Hyperbranched polyglycerol supported ruthenium catalysts for ring-closing metathesis Metallodendrimers Special Issue

Article abstract of DOI:10.1016/j.ica.2013.06.031

Immobilization of Hoveyda-Grubbs type I and II metathesis catalysts onto hyperbranched polyglycerol has been achieved via two distinct synthetic routes. The catalytic units were either placed at the periphery or the interior of the polymer and catalyst loadings between 0.166 and 0.517 mmol g^-1 were achieved. The activity of the catalysts in ring-closing metathesis of variously substituted dienes and the ruthenium leaching into the product was investigated. It was found that the supported second-generation Hoveyda-Grubbs catalyst displayed higher activity than the first-generation analogues and achieved high conversions in only two hours. After the first run, however, longer reaction times were required to reach a reasonable conversion. Ru-leaching was determined by ICP-MS and found to be in the range of 0.2-7.8% of the initial ruthenium content, which corresponds to 28-288 ppm of Ru in the products. Additional treatment with activated charcoal after dialysis was found to be beneficial for Ru-removal.

Full text of DOI:10.1016/j.ica.2013.06.031

Inorganica Chimica Acta 409 (2014) 179–184
Contents lists available at SciVerse ScienceDirect
Inorganica Chimica Acta
journal homepage: www.elsevier.com/locate/ica
Note
Hyperbranched polyglycerol supported ruthenium catalysts for
ring-closing metathesis
⇑
Venkatakrishnan S. Thengarai, Juliane Keilitz, Rainer Haag
Department of Chemistry and Biochemistry, Freie Universität Berlin, Takustraße 3, 14195 Berlin, Germany
a r t i c l e i n f o
a b s t r a c t
Article history:
Available online 28 June 2013
Immobilization of Hoveyda–Grubbs type I and II metathesis catalysts onto hyperbranched polyglycerol
has been achieved via two distinct synthetic routes. The catalytic units were either placed at the periph-
ery or the interior of the polymer and catalyst loadings between 0.166 and 0.517 mmol gÀ1 were
achieved. The activity of the catalysts in ring-closing metathesis of variously substituted dienes and
the ruthenium leaching into the product was investigated. It was found that the supported second-gen-
eration Hoveyda–Grubbs catalyst displayed higher activity than the first-generation analogues and
achieved high conversions in only two hours. After the first run, however, longer reaction times were
required to reach a reasonable conversion. Ru-leaching was determined by ICP-MS and found to be in
the range of 0.2–7.8% of the initial ruthenium content, which corresponds to 28–288 ppm of Ru in the
products. Additional treatment with activated charcoal after dialysis was found to be beneficial for
Ru-removal.
Metallodendrimers Special Issue
Keywords:
Ring-closing metathesis
Immobilization
Hyperbranched polyglycerol
Ruthenium leaching
Recycling
Ó 2013 Elsevier B.V. All rights reserved.
1. Introduction
2. Experimental and methods
Since its discovery [1], olefin metathesis has been significantly
developed and now plays an important role in organic synthesis
[2,3]. A number of well-defined ruthenium based metathesis cata-
lysts are known, which display very high activity in the metathesis
reactions of several organic substrates [4,5]. Removal of the ruthe-
nium residues from the final product is very tedious when homo-
geneous molecular catalysts are used and therefore is the subject
of extensive research [6,7]. One means to facilitate the separation
between catalyst and product is the immobilization of the homo-
geneous catalyst onto a polymeric scaffold which enables the use
of simple separation techniques [8]. In order to enhance catalyst
stability while retaining the high activity of the homogeneous
catalyst, there has been a continuing interest to immobilize homo-
geneous analogues onto various polymeric supports [9]. Our group
successfully employed hyperbranched polyglycerol (hPG) [10] as a
support for a variety of catalysts [11].
The importance of olefin metathesis in organic synthesis [3] and
the difficulties of the removal of metal residues from the products
[5,7] motivated us to investigate the ability of hPG to function as a
scaffold for metathesis catalysts with the long-term goal of an
application in continuous flow membrane reactors. Here we
present the synthesis and characterization of hPG-supported
metathesis catalysts, their performance in ring closing metathesis
(RCM) of differently substituted diolefins, and our results on the
Ru-leaching into the final products.
All reactions were performed under dry and air-free conditions
using standard Schlenk techniques or dry box procedures. Dry sol-
vents were taken from a solvent purification system MB SPS-800
from MBraun and were used after three vacuum-nitrogen cycles.
Hyperbranched polyglycerol has a Mw of 10 kDa and 13.5 mmol
OH per gram of polymer. 4-Isopropoxy-3-vinylbenzyl alcohol
[12], Grubbs II catalyst [13], and diethyl-allylmethallyl malonate
[14] were synthesized according to literature procedures. All other
chemicals were purchased from commercial sources and used as
received. Ultrafiltration was performed under Nitrogen (or Argon)
pressure in a solvent-resistant stirred cell with regenerated cellu-
lose membranes (molecular weight cut-off 1000 or 5000), both
from Millipore. Dialysis was performed with benzoylated cellulose
membrane tubing purchased from SIGMA Aldrich with a MWCO
1000. ICP-MS measurements were carried out on an Element 2
(Thermo Fisher) at low resolution (sample gas 0.863 L/min; plasma
power 1350 Watt). NMR spectra were recorded on a Bruker AMX
500, an Avance 700 instrument, or a Jeol 400 MHz instrument. IR
spectra were recorded as thin film on KBr or diamond anvil using
Avatar 320 FT-IR spectrometer. Ring closing metathesis (RCM)
reactions were performed in regular NMR tubes.
2.1. Synthesis of 4-isopropoxy-3-vinylbenzyl-2-bromoacetate (1)
To a solution of 4-isopropoxy-3-vinylbenzyl alcohol (0.25 g,
1.30 mmol) in THF (5 mL) at 0 °C, solid DMAP (0.16 g, 1.30 mmol,
1 equiv) was added in one portion and the mixture was stirred
⇑
Corresponding author. Tel.: +49 30 838 52633; fax: +49 30 838 53357.
E-mail address: haag@chemie.fu-berlin.de (R. Haag).
0020-1693/$ - see front matter Ó 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.ica.2013.06.031

180
V.S. Thengarai et al. / Inorganica Chimica Acta 409 (2014) 179–184
for 5 min. Bromoacetyl chloride (0.21 g, 1.30 mmol, 1 equiv) was
added at 0 °C over a period of 2 h. The cold bath was removed
and the reaction mixture was stirred at ambient temperature for
1 h. The reaction mixture was then cooled to 0 °C and quenched
with ice. It was transferred to a separating funnel, saturated
NaHCO3 and water (50 mL each) were added and the product
was extracted with dichloromethane (3 Â 50 mL). The organic ex-
tracts were washed with water, dried over Na2SO4, filtered and
dried in vacuo to obtain the product as a light yellow oil. Com-
pound 1 was used for the next step without further purification.
1H NMR (CDCl3, 400 MHz): d = 7.46 (d, J = 2.4 Hz, 1H, aromatic),
7.20 (dd, J = 8.0, 2.4 Hz, 1H, aromatic), 7.01 (dd, J = 6.8, 2.4 Hz,
1H, ArCHCH2), 6.84 (d, J = 8.8 Hz, 1H, aromatic), 5.73 (dd, J = 16.0,
1.6 Hz, 1H, ArCHCH2), 5.24 (dd, J = 11.2, 1.6 Hz, 1H, ArCHCH2),
5.12 (s, 1H, ArCH2OC(O)), 5.11 (s, 1H, ArCH2OC(O)), 4.53 (septet,
1H, ArOCHMe2), 3.85 (s, 1H, BrCH2C(O)), 3.84 (s, 1H, BrCH2C(O)),
1.33 (d, J = 6.0 Hz, 6H, ArOCHMe2). 13C NMR (CDCl3, 100 MHz):
d = 167.4, 155.5, 131.5, 129.6, 127.5, 127.4, 126.9, 114.7, 114.0,
(m, Ar–CH@CH2+ArCH2OC(O)), 4.48 (m, CHMe2), 4.00–3.20 (m br,
primary protons of polymer + polymer–OCH2C(O)), 1.25–1.15
(m br, CHMe2) ppm. IR (mC
@
cmÀ1): 1646 (m br), 1601 (m br).
O
2.4. Synthesis of hPG-(OAc)30%-supported ligand 4
The ligand was prepared by a similar procedure as ligand 3. A
well-dried sample of hPG-(OAc)30% 2 (0.10 g, 0.81 mmol OH) was
placed in a double-necked, round-bottom flask equipped with a
dropping funnel and dissolved in dry THF (10 mL). The solution
was then cooled to 0 °C. A solution of KOtBu (1.34 mL, 0.15 g,
1.3 mmol, 1.6 equiv; 1 M solution in THF) in THF (5 mL) was trans-
ferred into a dropping funnel and added dropwise to the cold THF
solution over 15 min resulting in the formation of a white suspen-
sion which was stirred for another 30 min at that temperature. The
above-prepared styrene isopropoxy derivative
1 (1.30 mmol,
1.60 equiv) was dissolved in THF (10 mL), transferred to the drop-
ping funnel, and added dropwise over 30 min to the white suspen-
sion at 0 °C. The resulting red–orange colored solution and a white
precipitate was allowed to warm to ambient temperature and stir-
red overnight. The reaction mixture was cooled to 0 °C and
quenched with ice. The volatiles were evaporated under vacuum
to obtain an orange-red residue. The residue was dissolved in
methanol, filtered through a cotton plug and subjected to ultrafil-
tration in the same solvent until the filtrates were colorless. The
supported ligand 4 was obtained as yellow sticky oil which was
further subjected to dialysis in methanol. Yield: 0.26 g (60%). 1H
NMR (CDCl3, 400 MHz): 7.45 (m br, aromatic), 7.15 (m br, aro-
matic), 7.00 (m br, Ar–CH@CH2), 6.80 (m br, aromatic), 5.75 (br,
Ar–CH@CH2), 5.10 (m br, Ar–CH@CH2 + ArCH2OC(O)), 4.48 (m,
CHMe2), 4.00–3.20 (m, primary protons of polymer + polymer–
71.0, 68.1, 41.1, 22.2 ppm. IR (mC
@
cmÀ1): 1742 (s), 1651 (vs).
O
2.2. Synthesis of hPG-(OAc)30%
2
A well-dried sample of hyperbranched polyglycerol (2.67 g,
36.10 mmol OH) was placed in a double-necked, round-bottom flask
equipped with a condenser and dissolved in dry pyridine (30 mL).
Acetic anhydride (1.1 mL, 11.91 mmol, 0.33 equiv per OH) was
added and the reaction mixture was heated to 55 °C for 18 h. After
cooling to ambient temperature, the mixture was cooled to 0 °C
and quenched by adding ice. The volatiles were evaporated by
cryo-distillation and the product was purified by dialysis using
methanol. The extent of acylation was determined from the relative
integral intensities of the acyl-methyl group protons (3H at
2.08 ppm; see below) against the combined integrals (5H, 5.25–
4.95 and 4.45–3.45 ppm; see below) from polyglycerol. Yield:
2.60 g (83%, degree of functionalization: 30%). 1H NMR (CD3OD,
500 MHz): d = 5.25–4.95 (m, 0.27, secondary protons of polymer),
4.45–3.45 (m, 14.79, primary protons of polymer), 2.08 (s br, 3H,
OC(O)CH3) ppm. 13C NMR (CD3OD, 125.8 MHz): d = 171.4 (s), 78.7
(s), 72.3 (s), 71.3 (s br), 71.2 (br), 69.8 (br), 68.3 (br), 65.6 (s), 63.2
OCH2C(O)), 1.25–1.15 (m br, CHMe2). IR (mC
@
cmÀ1): 1731 (s),
O
1645 (m sh), 1625 (m).
2.5. Synthesis of hPG-supported Hoveyda–Grubbs I catalyst 5
In a glove box, the hPG-supported ligand 3 (0.14 g, 94.85
lmol
styrenyl ligand) was suspended in dry dichloromethane (5 mL) in a
Schlenk flask. Solid Grubbs I catalyst (0.16 g, 0.19 mmol) was
added and the Schlenk flask was capped, taken out of the glove
box and the mixture was heated at 40 °C for 15 h. The reaction
mixture was then subjected to dialysis in dichloromethane to ob-
tain the supported catalyst 5 as a brownish-pink solid that was sta-
ble under air. Yield: 0.16 g (85%). 1H NMR (CDCl3, 400 MHz): 17.40
(s, br, Ru@CH), 7.75–7.30 (m, aromatic), 7.20–6.70 (m, aromatic),
6.60 (m br, aromatic), 5.25 (br, ArCH2OC(O)) 4.30 (m, CHMe2),
4.00–3.30 (m, primary protons of polymer + polymer-OCH2C(O)),
1.80–1.40 (m br, PCy3 protons + CHMe2). 31P NMR (CDCl3,
(s), 19.7 ppm. IR (mC@
O cmÀ1): 1731 (s), 1646 (s).
2.3. Synthesis of hPG-supported ligand 3
A well-dried sample of hyperbranched polyglycerol (0.40 g,
5.40 mmol OH) was placed in a double-necked, round-bottom flask
fitted with dropping funnel and dissolved in dry DMF (4 mL). The
solution was then cooled to 0 °C. A solution of KOtBu (0.24 g,
2.16 mmol, 0.4 equiv) in a 1:2 mixture of DMF/THF (6 mL) was
transferred into a dropping funnel and added drop-wise to the
cold DMF solution during 15 min leading to the formation of a
white suspension and was stirred for ½ h at that temperature.
The above-prepared styrene isopropoxy derivative 1 (345 mg,
1.10 mmol, 0.2 equiv) was dissolved in THF (15 mL), transferred
to the dropping funnel and added dropwise for 30 min to the white
suspension at 0 °C. The solution turned bright yellow and turned
more intense during the addition; a red–orange colored solution
and a white precipitate was obtained. The cold bath was removed
and the reaction mixture was stirred overnight. The reaction mix-
ture was cooled to 0 °C and quenched with ice. The volatiles were
evaporated under vacuum to obtain an orange-red residue, which
was dissolved in methanol, filtered through a cotton plug, and sub-
jected to ultrafiltration (MeOH) until the washings were colorless.
The hPG-supported ligand 3 was obtained as a yellow sticky
compound. Yield: 0.31 g (65%). 1H NMR (CDCl3, 400 MHz):
d = 7.37 (m br, aromatic), 7.10 (m br, aromatic), 6.93 (m br,
Ar–CH@CH2), 6.77 (m br, aromatic), 5.65 (m, Ar–CH@CH2), 5.13
162 MHz): 61.8 (s, br). IR (mC
@
cmÀ1): 1728 (s br), 1628 (m br).
O
ICP-MS: 0.517 mmol gÀ1 Ru, 0.572 mmol gÀ1 P; catalyst loading:
0.517 mmol gÀ1; degree of functionalization: 3.85%.
2.6. Synthesis of hPG-(OAc)30%-supported Hoveyda–Grubbs I catalyst 6
In a glove box, the hPG-supported ligand 4 (20 mg, 3.29
styrenyl ligand) was dissolved in dry dichloromethane (0.5 mL)
in a Schlenk flask. Solid Grubbs I catalyst (9 mg, 11.24 mol) was
lmol
l
added and the Schlenk flask was capped, taken out of the glove
box and the mixture was heated at 40 °C for 15 h. The reaction
mixture was then subjected to dialysis in dichloromethane to
obtain the air stable pinkish brown hPG-(OAc)30%-supported
catalyst 6. Yield: 19 mg (50%). 1H NMR (CDCl3, 400 MHz): 17.38
(s, br, Ru@CH), 7.80–7.40 (m, aromatic), 7.10–6.60 (m, aromatic),
7.10–6.60 (m br, aromatic), 5.10 (br, ArCH2OC(O)), 4.50
(m, CHMe2), 4.25–3.20 (m, primary protons of polymer + poly-
mer–OCH2C(O)), 2.05 (br, OC(O)CH3) 1.80–1.40 (m br, PCy3

V.S. Thengarai et al. / Inorganica Chimica Acta 409 (2014) 179–184
181
protons), 1.30–1.20 (m br, CHMe2). IR (mC
@
O cmÀ1): 1728 (s), 1683
(ROMBP) of glycidol and contains interior secondary monohydroxy
and terminal dihydroxy functional groups, which can be easily
converted by standard synthetic methods [10]. These different
types of hydroxyl groups offer a handle to distinguish between
the interior and exterior of the polymer by utilizing the appropri-
ate synthetic route.
(m), 1625 (m). ICP-MS: 0.166 mmol gÀ1 Ru, 0.162 mmol gÀ1 P; cat-
alyst loading: 0.166 mmol gÀ1; degree of functionalization: 1.23%.
2.7. Synthesis of hPG-(OAc)30%-supported Hoveyda–Grubbs II catalyst
7
The introduction of catalytic sites at the periphery can be
achieved by simply reacting hPG with the ligand 1 followed by me-
tal complexation. Since the outer hydroxyl groups are most easily
accessible, those will be functionalized primarily. In order to func-
tionalize the interior of hPG, the outer hydroxyl groups have to be
blocked before ligand coupling and metal complexation can take
place.
Several strategies have been employed for the immobilization
of ruthenium-based metathesis catalysts onto polymer supports,
among which immobilization via benzylidene has gained wide
attention [9,17–20]. In order to achieve a stable coupling, an ether
linkage was used to bind ligand 1 to hPG (Scheme 1).
In a glove box, the hPG-supported ligand 4 (25 mg, 4.11
styrenyl ligand) was dissolved in dry dichloromethane (0.5 mL)
in a Schlenk flask. Solid Grubbs II catalyst (12 mg, 14.05 mol)
lmol
l
was added and the Schlenk flask was capped, taken out of the glove
box and the mixture was heated to 40 °C for 15 h. The reaction
mixture was then subjected to dialysis in dichloromethane to ob-
tain the supported catalyst 7 as an air-stable bright green com-
pound. Yield: 25 mg (55%). 1H NMR (CDCl3, 400 MHz): 16.50 (s,
br, Ru@CH), 7.60–7.30 (m, aromatic), 7.10–6.80 (m, aromatic),
6.80–6.50 (m br, aromatic), 4.80–4.50 (m br, ArCH2-
OC(O) + CHMe2), 4.00–3.30 (m, primary protons of polymer + poly-
mer–OCH2C(O)), 2.50–2.15 (m br, mesityl-Me), 2.00 (br, OC(O)CH3)
For the synthesis of catalyst 5 with Grubbs I catalytic sites
mainly at the periphery, hPG (Mw 10 kDa, 13.5 mmol OH gÀ1
)
1.80–1.40 (m br, PCy3 protons), 1.30–1.20 (m br, CHMe2). IR (mC
@
O
cmÀ1): 1730 (s), 1632 (m). ICP-MS: 0.298 mmol gÀ1 Ru; catalyst
loading: 0.298 mmol gÀ1; degree of functionalization: 2.21%.
was first reacted with ligand 1 to obtain the hPG-supported ligand
3 with approximately 20% functionalized hydroxyl groups. Due to
their easy accessibility, the outer hydroxyl groups will react first.
Purification of 3 is achieved by simple ultrafiltration.
In order to obtain a catalyst with the catalytic sites in the inte-
rior of the polymer, the exterior hydroxyl groups have to be
blocked first, which is achieved by reaction with acetic anhydride.
hPG-(OAc)30% 2 was obtained with a degree of functionalization of
2.8. General procedure for ring-closing metathesis
In an NMR tube, the catalyst was dissolved in CDCl3. The sub-
strate was added, everything was mixed well and then placed in
a pre-heated oil bath at 55 °C for the indicated period of time.
The product formation and disappearance of the diene were mon-
itored by 1H NMR. The conversion of substrates S1 and S3 was
determined according to a literature procedure [15]. In the case
of S3, the conversion was determined by comparing the integrals
of the methylene protons of the product [4.48 (m, 2H), 4.36 (m,
2H)] [16] with those of the starting material [3.99 (m, 4H)].
about 30% and then reacted with ligand 1 to give hPG-(OAc)30%
-
supported ligand 4 in moderate yield (60%).
Complexation with Grubbs I or II proceeds via a similar reaction
to the one described for the analogous non-supported, homoge-
neous Hoveyda–Grubbs catalysts [21]. The treatment of hPG-sup-
ported ligand
dialysis gave the hPG-supported first-generation Hoveyda–Grubbs
catalyst 5 in good yield of 85%. Catalyst 5 was characterized by 1
3 with Grubbs I catalyst and purification by
H
and 31P NMR and IR spectroscopy. The 1H NMR of 5 shows a reso-
nance at 17.40 ppm for the benzylidene, which is in the range usu-
ally observed for first-generation Hoveyda–Grubbs catalysts [21].
Similarly, the 31P chemical shift of ruthenium-bound tricyclohexyl
phosphine is observed at 61.8 ppm [21]. The catalyst loading of 5
was analyzed by ICP-MS and it was found that 0.517 mmol gÀ1 of
metathesis catalyst were bound, which corresponds to a degree
3. Results and discussion
In order to understand the influence of the polymer on the cat-
alytic performance we sought to incorporate the catalytic sites
either predominantly at the periphery or at the interior of the poly-
mer (Fig. 1). Hyperbranched polyglycerol (hPG) can be easily ob-
tained by anionic ring-opening, multi-branching polymerization
Fig. 1. Schematic representation of the supported catalysts with the active catalytic sites located at the periphery (left) or the interior (right). The polyglycerol scaffold shows
an idealized structure of hPG.

182
V.S. Thengarai et al. / Inorganica Chimica Acta 409 (2014) 179–184
O
Cl
Br
+
OH
O
THF, DMAP
0 °C - r.t., 1 h
HO
OH
HO
hPG
hPG
(OAc)30%
hPG
2
O
DMF-THF
KOtBu
0 °C - r.t., 15 h
THF, KOtBu
0 °C - r.t., 15 h
O
Br
O
1
O
O
O
O
(OAc)30%
O
hPG
O
hPG
OH
O
O
3 (yield: 65%)
4 (yield: 60%; loading with OAc: 30%)
Grubbs I
DCM, 40 °C
Grubbs I/II
DCM, 40 °C
L
L
Cl
Cl
Cl
Cl
Ru
Ru
O
O
O
O
(OAc)30%
O
hPG
O
hPG
OH
O
O
L = PCy3 (yield: 85%; loading: 0.517 mmol·g-1, 3.85%)
L = PCy3 (yield: 50%; loading: 0.166 mmol·g-1, 1.23%)
5
6
L = NHC 7 (yield: 55%; loading: 0.298 mmol·g-1, 2.21%)
Scheme 1. Synthesis of hPG-supported ligands 3 and 4 and catalysts 5–7.
of functionalization of 3.85%. It is worth mentioning that the ruthe-
nium to phosphorus ratio in the supported catalyst is 0.9 which
confirms the nature of the supported catalyst.
Catalyst 6 was obtained in moderate yield (50%) by the same
synthetic procedure and the 1H NMR and IR spectroscopic data is
similar to that of catalyst 5. However, since the catalytic units in
6 are bound to the interior hydroxyl groups of hPG, the loading
3.1. Catalytic ring-closing metathesis
With the supported catalysts 5–7 in hand, we investigated their
activity in ring-closing metathesis reactions (Scheme 2). Initial
experiments were performed on the benchmark substrate diethyl-
diallyl malonate (S1). Substrate S2 was chosen to investigate the
tolerance of the supported catalysts towards N-heterocycles and
substrate S3 was chosen to investigate the efficacy of the catalysts
in the synthesis of tri-substituted cycloalkenes. The results are
shown in Table 2.
As can be seen from Table 2, substrate S1 undergoes quantita-
tive conversion to the corresponding cycloalkene P1 in the pres-
ence of all three catalysts 5–7 (entries 1–3, 1st run). However,
the phosphine-free catalyst 7 is obviously significantly more active
is significantly reduced. By ICP-MS
a catalyst loading of
0.166 mmol gÀ1 is measured, which is approximately three times
lower than the loading of 5. The ruthenium to phosphorus ratio
in 6 (see Table 1) is 1.02, again confirming the nature of the sup-
ported catalyst.
The second generation of metathesis catalysts bear strongly
nucleophilic N-heterocyclic carbenes and is known to be highly ac-
tive in ring-closing metathesis [5,17,22,23]. Since the removal of
phosphorus compounds from organic products is often difficult,
the use of these phosphine-free catalysts is attractive. In order to
support the second-generation catalyst, hPG-supported ligand 4
is reacted with Grubbs II to afford the hPG-supported Hoveyda–
Grubbs II catalyst 7 in moderate yield (55%). The benzylidene pro-
ton in 7 is observed at 16.50 ppm in the 1H NMR spectrum. This
chemical shift is in agreement with non-supported catalysts
[17,22]. The loading of the catalyst on the polymer was found to
be 0.298 mmol gÀ1, which corresponds to a degree of functionali-
zation of 2.21%.
EtO2C CO2Et
EtO2C CO2Et
cat.
CDCl3, 55 °C
S1
P1
COCF3
N
COCF3
N
cat.
CDCl3, 55 °C
S2
P2
Table 1
EtO2C CO2Et
Ruthenium and phosphorus content and the loading of catalysts 5–7.a
EtO2C CO2Et
cat.
Entry
Catalyst
Ru (mmol gÀ1
)
P (mmol gÀ1
)
DF (%)
Ru/P
CDCl3, 55 °C
1
2
3
5
6
7
0.517
0.166
0.298
0.572
0.162
0.184
3.85
1.23
2.21
0.90
1.02
1.62
S3
P3
Scheme 2. Catalytic ring-closing metathesis reactions of substrates S1–S3 using
hPG-supported catalysts 5–7.
a
Results obtained from ICP-MS measurements.

V.S. Thengarai et al. / Inorganica Chimica Acta 409 (2014) 179–184
183
Table 2
Results of ring-closing metathesis reactions of substrates S1–S3 using catalysts 5–7.
Entry
Subst.
Cat.
S/Ca
Conversion (%)b (time (h))
TON
Ru-leachingc
Run 1
Run 2
Run 3
(1st run)
(ppm) ((%))
1
2
3
4
5
6
7
S1
S1
S1
S2
S2
S2
S3
5
6
7
5
6
7
5
20
77
47
55
105
60
15
99 (20)
97 (17)
>99 (2)
87 (13)
>99 (17)
86 (2)
63 (20)
80 (39)
85 (17)
91 (2)
28 (20)
37 (20)
82 (18)
59 (2)
63 (20)
34 (18)
–
–
–
20
75
47
48
104
52
9
12
90
36
283 (2.4)
288 (7.3)
224 (7.8)
150 (5.2)
2884 (98.9)d
134 (6.0)
–
64 (0.2)
173 (6.0)
28 (0.8)e
83 (20)
75 (66)
–
25 (3)
–
–
–
–
8
9
S3
S3
6
7
106
40
33 (20)
80 (18)
63 (3)
a
b
c
Valid for the first run.
Determined from the relative integral intensities in the crude 1H NMR spectrum.
Results obtained after the last run by ICP-MS after purification by dialysis.
d
Since the leaching accounts to 99% of the initial Ru-content of the reaction mixture, it is most likely that a leak in the dialysis tube during purification is responsible for
this leaching value.
e
Results obtained by ICP-MS after purification by dialysis and subsequent treatment with activated charcoal.
(entry 3) affording the cycloalkene P1 in less time than required
with catalysts 5and 6 (entries 1, 2). Similarly, catalysts 5–7 also
transform substrate S2 to P2 with high conversions (entries 4–6,
1st run). Again, catalyst 7 displays a higher activity, giving 86%
conversion in two hours. In the case of the sterically demanding
substrate S3 a significant difference in activity between the three
catalysts is observed (entries 7–9). The overall activity trend is
similar to the one observed for substrates S1 and S2: catalyst 7 dis-
plays superior activity compared to 5 and 6, affording 91% conver-
sion in only two hours. Catalyst 5 shows poor activity in the
formation of the tri-substituted olefin P3, requiring 39 h to achieve
80% conversion of S3 whilecatalyst 6 needs only 17 h to reach 85%
conversion. However, the activity of 7 is comparable to those of
corresponding non-supported catalysts [17,21,22] and other sec-
ond generation catalysts [23]. For substrates S1 and S2 the mono-
meric second generation metathesis catalyst usually achieves up to
99% yield within a few hours reaction time [23a,b]. In comparison
to other supported second-generation Hoveyda–Grubbs catalysts,
7 shows comparable activity [19,20,24–26].
ranges from 28 to about 288 ppm. Purification by dialysis was able
to reduce the amount of ruthenium to as low as 64 ppm. However,
supplementing dialysis with activated charcoal treatment reduced
theruthenium level to as low as 28 ppm. When the leaching values
are calculated in per cent of the original amount of Ru used, values
between 0.2 and 7.8% are obtained. The amount of ruthenium
leaching found in this work is comparable to the leaching found
for other supported metathesis catalysts immobilized on polysty-
rene [27], poly(ethylene glycol) [28], poly(ethylene) [26],
poly(divinyl benzene) [29], porous silica [25,30], silica-based cat-
ionic exchange resins [20], or sol–gel monoliths [12].
4. Conclusion
In conclusion we successfully synthesized hPG-supported
metathesis catalysts 5–7 with the catalytic moieties either located
at the periphery or interior of the dendritic polymer. They are quite
robust and air stable at ambient temperature for at least six weeks.
The second-generation catalyst exhibiting an N-heterocyclic car-
bene ligand is significantly more active than the immobilized first
generation Hoveyda–Grubbs catalysts, leading to the formation of
both di- and trisubstituted olefins in good to high conversions. The
analysis of ruthenium levels in the final products shows a ruthe-
nium contamination in the range of 64–288 ppm which corre-
sponds to a loss of ruthenium of 0.2–7.8% based on the initial
metal content. A further reduction of the contamination was
achieved by complementary treatment with activated charcoal
after dialysis which reduced the Ru-content of the product to
28 ppm.
3.2. Catalyst recycling and ruthenium leaching into theproduct
Initial experiments on catalyst recycling were performed with
the following procedure: the organic product was separated from
the supported catalyst by dialysis and additional ligand 4 was
added to re-form the catalyst. However, upon addition of another
equivalent of substrate and subjection to the standard reaction
conditions, no conversion was observed.
Therefore, we decided to use sequential addition of substrate
without intermediate product purification to increase the total
turnover numbers and to determine the overall activity of 5–7.
After the first addition of the substrate and heating, the conversion
was monitored by 1H NMR every two hours to ensure complete
conversion. Upon complete consumption of the starting material,
another equivalent of substrate was added and the NMR tube
was heated again. The reactions wereusually stopped after two
or three runs and the mixture was then subjected to dialysis to iso-
late the ring-closing metathesis product.
Acknowledgements
The authors thank Mr. Thomas Wons for ICP-MS measurements.
This work is part of the Cluster of Excellence ‘‘Unifying Concepts in
Catalysis.’’ Financial support by the Deutsche Forschungsgemeins-
chaft (DFG) (EXC 314) is gratefully acknowledged. We are grateful
to Dr. Pamela Winchester for proofreading this manuscript.
Two protocols were employed for ruthenium removal: (a) dial-
ysis only and (b) dialysis followed by treatment with activated
charcoal for 12 h [16]. The organic product obtained after dialysis
was found to be free of polymer by 1H NMR. However, the presence
of trace amounts of ruthenium was determined by inductively cou-
pled plasma mass spectrometry (ICP-MS). The results are shown in
Table 2. The amount of ruthenium found in the organic products
References
[1] R.H. Grubbs, Handbook of Metathesis, first ed., vol. 1–3, 2003, Weinheim,
Wiley-VCH.
[2] M. Schuster, S. Blechert, Angew. Chem., Int. Ed. Engl. 36 (1997) 2036.
[3] (a) A.H. Hoveyda, S.J. Malcolmson, S.J. Meek, A.R. Zhugralin, Angew. Chem., Int.
Ed. 49 (2010) 34;
(b) H.M.A. Hassan, Chem. Commun. 46 (2010) 9100;

184
V.S. Thengarai et al. / Inorganica Chimica Acta 409 (2014) 179–184
(c) U.H.F. Bunz, D. Mäker, M. Porz, Macromol. Rapid Commun. 33 (2011) 886;
(d) S. Chikkali, S. Mecking, Angew. Chem., Int. Ed. 51 (2012) 5802.
[16] W.A. Nugent, J. Feldman, J.C. Calabrese, J. Am. Chem. Soc. 117 (1995) 8992.
[17] S.B. Garber, J.S. Kingsbury, B.L. Gray, A.H. Hoveyda, J. Am. Chem. Soc. 122
(2000) 8168.
[18] C. Cope´ret, J.-M. Basset, Adv. Synth. Catal. 349 (2006) 78.
[19] A. Keraani, M. Rabiller-Baudry, C. Fischmeister, C. Bruneau, Catal. Today 156
(2010) 268.
[20] E. Borré, M. Rouen, I. Laurent, M. Magrez, F. Caijo, C. Crévisy, W. Solodenko, L.
Toupet, R. Frankfurter, C. Vogt, A. Kirschning, M. Mauduit, Chem. Eur. J. 18
(2012) 16369.
[4] (a) S.P. Nolan, H. Clavier, Chem. Soc. Rev. 39 (2010) 3305;
(b) S. Kress, S. Blechert, Chem. Soc. Rev. 41 (2012) 4389.
[5] G.C. Vougioukalakis, R.H. Grubbs, Chem. Rev. 110 (2010) 1746.
[6] H.D. Maynard, R.H. Grubbs, Tetrahedron Lett. 40 (1999) 4137.
[7] G.C. Vougioukalakis, Chem. Eur. J. 18 (2012) 8868.
[8] C. Tzschucke, C. Markert, W. Bannwarth, A. Hebel, S. Roller, R. Haag, Angew.
Chem., Int. Ed. 41 (2002) 3694.
[9] (a) H. Clavier, K. Grela, A. Kirschning, M. Mauduit, S.P. Nolan, Angew. Chem.,
Int. Ed. 46 (2007) 6786;
[21] J.S. Kingsbury, J.P.A. Harrity, P.J. Bonitatebus, A.H. Hoveyda, J. Am. Chem. Soc.
121 (1999) 791.
(b) M.R. Buchmeiser, Chem. Rev. 109 (2009) 303.
[10] (a) A. Sunder, R. Hanselmann, H. Frey, R. Mülhaupt, Macromolecules 32 (1999)
4240;
(b) A. Sunder, R. Mülhaupt, R. Haag, H. Frey, Macromolecules 33 (2000) 253.
[11] (a) R. Haag, S. Roller, Polymeric Supports for the Immobilisation of Catalysts, in
Immobilized Catalysts – Solid Phases, Immobilization and Applications, A.
Kirschning, Editor. 2004, Springer, p. 1.;
[22] S. Gessler, S. Randl, S. Blechert, Tetrahedron Lett. 41 (2000) 9973.
[23] (a) H. Wakamatsu, S. Blechert, Angew. Chem., Int. Ed. 41 (2002) 794;
(b) H. Wakamatsu, S. Blechert, Angew. Chem., Int. Ed. 41 (2002) 2403;
(c) K. Grela, S. Harutyunyan, A. Michrowska, Angew. Chem., Int. Ed. 41 (2002)
4038;
(d) P.S. Kumar, K. Wurst, M.R. Buchmeiser, Organometallics 28 (2009) 1785.
[24] B. Marciniec, S. Rogalski, M.J. Potrzebowski, C. Pietraszuk, ChemCatChem 3
(2011) 904.
(b) J. Keilitz, R. Haag, Eur. J. Org. Chem. (2009) 3272;
(c) J. Keilitz, S. Nowag, J.-D. Marty, R. Haag, Adv. Synth. Catal. 352 (2010) 1503;
(d) J. Dimroth, J. Keilitz, U. Schedler, R. Schomäcker, R. Haag, Adv. Synth. Catal.
352 (2010) 2497;
[25] H. Yang, Z. Ma, Y. Wang, Y. Wang, L. Fang, Chem. Commun. 46 (2010) 8659.
[26] C. Hobbs, Y.-C. Yang, J. Ling, S. Nicola, H.-L. Su, H.S. Bazzi, D.E. Bergbreiter, Org.
Lett. 13 (2011) 3904.
(e) F. Mummy, R. Haag, Synlett 23 (2012) 2672;
(f) A. Berkessel, J. Krämer, F. Mummy, J.-M. Neudörfl, R. Haag, Angew. Chem.,
Int. Ed. 52 (2013) 739.
[27] P. Nieczypor, W. Buchowicz, W.J.N. Meester, F.P.J.T. Rutjes, J.C. Mol,
Tetrahedron Lett. 42 (2001) 7103.
[28] J.P. Gallivan, J.P. Gordan, R.H. Grubbs, Tetrahedron Lett. 46 (2005) 2577.
[29] L. Jafarpour, M.-P. Heck, C. Baylon, H.M. Lee, C. Mioskowski, S.P. Nolan,
Organometallics 21 (2002) 671.
[12] J.S. Kingsbury, S.B. Garber, J.M. Giftos, B.L. Gray, M.M. Okamato, R.A. Farrer, J.T.
Fourkas, A.H. Hoveyda, Angew. Chem., Int. Ed. 40 (2001) 4251.
[13] M. Scholl, S. Ding, C.W. Lee, R.H. Grubbs, Org. Lett. 1 (1999) 953.
[14] T.A. Kirkland, R.H. Grubbs, J. Org. Chem. 62 (1997) 7310.
[15] T. Ritter, A. Hejl, A.G. Wenzel, T.W. Funk, R.H. Grubbs, Organometallics 25
(2006) 5740.
ˇ
ˇ
ˇ
[30] D. Bek, H. Balcar, N. Zilková, A. Zukal, M. Horácek, J. Cejka, ACS Catal. 1 (2011)
709.