16 h. 8.6 mL (8.02 g, 38.5 mmol) of tetraethyl-
orthosilane (TEOS) was then added and the mixture stirred for
5 h at 40 ◦C. Finally, 0.5 mL (0.525 g, 2.8 mmol, 6.8 mol%)
of (3-mercaptopropyl)-trimethoxysilane (MPTMS) was added
and the reaction mixture stirred for 19 h at 40 ◦C. The pow-
2. Experimental
2.1. General experimental
Reagents were purchased or obtained from the following
suppliers: Pluronic 123 was donated by BASF. TEOS (tetra-
ethylorthosilicate), mercaptopropyl trimethoxysilane (MP-
TMS), aminopropyl triethoxysilane (APTES), para-bromo-
acetophenone (BrPhAc), potassium carbonate, DMF, 4-bromo-
benzoyl chloride, N-methyl morpholine and para-dimethoxy-
benzene were obtained from Aldrich. Phenylboronic acid, pina-
col ester was prepared from phenylboronic acid, obtained from
der was then hydrothermally treated at 80
stirring. The solid was recovered by filtration and washed with
◦C for 48 h with no

J.D. Webb et al. / Journal of Catalysis 252 (2007) 97–109
99
water. The surfactant P123 was then removed by Soxhlet ex-
traction with ethanol for 5 days. The solid was then filtered and
dried under vacuum. SBA-15-SH(cc) (BET SA: 790 m2/g, BJH
absorption pore diameter: 69 Å, pore volume: 0.945 mL/g).
Sulfur content was determined by elemental analysis as noted
above.
114.77 mg (0.511 mmol) of Pd(OAc)2 was dissolved in
75 mL of dry THF. This was stirred under an argon atmosphere
for 15 min to ensure complete dissolution. SBA-15-SH(cc)
1.5157 g (1.03 mmol S) was then added to the solution and
the mixture stirred under argon for 1 h at room temperature.
The catalyst was filtered using a sintered glass funnel, scraped
into a vial and dried overnight under high vacuum. The filtrate
was collected and sent for ICPMS (which is able to detect Pd
at parts per trillion levels) analysis to determine palladium con-
tent. The residual palladium in the solution was found to be
<0.03% of that initially added. SBA-15-SH(cc)·Pd (BET SA:
662 m2/g, BJH absorption pore diameter: 42 Å, pore volume:
0.688 mL/g).
HCl (20 mL, 3 times) followed by 0.02 M aqueous K2CO3
(20 mL, twice) and rinsed with distilled water (40 mL) and
ethanol (40 mL). The resulting solid was washed with excess
dichloromethane and dried at room temperature in air, yielding
1.14 g of a white powder. CP-MAS 13C NMR (ppm), 233.7,
167.2, 130.5, 58.782, 42.0, 35.6, 30.2, 21.7, 16.7, 9.5. Calcu-
lated elemental analysis; weight %C = 10.38 (8.65 mmol/g),
%H = 1.32 (13.2 mmol/g), %N = 1.51 (1.07 mmol/g), %Br
= 5.60 (0.7 mmol/g).
Alternatively, 4-bromobenzamide 3-propyltriethoxysilane
(8) was prepared for use in the three-phase test by mixing
4-bromobenzoyl chloride (2.01 g, 9.1 mmol), N-methyl mor-
pholine (1.00 mL, 9.1 mmol) and THF (50 mL) in a round-
bottomed flask and cooled to −15 ◦C. 3-Aminopropyl tri-
ethoxysilane (2.14 mL, 9.1 mmol) in THF (10 mL) was added
dropwise to the cooled solution. The resulting mixture was
warmed to room temperature (2 h). After that time, the solid
was removed via filtration and the solvent was removed under
vacuum at room temperature. The product was obtained as 3.5 g
1
of a beige solid. H NMR (ppm), 7.8 (1H, t), 7.6 (2H, d), 7.2
2.3. General procedure for the Suzuki–Miyaura coupling
reaction
(2H, d), 3.5 (6H, q), 3.2 (2H, m), 1.5 (2H, m), 1.0 (9H, t), 0.3
(2H, m).
4-Bromobenzamide propyl triethoxysilane (3.5 g, 8.6 mmol)
thus obtained, was mixed with pyridine (0.912 mL, 11.5 mmol)
and added dropwise to a suspension of SiO2 (1 g) in dry toluene
(3 mL), under argon. The resulting mixture was refluxed for
24 h. After that time, the suspension was filtered and Soxhlet
extracted with dichloromethane for 24 h. The resulting solid
was dried under vacuum at room temperature, giving 1.6 g
of a white powder. CP-MAS 13C NMR (ppm), 223.8, 167.0,
130.4, 127.9, 59.2, 42.2, 22.3, 17.3, 9.4. CP-MAS 29Si (ppm),
−50.0, −52.9, −58.0, −68.3, −103.2, −112.1. Calculated el-
emental analysis; weight %C = 12.72 (10.59 mmol/g), %H =
1.41 (14.1 mmol/g), %N = 1.53 (1.08 mmol/g), %Br = 7.91
(0.98 mmol/g).
Pd-SH-silica (1 mol% Pd), p-bromoacetophenone (50 mg,
0.25 mmol), K2CO3 (70 mg, 0.50 mmol), PhBpin (pinacol
ester of phenyl boronic acid, 75 mg, 0.375 mmol) and 1,4-
dimethoxybenzene (35 mg, 0.253 mmol) as internal standard
were weighed into a sealed tube with a stir bar and flushed
with Ar. Dry DMF (2.5 mL), degassed by bubbling Ar, and
distilled H2O (0.125 mL) were added via syringe. The solu-
tion was heated at 80 ◦C with stirring while monitoring reaction
progress by GC-FID.
Recycling studies were carried out following the general
Suzuki reaction procedure except scaled up so that either 50
or 100 mg of SBA-15-SH(g)-Pd were used. Following reac-
tion/recycle the supernatant was collected for ICP-MS analysis
and the recovered catalyst washed with copious amounts of
EtOAc and H2O to remove residual salts and organics. Each
recycle was performed by weighing the recovered washed cat-
alyst and calculating the amount of reagents necessary for a
1 mol% Pd reaction, it was assumed that >99% of Pd was
retained by the material as validated by ICP-MS. Nitrogen ad-
sorption isotherms of the recovered catalyst are given in Figs.
S-6 and S-7 of Supporting information. Data generated from
these plots are given in Table S-1. Powder XRD data for SBA-
15-SH(g)·Pd before and after reaction are given in Fig. S-5 and
13C CPMAS NMR data are shown in Fig. S-8. Fig. S-9 gives
the nanoparticle distribution observed by TEM for catalysis run
in argon.
3. Results
3.1. Reaction conditions
Our study began by assessing the effect of several vari-
ables on catalyst activity and leaching. Two of the key features
which emerged from this study are the method of incorpora-
tion of the thiol and the ratio of thiol to Pd. In addition to
SBA-15, amorphous silica was examined as a cost-effective al-
ternative to this ordered silicate. The Suzuki–Miyaura reaction
between 4-bromoacetophenone and the pinacol ester of phenyl
boronic acid were chosen as the test reaction [64] (Eq. (1)).
Potassium carbonate was employed as the base and the solvent
was either mixtures of DMF and water, or pure water. Controls
showed that no reaction occurred in the absence of added cat-
alyst. Under these conditions, amorphous silica modified with
mercaptopropyl trialkoxysilane and treated with Pd is not only
catalytically active, but displays activity comparable to that of
SBA-15-SH supported Pd (Supporting information, Fig. S-4).
2.4. Experimental procedure for the three-phase test
4-Bromobenzoyl chloride (1.15 g, 5.25 mmol), 3-amino-
propyl trimethoxysilane-modified silica (1.00 g, Aldrich), pyri-
dine (404 µL, 5 mmol) and dry THF (10 mL) were stirred in
a round bottom flask under a nitrogen atmosphere at 40 ◦C for
12 h. The suspension was filtered and washed with 5% (v/v)

100
J.D. Webb et al. / Journal of Catalysis 252 (2007) 97–109
Table 1
Table 2
a
Catalytic activity and leaching as a function of S to Pd ratio
Pd leaching in solution relative to support, atmosphere and method of thiol
functionalization
Entry
Material
S:Pd
Yield (%)
Leaching (%) [67]
a
Entry
Catalyst support
Conditions
Yield
Pd leaching (%)
1
2
3
4
5
SiO -SH·Pd
1:1
2:1
3:1
4:1
8:1
86
98
72
25
0
0.3
2
b
SiO -SH·Pd
0.06
0.02
0.03
1
2
3
4
SBA-15-SH(g)
SBA-15-SH(g)
SBA-15-SH(cc)
SBA-15-SH(cc)
Air, 8 h
98–99
98–99
90–98
90–95
0.32 0.05
0.12 0.05
0.17 0.09
0.22 0.09
2
SiO -SH·Pd
Argon, 4 h
Air, 18 h
Argon, 4 h
2
SiO -SH·Pd
2
b
SiO -SH·Pd
n.e.
2
a
a
Reaction conditions: 0.75 mmol PhBpin, 0.51 mmol BrAcPh (4-bromo-
Pd leaching is given as a of added Pd (see Ref. [67]). In all cases less than
acetophenone), 1 mol% SiO -SH·Pd, 1.02 mmol K CO , 0.51 mmol p-di-
0.5 ppm Pd is observed.
b
2
2
3
methoxybenzene as internal standard, 20/1 DMF/H O (0.l M in BrAcPh),
16 biphenyl observed (1.5 equivalents of PhB(OH) , 1 equivalent of ArBr).
2
2
◦
80 C, 18 h.
b
Not examined.
(1)
As previously reported [63,64], low levels of leaching were
observed in the solution at the completion of the reaction. At
the standard 2:1 ratio of S:Pd, the Pd concentration in solu-
tion at the end of the reaction was generally less than 0.2 ppm,
and often as low as several ppb. This corresponds to between
0.05 and 0.3% of the initially added Pd (Table 1) [67]. At a
1:1 ratio of sulfur to Pd, catalysts displayed increased leach-
ing (compare entries 1 and 2). As the ratio of sulfur to Pd
increased, the catalytic activity decreased (see Fig. S-5 in Sup-
porting information) such that at ratios of 8:1 or greater, no
catalytic activity was observed (Table 1). We chose 2:1 as
the optimum S:Pd ratio since scavenging studies showed that
this ratio was the most effective at removing Pd from so-
lution [64] and since it gave the most active catalyst. Un-
der these conditions, we examined the reactivity and leaching
of the SBA-15-derived materials prepared by grafting (SBA-
15-SH(g)) and by sol–gel co-condensation (SBA-15-SH(cc))
methods for the incorporation of the mercaptopropyl function-
ality. The effect of the atmosphere of the reaction was also
examined.
The reactions can be performed in air or even oxygen, al-
though better results are obtained in argon in terms of rate of re-
action, turnover and leaching. In air, reactions require between
6 and 8 h to go to completion, while in argon, they are gener-
ally complete in 4 h. As expected, reactions run in air display
small amounts of homocoupling of the boronate ester, which
continues to be observed after the Suzuki–Miyaura reaction is
complete until all of the excess boronate ester is consumed (Ta-
ble 2).
(A)
(B)
Fig. 1. Effect of recycling on catalyst yield (A) and leaching (B).
3.2. Recyclability
In addition to the total leaching and overall catalytic ac-
tivity, recyclability is obviously a critical feature of supported
catalysts. Employing SBA-15-SH(g) in which the thiol was
introduced by grafting on a previously synthesized material,
we found that the material displayed minimal loss of activ-
ity after 4 runs (Fig. 1). In contrast, catalysts prepared using
a co-condensation method for thiol incorporation [68a], were
quickly deactivated [68b]. Interestingly, this drop in activity oc-
curred concomitantly with an increase in leaching. As will be
shown subsequently, the loss of activity with the co-condensed
catalyst was coincident with a loss of order of the support (vide
infra).
Amorphous silica-supported catalysts had intermediate be-
havior in terms of recyclability and leaching, with activity drop-
ping off after the fourth use. In this case, 18 h was necessary for
90% conversion, and then eventually only 45% conversion ob-
tained on the fifth run. Again, loss of activity appeared to be
coincident with increased leaching, although Pd leaching was
The turnover numbers of the various catalysts were also as-
sessed in air and argon and were found to be higher in all
cases in argon. SBA-15-SH(g)·Pd reacted with a total turnover
number of >10,000 in argon compared to 5100 in air. This cor-
responds to a reaction time of 24 h for 97% yield at a loading
of 0.009 mol%. By comparison, amorphous silica reacted with
a total turnover number of 6800 in argon and 5100 in air.

J.D. Webb et al. / Journal of Catalysis 252 (2007) 97–109
101
Table 3
Recycling studies
a
Entry
Pd catalyst
Run
Yield
[time]
Pd leaching (%)
[ppm]
1
2
3
4
5
6
7
8
9
SBA-15-SH(g)·Pd
SBA-15-SH(g)·Pd
SBA-15-SH(g)·Pd
SBA-15-SH(g)·Pd
SBA-15-SH(g)·Pd
1
2
3
4
5
1
2
3
4
5
99 [4 h]
99 [4 h]
97 [4 h]
91 [8 h]
97 [8 h]
98 [4 h]
90 [4 h]
91 [4 h]
90 [18 h]
45 [18 h]
0.036 [0.011]
0.008 [0.006]
0.032 [0.018]
0.26 [0.11]
0.59 [0.13]
–
0.001 [0.003]
0.008 [0.005]
0.04 [0.017]
0.2 [0.072]
SiO -15-SH·Pd
2
SiO -15-SH·Pd
2
SiO -15-SH·Pd
2
SiO -15-SH·Pd
2
10
SiO -15-SH·Pd
2
a
See Section 2 for details.
still relatively low (0.2% of the initially added catalyst, corre-
sponding to 0.07 ppm Pd in solution, Table 3).
Fig. 2. Treatment of various materials with solvent and base.
3.3. Catalyst integrity and leaching
Catalysts which were employed in reactions run under an air
atmosphere showed no evidence of Pd nanoparticles either by
TEM (Fig. 5) or XRD, although the latter technique may not
be useful at the low concentrations of Pd(0) which are likely
present on the support. Again the high order maintained in the
catalyst after the reaction is especially evident in the electron
diffraction (inset).
Although the microscopy results are generally satisfying in
terms of the retention of order of the material, we also examined
the supported catalyst by physisorption analysis after reaction
since this is a bulk-characterization technique. As shown in
Figs. S-6 and S-7 (Supporting information), there is a moder-
ate loss of order after each run, in which the material is exposed
to aqueous base at 80 ◦C for 4 h [70]. Fig. 6A shows fresh cat-
alyst before use and the same catalyst after it has been used
three times. Similar studies with the co-condensed material il-
lustrated the dramatic difference between the two methods of
grafting. As shown in Fig. 6B, the material prepared by co-
condensation is dramatically more sensitive, having lost a sig-
nificant amount of order after only one run. As noted previously,
the two materials have similar overall thiol loading as deter-
mined by elemental analysis, indicating that the effect observed
is related to the method of incorporation [71].
Although the hydrothermal stability of SBA-15 has been
well documented [69], the conditions of the Suzuki–Miyaura
reaction are significantly harsher than simply hydrothermal
treatment since the reaction is performed with aqueous base.
Considering that the wall sizes in SBA-15 are on the order of
3 nm, dissolution/decomposition of the support under the reac-
tion conditions is a concern.
Thus the various catalysts were treated with DMF and water
at 80 ◦C for varying times with and without base. As shown
in Fig. 2, the leaching after 24 h is on the order of 0.05%
(0.013 ppm) for all three catalysts. This value does increase
upon the inclusion of base (potassium carbonate) in the mix-
ture, but not substantially. Consistent with the recycling study,
the co-condensed catalyst showed greater sensitivity to base.
We attribute these results to the greater protective effect of the
grafting conditions, in which the surface is rendered more hy-
drophobic than the co-condensed material, where greater por-
tions of unprotected silica are exposed to the reaction media.
3.4. Catalyst support
In addition to assessing the loss of Pd from the support
upon treatment with base, it is obviously crucial to determine
whether or not the support itself survives the reaction condi-
tions, a fact which becomes more important as the cost of the
support increases. The integrity of the silicate was assessed in
several ways. Transmission electron microscopy of catalyst re-
covered from the reaction showed that the overall structure was
largely maintained (Figs. 3A and 3B (before use) and Figs. 4A
and 4B (after use)), although some destruction of the frame-
work porosity is evident at the edges of the particles. Similarly
to Shimizu et al. [63], we observed Pd nanoparticles on the sur-
face (Fig. 4A), none of which were larger than the pores (6 nm).
The distribution of nanoparticles is shown in Fig. S-8 (Sup-
porting information), which demonstrates that the nanoparticles
range in size from 1 to 6 nm, with the majority in the 3–5 nm
range.
3.5. Heterogeneity tests
In the initial study of MPTMS-modified SBA-15 as a cat-
alyst support for the coupling of boronic acids [64], the het-
erogeneity of the catalyst was assessed by two of the standard
techniques in the area, three-phase tests and a hot filtration
test [52]. In the hot filtration test, a portion of the reaction mix-
ture is filtered from the suspended catalyst mid-way through the
reaction, and the filtrate is tested for catalytic activity. The ob-
servation of activity in the filtrate is indicative of soluble Pd
present in solution. Since catalyst redeposition on support upon
partial cooling of a solution can occur prior to or during filtra-
tion, a negative test is not sufficient to declare a catalyst to be
truly heterogeneous [19]. Also since many Pd-catalyzed cou-
pling reactions employ sparingly soluble reagents and starting

102
J.D. Webb et al. / Journal of Catalysis 252 (2007) 97–109
(A)
(B)
Fig. 3. Different views of SBA-15-SH·Pd before use.
materials (such as boronic acids), it is essential to ensure that
these are exposed to the filtrate and are not removed during fil-
tration. In the initial study, approximately 5% reactivity in the
filtered phase was detected, indicating that some catalytically
active Pd was being released from the surface.
However, due to the difficulties noted with the hot filtration
test, a three-phase test was performed, which is believed to be
one of the most definitive tests for heterogeneity since it actu-
ally tests for the presence of active, catalytically competent Pd
in solution [62,72]. This test was developed in the late 70’s by
Rebek and Zimmerman to test for the presence of reactive in-
termediates [72]. The concept is that reactive intermediates in
solution can be detected by allowing them to react with an im-
mobilized substrate which can later be filtered and analyzed.
The test is performed by immobilization of an aryl halide and
addition of solution phase reagents along with the “heteroge-
neous” catalyst in question [18,62]. After reaction for a pre-
scribed amount of time, the immobilized substrate is cleaved
from the second support to determine whether or not any reac-
tive catalyst was present in solution. The test is considered to
be an indication of the release of active Pd from support, since
the supported reagent and supported catalyst are unlikely to re-
act together. However, there are pitfalls associated with this test
as well. As elegantly demonstrated by Davies and co-workers
at Merck, it is important to add a soluble aryl halide, since it
is believed that oxidative addition with this soluble reagent is
the main pathway for bringing Pd into solution [18]. Moreover,
the presence of soluble aryl halide provides an opportunity to
ensure that an active catalyst is in fact generated since the cou-
pling of the two soluble species can be detected and quantified.
Thus, we carried out the three-phase test with grafted and
sol–gel-modified mesoporous materials and also amorphous
silica-based catalysts. Supported aryl bromide 4 was prepared
by coupling of 4-bromobenzoyl chloride with aminopropyl

J.D. Webb et al. / Journal of Catalysis 252 (2007) 97–109
103
(A)
(B)
Fig. 4. SBA-15-SH·Pd after use (argon, 4 h).
(A)
(B)
Fig. 5. TEM image of SBA-15-SH·Pd after reaction in air.

104
J.D. Webb et al. / Journal of Catalysis 252 (2007) 97–109
triethoxysilane-modified silica according to the method of
Corma et al. Bromoacetophenone (1) and phenyl boronic acid
(2) (in excess) are dissolved in the solution. Finally, the immo-
bilized catalyst is the third phase (Scheme 1). In order to test
for the presence of catalytically active Pd in solution, supported
halide 4 and soluble halide 1 are reacted with phenyl boronic
acid under typical reaction conditions. The soluble fraction is
monitored for the presence of 4-phenylacetophenone (3) which
indicates the generation of an active catalyst. The silica sup-
ported reagent is then separated from the solution, and the
amide bond cleaved with base. The critical species in this ex-
periment is compound 6, whose presence gives an indication of
the amount of active Pd that has left the surface of SBA-15 and
has reached the solid-phase reagent 4.
As shown in Table 4, all three materials gave results consis-
tent with our previous study, namely at reaction times similar to
those normally employed, low levels of 6 were observed, gener-
ally on the order of 2–10% reaction on support. Since the S:Pd
ratio has been proposed to be an important feature in the func-
tional heterogeneity of the material, the three-phase test was
performed on a material prepared with different ratios of Pd to
sulfur. At a S:Pd ratio of 1:1, greatly increased reaction on sup-
port was observed, entry 5, indicating that the three-phase test
was at least qualitatively assessing the escape of catalytically
competent Pd. Alternatively, increasing the S:Pd ratio from 2:1
to almost 3:1 decreased the yield of the homogeneous reaction
compared to the material that had a ratio of only 2:1, while
the ratio of bromobenzoic acid to phenyl benzoic acid ratio re-
mained high (entry 4, 94:6). Higher reaction on support was
also observed for the 2:1 materials when the reaction times were
extended (entries 6 and 7). In addition, greater reaction on sup-
port is observed when the reaction is run in air (entry 8). Finally,
in order to ensure that the reaction is not starved of boronic
acid, the three-phase test was repeated with an even greater ex-
cess (three equivalents) of boronic acid (entry 9). It is clear that
the extent of reaction on support increases, implying that com-
petition between the soluble aryl halide and the supported aryl
halide favors reaction in solution at the expense of reaction on
support. This is an important feature of the three-phase test that
needs to be considered when its results are used to assess cata-
lyst heterogeneity.
(A)
(B)
Fig. 6. Nitrogen physisorption analysis of SBA-15-SH(g)·Pd (A) before use
(blue) and after three reactions (left); and of SBA-15-SH(cc)·Pd (B) before use
and after one run in argon.
One other feature from Table 4 which deserves comment
is with regard to the overall yield of compound 3, resulting
from coupling in solution. In virtually all of the cases where
Scheme 1. Application of the three-phase test to the Suzuki–Miyaura reaction.

J.D. Webb et al. / Journal of Catalysis 252 (2007) 97–109
105
Table 4
showed a higher level of reaction on support indicating greater
leaching or poorer capture of catalytically active material. Also
consistent with the results previously obtained, the leaching be-
comes more evident as the reaction is left to a longer time, such
that at 24 h with the grafted mesoporous catalyst, substantial
(40%) reaction was observed on the supported aryl halide (en-
try 4).
Thus the three-phase data, when repeated under more forc-
ing conditions, give a picture of a catalyst that does show ac-
tivity in solution at varying levels depending on the conditions.
In order to benchmark the catalyst vs other known systems, we
attempted the reaction with Pd/C under our now standard con-
ditions. This unfortunately was not successful since Pd/C was
not an effective catalyst in water (1% yield after 5 h).
Three-phase test for assessment of the presence of catalytically active Pd in
solution
a
Entry Catalyst
S:Pd Reaction
Yield of 3 5:6
(%)
ratio
conditions
1
2
3
4
5
6
7
8
9
SBA-15-SH(g)·Pd
2:1
2:1
2:1
3:1
1:1
2:1
2:1
2:1
2:1
5 h, argon
5 h, argon
5 h, argon
5 h, argon
5 h, argon
13 h, argon
20 h, argon
20 h, air
48
42
94
45
93
55
42
56
98:2
SBA-15-SH(cc)·Pd
90:10
90:10
94:6
65:35
82:18
85:15
58:42
73:27
SiO -SH·Pd
2
SiO -SH·Pd
2
SiO -SH·Pd
2
SBA-15-SH(g)·Pd
SBA-15-SH(cc)·Pd
SBA-15-SH(cc)·Pd
SBA-15-SH(cc)·Pd
b
5 h, air, excess 69
PhB(OH)
2
a
Reaction conditions: 0.75 mmol PhBPin, 0.51 mmol BrAcPh (4-bromo-
Therefore, we assessed the heterogeneity of the catalyst in
other solvents, particularly in combinations of DMF and water
as this is a commonly employed solvent for Suzuki–Miyaura re-
actions, and was employed for many of the reactions described
in this study. In this case, the reaction was much more sensitive
to the presence of the supported aryl halide and initially no con-
version was observed in either solution or on support. However,
the use of the amide 8 along with gentle stirring to prevent me-
chanical decomposition of the catalyst did give a system which
was catalytically competent, providing compound 3 in reason-
able yields. Remarkably, in terms of the three-phase test, the
trends observed in DMF/water were quite different from those
observed in water. Again the yields of the homogeneous reac-
tion were moderate to good, but the yield on support did not
increase as expected, even after extended time. Thus using the
grafted catalyst, a 74% yield of 3 was obtained after 5 h and a
94:6 ratio of 4:5 after hydrolysis. Extending the reaction time
to 24 h increased the yield of 3 to 90%, however the ratio of
4:5 remained at 94:6. Since this was inconsistent with the re-
sults we obtained in water, where increased time always led to
increased reaction on support, we carried out the reaction with
Pd/C, which is known to act via leached Pd species. Again we
found only low levels of reaction on support (94:6) indicating
that in DMF/water, the three-phase test is not meaningful for
this system, since Pd/C is known to leach active Pd.
acetophenone), 1 mol% SiO -SH·Pd, 1.02 mmol K CO , 0.51 mmol p-di-
2
2
3
methoxybenzene as internal standard, 20/1 DMF/H O (0.1 M in BrAcPh),
2
◦
80 C.
b
Reaction performed with 3 equivalents of boronic acid.
Table 5
a
Three-phase test using supported amide 8
Entry
Catalyst
S:Pd
ratio
Reaction
time (h)
Yield of 3
(%)
5:6
1
2
3
4
SBA-15-SH(g)·Pd
SBA-15-SH(cc)·Pd
2:1
2:1
2:1
2:1
5
5
5
74
63
95
90
94:6
94:6
60:40
60:40
SiO -SH·Pd
2
SBA-15-SH(g)·Pd
See Section 2 for details.
24
a
mesoporous silica is employed as the support, lower yields are
observed in the presence of the supported aryl halide than they
are in its absence. This led us to consider the possibility that any
free amines on the surface of the supported aryl halide were
acting alternatively as a sink for Pd that is released from the
support, or as a poison for active soluble Pd. Based on elemen-
tal analysis, the supported amide reaction was incomplete and
ca. 0.3 mmol/g of free NH2 groups were present on the sup-
port. Because of the small amount of Pd that is added and the
large amount of the supported reagent, the ratio of free amine
to Pd was 35:1.
Searching for an alternative method to assess the cat-
alytic activity, we carried out the reaction in the presence of
Quadrapure-TU and polyvinylpyridine, two of the most com-
monly employed poisons. In addition, the effect of adding
aminopropyl-modified silica was also examined, as it mim-
icked the effect of the three-phase material. For reactions in
DMF/water, the addition of Quadrapure-TU had a moder-
ate effect, slowing but not stopping the reaction as shown in
Fig. 7. Interestingly, with the exception of PL-thiourea, the
other scavengers took approximately 5 h before the reaction
was completely stopped, although the initial rate of reaction
was decreased. As shown by monitoring the reaction at ex-
tended times, the reaction continued to occur with PVP and
Quadrapure-TU, although at a greatly diminished rate. This is
likely attributable to slow release of Pd from the surface of
these scavengers. Similar results were observed for the grafted
catalyst (see Supporting information). With this catalyst, the
addition of silica-supported amines and thiols [73] were also
Considering these facts, a new supported reagent was pre-
pared such that no free amines would be on the surface (Eq. (2)).
Concurrent with this study, we assessed the ability of a variety
of poisons to inhibit the reaction (vide infra).
(2)
Using the new supported aryl halide, we were able to achieve
higher catalytic activity in the soluble portion of the three-phase
test (Table 5). Interestingly, the amorphous silica catalyst again

106
J.D. Webb et al. / Journal of Catalysis 252 (2007) 97–109
Fig. 7. Effect of various polymeric poisons on the activity of SBA-15-SH(cc)·Pd.
with the 1:1 catalyst compared with the 3:1 catalyst which is in
some ways counterintuitive based on the fact that greater leach-
ing and solution activity is undoubtedly occurring for the 1:1
S:Pd catalyst.
However, based on the kinetics of the scavenging, the correct
interpretation of these results is likely that a greater degree of
reaction has occurred in solution for the 1:1 catalyst before the
scavenger has a chance to become fully active.
4. Discussion
The development of Pd catalysts that do not lead to contam-
ination of organic products at the end of a reaction is an impor-
tant advance in the preparation of pharmaceutical compounds
because of the great difficulty associated with removing Pd
from these often highly functionalized compounds [15]. How
these catalysts actually function is another question since many
have been shown to act more correctly as reservoirs for soluble
Pd, which may be solubilized as the oxidative addition prod-
uct or as small Pd nanoclusters [74,75]. A complicating factor
in the analysis of the heterogeneity of Pd catalysts is that the
reaction which is generally employed as a test is the Mizoroki–
Heck reaction, which is often run at temperatures as high as
140 ◦C [39]. Considering the large number of Pd complexes
that function as asymmetric catalysts for this reaction [76,77]
and others, the possibility of preparing a supported catalyst that
remains bound to the surface seems to be only a question of
choosing the correct ligand and conditions. However, the high
activity of leached Pd [74] remains a problem and thus the ques-
tion of whether any activity can be ascribed to surface-bound Pd
for catalysts reported to date is questionable.
Fig. 8. Effect of Quadrapure-TU on the reaction of the amorphous silica pre-
pared at different sulfur to Pd ratios.
attempted at ratios of 35:1 (heteroatom:Pd) in order to mimic
the amount of unreacted amine present on the surface of the
supported aryl halide. These tests resulted in a complete cessa-
tion of catalytic activity, confirming the scavenging capability
of these materials, and the likelihood that the reaction was char-
acterized by a release and catch mechanism. This analysis also
explains the results obtained with respect to the thiol:Pd ratio
as described in Table 1. At ratios such as 8:1, Pd is held more
efficiently on the support and is then not able to escape into
solution.
Examining the effect of Quadrapure-TU on the reaction of
the amorphous silica prepared at different sulfur to Pd ratios,
provided an interesting trend (Fig. 8). The yield of the desired
product was assessed after 300 min, a time at which the un-
poisoned reaction is usually complete. As shown in Fig. 8, a sig-
nificantly greater yield was observed for the reaction performed
Several tests have been developed to assess the activity of
surface vs homogeneous Pd. In fact, most of these tests deter-
mine only whether activity is present in solution, not whether

J.D. Webb et al. / Journal of Catalysis 252 (2007) 97–109
107
active species are present on support. One such test is the hot-
filtration test, where activity of a solution after removal of the
supposedly supported catalyst is examined. The presence of
high levels of activity is then indicative of the presence of sol-
uble Pd species. However, this test can be complicated by the
fast redeposition of Pd on the surface which means that the ob-
servation of no activity in solution is not an indicator of surface
catalysis. One of the tests which proved useful in discerning
whether or not any activity could be attributed to surface-bound
Pd in the case of pincer ligands is the polymeric-poison chal-
lenge test. In this approach, an insoluble poison is added to the
reaction to trap any reactive Pd species, and then cessation of
reaction is taken as evidence that the reaction was indeed cat-
alyzed by leached Pd. This test also has some difficulties, since
the polymeric poisons have been shown to bind to surfaces and
inhibit reaction merely by preventing access to the surface [34],
although the magnitude of this effect can be assessed in part by
examining the effect of added polystyrene on the reaction. In
addition, the quantity of the poison that is needed to inhibit the
reaction may need to be optimized. Most importantly, however,
for catalysts in which there is some equilibrium between solu-
ble and bound Pd, these poisons may serve only to scavenge Pd
from the support and hold it in an inactive form. Thus cessation
of activity may not differentiate between the presence of active
Pd in solution and the mere presence of Pd in solution. A better
test of whether persistent catalytically active material is present
in solution has been believed to be the three-phase test.
have given divergent results. Thus the addition of poisons to the
reaction mixture slows or stops the solution reaction while the
three-phase test shows the presence of only small amounts of
active Pd in solution. This led us to examine the three-phase
test in more detail. As described in the results section, the three-
phase test does give an accurate picture of the reaction, but only
under appropriate conditions. Key features to consider are that
the reaction between the supported reagent and leached Pd oc-
curs at a decreased rate compared with reaction between Pd
and soluble reagents. Thus, in order to observe reaction on sup-
port, it is important to run reactions to extended times/higher
conversions, and also to increase the amount of coupling part-
ner (in this case boronic acid) to ensure the availability of this
reagent for coupling on support. Finally, it is instructive to
probe the three-phase test with an appropriate homogeneous
catalyst, or, as we demonstrated, with a heterogeneous catalyst
that is known to leach Pd. In our case, we employed Pd/C to test
the effectiveness of the three-phase test which showed that the
test is unreliable in DMF/water, possibly due to differential sol-
vation of the various surfaces involved. Although this did not
prove feasible in water, the response of the three-phase test to
imply that
prolonged reaction times and increases in PhB(OH)2
the test is effective in this medium, and that it provides at least
a qualitative measure of the amount of active Pd that leaches
from the surface.
5. Conclusions
In this test, an aryl halide is tethered to the surface of a sup-
port and conversion on this support is deemed to result from the
loss of catalytically active Pd from the surface. In this way, the
three-phase test is superior to the hot filtration test, since the
supported aryl halide is present at the time of the reaction, no
filtration step is necessary. However, a fact which is often ig-
nored is that in its original inception, the three-phase test was
designed to capture intermediates from solution which had a
certain lifetime. In the case of Pd catalysis, the Pd species must
be sufficiently long-lived to diffuse from the surface on which it
was originally supported and reach the supported reagent. Other
factors which need to be considered are solvation of the two
surfaces which may inhibit reaction at the surface, and the in-
herently slower rates of reaction of insoluble reagents. This will
likely change with the nature of the Pd species involved, par-
ticularly as a function of the aryl halide since the ease of the
individual steps in the catalytic cycle will control the nature of
the most abundant catalytic species (MACS).
For example, based on the work of Reetz, the oxidatively
added Pd(II) species is likely the MACS when aryl iodides are
employed as the substrate [75]. With aryl bromides, there is ev-
idence that the transmetallation step is also the rate determining
step [28], and with aryl chlorides, a completely different step
controls the nature of the resting state of the catalyst [78]. The
solubility and the rate of deposition of the species that is most
abundant will then have a significant impact on the observation
of soluble Pd. With few exceptions [28,39,50,75], this point has
not received much attention in the literature.
In conclusion, the recyclability, stability and heterogeneity
of thiol-modified silicates have been examined. It was shown
that depending on the reaction conditions, thiol-modified ma-
terials which have been loaded with Pd can be highly active
and recyclable catalysts. The method of introducing the thiol is
critical to maintain structural integrity, since catalysts prepared
by co-condensation undergo rapid reduction in surface area and
mesoporosity upon exposure to aqueous base. This loss of order
is coincident with a decrease of catalytic activity. In contrast,
the material prepared by grafting of the thiol on the surface of
pre-formed support retains most of its activity and structure up
to 5 runs, although slight decreases in both are noted. Com-
pared with amorphous silica, this particular catalyst is the most
recyclable among those examined.
The effectiveness of the various heterogeneity tests de-
scribed in the literature has also been probed, and it was demon-
strated that like many other heterogeneity tests, both the three-
phase test and the poison-challenge test need to be interpreted
carefully. In particular, it is important to validate the three-phase
test with catalysts known to leach Pd, or with conditions known
to promote leaching. These results, in concert with poison-
challenge tests can provide at least a qualitative picture of cat-
alyst activity and heterogeneity. In the case of thiol-modified
mesoporous silicates, they indicate that the material likely func-
tions in a release and catch mechanism due to the strong binding
of the Pd to the thiolated surface and the effectiveness of this
material as a scavenger for Pd.
Interestingly, in the case of mercaptopropyl trialkoxysilane-
modified silicates as supports for Pd, various heterogeneity tests

108
J.D. Webb et al. / Journal of Catalysis 252 (2007) 97–109
Acknowledgments
[29] K. Kohler, W. Kleist, S.S. Prockl, Inorg. Chem. 46 (2007) 1876.
[30] C.A. Jaska, T.J. Clark, S.B. Clendenning, D. Grozea, A. Turak, Z.H. Lu,
I. Manners, J. Am. Chem. Soc. 127 (2005) 5116.
[31] C.M. Hagen, J.A. Widegren, P.M. Maitlis, R.G. Finke, J. Am. Chem.
Soc. 127 (2005) 4423.
[32] D.E. Bergbreiter, Chem. Rev. 102 (2002) 3345.
[33] M.H. Guino, K.K. Hill, Chem. Soc. Rev. 36 (2007) 608.
[34] M. Weck, C.W. Jones, Inorg. Chem. 46 (2007) 1865.
[35] W.J. Sommer, M. Weck, Coord. Chem. Rev. 251 (2007) 860.
[36] M. Beller, H. Fischer, W.A. Herrmann, K. Ofele, C. Brossmer, Angew.
Chem. Int. Ed. 34 (1995) 1848.
Johnson and Johnson is acknowledged for support of this
research in terms of a Focused Giving Grant. The Natural Sci-
ences and Engineering Research Council of Canada (NSERC)
is acknowledged for support of this research in terms of oper-
ating and equipment grants to C.M.C., a postgraduate doctoral
scholarship to K.M., postgraduate and Canada doctoral schol-
arships to J.W.
[37] M. Ohff, A. Ohff, D. Milstein, Chem. Commun. (1999) 357.
[38] D.A. Alonso, C. Najera, M.C. Pacheco, Adv. Synth. Catal. 344 (2002) 172.
[39] J.G. de Vries, Dalton Trans. (2006) 421.
Supporting information
[40] I.P. Beletskaya, A.N. Kashin, N.B. Karlstedt, A.V. Mitin, A.V. Cheprakov,
G.M. Kazankov, J. Organomet. Chem. 622 (2001) 89.
[41] J. Louie, J.F. Hartwig, Angew. Chem. Int. Ed. 35 (1996) 2359.
[42] C.S. Consorti, G. Ebeling, F.R. Flores, F. Rominger, J. Dupont, Adv.
Synth. Catal. 346 (2004) 617.
[43] M.R. Eberhard, Org. Lett. 6 (2004) 2125.
[44] D. Olsson, P. Nilsson, M. El Masnaouy, O.F. Wendt, Dalton Trans. (2005)
1924.
[45] W.J. Sommer, K.Q. Yu, J.S. Sears, Y.Y. Ji, X.L. Zheng, R.J. Davis, C.D.
Sherrill, C.W. Jones, M. Weck, Organometallics 24 (2005) 4351.
[46] K.Q. Yu, W. Sommer, J.M. Richardson, M. Weck, C.W. Jones, Adv. Synth.
Catal. 347 (2005) 161.
Characterization data (physisorption analysis, pXRD analy-
sis) of SBA-15-SH·Pd before and after use, conversion vs time
plots for the reaction of amorphous silica at different S:Pd ra-
tios, for SBA-15-SH(g) and for these two materials in the pres-
ence of various poisons can be found in the online version of
this article.
Please visit DOI: 10.1016/j.jcat.2007.09.007.
References
[47] K.Q. Yu, W. Sommer, M. Weck, C.W. Jones, J. Catal. 226 (2004) 101.
[48] W.A. Herrmann, C. Brossmer, K. Ofele, M. Beller, H. Fischer, J.
Organomet. Chem. 491 (1995) C1.
[49] M. Nowotny, U. Hanefeld, H. van Koningsveld, T. Maschmeyer, Chem.
Commun. (2000) 1877.
[50] C.S. Consorti, F.R. Flores, J. Dupont, J. Am. Chem. Soc. 127 (2005)
12054.
[51] D.E. Bergbreiter, P.L. Osburn, J.D. Frels, Adv. Synth. Catal. 347 (2005)
172.
[1] L. Yin, J. Liebscher, Chem. Rev. 107 (2007) 133.
[2] J.M. Thomas, R. Raja, J. Organomet. Chem. 689 (2004) 4110.
[3] C. Gonzalez-Arellano, A. Corma, M. Iglesias, F. Sanchez, Adv. Synth.
Catal. 346 (2004) 1758.
[4] C. Baleizao, A. Corma, H. Garcia, A. Leyva, J. Org. Chem. 69 (2004) 439.
[5] Q.M. Yang, S. Ma, J. Li, F. Xiao, H. Xiong, Chem. Commun. (2006) 2495.
[6] J. Demel, J. Cejka, S. Bakardjieva, P. Stepnicka, J. Mol. Catal. A
Chem. 263 (2007) 259.
[7] D. Astruc, Inorg. Chem. 46 (2007) 1884.
[52] J.A. Widegren, R.G. Finke, J. Mol. Catal. A Chem. 198 (2003) 317.
[53] R. Akiyama, S. Kobayashi, J. Am. Chem. Soc. 125 (2003) 3412.
[54] H. Hagio, M. Sugiura, S. Kobayashi, Org. Lett. 8 (2006) 375.
[55] W. Solodenko, H.L. Wen, S. Leue, F. Stuhlmann, G. Sourkouni-Argirusi,
G. Jas, H. Schonfeld, U. Kunz, A. Kirschning, Eur. J. Org. Chem. (2004)
3601.
[8] C.P.W. Mehnert, D.W. Weaver, J.Y. Ying, J. Am. Chem. Soc. 120 (1998)
12289.
[9] J.Y. Ying, C.P. Mehnert, M.S. Wong, Angew. Chem. Int. Ed. 38 (1999) 56.
[10] A. Corma, Chem. Rev. 97 (1997) 2373.
[11] J.G. de Vries, Can. J. Chem. (2001) 79.
[12] V. Farina, Adv. Synth. Catal. 346 (2004) 1553.
[13] F.X. Felpin, T. Ayad, S. Mitra, Eur. J. Org. Chem. (2006) 2679.
[14] M. Seki, Synthesis (2006) 2975.
[56] C. Thiot, M. Schmutz, A. Wagner, C. Mioskowski, Angew. Chem. Int.
Ed. 45 (2006) 2868.
[57] M.D. Smith, A.F. Stephan, C. Ramarao, P.E. Brennan, S.V. Ley, Chem.
Commun. (2003) 2652.
[58] S.P. Andrews, A.F. Stepan, H. Tanaka, S.V. Ley, M.D. Smith, Adv. Synth.
Catal. 347 (2005) 647.
[15] C.E. Garrett, K. Prasad, Adv. Synth. Catal. 346 (2004) 889.
[16] V.W. Rosso, D.A. Lust, P.J. Bernot, J.A. Grosso, S.P. Modi, A. Rusow-
icz, T.C. Sedergran, J.H. Simpson, S.K. Srivastava, M.J. Humora, N.G.
Anderson, Org. Proc. Res. Dev. 1 (1997) 311.
[59] N. Ren, Y.H. Yang, Y.H. Zhang, Q.R. Wang, Y. Tang, J. Catal. 246 (2007)
215.
[60] C.K.Y. Lee, A.B. Holmes, S.V. Ley, I.F. McConvey, B. Al-Duri, G.A.
Leeke, R.C.D. Santos, J.P.K. Seville, Chem. Commun. (2005) 2175;
I.R. Baxendale, C.M. Griffiths-Jones, S.V. Ley, G.K. Tranmer, Chem. Eur.
J. 12 (2006) 4407.
[17] K. Konigsberger, G.P. Chen, R.R. Wu, M.J. Girgis, K. Prasad, O. Repic,
T.J. Blacklock, Org. Proc. Res. Dev. 7 (2003) 733.
[18] I.W. Davies, L. Matty, D.L. Hughes, P.J. Reider, J. Am. Chem. Soc. 123
(2001) 10139.
[19] N.T.S. Phan, M. Van Der Sluys, C.W. Jones, Adv. Synth. Catal. 348 (2006)
609.
[61] C. Ramarao, S.V. Ley, S.C. Smith, I.M. Shirley, N. DeAlmeida, Chem.
Commun. (2002) 1132.
[62] S.J. Broadwater, D.T. McQuade, J. Org. Chem. 71 (2006) 2134.
[63] K. Shimizu, S. Koizumi, T. Hatamachi, H. Yoshida, S. Komai, T. Kodama,
Y. Kitayama, J. Catal. 228 (2004) 141.
[20] F.Y. Zhao, K. Murakami, M. Shirai, M. Arai, J. Catal. 194 (2000) 479.
[21] F.Y. Zhao, M. Shirai, M. Arai, J. Mol. Catal. A Chem. 154 (2000) 39.
[22] F.Y. Zhao, M. Shirai, Y. Ikushima, M. Arai, J. Mol. Catal. A Chem. 180
(2002) 211.
[23] R.G. Heidenreich, K. Kohler, J.G.E. Krauter, J. Pietsch, Synlett (2002)
1118.
[24] R.G. Heidenreich, E.G.E. Krauter, J. Pietsch, K. Kohler, J. Mol. Catal. A
Chem. 182 (2002) 499.
[25] M. Lysen, K. Kohler, Synthesis (2006) 692.
[64] (a) C.M. Crudden, M. Sateesh, R. Lewis, J. Am. Chem. Soc. 127 (2005)
10045;
(b) C.M. Crudden, K. McEleney, S.L. MacQuarrie, A. Blanc, M. Sateesh,
J.D. Webb, Pure Appl. Chem. 79 (2007) 247.
[26] M. Lysen, K. Kohler, Synlett (2005) 1671.
[27] S.S. Prockl, W. Kleist, M.A. Gruber, K. Kohler, Angew. Chem. Int. Ed. 43
(2004) 1881.
[28] D.A. Conlon, B. Pipik, S. Ferdinand, C.R. LeBlond, J.R. Sowa, B. Izzo, P.
Collins, G.J. Ho, J.M. Williams, Y.J. Shi, Y.K. Sun, Adv. Synth. Catal. 345
(2003) 931.
[65] R.B. Bedford, U.G. Singh, R.I. Walton, R.T. Williams, S.A. Davis, Chem.
Mater. 17 (2005) 701.
[66] (a) D. Zhao, J. Feng, Q. Huo, N. Melosh, G.H. Fredrickson, B.F. Chmelka,
G.D. Stucky, Science 279 (1998) 548;
(b) D. Zhao, Q. Huo, J. Feng, B.F. Chmelka, G.D. Stucky, J. Am. Chem.
Soc. 120 (1998) 6024.

J.D. Webb et al. / Journal of Catalysis 252 (2007) 97–109
109
[67] Since the concentration of Pd quoted in ppm/ppb is dependent on the
amount of Pd initially added (lower levels at lower catalytic loadings) and
the volume of solvent, Pd leaching will be quoted as a % of the initially
added Pd.
protective layer has been shown to dramatically improve the stability of
this material (K. McEleney, C.M. Crudden, manuscript in preparation).
[72] (a) J. Rebek, F. Gavina, J. Am. Chem. Soc. 97 (1975) 3453;
(b) J. Rebek, F. Gavina, J. Am. Chem. Soc. 96 (1974) 7112.
[73] This test has been previously reported with amorphous silica supported
thiol as a test of heterogeneity by M.V. Parisien, D. Valette, K. Fagnou, J.
Org. Chem. 70 (2005) 7578, and recently by J.M. Richardson, C.W. Jones,
J. Catal. 251 (2007) 80.
[74] A.H.M. de Vries, J.M.C.A. Mulders, J.H.M. Mommers, H.J.W. Hender-
ickx, J.G. de Vries, Org. Lett. 5 (2003) 3285.
[75] M.T. Reetz, E. Westermann, Angew. Chem. Int. Ed. 39 (2000) 165.
[76] A.B. Dounay, L.E. Overman, Chem. Rev. 103 (2003) 2945.
[77] M. Shibasaki, E.M. Vogl, T. Ohshima, Adv. Synth. Catal. 346 (2004) 1533.
[78] F. Barrios-Landeros, J.F. Hartwig, J. Am. Chem. Soc. 127 (2005) 6944.
[68] (a) The thiol content in both materials was determined by elemental analy-
sis. Only materials with the same % thiol were compared;
(b) Note that these results were performed in air. The co-condensation-
derived catalyst displays even less stability in argon.
[69] F. Zhang, Y. Yan, H. Yang, Y. Meng, C. Yu, B. Tu, D. Zhao, J. Phys. Chem.
B 109 (2005) 8723–8732.
[70] The catalysts are also somewhat sensitive to mechanical agitation.
[71] This effect is likely due to the protective effects of the grafting layer in-
sulating the silica against attack by the aqueous base. The addition of a