A study into Stille cross-coupling reaction mediated by palladium catalysts deposited over siliceous supports bearing N-donor groups at the surface

Article abstract of DOI:10.1002/aoc.2987

Mesoporous molecular sieve SBA-15 and conventional silica gel were grafted with 3-[2-(diethylamino)ethylamino]propyl groups and used as supports for the preparation of deposited palladium catalysts, which were subsequently evaluated in Stille cross-coupling reaction of aryl bromides. The highest conversions were achieved with the metal-saturated catalyst resulting from the modified SBA-15. Influence of various reaction parameters (solvent, additive, temperature, etc.) on the overall yield and selectivity was studied. Copyright

Full text of DOI:10.1002/aoc.2987

Full Paper
Received: 2 January 2013
Revised: 28 February 2013
Accepted: 28 February 2013
Published online in Wiley Online Library: 29 April 2013
(wileyonlinelibrary.com) DOI 10.1002/aoc.2987
A study into Stille cross-coupling reaction
mediated by palladium catalysts deposited
over siliceous supports bearing N-donor
groups at the surface
a,b
b
a
ꢀ
ꢀ
ꢀ
ꢀ
Miloslav Semler , Jirí Cejka and Petr Štepnicka *
Mesoporous molecular sieve SBA-15 and conventional silica gel were grafted with 3-[2-(diethylamino)ethylamino]propyl
groups and used as supports for the preparation of deposited palladium catalysts, which were subsequently evaluated in Stille
cross-coupling reaction of aryl bromides. The highest conversions were achieved with the metal-saturated catalyst resulting
from the modified SBA-15. Influence of various reaction parameters (solvent, additive, temperature, etc.) on the overall yield
and selectivity was studied. Copyright © 2013 John Wiley & Sons, Ltd.
Keywords: palladium; supported catalysts; Stille reaction; X-ray diffraction; TEM measurements
extends our work toward the palladium-catalyzed Stille reaction,
which is, despite the generally high toxicity of organotin com-
pounds, widely used owing to its high functional group tolerance
and complementary character to other cross-coupling reactions.
Introduction
Palladium-catalyzed cross-coupling reactions (e.g. Suzuki–Miyaura,[1]
Heck–Mizoroki[2] and Stille[3] reactions) are frequently employed in
the large-scale synthesis of fine chemicals and pharmaceutically ac-
tive substances.[4] In view of their manifold practical applications,
numerous attempts have been made to develop highly active
and reusable noble metal catalysts for these reactions, typically
those based on palladium immobilized over siliceous supports,
polymers and other inert materials.[5] Silicate and aluminosilicate
supports represented by conventional (amorphous) silica gel,
mesoporous molecular sieves and zeolites proved to be particularly
attractive for this purpose owing to their high structural variability,
relatively easy functional modification as well as thermal and
chemical stability.[6]
It should be noted that several reports are available in the
literature describing the use of Pd catalysts deposited over
siliceous supports in model Stille reactions. For instance, Cai and
coworkers[12] described the use of Pd catalysts heterogenized on
3-mercaptopropylated MCM-41, and Jin et al. reported on Stille
coupling mediated by a Pd catalyst deposited on silica gel modified
with chelating N,O-donor groups.[13] Carbonylative variant of the
Stille reaction was performed with analogous catalysts bearing P,
N-donor groups.[14] In our contribution we focus on model
reactions, paying particular attention to the influence of various
reaction parameters on the reaction outcome.
The deposition of the active metal component is typically
performed by means of attached donor groups (i.e. by coordination
of various donor groups attached to the support’s surface) or
without these ligating moieties. The latter approach based on
in-synthesis[7] or post-synthesis palladation (and subsequent
reduction to Pd(0) if appropriate) affords materials showing
typically worse catalytic results than those resulting by the
ligand-assisted metal deposition. The other approach requires
the presence of donor groups suitable for palladium coordina-
tion, which can be introduced either during the synthesis of
the support or by post-synthetic modifications of the surface of
the parent solid material. The donor-modified supports are
subsequently palladated and can be applied in catalysis (directly
or after reduction of Pd(II) to Pd(0)).[8] Covalent anchoring of the
donor groups is usually achieved via the reaction of the support
with appropriate alkoxysilanes, followed by further reactions
if required.[9]
Results and Discussion
Preparation and Characterization of the Catalysts
Catalysts for this study were prepared using two different
supports: the all-siliceous mesoporous molecular sieve SBA-15
(1 in Scheme 1)[11] and conventional, chromatography-grade
silica gel (size fraction 63–200 mm). The surfaces of these
supports were first functionalized with 3-chloropropyl groups
ꢀ
ꢀ
*
Correspondence to: P. Štepnicka, Department of Inorganic Chemistry, Faculty
of Science, Charles University in Prague, Hlavova 2030, 128 40 Prague, Czech
Republic. E-mail: stepnic@natur.cuni.cz
a Department of Inorganic Chemistry, Faculty of Science, Charles University in
Prague, 128 40 Prague, Czech Republic,
In our work, we used N-donor groups mostly because of their
stability, easy accessibility and favorable coordination proper-
ties. Materials of this type were previously tested in C–C bond-
forming and dehalogenation reactions.[10,11] This contribution
b J. Heyrovský Institute of Physical Chemistry, Academy of Sciences of the Czech
Republic, v.v.i., 182 23 Prague, Czech Republic
Appl. Organometal. Chem. 2013, 27, 353–360
Copyright © 2013 John Wiley & Sons, Ltd.

M. Semler et al.
Scheme 1. Preparation of Pd catalysts 6 and 7. Legend: i, (3-chloropropyl)triethoxysilane in toluene, refluxing (repeated twice); ii, treatment with
Et2NCH2CH2NH2 in toluene and then with ammonia in ethanol–water; iii, Pd(OAc)2 in dichloromethane.
via treatment with an excess of (3-chloropropyl)triethoxysilane
in refluxing toluene. This procedure was repeated twice to ensure
a full coverage of the accessible surface with the modifying
groups. The 3-chloropropyl groups were subsequently converted
to 3-[2-(diethylamino)ethylamino]propyl ones by alkylation with
N,N-diethyl-1,2-diaminoethane and subsequent neutralization of
the intermediate ammonium salt with ammonia in an ethanol–
water solution to give materials 4 and 5.
In the last step, the aminated supports were loaded with palla-
dium using palladium(II) acetate dissolved in dichloromethane.
Two catalysts were prepared from each modified support: one
containing ~0.09 mmol Pd per gram of the support (6a and 7a)
and its metal-saturated counterpart (6b and 7b). Considering that
catalysts 6a and 7a resulted upon the addition of 0.1 mmol Pd
(OAc)2 per gram of the support, the amount of palladium in the
final products indicates that virtually all metal component was
anchored during the metalation step. Owing to the relatively low
metal loading, materials 6a and 7a contain an excess of amine
groups, the N/Pd ratios being 7.61 for 6a and 10.20 for 7a,
containing more functional groups. The metal-saturated catalysts
differed significantly in the amount of deposited palladium
(6b: 0.31 mmol Pd gÀ1; 7b: 0.45 mmol Pd gÀ1), which is indeed in
accordance with the different amounts of the grafted donor
groups. On the other hand, both metal-saturated catalysts exerted
similar N/Pd values (1.97 for 6b, and 1.87 for 7b), approaching the
limiting value of 2 corresponding to one palladium atom per one
potentially chelating Et2NCH2CH2N(CH2)3Siꢀ donor group.
The most intense diffraction line at 2θ = 0.87ꢁ was attributed to
diffraction (0 0 1), whereas the relatively weaker ones at 1.47ꢁ and
1.64ꢁ could be assigned to reflections (1 1 0) and (2 0 0), respec-
tively. The XRD patterns, which remained after modification practi-
cally unaffected except for an insignificant shift of the diffraction
peaks, further suggested that the stepwise modification of the solid
support does not affect internal structure of the crystalline fraction.
In addition, the XRD patterns corroborated the absence of
nanosized metallic palladium particles in catalysts 6 and 7 by show-
ing no diffraction peaks in the region of palladium (1 1 1) diffrac-
tion at 2θ ꢂ 40ꢁ.
Adsorption–desorption isotherms of parent SBA-15 and catalyst
6a are shown in Fig. 3 and the calculated textural parameters are
given in Table 1. The isotherms of SBA-15-based materials
displayed the characteristic hysteresis loop typical for mesoporous
materials with cylindrical pores.[16] The fact that the adsorption and
desorption branches did not overlap in the high-pressure region
(p/p0 ~0.8–1) indicates some structural defects and, perhaps, the
presence of pores of different size. In the series of SBA-15 type
materials, the calculated BET (Brunauer–Emmett–Teller) surfaces de-
creased along the reaction sequence, particularly during the
amination step which, indeed, can reflect
a
partial
decomposition of the support (base hydrolysis). As a result, the
surface areas in catalysts 6 as the final products were compara-
ble to those of 7 prepared from amorphous silica. On the other
hand, the average pore size determined according to the BJH
(Barrett–Joyner–Halenda) method[17] remained practically constant
(~ 8 nm) for all SBA-15 type materials.
The prepared materials were characterized by elemental
analysis (C, H, N and Cl by conventional combustion method,
Pd by optical emission spectroscopy), transmission electron
microscopy (TEM), powder X-ray diffraction (XRD), IR spectroscopy
and by nitrogen adsorption isotherms. The calculated textural
parameters and XRD data are summarized in Table 1.
TEM micrographs of material 6b (Fig. 1) revealed the
well-ordered porous structure of the siliceous support and the
absence of nanosized metallic particles, which is in accordance with
the powder XRD measurements (Fig. 2). The pattern observed for
the parent SBA-15 (material 1) consisted of three characteristic
diffraction lines, typical for hexagonal SBA-15 type material.[15]
Catalytic Tests
The initial catalytic experiments aimed at surveying the influ-
ence of the reaction conditions were performed for the coupling
of phenyltributyltin (8) with 4-bromotoluene (9) as this reaction
allows one to distinguish between the cross-coupling
(4-methylbiphenyl; 10) and homocoupling products (biphenyl 11;
Scheme 2). First, the series of catalysts 6 and 7 at 0.5 mol%
palladium loading was evaluated in this reaction performed in
N,N-dimethylformamide (DMF) at 80ꢁC. The results achieved
Table 1. Summary of X-ray diffraction patterns and textural parameters
Material
2θ (ꢁ)
BET surface area
BJH pore volume
Average pore
diameter (nm)
(m2 gÀ1
)
(cm3 gÀ1
)
1
0.87, 1.47, 1.64
0.86, 1.46, 1.57
0.82, 1.42, 1.59
0.82, 1.42, 1.56
0.84, 1.45, 1.58
—
678
579
297
311
295
347
312
0.97
0.92
0.86
0.83
0.81
0.46
0.40
7.7
7.7
8.2
7.9
8.0
4.7
4.6
2
4
6a
6b
7a
7b
—
wileyonlinelibrary.com/journal/aoc
Copyright © 2013 John Wiley & Sons, Ltd.
Appl. Organometal. Chem. 2013, 27, 353–360

Deposited Pd catalysts
Figure 1. Representative TEM images of catalyst 6b.
Scheme 2. The model Stille coupling of phenyltributyltin (8) with
4-bromotoluene (9) to give 4-methylbiphenyl (10).
Figure 2. Small-angle X-ray diffraction patterns of materials 1 (solid line)
and 4 (dashed line) and catalyst 6b (dotted line).
Figure 3. BET adsorption isotherms for the parent SBA-15 (1; circles) and
material 6a (triangles). Filled symbols, adsorption; empty symbols, desorption.
with catalysts 6a, 6b and 7b are presented graphically in Fig. 4.
Note: since biphenyl is formed from two molecules of stannane
8, the gas chromatographic (GC) yield expressing the content
of 11 in the reaction mixture was multiplied by two to represent
to the real chemical yield. The data in the figures thus correspond
to the relative ‘rates’ of the two parallel reactions.
The best catalytic results were obtained with the metal-saturated
catalyst obtained from SBA-15-type support, material 6b. This
catalyst achieved nearly 20% yield of the coupling product and
Figure 4. Kinetic profiles for the model Stille coupling mediated by catalysts
6a, 6b and 7b. Material 7a showed no catalytic activity and was therefore
omitted from this figure. Black bars indicate the yield of 4-methylbiphenyl
(10); gray bars indicate the yield of biphenyl (homocoupling product). The bars
are missing in cases where no product was detected.
Appl. Organometal. Chem. 2013, 27, 353–360
Copyright © 2013 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/aoc

M. Semler et al.
~24% yield of biphenyl in 24 h. The related material 6a, having a
lower metal loading, performed considerably worse, affording
the coupling product in only 6% yield after 24 h. Unlike its
metal-saturated counterpart, this catalyst also showed a significant
induction period (~4 h). Catalytic results obtained with 7b resulting
from silica gel were slightly worse than for the analogous material
6b, which contains less palladium per gram of the support than
7b. On the other hand, the metal-unsaturated material 7a
showed, rather unexpectedly, no detectable catalytic activity.
Considering these results, it appears likely that the catalytic
performance is determined mainly by the relative proportion
between the anchored palladium and the amount of the N-donor
moieties (i.e. by the N/Pd ratio). Whereas the metal-
unsaturated catalysts showing high N/Pd ratios exerted rela-
tively low catalytic activity (note: material 7a proved to be inac-
tive), the best results were achieved with the metal-saturated
catalyst 6b, showing a relatively high N/Pd ratio (cf. the N/Pd
ratios: 6a, 7.61; 6b, 1.97; 7a, 10.2; and 7b, 1.87) but a lower
number of anchoring groups than in 7b.
The most reactive catalyst, 6b, was subsequently tested in
different solvents. The reactions performed at 80ꢁC and with
0.5 mol% of palladium in tri-n-butylamine (it should be noted
that Vedejs et al.[18] noted an increase in reactivity upon using
an intramolecularly N-coordinated organotin reagent whereas,
in our case, the amine solvent as an ‘external’ donor stopped
the reaction) and propionitrile did not proceed at all (neither
10 nor biphenyl were detected by GC analysis in the reaction
mixtures after 24 h), while those in n-butanol, di-n-butylether
and toluene achieved 2–5% yields of 10 after 24 h. A similar
reaction in dioxane gave the coupling product in 7% yield. It is
noteworthy that no homocoupling product 11 was detected in
the final reaction mixtures obtained in n-butanol, dioxane and
toluene. Finally, a result comparable to that achieved in DMF
was obtained with N,N-dimethylacetamide (26% GC yield of
the coupling product 10 and ~16% yield of 11).
Figure 5. Influence of fluoride additives on the course of the coupling
reaction. Black bars represent the yield of the coupling product 10, gray bars
that of biphenyl 11. Conditions: PhSnBu3 (1.0 mmol ), 4-bromotoluene
(1.0 mmol), additive (2.0 mmol), mesitylene (1.0 mmol; internal standard)
and catalyst 6b (0.5 mol% Pd), reaction in DMF or DMA (4 ml) at 80ꢁC for
24 h. TBAF, [Bu4N]F.3H2O.
and an increase in the amount of 11 (GC yields: ~32% in DMF
and 33% in DMA).
Influence of the reaction temperature was assessed next for
the model coupling reaction performed in the presence of KF,
which afforded the highest conversions and selectivity among
the additives tested. The results summarized in Fig. 6 indicate
that the best yields of the coupling product 10 are obtained at
80ꢁC, while the selectivity (i.e. the 10/[10 + 11] ratio) increases
with increasing reaction temperature owing to a diminished
formation of the homocoupling product 11 in both DMF and DMA.
The selectivity of the reaction did not change upon changing the
amount of catalyst 6b in the reaction mixture. Upon increasing
the amount of Pd in the reaction mixture from 0.1% to 0.2%,
the yield of the coupling product 10 increased from ~ 20% to
28% (after 1 h at 80ꢁC). Surprisingly, a further increase in the
amount of the catalyst up to 1.0% increased the yield of 10 only
slightly (Fig. 7). This in turn results in a pronounced change in
the overall turnover number (TON = mmol 10/mmol Pd), which
decreases with increasing metal loading (Fig. 7).
In an attempt at increasing the amount of the desired
heterocoupling product, the reaction was also performed with
an excess of phenyltributyltin (2 equiv. with respect to 9; KF
additive) under otherwise identical conditions in DMF. Indeed,
the addition of phenyltributyltin led to an increase in the
amount in 10 in the reaction mixture (42% after 1 h, 48% after
6 h). When the same amount of PhSnBu3 was added sequentially
in two portions (i.e. one equivalent to the initial reaction mixture
followed by another equivalent after heating for 6 h), the reaction
outcome practically did not change (cf. 35% yield of 10 after 12 h).
This observation suggesting that the catalyst loses its activity
dramatically during the course of the catalytic run and also the
large amount of side products present in the solid phase (harmful
The Stille reaction is known to be affected by fluoride salts,[19]
yet their mode of action has not been fully clarified. One of the
hypotheses attributes the better yields achieved in the presence
of fluoride reagents to the formation of insoluble organotin
fluorides, which helps in shifting the reaction equilibrium toward
the side of the products.[20] Another explanation takes into the
account a labilization (increased reactivity) of the Sn–C bond
upon the formation of pentacoordinate tin centres from the
starting stannanes and fluoride ions.[21]
The results obtained after addition of alkali metal fluorides or
tetrabutylammonium fluoride trihydrate (TBAF; 2 equiv. with
respect to 8 and 9) to our catalytic system are presented in Fig. 5.
In the case of reaction carried out in DMF, the addition of sodium,
potassium and cesium fluorides improved the overall catalytic
performance, resulting in higher yields of the coupling product
10 and even the undesired biphenyl 11. A slight decrease in
the yields of both products was observed for LiF, while the addi-
tion of TBAF, which is fully soluble in the reaction mixture, practi-
cally stopped the coupling reaction. A similar situation was seen
in N,N-dimethylacetamide (DMA) except that even LiF improved
(albeit slightly) the reaction outcome in terms of the yields. On the
other hand, KF afforded the best results in both solvents.
When CuI was employed as an additive (10 mol%; other condi-
tions unchanged) to the reaction system, the reaction became
less selective to the desired heterocoupling product 10, as
evidenced by the decrease in the amount of 10 in the reaction
mixture (GC yields: ~10% in DMF and 12% in DMA after 24 h)
wileyonlinelibrary.com/journal/aoc
Copyright © 2013 John Wiley & Sons, Ltd.
Appl. Organometal. Chem. 2013, 27, 353–360

Deposited Pd catalysts
heterogenized catalysts have been devised; in the current work,
we restricted ourselves (also in view of the catalyst stability) to
the ‘hot filtration’ and mercury poisoning tests; see selected
references)[22] simple hot filtration and mercury poisoning tests
were carried out with the representative catalyst 6b. Both these
tests suggested the heterogeneous nature of the catalyzed
process. In the first case, the reaction was allowed to proceed
under the usual conditions for 1 h before the reaction mixture
was centrifuged at 4000 rpm for 5 min and the liquid phase
was filtered while hot through a 0.45 mm PTFE syringe filter.
The filtrate was warmed again and its composition was monitored
by GC analysis. The results are presented in Fig. 8. Whereas the
non-filtered reaction mixture showed the expected kinetic
profile, the removal of the solid phase resulted in marked
decrease in the catalytic activity. In the second case, mercury
(~0.2 ml) was added to the reaction mixture prepared as
described in the Experimental section (see also caption to Fig. 8).
GC analysis performed after reacting at 80ꢁC for 6 h revealed
only ~3% yield of 10 and 1% yield of 11. Theses values are at
the limit of reliable quantification and may well reflect the fact
that the mercury (catalytic poison) was added as the last compo-
nent to the pre-heated reaction mixture.
In addition, the catalyst recovered after a catalytic run
was subjected to powder XRD and TEM analysis. The TEM images
indicated loss of the periodical internal structure of the
mesoporous support. In addition, the micrographs showed dark
oval spots attributable to metallic particles that are presumably
formed via aggregation of fine metallic particles resulting
primarily via reduction of the deposited Pd(II) precursor.
Similarly, the X-ray diffraction patterns of the recovered material
lack the characteristic peaks due to SBA-15 support at low
diffraction angles. Instead, a strong sharp peak is seen at 2θ=6.77ꢁ
accompanied by a weaker peak at 2θ = 9.46ꢁ. A search in
the Powder Data File database showed good agreement of
these values with those reported for ITQ-14 (card no. 01-070-7350)
and VPI-8 (card no. 00-056-0626). These data clearly suggest a
massive structural change of the support to afford other (perhaps
largely amorphous) siliceous phases, very likely due to an action
of KF additive on the parent mesoporous sieve (Fig. 9).
Figure 6. Influence of the reaction temperature on the yield of the coupling
reaction. Black bars represent the yield of the coupling product 10, gray bars
that of biphenyl 11. Conditions: PhSnBu3 (1.0 mmol ), 4-bromotoluene
(1.0 mmol), KF (2.0 mmol), mesitylene (1.0 mmol) and catalyst 6b
(0.5 mol% Pd), reaction in DMF or DMA (4 ml) at 80ꢁC for 24 h.
organotin compounds, base and other additives) led us to abandon
any recyclation tests.
In order to determine the nature of the catalyst (note that
a number of approaches toward establishing the nature of
Finally, catalyst 6b was utilized in reactions of different
substrates under optimized conditions. The results summarized
in Table 2 indicate that the catalyst promotes the reaction of 8 with
Figure 7. Dependence of the yield of the coupling product 10 and TON
on the loading of catalyst 6b after 1 h of reaction. Conditions: PhSnBu3
(1.0 mmol ), 4-bromotoluene (1.0 mmol), KF (2.0 mmol), mesitylene
(1.0 mmol) and catalyst 6b, reaction in DMF (4 ml) at 80ꢁC for 1 h.
Figure 8. Results of the filtration test with catalyst 6b. Conditions: PhSnBu3
(1.0 mmol ), 4-bromotoluene (1.0 mmol), KF (2.0 mmol), mesitylene
(1.0 mmol) and catalyst 6b (0.5 mol% Pd), reaction in DMF (4 ml) at 80ꢁC.
Appl. Organometal. Chem. 2013, 27, 353–360
Copyright © 2013 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/aoc

M. Semler et al.
Figure 9. TEM images of catalyst 6b recovered from a catalytic experiment in the presence of KF.
Table 2. Results of the coupling reactions with other substratesa
Entry
Halide
Stannane
Product
NMR yield (%)
1
2
3
4
5
6
7
8
4-Bromotoluene
4-Bromanisole
4-Chlorotoluene
4-Iodotoluene
4-Bromotoluene
Benzylbromide
4-Bromanisole
4-Bromanisole
Bu3SnPh
10
4-Methoxybiphenyl
10
32
42
0
Bu3SnPh
Bu3SnPh
Bu3SnPh
10
61
14
63
0
SnPh4
10
Bu3SnPh
Diphenylmethane
4-Vinylanisole
4-Allylanisole
Bu3SnCH = CH2
Bu3SnCH2CH = CH2
0
aConditions: stannane (1.0 mmol ), halide (1.0 mmol), KF (2.0 mmol), mesitylene (1.0 mmol) and catalyst 6b (0.5 mol.% Pd), reaction in DMF (4 ml)
at 80ꢁC for 6 h.
4-bromotoluene and 4-bromoanisole (deactivated substrates) to a
similar extent (entries 1 and 2). Replacing 4-bromotoluene with
Experimental
Materials and Methods
the corresponding iodide expectedly resulted in a higher yield
of the coupling product, while no appreciable reaction was
observed for 4-chlorotoluene (entries 3 and 4). Replacement of
8 with tetraphenyltin (entry 5) in the model reaction also resulted
in a lower conversion. Good results were achieved in the reaction
of 8 with benzylbromide, which provided diphenylmethane in
a 63% yield after 6 h (entry 6). On the other hand, catalyst was
not able to promote the coupling of vinyl- and allyl-stannanes
(entries 7 and 8).
Toluene and dichloromethane (Lachner) were dried by standing
over sodium metal and anhydrous potassium carbonate,
respectively, and were freshly distilled under argon prior to
use. Other chemicals and solvents (DMF) were used as received
from commercial suppliers (Lachner, Sigma-Aldrich, Fluka).
Powder XRD patterns were recorded with a Bruker D8
diffractometer using graphite monochromated Cu Ka radiation
(1.5412 Å) and a position-sensitive detector Våntec-1 in Bragg–
Brentano arrangement. Nitrogen adsorption isotherms were
determined with a Micromeritics 2020 instrument at À196ꢁC. The
samples were degassed by heating to 130ꢁC and then evacuated
until a pressure of 1 mPa was attained (at least for 24 h) prior
Conclusions
1
The results presented in this paper indicate that palladium
catalysts deposited over siliceous supports bearing N-donor
groups at the surface can be employed in the Stille coupling of
aryl bromides with phenyltributyltin. However, careful attention
has to be paid to both optimization of the catalyst design and
catalytic conditions as the catalyst performance varies greatly
with the amount of N-donor groups and the N/Pd ratio as well
as with the reaction conditions (temperature, solvent, possible
additives, etc.). The catalysts are heterogeneous in nature and
probably serve as reservoirs of active metal for the reaction
proceeding in the liquid phase. The observed reaction profiles,
which typically level off during the course of the reaction,
suggest that the catalysts lose their activity during the catalytic
run, probably due to aggregation of the metal particles or
blocking of the support surface.
to measurement. H NMR spectra were measured for solutions
in CDCl3 on a Varian VNMRS 300 spectrometer operating at
299.94 MHz. The conversion was determined by a comparison
of integral intensities of the signals due to the product (methoxy
or benzylic CH2 protons) and the signal of C6Me6 used as an
internal standard.
The content of palladium was determined by optical emission
spectroscopy with inductively coupled plasma (ICP-OES) on an
IRIS Interpid II instrument (Thermo Electron) using axial plasma
and an ultrasonic nebulizer CETAC U-5000AT + (wavelength
used for spectrophotometric analysis: 324.270 nm). The samples
were dissolved in a mixture of concentrated HNO3 and HF (2:3;
both Suprapure, Merck; 50ꢁC/15 min) and the solutions were di-
luted with redistilled water. The content of C, H, N and Cl were
determined by conventional combustion analysis.
wileyonlinelibrary.com/journal/aoc
Copyright © 2013 John Wiley & Sons, Ltd.
Appl. Organometal. Chem. 2013, 27, 353–360

Deposited Pd catalysts
TEM images were obtained with a JEOL 2010 transmission elec-
tron microscope (point resolution 0.19 nm, acceleration voltage
160 kV). The samples were dispersed in ethanol and
ultrasonicated for 2 min. The resulting suspensions were applied
to a copper grid covered with holey carbon membrane and
allowed to dry in the air.
argon at 100ꢁC for 39 h. The solid product was filtered off, washed
with toluene (100 ml) and ethanol (400 ml in several portions) and
left to dry in the air (1 day). The solid converted to free amine form
by stirring with a mixture of ethanol and 25% aqueous ammonia
(3:1, 150 mmol) for 2 h, filtered, washed with ethanol until the
washings were neutral and dried in the air to give material 4.
Material 5 was prepared similarly from chloropropylated silica 3.
Catalyst Preparation
Characterization data for 4
Elemental analysis (mmol gÀ1): C 3.73, H 15.9, N 0.68, Cl below
~0.05. Powder XRD, 2θ (deg.): 0.83 (s), 1.42 (m), 1.52 (sh). FT-IR,
nmax (cmÀ1): 3330 (br w), 1633 (w), 1473 (vw), 1175 (sh), 1078
(s), 962 (w), 802 (w), 558 (sh), 463 (s).
Mesoporous molecular sieve SBA-15 (1) was prepared in analogy
to the literature procedure[23] as follows. A polyethylene vessel
was charged successively with triblock copolymer Pluronic
P123 ((ethylene oxide)20(propylene oxide)70(ethylene oxide)20
from Sigma-Aldrich; 10 g), Na2SiO3.9H2O (29.7 g, 0.10 mmol)
and deionized water (256 ml). The resulting mixture was stirred
until the solid reagents completely dissolved (4 days) and then
warmed to 45ꢁC in a water bath. Concentrated aqueous HCl
(65 ml 35%, ~0.75 mol) was introduced and the resulting mixture
was stirred at 45ꢁC for another 90 min. It was then reacted at
90ꢁC without stirring and under autogenous pressure for 7 days.
The separated solid was filtered off, washed with deionized
water until the washings were neutral and twice with ethanol,
and dried in air. The resultant solid was extracted with ethanol
in Soxhlet apparatus in order to remove the polymeric template
and dried in air once again (2 days). Prior to the subsequent
functionalization, the solid mesoporous sieve was calcined at
550ꢁC for 6 h (the temperature was raised from room temperature
to 550ꢁC with an increase of 1ꢁC per minute and then
maintained at 550 ꢁC for 6 h).
Characterization data for 5
Elemental analysis (mmol gÀ1): C 5.23, H 11.0, N 0.93, Cl below
~0.09. FT-IR, nmax (cmÀ1): 3200 (br m), 2980 (w), 2942 (w), 2885
(vw), 2828 (vw), 1986 (w), 1866 (w), 1638 (w), 1457 (w), 1179
(sh), 1077 (s), 963 (sh), 796 (m), 555 (sh), 457 (s).
For the preparation of the metal-unsaturated catalysts 6a and
7a, a solution of palladium(II) acetate (0.112 g, 0.1 mmol per g
of the support) in dry dichloromethane (20 ml; the solution was
filtered through a 0.45 mm syringe filter prior to the addition)
was added to the respective aminated support (5.0 g of 3 or 4).
The mixture was diluted with dichloromethane (100 ml) and
stirred under argon for 2 h. The solid was filtered off, washed
carefully with dichloromethane and, finally, dried in air.
Characterization data for 6a
Elemental analysis (mmol gÀ1): C 3.74, H 15.1, N 0.67, Cl 0.17; Pd
0.088. Powder XRD, 2θ (deg.): 0.82 (s), 1.42 (m), 1.59 (sh). BET
surface area: 311 m2 gÀ1. FT-IR, nmax (cmÀ1): 3340 (br w), 1632
(w), 1552 (w), 1410 (w), 1172 (sh), 1084 (s), 964 (w), 802 (w), 560
(sh), 456 (s).
Characterization data for 1
Powder XRD, 2θ (deg.): 0.87 (s), 1.47 (m), 1.64 (sh). BET surface
area: 678 m2
g
À1. FT-IR, nmax (cmÀ1): 3745 (vs), 3330 (br w),
1976 (vw), 1866 (vw), 1634 (w), 1164(sh), 1066 (s), 962 (m), 804
(m), 572 (w), 460 (m).
Characterization data for 7a
Neat (3-chloropropyl)triethoxysilane (13.5 g) was added to
a suspension of calcined SBA-15 (1) in dry toluene (16.2 g in
400 ml of the solvent). After the resulting mixture was heated at re-
flux under argon for 16 h, the solid was filtered off and washed suc-
cessively with toluene (2 Â ~50 ml), acetone and pentane (~50 ml
each). The resulting material was dried first in the air and then at
120ꢁC under vacuum (~10 torr) for 3 h. The silylation procedure
was repeated once again (13.5 g (3-chloropropyl)triethoxysilane
in 300 ml toluene; isolation as above) to afford material 2.
Elemental analysis (mmol gÀ1): C 5.20, H 11.6, N 0.89, Cl 0.25; Pd
0.087. BET surface area: 347 m2 gÀ1. FT-IR, nmax (cmÀ1): 3250 (br
w), 2974 (w), 2946 (w), 2886 (w), 2825 (vw), 1996 (vw), 1860 (w),
1632 (w), 1568 (w), 1464 (w), 1400 (w), 1179 (sh), 1074 (s), 997
(sh), 796 (w), 562 (sh), 457 (s).
The corresponding metal-saturated catalysts 6b and 7b were
prepared similarly using an excess of palladium(II) acetate to
ensure saturation of the support with the metal component
(0.561 g (2.5 mmol) Pd(OAc)2 per 5.0 g 3, and 0.898 g (4 mmol)
Pd(OAc)2 per 5.0 g 4). Unanchored palladium(II) acetate was
removed by washing with a copious amount of dichloromethane
(1 l in several portions).
Characterization data for 2
Elemental analysis (mmol gÀ1): C 2.56, H 11.6, Cl 0.30. Powder
XRD, 2θ (deg.): 0.85 (s), 1.46 (m), 1.63 (sh). FT-IR, nmax (cmÀ1):
3745 (w), 3308 (br w), 1964 (vw), 1866 (vw), 1632 (w), 1184 (sh),
1074 (s), 958 (m), 808 (w), 460 (m).
Characterization data for 6b
Elemental analysis (mmol gÀ1): C 4.45, H 14.2, N 0.61, Cl 0.16; Pd 0.31.
Powder XRD, 2θ (deg.): 0.84 (s), 1.44 (m), 1.58 (sh). BET surface area:
295 m2 gÀ1. FT-IR, nmax (cmÀ1): 3253 (br w), 1634 (vw), 1557 (w),
1428 (w), 1166 (sh), 1085 (s), 961 (w), 800 (w), 558 (w), 460 (s).
Material 3 was obtained similarly from chromatographic silica
gel (63–200 mm, Fluka). The support was first calcined and then
twice silylated as specified above (26.1 g dried silica gel; 22.1 g
(3-chloropropyl)triethoxysilane in 1000 and 200 ml dry toluene
for the first and second silylation steps, respectively).
Characterization data for 7b
Characterization data for 3
Elemental analysis (mmol gÀ1): C 6.35, H 14.53, N 0.84, Cl 0.14; Pd
0.45. BET surface area: 312 m2 gÀ1. FT-IR, nmax (cmÀ1): 3200 (br w),
2978 (w), 2940 (w), 2882 (w), 1996 (w), 1866 (w), 1574 (m), 1413
(w), 1352 (w), 1179 (sh), 1081 (s), 966 (sh), 803 (m), 690 (w), 562
(w), 450 (s).
Elemental analysis (mmol g–1): C 4.47, H 7.29, Cl 0.55.
In the next step, the chloropropylated supports were aminated
with N,N-diethyl-1,2-diaminoethane. A mixture of material 2,
Et2NCH2CH2NH2 (17.4 g) and toluene (200 ml) was stirred under
Appl. Organometal. Chem. 2013, 27, 353–360
Copyright © 2013 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/aoc

M. Semler et al.
Catalytic Tests
References
Initial kinetic experiments were performed with a parallel batch
reactor (Heidolph Synthesis I). The progress of the reactions
was monitored by GC using an Agilent 6850 chromatograph
equipped with a DB-5 column (internal diameter 0.18 mm, film
thickness 0.18 mm, length 20 m). The reaction mixtures prepared
by mixing the appropriate Pd catalyst (0.5 mol% Pd), halide
(1.0 mmol), organotin reagent (PhSnBu3, 1.0 mmol), mesitylene
(1.0 mmol) and solvent (4 ml) were heated (usually for 80ꢁC)
and continuously stirred. Aliquots were withdrawn at regular
intervals and centrifuged at 4000 rpm for 5 min prior to GC
analysis. For the filtrate test, the amount of reactants was
doubled and the reaction mixture was centrifuged and its half
was filtered through a 0.45 mm PTFE syringe filter while hot to
avoid readsorption of the metal catalyst on to the solid phase.
In the case of preparative batch reactions performed with
catalyst 6b, the reaction mixture contained the substrate
(1.0 mmol 4-bromoanisole or benzyl bromide), organotin reagent
(1.1 mmol), the catalyst (16.2 mg 6b, 0.5 mol% Pd),
hexamethylbenzene (internal standard, 1/6 mmol), KF (2.0 mmol)
and dry DMF (4 ml). The reaction was allowed to proceed at 80ꢁC
for 6 h with continuous stirring and, after cooling to room temper-
ature, was terminated by addition of 1 M aqueous KF (10 ml). The
products were extracted with pentane (10 ml). The organic layer
was dried over magnesium sulfate, filtered and evaporated, leaving
a residue, which was analyzed by 1H NMR spectroscopy.
[1] a) N. Miyaura, K. Yamada, A. Suzuki, Tetrahedron Lett. 1979, 20, 3437;
b) N. Miyaura, A. Suzuki, J. Chem. Soc., Chem. Commun. 1979, 866.
[2] a) R. F. Heck, J. P. Nolley, J. Org. Chem. 1972, 37, 2320; b) T. Mizoroki,
K. Mori, A. Ozaki, Bull. Chem. Soc. Jpn. 1971, 44, 581.
[3] a) D. Milstein, J. K. Stille, J. Am. Chem. Soc. 1978, 100, 3636; b) J. K. Stille,
Angew. Chem. Int. Ed. 1986, 25, 508.
[4] C. Torborg, M. Beller, Adv. Synth. Catal. 2009, 351, 3027.
[5] L. Yin, J. Liebscher, Chem. Rev. 2007, 107, 133.
[6] F. Schüth, Annu. Rev. Mat. Res. 2005, 35, 209.
[7] S. Jana, S. Haldar, S. Koner, Tetrahedron Lett. 2009, 50, 4820.
ꢀ
ꢀ
[8] M. Capka, J. Hetflejs, Collect. Czech. Chem. Commun. 1974, 39, 154.
[9] V. Polshettiwar, C. Len, A. Fihri, Coord. Chem. Rev. 2009, 253, 2599.
ꢀ
ꢀ
ꢀ
ꢀ
[10] a) J. Demel, M. Lamac, J. Cejka, P. Štepnicka, ChemSusChem 2009, 2,
ꢀ
ꢀ
ꢀ
442; b) P. Štepnicka, M. Semler, J. Demel, A. Zukal, J. Cejka, J. Mol.
Catal. A: Chem. 2011, 341, 97.
ꢀ
ꢀ
ꢀ
[11] J. Demel, J. Cejka, P. Štepnicka, J. Mol. Catal. A: Chem. 2010, 329, 13.
[12] H. Zhao, G. Zheng, W. Hao, M. Cai, Appl. Organomet. Chem. 2010, 24, 92.
[13] D.-H. Lee, J.-Y. Jung, M.-J. Jin, Green Chem. 2010, 12, 2024.
[14] a) M. Cai, G. Zheng, G. Ding, Green Chem. 2009, 11, 1687; b) W. Hao,
J. Sha, S. Sheng, M. Cai, J. Mol. Catal. A: Chem. 2009, 298, 94.
[15] D. Zhao, Q. Huo, J. Feng, B. F. Chmelka, G. D. Stucky, J. Am. Chem. Soc.
1998, 120, 6024.
[16] K. S. W. Sing, D. H. Everest, R. A. W. Haul, L. Moscou, R. A. Pieroti,
J. Rouquérol, T. Siemieniewska, Pure Appl. Chem. 1985, 57, 603.
[17] E. P. Barrett, L. G. Joyner, P. P. Halenda, J. Am. Chem. Soc. 1951,
73, 373.
[18] E. Vedejs, A. R. Haight, W. O. Moss, J. Am. Chem. Soc. 1992, 114, 6556.
[19] P. Espinet, A. M. Echavarren, Angew. Chem. Int. Ed. 2004, 43, 4704.
[20] S. P. H. Mee, V. Lee, J. E. Baldwin, Angew. Chem. Int. Ed. 2004,
43, 1132.
[21] E. Fouquet, M. Pereyre, A. L. Rodriguez, J. Org. Chem. 1997, 62, 5242.
[22] a) D. Astruc, Inorg. Chem. 2007, 46, 1884; b) C. M. Crudden, M.
Sateesh, R. Lewis, J. Am. Chem. Soc. 2005, 127, 10045; c) L. Yin, J.
Liebscher, Chem. Rev. 2007, 107, 133; d) A. F. Schmidt, A. Al-Halaiqa,
V. V. Smirnov, Kinet. Catal. 2008, 49, 395; e) S. Jana, B. Dutta, R. Bera,
S. Koner, Inorg. Chem. 2008, 47, 5512.
Acknowledgments
Financial support from the Czech Science Foundation (project
no. 104/09/0561) and the Grant Agency of Charles University in
Prague (project no. 639512) is gratefully acknowledged.
ꢀ
ꢀ
ꢀ
[23] J. Demel, Sujandi, S.-E. Park, J. Cejka, P. Štepnicka, J. Mol. Catal. A:
Chem. 2009, 302, 28.
wileyonlinelibrary.com/journal/aoc
Copyright © 2013 John Wiley & Sons, Ltd.
Appl. Organometal. Chem. 2013, 27, 353–360