polymers
Review
Supported Catalysts Useful in Ring-Closing
Metathesis, Cross Metathesis, and Ring-Opening
Metathesis Polymerization
Jakkrit Suriboot 1 , Hassan S. Bazzi 2 and David E. Bergbreiter 1, *
1
2
*
Department of Chemistry, Texas A & M University, College Station, TX 77840, USA;
[email protected]
Department of Chemistry, Texas A & M University at Qatar, P.O. Box 23874, Doha, Qatar; [email protected]
Correspondence: [email protected]; Tel.: +1-979-845-3437
Academic Editor: Changle Chen
Received: 17 March 2016; Accepted: 7 April 2016; Published: 12 April 2016
Abstract: Ruthenium and molybdenum catalysts are widely used in synthesis of both small molecules
and macromolecules. While major developments have led to new increasingly active catalysts that
have high functional group compatibility and stereoselectivity, catalyst/product separation, catalyst
recycling, and/or catalyst residue/product separation remain an issue in some applications of these
catalysts. This review highlights some of the history of efforts to address these problems, first
discussing the problem in the context of reactions like ring-closing metathesis and cross metathesis
catalysis used in the synthesis of low molecular weight compounds. It then discusses in more detail
progress in dealing with these issues in ring opening metathesis polymerization chemistry. Such
approaches depend on a biphasic solid/liquid or liquid separation and can use either always biphasic
or sometimes biphasic systems and approaches to this problem using insoluble inorganic supports,
insoluble crosslinked polymeric organic supports, soluble polymeric supports, ionic liquids and
fluorous phases are discussed.
Keywords: ROMP; supported catalysts; olefin metathesis; green chemistry
1. Introduction
The use of metal complexes in catalysis has become a standard practice in organic synthesis.
Olefin metathesis is among these various catalytic reactions. The carbon-carbon double bond
rearrangements effected by olefin metathesis have developed into one of the most powerful tools in
organic synthesis due to the facility with which this chemistry constructs carbon-carbon double bonds
and the compatibility of this reaction with other functionalities [1,2]. The search for novel applications
of olefin metathesis reactions and study of improvements of the catalysts’ reactivity, stability, and
selectivity have been extremely active fields since the breakthrough introduction of the current
two most recognized alkylidene-types of catalyst families of ruthenium and molybdenum based
complexes introduced in the 1980s by the Grubbs and Schrock groups, respectively (Figure 1) [3,4].
Examples of these catalysts are bis(tricyclohexylphosphine)benzylidine ruthenium(II) dichloride
(a Grubbs 1st generation catalyst) 1 and 2,6-diisopropylphenylimidoneophylidene molybdenum(IV)
bis(hexafluoro-t-butoxide) (a Schrock catalyst) 4. A critical aspect of the development of these families
of catalysts is the fact that the properties of these olefin metathesis catalysts can be modified by
alteration of the organic ligands in these ruthenium and molybdenum complexes. The success
and impact of such modifications have been the subject of many studies that have led to a variety
of organic ligands for both of the catalyst families (Figure 1). For example, in the case of Ru
metathesis catalysts, these studies led to a variety of N-heterocyclic carbene complexes including
Polymers 2016, 8, 140; doi:10.3390/polym8040140
www.mdpi.com/journal/polymers
Polymers 2016, 8, 140
2 of 23
Polymers 2016, 8, 140
2 of 22
(1,3-bis-(2,4,6-trimethylphenyl)-2-imidazolidinylidene)-dichloro(o-isopropoxyphenylmethylene)
ruthenium
ruthenium (an example of a Hoveyda-Grubbs 2nd generation catalyst) 2 and [1,3-bis(2,4,6(an
example
of
a
Hoveyda-Grubbs
2nd
generation
catalyst)
2
and
[1,3-bis(2,4,6-trimethylphenyl)trimethylphenyl)-2-imidazolidinylidene]dichloro-(phenylmethylene)bis(3-bromopyridine) ruthenium(II)
2-imidazolidinylidene]dichloro-(phenylmethylene)bis(3-bromopyridine)
ruthenium(II)
(a Grubbs
(a
Grubbs 3rd generation catalyst) 3. Similar studies that use ligands to influence
catalyst utility
have
3rd
generation
catalyst)
3.
Similar
studies
that
use
ligands
to
influence
catalyst
utility
have
been
carried
been carried out with Mo catalysts leading, for example, to more useful chiral catalysts like
2,6out with Mo catalysts leading, for example, to more useful chiral catalysts like 2,6-diisopropylphenylimidodiisopropylphenylimido-neophylidene-[(S)-(−)-BIPHEN]-molybdenum(IV)
catalysts (an (S)-Schrockneophylidene-[(S)-(´)-BIPHEN]-molybdenum(IV)
catalysts (an (S)-Schrock-Hoveyda catalyst) 5.
Hoveyda
catalyst) 5.
Figure
1–5.
Figure 1.
1. Examples
Examples of
of olefin
olefin metathesis
metathesis catalysts
catalysts 1–5.
Although these studies have led to newer versions of olefin metathesis catalysts that have
Although these studies have led to newer versions of olefin metathesis catalysts that have
improved reactivity, air and moisture stability, functional groups tolerance, and stereoselectivity,
improved reactivity, air and moisture stability, functional groups tolerance, and stereoselectivity,
challenges still remain in using these transition metal complexes. Challenges that remain include: (i)
challenges still remain in using these transition metal complexes. Challenges that remain include:
the high cost of the transition metal complexes used as catalysts or precatalysts; (ii) the cost and/or
(i) the high cost of the transition metal complexes used as catalysts or precatalysts; (ii) the cost and/or
tediousness of the ligand syntheses; and (iii) the potential environmental toxicological or practical
tediousness of the ligand syntheses; and (iii) the potential environmental toxicological or practical
concerns that ensue when there is significant metal or ligand contamination in the product. This last
concerns that ensue when there is significant metal or ligand contamination in the product. This last
issue is a green chemistry issue in that additional process steps have to be used to sequester or remove
issue is a green chemistry issue in that additional process steps have to be used to sequester or remove
metals or ligands from products. Such steps increase overall process cost and can be an
metals or ligands from products. Such steps increase overall process cost and can be an environmental
environmental issue because of the waste that is typically generated in such catalyst/ligand/product
issue because of the waste that is typically generated in such catalyst/ligand/product separation steps.
separation steps. The last issue is especially critical in pharmaceutical industry and in materials
The last issue is especially critical in pharmaceutical industry and in materials synthesis. In the case of
synthesis. In the case of pharmaceuticals, the acceptable level of ruthenium content in products
pharmaceuticals, the acceptable level of ruthenium content in products should be less than 10 ppm in
should be less than 10 ppm in the final compound [5–7]. An efficient separation of ruthenium
the final compound [5–7]. An efficient separation of ruthenium impurities is also important in the case
impurities is also important in the case of polymeric materials used in electronic or device
of polymeric materials used in electronic or device applications. In addition, other issues can arise. For
applications. In addition, other issues can arise. For example, in some syntheses, an efficient
example, in some syntheses, an efficient separation of ruthenium impurities is important as ruthenium
separation of ruthenium impurities is important as ruthenium residues can lead to undesirable sideresidues can lead to undesirable side-reactions like hydrogenation or alkene isomerization in reaction
reactions like hydrogenation or alkene isomerization in reaction products [8,9].
products [8,9].
There are several ways to approach the three issues noted above. Perhaps the simplest approach
There are several ways to approach the three issues noted above. Perhaps the simplest approach is
is to simply reduce the amount of catalyst that is used in the reaction. Originally, ring-closing
to simply reduce the amount of catalyst that is used in the reaction. Originally, ring-closing metathesis
metathesis (RCM) reactions required catalyst loadings that were 1–5 mol % or greater depending on
(RCM) reactions required catalyst loadings that were 1–5 mol % or greater depending on specific
specific application. However, this level of catalyst loadings seems to be an overestimate. For
application. However, this level of catalyst loadings seems to be an overestimate. For example,
example, according to the studies by Mol and Dinger in 2002, they found that catalyst loadings could
according to the studies by Mol and Dinger in 2002, they found that catalyst loadings could be much
be much lower in some cases. Their work shows that in suitable cases Ru-catalyzed olefin metathesis
lower in some cases. Their work shows that in suitable cases Ru-catalyzed olefin metathesis can be
can be carried out effectively with catalyst loadings that are several orders of magnitude lower than
carried out effectively with catalyst loadings that are several orders of magnitude lower than normally
normally reported [10]. In suitable cases, the effective turnover number for the Ru catalysts can be as
reported [10]. In suitable cases, the effective turnover number for the Ru catalysts can be as high as
high as 600,000. While these studies focused on achieving the maximum turnover number, the
reactions with very low catalyst loadings in some cases never reached 100% conversion. Nonetheless,
Polymers 2016, 8, 140
3 of 23
600,000. While these studies focused on achieving the maximum turnover number, the reactions with
very low catalyst loadings in some cases never reached 100% conversion. Nonetheless, their work
demonstrated the potential of lower mol % loadings of Ru catalysts, a potential that has been realized
by others such as the Grubbs group who reported success in reducing the catalyst loadings to be as
low as 25 ppm in ring-closing metathesis of diethyl diallylmalonate, by employing more sterically
demanding and reactive Ru-based catalysts [7].
An alternative strategy to address the issue of separation of transition metal complexes and
products in metathesis that has been also used in other transition metal catalyzed chemistry is to
design the catalyst and its ligands such that both can be efficiently separated, recovered, and recycled.
If such separations recover an active catalyst, a reuse of this catalyst can in effect increase turnover
numbers. In any case, a simpler way to separate catalyst residues from products can minimize the
effects of catalyst or ligand contamination in products.
Various methods to separate, recover and recycle the catalyst have been developed. The most
classical approach is to design a catalyst such that it is coupled to an always insoluble support
such as silica gel or a cross-linked and thus insoluble polymer. Such an approach leads to so-called
heterogeneized catalysts on supports that are always insoluble before, during and after a reaction. A
second approach is to design a catalyst such that it has some sort of phase selective solubility. Examples
of this second approach have involved ligands that contain ionic liquid compatible functionality, or
fluorocarbon groups that make metathesis catalysts soluble in fluorocarbon solvents, or soluble
polymers that lead to catalysts whose solubility mirrors that of the polymer to which they are attached.
Unlike cases where catalyst immobilization involves an insoluble support, these catalysts can typically
be characterized by conventional spectroscopy. A further advantage of each of these approaches is that
the catalysts that can be used as homogeneous catalysts avoiding potential problems of heterogeneous
catalysts. However, in each of these three cases, some sort of biphasic separation is required after a
reaction to separate the tagged catalyst and ligands from the products. Such separations can involve a
simple filtration in the case of heterogeneous supports. Soluble supports usually require some sort of
liquid/liquid separation or extraction to separate catalysts and products. The discussion below briefly
summarizes some examples where heterogeneous and homogeneous supports are used in metathesis.
It then describes in more comprehensive detail examples where phase separable metathesis catalysts
are used in ring-opening metathesis polymerization (ROMP).
2. Heterogeneous Supported Olefin Metathesis Catalysts
The use of a heterogeneous insoluble support is among the oldest and most widely used tool
to effect separation and isolation of catalysts from the products. The first insoluble organic support
explored by the scientific community that was later used for homogeneous catalysts was based on a
paper that was published by Merrifield—work that subsequently earned Merrifield a Nobel Prize [11].
While Merrifield’s work was directed at using cross-linked polystyrene resins (Merrifield’s resin) in
peptide and nucleotide synthesis [11], others recognized that the physical separation of a growing
peptide bound to an insoluble organic support he described had potential applications in separation
of homogeneous catalysts and products and that if catalysts had stable stationary states that such
catalysts could be recycled. Thus, this discovery of Merrifield that focused on peptide synthesis led
to many studies using similar insoluble polymeric materials as supports for homogeneous catalysts.
As is true in peptide synthesis, catalysts immobilized on these divinylbenzene (DVB)-cross-linked
polystyrene supports have the principle advantage of allowing for separation of catalysts and their
ligands from a solution of the product in the solvent phase of a reaction mixture via simple filtration.
In some cases, heterogeneous supports can also improve catalyst stability and prevent bimolecular
decomposition pathways via a phenomenon known as site isolation [12–15]. However, in other cases,
the heterogeneity of an otherwise homogeneous catalyst can change activity in undesirable ways. If
the goal of immobilization is to replace a homogeneous catalyst with a separable catalyst with the
same activity, this is undesirable.
Polymers 2016, 8, 140
4 of 23
The discussion below first describes in a general way some examples where olefin metathesis
catalysts have been immobilized on always phase separated supports (i.e., insoluble inorganic
or cross-linked polymer supports). It then describes similar chemistry that uses sometimes
phase-separated supports (i.e., biphasic catalysis or soluble polymers that can be phase separated after
a catalytic reaction). Finally, it goes on to discuss in more detail issues associated with the specific
example of immobilized ring-opening metathesis polymerization (ROMP) catalysis.
Ru metathesis catalysts like those shown in Figure 2 have been immobilized on divinylbenzene
cross-linked polystyrene (DVB-PS). An early example of such catalyst was 6 which was first reported
by Grubbs and Nguyen in 1995 [16]. In this example, they used an established type of supported
ligand—a phosphine modified DVB-PS. This supported catalyst had extended lifetimes. This was
ascribed to the reduced diffusivity of the catalyst molecules on the polystyrene support, which prevents
a decomposition pathway that occurs via a bimolecular reaction. However, while 6 was recycled
three times in a metathesis reaction of cis-2-pentene to form cross metathesis products of 3-hexene
and 2-butene, the catalyst lost 20% of its activity after each cycle. This catalyst also had a modest
activity that was ascribed to (i) incomplete substitution of phosphine; (ii) the diffusion limit of olefin
into the cavities of cross-linked DVB-PS support; and (iii) the local high concentration of phosphine
on the support. The amount of Ru leaching in the products was not analyzed. However, given that
the Ru complex is immobilized by a labile phosphine ligand, given that the Ru and phosphine ligand
dissociate in the reaction, and that the recovered polymer-bound catalyst loses significant activity
cycle to cycle, it is likely that there is a significant amount of Ru leaching. Several years later, Barrett
and co-workers described an alternative scheme that also used DVB-PS as a support. In this case,
they immobilized a Ru pre-catalyst onto DVB-PS using a benzylidine group. In this second approach,
Barrett introduced a new concept called the “boomerang” effect [17]. This was a quite different scheme
for use of a heterogeneous-supported catalyst. In this case a pre-catalyst 7 was synthesized by a
metathesis reaction, shaking Grubbs 1st generation catalyst and a vinyl-substituted DVB-PS derivative
for 2 h in dichloromethane. The resulting benzylidene-immobilized pre-catalyst 7 is immobilized
on the insoluble resin by a strong bond and the immobilized pre-catalyst was isolated by filtration.
According to their report, 7 was indefinitely stable under normal atmospheric conditions with no
loss of activity. In the application of 7, a so-called boomerang reaction was involved. In this scheme,
the Ru initially undergoes a metathesis reaction with an alkene of a ring-closing metathesis (RCM)
substrate like diethyl diallylmalonate. The resulting metallocycle then forms a new Ru alkylidene
and the original vinyl-substituted DVB-PS. The alkylidene complex goes into solution, becoming
a homogeneous catalyst effecting ring-closing metathesis (RCM) on the rest of the substrate. Then,
when the substrate is consumed, the authors postulate that the remaining soluble alkylidene reacts
with the vinyl-substituted DVB-PS. This leads to a recapture of the Ru complex by the resin by a
strong bond after the completion of the reaction. This behavior of the ruthenium catalyst where it
comes off the resin and returns later was likened to the action of a “boomerang”. In this work, the
catalyst was recycled up to three times in an RCM reaction of diethyl diallylmalonate, with addition
of 3,3-dimethyl-1-butene (9 mol % based on diethyl diallylmalonate), with Ru contamination in the
product of 500 ppm. Catalyst recycling was accomplished by simple filtration of the solid-supported
catalyst and the product was isolated from the filtrate by evaporation of the solvent.
Following Grubbs’ and Barrett’s work on polymer-supported catalysts and the other studies of
homogeneous Ru metathesis catalysts that had shown the utility of Ru-based alkylidene catalysts that
contained N-heterocyclic carbene (NHC) ligands, Blechert and co-workers described the synthesis
of Grubbs 2nd generation catalysts like 8 that were immobilized on DVB-PS via an N-heterocyclic
carbene ligand [18]. An advantage of this approach using N-heterocyclic carbene (NHC) ligands to
immobilize a Ru complex is that an NHC ligand is a stronger sigma donor than phosphine ligands.
Unlike phosphine ligands like those in 7 that are thought to dissociate from the ruthenium center to
initiate catalysis [19], the NHC ligand should remain bound to the ruthenium center before, during,
and after the metathesis reaction. Thus, leaching of Ru from the resin was expected to be reduced [18].
Polymers 2016, 8, 140
5 of 23
In the first example of this chemistry, the solid-supported catalyst 8 was prepared by ligand exchange
between
a DVB-PS-supported
N-heterocyclic carbene ligand and the phosphine ligand on Grubbs
Polymers
2016,
8, 140
5 of 522of 22
Polymers
2016,
8, 140
1st generation catalyst 1. The resulting DVB-PS supported heterogeneous ruthenium catalyst was
5 of 22
characterized
by 31 PbyNMR
anduse
IR 8of
spectroscopy.
The
general
reactivity
of
8 wasRCM
demonstrated
by
waswas
demonstrated
thethe
use
of
in8 various
metathesis
reactions
including
andand
yne-ene
demonstrated
by
in various
metathesis
reactions
including
RCM
yne-ene
the
use
of
8
in
various
metathesis
reactions
including
RCM
and
yne-ene
metathesis
reactions.
In
the
reactions.
In the
casecase
of ring-closing
metathesis
reaction
of diethyl
diallylmalonate,
up up
to to
s including RCM metathesis
and
yne-ene
metathesis
reactions.
In the
of ring-closing
metathesis
reaction
of diethyl
diallylmalonate,
case
of
ring-closing
metathesis
reaction
of
diethyl
diallylmalonate,
up
to
four
cycles
of
cyclization
of
four
cycles
of
cyclization
of
diethyl
diallylmalonate
to
4,4-dicarboethoxycyclopentene
could
be
f diethyl diallylmalonate,
to of cyclization of diethyl diallylmalonate to 4,4-dicarboethoxycyclopentene could be
four up
cycles
diethyl
diallylmalonate
toreports
4,4-dicarboethoxycyclopentene
could
be effected
using
8.
These
reports
effected
using
8. These
reports
stated
thatthat
thethe
products
were
obtained
as colorless
solids
or oils.
While
oethoxycyclopentene
could
be
effected
using
8. These
stated
products
were
obtained
as colorless
solids
or oils.
While
stated
that
the
products
were
obtained
as
colorless
solids
or
oils.
While
this
visual
assay
of
Ru
this
visual
assay
of
Ru
leaching
is
promising,
a
more
quantitative
assay
of
the
amount
of
Ru
leaching
as colorless solids or oils.
While assay of Ru leaching is promising, a more quantitative assay of the amount of Ru leaching
this visual
leaching
isreported.
promising,
aThis
more
quantitative
assay of extended
theextended
amount
wasthe
not
reported.
This
was
not
This
work
waswas
subsequently
inofainRu
report
where
Blechert
group
ay of the amount of
Ru
leaching
was
not
reported.
work
subsequently
a leaching
report
where
the
Blechert
group
work
was
subsequently
extended
in
a
report
where
the
Blechert
group
reported
solid-supported
Ruand
reported
solid-supported
Ru Ru
metathesis
catalysts
likelike
9 and
10 that
hadhad
high
stability,
activity,
and
port where the Blechert
group
reported
solid-supported
metathesis
catalysts
9 and
10 that
high
stability,
activity,
catalysts
like
9 andexperiments
10experiments
that had high
stability,
activity,
andboth
recyclability
[20].
In
recycling
recyclability
[20].
In recycling
using
these
two
catalysts,
catalysts
were
recycled
up up
had high stability,metathesis
activity,
and
recyclability
[20].
In recycling
using
these
two
catalysts,
both
catalysts
were
recycled
experiments
using
these
two
catalysts,
both
catalysts
were
recycled
up
to
four
times
in
RCM
reaction
of
torecycled
four
times
reaction
of diallyltosylamide.
However,
thethe
ability
to catalyze
cross
metathesis
both catalysts were
upin RCM
to four
times
in RCM
reaction
of diallyltosylamide.
However,
ability
to catalyze
cross
metathesis
diallyltosylamide.
However,
the
ability
to
catalyze
cross
metathesis
(CM)
reactions
of
complex
9
was
(CM)
reactions
of complex
9 was
showed
to be
higher
than
thatthat
of complex
10. 10.
ForFor
example,
thethe
lity to catalyze cross
metathesis
(CM)
reactions
of complex
9 was
showed
to be
higher
than
of complex
example,
showed
to
be
higher
than
that
of
complex
10.
For
example,
the
conversion
of
CM
between
acrylonitrile
conversion
of
CM
between
acrylonitrile
and
4-pentenyl
benzoate
catalyzed
by
9
was
98%
in
12
complex 10. For example,
the of CM between acrylonitrile and 4-pentenyl benzoate catalyzed by 9 was 98% in h
conversion
12 h
and
4-pentenyl
benzoate
catalyzed
by
9inwas
in 12
while
the
conversion
was only
15%
insuperior
12 h
the
conversion
waswas
only
15%15%
in 12
h12inh98%
the
case
ofh 10.
TheThe
authors
suggested
thatthat
the
superior
atalyzed by 9 waswhile
98%
in 12
hconversion
while
the
only
in the
case
of 10.
authors
suggested
the
the
case
10.
The authors
suggested
that
the
superior
activity
of dissociate
complexin9 owe
to thebecoming
ability
of
activity
of ofof
complex
9 owe
to to
thethe
ability
of of
the
catalyst
to to
dissociate
solution,
ors suggested thatin
the
superior
activity
complex
9 owe
ability
the
catalyst
in
solution,
becoming
the
catalyst
to
dissociate
in
solution,
becoming
homogeneous
active
species,
unlike
complex
10.
The
homogeneous
active
species,
unlike
complex
10.
The
leaching
of
Ru
in
the
products
was
not
sociate in solution, homogeneous
becoming
active species, unlike complex 10. The leaching of Ru in the products was not
leaching
Ru in the products was not described.
described.
Ru in the products
wasofnot
described.
Figure
2.
of of
DVB-PS-supported
( ((= divinylbenzene
cross-linked
polystyrene
or oror=
Figure
2.Examples
Examples
of DVB-PS-supported
DVB-PS-supported
cross-linked
polystyrene
Figure
2. Examples
= divinylbenzene
divinylbenzene
cross-linked
polystyrene
cross-linked
4-vinylpyridine)
olefin
metathesis
catalysts
6–11.
cross-linked
4-vinylpyridine)
olefin
metathesis
catalysts
6–11.
s-linked polystyrene or divinylbenzene
==divinylbenzene
divinylbenzene
cross-linked
4-vinylpyridine)
olefin
metathesis
catalysts
6–11.
s 6–11.
=
The invention
of new
highly
active
ruthenium
olefin
metathesis
catalyst
pyridine-ligated
Grubbs
invention
of new
highly
active
ruthenium
olefin
metathesis
catalyst
pyridine-ligated
Grubbs
TheThe
invention
of new
highly
active
ruthenium
olefin
metathesis
catalyst
pyridine-ligated
Grubbs
3rd
generation
catalysts
has
led
to
a
new
type
of
Ru
catalyst
which
can
catalyze
cross
metathesis
atalyst pyridine-ligated
Grubbs
3rd
generation
catalysts
has
led
to
a
new
type
of
Ru
catalyst
which
can
catalyze
cross
metathesis
3rd generation catalysts has led to a new type of Ru catalyst which can catalyze cross metathesis
reactions
of a broader
range
of substrates
and afford
polymers
with narrow
polydispersity
(PDI) by by
h can catalyze cross
metathesis
reactions
a broader
range
of substrates
afford
polymers
narrow
polydispersity
reactions
of aof
broader
range
of substrates
and and
afford
polymers
withwith
narrow
polydispersity
(PDI)(PDI)
by a
a
ring-opening
metathesis
polymerization
(ROMP).
Grela
and
Kirschning
reported
the
first
example
narrow polydispersity
(PDI)
by
a
ring-opening
metathesis
polymerization
(ROMP).
Grela
and
Kirschning
reported
the
first
example
ring-opening metathesis polymerization (ROMP). Grela and Kirschning reported the first example of a
of
aofsolid-supported
version of this
catalyst
in 2005
[21].
Their
report
noted
thethe
possibility
of using
ning reported the solid-supported
first
example
a solid-supported
of this
catalyst
2005
[21].
Their
report
noted
possibility
using
versionversion
of this catalyst
in 2005 in
[21].
Their
report
noted
the
possibility
of usingofthis
thisthis
supported
catalyst
in aincontinuous
flow
process
duedue
to the
ability
to reload
thethe
catalytic
species
t noted the possibility
of
using
supported
catalyst
a
continuous
flow
process
to
the
ability
to
reload
catalytic
species
supported catalyst in a continuous flow process due to the ability to reload the catalytic species onto
onto the
same
solid
support.
Thus, even
if leaching
were
to occur,
thethe
catalyst could
be be
easily
be be
y to reload the catalytic
species
the
same
solid
support.
even
if leaching
were
to occur,
easily
the onto
same
solid
support.
Thus,
evenThus,
if leaching
were
to occur,
the catalyst
could catalyst
be easilycould
be regenerated.
regenerated.
The
recyclability
of
11
was
tested
in
a
ring-closing
metathesis
reaction
of
diethyl
the catalyst couldThe
beregenerated.
easily be The
was tested metathesis
in a ring-closing
of diethyl
recyclability
of 11recyclability
was tested of
in a11ring-closing
reactionmetathesis
of diethyl reaction
diallylmalonate
diallylmalonate
at
110
°C.
This
solid-supported
catalyst
showed
activity
up
to
5
cycles
but
a
decrease
metathesis reaction
of
diethyl
diallylmalonate
at 110 °C. Thiscatalyst
solid-supported
catalyst up
showed
activitybut
upatodecrease
5 cycles but
a decrease
˝ C. This solid-supported
at 110
showed activity
to 5 cycles
in product
in product
yield
waswas
noted
in each
subsequent
cycle.
TheThe
authors
ascribed
thisthis
to the
thermal
ity up to 5 cycles but
a
decrease
in
product
yield
noted
in
each
subsequent
cycle.
authors
ascribed
to
the
thermal
yield was noted in each subsequent cycle. The authors ascribed this to the thermal instability of
11
instability
of 11
or
leaching
of Ru
from
thethe
weakly
bound
pyridine
ligands.
However,
thethe
authors
s ascribed this toor
the
thermal
instability
of
11
or
leaching
of
Ru
from
weakly
bound
pyridine
ligands.
However,
authors
leaching of Ru from the weakly bound pyridine ligands. However, the authors were able to show
were
able
to show
thatthat
11 could
indeed
be reactivated
by washing/Ru
re-addition
protocol
(1 M
ligands. However,
the
authors
were
able
to show
11 could
indeed
be reactivated
by washing/Ru
re-addition
protocol
(1 HCl,
M HCl,
1
M
NaOH,
H
2
O,
MeOH,
toluene,
then
addition
of
3).
Ru
contamination
in
the
products
was
notnot
re-addition protocol1(1MMNaOH,
HCl, H2O, MeOH, toluene, then addition of 3). Ru contamination in the products was
mentioned.
nation in the products
was not
mentioned.
TheThe
other
general
strategy
for for
catalyst
immobilization
on insoluble
supports
commonly
is toisuse
other
general
strategy
catalyst
immobilization
on insoluble
supports
commonly
to use
inorganic
supports.
Silica-based
materials
are
the
most
common
examples
of
these
sorts
of solid
ble supports commonly
is to usesupports. Silica-based materials are the most common examples of these sorts
inorganic
of solid
Polymers 2016, 8, 140
6 of 23
that 11 could indeed be reactivated by washing/Ru re-addition protocol (1 M HCl, 1 M NaOH, H2 O,
MeOH, toluene, then addition of 3). Ru contamination in the products was not mentioned.
The other general strategy for catalyst immobilization on insoluble supports commonly is to
use inorganic supports. Silica-based materials are the most common examples of these sorts of solid
supports. Select examples of Ru metathesis catalysts immobilized on silica are shown in Figure 3.
Fürstner and co-workers reported the synthesis of an immobilized Grubbs 2nd generation ruthenium
complex on silica gel using hydroxyalkyl groups on N-heterocyclic carbene ligand 12 [22]. This
immobilized catalyst was successfully recycled in RCM reaction of diethyl diallylmalonate through
three cycles. The catalyst 12 was separated from the RCM product by simple filtration. The crude
product was isolated as colorless liquid. The authors did report a detailed analysis of Ru leaching
noting that the products had ca. 250 ppm of Ru contamination based on ICP-MS analysis. Grubbs
and co-workers have also described other versions of silica-supported olefin metathesis ruthenium
catalysts 13 and 14 [23]. The silica-supported complexes 13 and 14 were competent catalysts in RCM
and CM reactions with reactivity that was comparable to that of their homogeneous analogs. The
catalytic activity toward an RCM reaction of diethyl diallylmalonate of catalyst 14 (31% conversion
after 10 min) was shown to have slightly lower than the catalytic activity of 13 (67% conversion after
10 min). The authors also showed that immobilized ruthenium catalyst 13 can be recycled up to
eight times in RCM reaction of allyl(2-methyl-2-propenyl)malonate with conversions of substrate
to product in the 60%–80% range when reaction time was 2 h, conversions that could reach 100%
with reaction times of 12 h. The authors suggested that these catalysts improve recyclability and
eliminate issues associated with the decomposition of the ruthenium complex via bimolecular pathway.
Such immobilized ruthenium catalysts on silica support have less intermolecular activity between the
catalysts—the same phenomena reported earlier by Grubbs’ group for site isolated DVB-PS supported
species [16]. The analysis of the Ru leaching in this case would seem to support this claim. In this
example, the ruthenium leaching studied by ICP-MS revealed the contamination level in products to
be less than 5 ppb for those prepared by both 13 and 14. This is especially notable since it is by several
orders of magnitude the lowest Ru leaching ever reported. Ying group also described using click
chemistry for the immobilization of Hoveyda-Grubbs type complexes on nanoporous silica 15 [24].
The catalyst they prepared exhibits good activity and stability as well as recyclability. In addition,
these authors demonstrated that this catalyst can be used in a circulating flow reactor. The catalyst was
reused in RCM reaction of diethyl diallylmalonate over 8 times with overall conversions of 90% with
Ru leaching levels of 11.3 ppm at the first 60 min and 1.6 ppm at 180 min based on ICP-MS analysis
of the isolated products. Yet another example of silica-bound Ru metathesis catalysts was reported
by Balcar and co-workers who used commercially available molecular sieves as supports for the Ru
catalyst 16 [25]. These SBA-15 molecular sieves possess several advantages including high surface
area, narrow pore size distribution, and high thermal and mechanical stability. The solid-supported
ruthenium complex 16 on such sieves was shown to be competent as an RCM catalyst using diethyl
diallylmalonate as a substrate. The leaching of Ru into RCM product was found to be as low as 17 ppm.
However, attempts to recycle this catalyst were unsuccessful, with the conversion reaching 90% for
only two cycles. More recently, Monge-Marcet and co-workers also reported a synthesis of recyclable
silica-supported Hoveyda-Grubbs type complex using an NHC ligand 17 [26]. This catalyst proved to
be recyclable for RCM reaction of diethyl diallylmalonate with Ru contamination of 221 ppm found in
the product based on ICP-MS of products from the first cycle. Ru leaching in subsequent cycles was
not measured but would likely be less as Ru leaching in the first or second cycles of a catalytic reaction
could reflect undetectable issues in the synthesis of a catalyst. If a catalyst were only 99% pure (pure
by most analyses used in synthesis), the 1% impurity would represent most or all of the observed Ru
leaching but would not be relevant in later cycles.
Polymers 2016, 8, 140
7 of 23
Polymers 2016, 8, 140
7 of 22
Figure 3.
3. Examples
of silica-supported
silica-supported olefin
olefin metathesis
metathesis catalysts
catalysts 12–17.
12–17.
Figure
Examples of
Schrock’s molybdenum olefin metathesis catalysts have also been supported on always insoluble
Schrock’s molybdenum olefin metathesis catalysts have also been supported on always insoluble
supports. Some examples of such silica-supported catalysts supports are shown in Figure 4. For
supports. Some examples of such silica-supported catalysts supports are shown in Figure 4. For
example, Schrock and co-workers described the synthesis of well-defined surface immobilized
example, Schrock and co-workers described the synthesis of well-defined surface immobilized catalysts
catalysts 18 and 19 [27]. These two silica-supported catalysts showed very similar activities in cross
18 and 19 [27]. These two silica-supported catalysts showed very similar activities in cross metathesis
metathesis reaction of ethyl oleate (EO) with TOF (turnover frequency measured after 5 min of
reaction of ethyl oleate (EO) with TOF (turnover frequency measured after 5 min of reaction expressed
reaction expressed in mol of substrate converted per mol of Mo per sec) of 0.04 and 0.03, respectively.
in mol of substrate converted per mol of Mo per sec) of 0.04 and 0.03, respectively. However, the
However, the silica-supported 18 was reportedly more stable than 19, a difference that was ascribed
silica-supported 18 was reportedly more stable than 19, a difference that was ascribed to the site
to the site isolation of metal complexes on the silica support [28]. The stability of 18 and 19 were
isolation of metal complexes on the silica support [28]. The stability of 18 and 19 were determined by
determined by the time needed to reach equilibrium in the CM reaction of ethyl oleate (i.e., around
the time needed to reach equilibrium in the CM reaction of ethyl oleate (i.e., around 50% conversion),
50% conversion), which was 1 and 24 h, respectively (the longer the time was presumed to reflect a
which was 1 and 24 h, respectively (the longer the time was presumed to reflect a faster decomposition
faster decomposition rate for the catalyst). Shortly later, Schrock and co-workers developed more
rate for the catalyst). Shortly later, Schrock and co-workers developed more active, stable, and selective
active, stable, and selective silica supported molybdenum olefin metathesis catalyst 20 [29]. The
silica supported molybdenum olefin metathesis catalyst 20 [29]. The increase in reactivity was achieved
increase in reactivity was achieved by replacing one imido group with a siloxy group from the
by replacing one imido group with a siloxy group from the surface. Keeping one remaining imido
surface. Keeping one remaining imido ligand enhanced the stability of the molybdenum catalyst.
ligand enhanced the stability of the molybdenum catalyst. This catalyst was also tested its activity
This catalyst was also tested its activity in CM of ethyl oleate. The results showed that the TOF of 20
in CM of ethyl oleate. The results showed that the TOF of 20 was 0.15 and the time needed to reach
was 0.15 and the time needed to reach reaction equilibrium was only 10 min. More recently, Schrock
reaction equilibrium was only 10 min. More recently, Schrock group described the silica immobilized
group described the silica immobilized molybdenum alkene metathesis catalyst 21 that showed
molybdenum alkene metathesis catalyst 21 that showed enhancement in metathesis activity compared
enhancement in metathesis activity compared to its low molecular weight analog in CM reactions of
to its low molecular weight analog in CM reactions of ethyl oleate. The TOFs of 21 and its low
ethyl oleate. The TOFs of 21 and its low molecular weight analog were 0.07 and 0.01, respectively
molecular weight analog were 0.07 and 0.01, respectively [30]. None of these reports reported an
[30]. None of these reports reported an analysis of Mo leaching.
analysis of Mo leaching.
Polymers 2016, 8, 140
8 of 22
Polymers 2016, 8, 140
Polymers 2016, 8, 140
8 of 23
8 of 22
Figure 4. Examples of silica-supported olefin metathesis catalysts 18–21.
The first recyclable supported chiral olefin metathesis catalyst was reported by Hoveyda and
Figure 4. Examples of silica-supported olefin metathesis catalysts 18–21.
Schrock in 2002 [31].
Examples
of these
catalysts are olefin
shown
in Figure
5. The18–21.
polystyrene-supported
Figure
4. Examples
of silica-supported
metathesis
catalysts
catalyst 22 showed similar activity to its homogeneous analog both in terms of yield and
recyclable
supportedchiral
chiral olefin
olefin metathesis
by by
Hoveyda
andand
TheThe
firstfirst
recyclable
supported
metathesis
catalystwas
wasreported
reported
enantioselectivity
in asymmetric
ring-opening
cross catalyst
metathesis
reaction
of Hoveyda
(1R,4S,7R)-7Schrock
in
2002
[31].
Examples
of
these
catalysts
are
shown
in
Figure
5.
The
polystyrene-supported
catalyst
Schrock in 2002 [31]. Examples of these
are shown
in Figure
5. The
(benzyloxy)bicyclo[2.2.1]hept-2-ene
31 catalysts
and styrene
to yield
product
32 polystyrene-supported
(92% yield and 98%
22 showed similar activity to its homogeneous analog both in terms of yield and enantioselectivity in
catalyst
22
showed
similar
activity
to
its
homogeneous
analog
both
in
of three
yield times
and
enantiomeric excess (ee)) (Scheme 1). The polymer-supported complex can beterms
recycled
asymmetric ring-opening cross metathesis reaction of (1R,4S,7R)-7-(benzyloxy)bicyclo[2.2.1]hept-2-ene
enantioselectivity
in asymmetric
ring-opening
cross However,
metathesisthere
reaction
of (1R,4S,7R)-7but
conversion
significantly
dropped
in yield
the third
cycle.
was(Scheme
little
difference
31 and styrene
to yield product
32 (92%
and 98%
enantiomeric excess
(ee))
1). The in
(benzyloxy)bicyclo[2.2.1]hept-2-ene
31
and
styrene
to
yield
product
32
(92%
yield
and 98%of
enantioselectivity
in the
product
the
three cycles.
Thisbut
catalyst
also significantly
affords gooddropped
recoverability
polymer-supported
complex
caninbe
recycled
three times
conversion
in the
enantiomeric
excess
(ee))
(Scheme
1).
The
polymer-supported
complex
can
three
times
the
catalyst.
The
reported
leaching
of difference
molybdenum
in the product
3% be
of recycled
the
charged
catalyst.
third cycle.
However,
there
was little
in enantioselectivity
in was
the product
in the
three
cycles.
but This
conversion
significantly
dropped
in theofthird
cycle.
However,
there
was
Three
years
lateralso
Hoveyda
Schrock
described
immobilized
catalysts
using
bothdifference
polystyrenecatalyst
affords and
good
recoverability
thenew
catalyst.
The
reported
leaching
of little
molybdenum
in in
enantioselectivity
in
the
product
in
the
three
cycles.
This
catalyst
also
affords
good
recoverability
product was 3% of thesupports,
charged catalyst.
Three
years respectively
later Hoveyda[32].
and These
Schrock
described new of
and the
polynorbornene-based
23–26 and
27–29,
polymer-supported
the immobilized
catalyst.
reported
leaching
ofring-opening
molybdenum
in the
product was
3% ofofthe
charged
catalyst.
both polystyreneand polynorbornene-based
supports,
23–26
and substrates
27–29,
catalysts
can The
be catalysts
used
in using
asymmetric
cross
metathesis
reaction
the
same
Three
yearsearlier
later[32].
Hoveyda
and Schrock
newcan
immobilized
catalysts
using
polystyrenerespectively
These
catalysts
bethe
used
in asymmetric
ring-opening
cross
mentioned
and
the polymer-supported
catalysts
candescribed
be separated
from
reaction
mixture
byboth
simple
filtration.
metathesis
reaction
of
the
same
substrates
mentioned
earlier
and
the
catalysts
can
be
separated
from and
and
polynorbornene-based
supports,
23–26
and
27–29,
respectively
[32].
These
polymer-supported
The leaching of molybdenum in the product was found to be as low as 1%, with reactivity
the
reaction
mixture
by
simple
filtration.
The
leaching
of
molybdenum
in
the
product
was
found
to
catalysts can be used
in asymmetric
ring-openingcounterparts.
cross metathesis
of of
thea same
substrates
enantioselectivity
similar
to their homogeneous
The reaction
synthesis
polynorbornene
be as low
as 1%,and
with
reactivity
similar
their homogeneous
counterparts.
mentioned
earlier
the
catalystsand
canenantioselectivity
be separated
from
the to
reaction
filtration.
monolith-supported
Schrock-type
catalyst
30 was also
reported
by themixture
groupsby
ofsimple
Buchmeiser
and
The
synthesis
of
a
polynorbornene
monolith-supported
Schrock-type
catalyst
30
was
also
reportedand
The
leaching
of
molybdenum
in
the
product
was
found
to
be
as
low
as
1%,
with
reactivity
Fürstner
[33].
The
monolith-supported
chiral
catalyst
30
was
used
with
excellent
product
yields
and
by the groups of
Buchmeiser
and
Fürstner [33]. counterparts.
The monolith-supported
chiral
30 was
enantioselectivity
similar
to their
homogeneous
The synthesis
of catalyst
a polynorbornene
with
excellent
results
inproduct
term
ofyields
recovery
theexcellent
catalyst results
in asymmetric
reactions
3-(allyloxy)used
with excellent
and of
with
in term ofRCM
recovery
of theof
catalyst
in
monolith-supported Schrock-type catalyst 30 was also reported by the groups of Buchmeiser and
2,4-dimethylpenta-1,4-diene
its derivatives. In most cases, the reaction
proceeded In
with
yields
asymmetric RCM reactionsand
of 3-(allyloxy)-2,4-dimethylpenta-1,4-diene
and its derivatives.
most
Fürstner [33]. The monolith-supported chiral catalyst 30 was used with excellent product yields and
that cases,
exceeded
99% with
the molybdenum
contamination
in with
the products
being less
than 2% in all
the reaction
proceeded
with yields that
exceeded 99%
the molybdenum
contamination
with excellent results in term of recovery of the catalyst in asymmetric RCM reactions of 3-(allyloxy)in The
the products
being less than
all cases. The
was
also comparable
to its
cases.
enantioselectivity
was 2%
alsoincomparable
toenantioselectivity
its homogeneous
analog
with slightly
lower
2,4-dimethylpenta-1,4-diene
and
its
derivatives.
In
most
cases,
the
reaction
proceeded
with
yields
homogeneous
analog with slightly lower enantiomeric excesses.
enantiomeric
excesses.
that exceeded 99% with the molybdenum contamination in the products being less than 2% in all
cases. The enantioselectivity was also comparable to its homogeneous analog with slightly lower
enantiomeric excesses.
Scheme
1. Asymmetric
Ring-Opening
MetathesisReaction
Reaction
catalyzed
by 22.
Scheme
1. Asymmetric
Ring-OpeningCross
Cross Metathesis
of of
31 31
catalyzed
by 22.
Scheme 1. Asymmetric Ring-Opening Cross Metathesis Reaction of 31 catalyzed by 22.
Polymers 2016, 8, 140
Polymers 2016, 8, 140
9 of 23
9 of 22
iPr
tBu
tBu
N
O Mo
O
Me
Me
iPr
R
N
O Mo
O
Me
Me
tBu
tBu
22; R = Ph
23; R = Me
24
R
tBu
R
tBu
N
O Mo
O
R
N
O Mo
O
Me
Me
R'
tBu
tBu
N
O Mo
O
tBu
29
R
tBu
27; R = iPr
28; R = Cl
25; R = iPr; R' = Ph
26; R = Cl; R' = Me
Me
Me
Ph
n
iPr
tBu
O
N
O Mo
O
Me
Me
O
iPr
Ph
tBu
30
n
Schrock
catalysts
22–26 and
polynorbornene-based-supported
Figure 5.5.Polystyrene-based-supported
Polystyrene-based-supported
Schrock
catalysts
22–26
and polynorbornene-basedSchrock catalysts
27–30.
supported
Schrock
catalysts 27–30.
3. Homogeneous
HomogeneousSupported
SupportedOlefin
Olefin Metathesis
Metathesis Catalysts
Catalysts
Up to this
this point,
point,this
thisdiscussion
discussionhas
hasfocused
focusedononthe
the
immobilization
olefin
metathesis
catalysts
immobilization
of of
olefin
metathesis
catalysts
on
on
heterogeneous
supports.
While
thisbeen
has abeen
a common
technique
toand
isolate
andthe
recycle
the
heterogeneous
supports.
While
this has
common
technique
to isolate
recycle
catalysts,
catalysts,
is not
the only
possible
scheme product
for catalyst
product
separation
and catalyst
recycling.
it is not theit only
possible
scheme
for catalyst
separation
and
catalyst recycling.
Soluble
phase
Soluble
phase
tag in
methods
used inchemistry
combinatorial
chemistry
and in peptide
synthesis
have also as
been
tag methods
used
combinatorial
and in
peptide synthesis
have also
been developed
an
developed
alternative
for separation
process between
catalysts and
the products.
these
alternative as
toolanfor
separationtool
process
between catalysts
and the products.
In these
cases, the In
strategy
cases,
thetwo
strategy
is to phases
use twotodifferent
to recover
catalysts.
Thetwo
presence
ofcan
the result
two phases
is to use
different
recover phases
catalysts.
The presence
of the
phases
from
can
result
from
perturbation
of
a
homogeneous
reaction
mixture
or
can
involve
two
liquid
phases.
perturbation of a homogeneous reaction mixture or can involve two liquid phases. Thus, the supports
Thus,
supportsdofor
this
technique
notmacromolecules—small
always have to be macromolecules—small
molecules
for thisthe
technique
not
always
have do
to be
molecules like fluorous
tags or
like
ionic to
tags
cancatalyst/product
also be used to effect
catalyst/product
and catalyst
ionicfluorous
tags can tags
also or
be used
effect
separation
and catalystseparation
recycling. Such
soluble
recycling.
Suchpotential
soluble advantages
supports have
potential
advantages
in of
that
can avoid
some of the
supports have
in that
they can
avoid some
thethey
problems
of heterogeneized
problems
of heterogeneized
catalysts that
include
complicated
analyses
and the
observation
catalysts that
include more complicated
analyses
andmore
the observation
that
the reactivity
and
selectivity
that
the
reactivity
and
selectivity
of
an
immobilized
catalyst
differ
from
what
is
seen
an
of an immobilized catalyst differ from what is seen with an optimized homogeneous catalystwith
analog.
optimized
homogeneous
These problems
reflect the
fact that
the isadvantage
These
problems
reflect thecatalyst
fact thatanalog.
the advantage
of heterogeneous
catalysts,
which
their ease of
heterogeneous
which
is theirisease
of separation
the reaction
mixture,
is an
issue but
not
separation fromcatalysts,
the reaction
mixture,
an issue
not just atfrom
the separation
step
after the
reaction
just
at
the
separation
step
after
the
reaction
but
also
during
the
reaction
and
during
the
catalyst
also during the reaction and during the catalyst synthesis. The strategy in which phase tag that is
synthesis.
The strategy
which
phase
tagcatalyst
that is soluble
solublesupports
is used can
avoid tags
this allow
issue in
catalyst
soluble is used
can avoidinthis
issue
in that
or phase
thethat
catalysts
to
soluble supports or phase tags allow the catalysts to be synthesized and characterized in
homogeneous solution and allow them to carry out their reaction as homogenous catalysts. Such tags
are only used to effect separation after the reaction is complete.
Polymers 2016, 8, 140
10 of 23
Polymers 2016, 8, 140
10 of 22
be synthesized and characterized in homogeneous solution and allow them to carry out their reaction
as homogenous
Such tags
are use
onlyofused
effect(IL)
separation
after the
reaction
is complete.
An examplecatalysts.
of this approach
is the
ionictoliquid
immobilized
catalysts
(Figure
6). Ionic
Anare
example
of this
approach
theuseful
use ofbecause
ionic liquid
(IL)unique
immobilized
catalysts
(Figure
6).
liquids
alternative
solvents
thatisare
of their
properties
including
nonIonic
liquids
are
alternative
solvents
that
are
useful
because
of
their
unique
properties
including
volatility, high stability, and good recyclability [34]. These alternative solvents are immiscible with
non-volatility,
high stability,
recyclability
[34]. These
alternative
solventsphase.
are immiscible
most
organic solvents.
Thus, and
they good
can be
used in catalytic
reactions
as a recyclable
Buijsman
withco-workers
most organic
solvents.
Thus,
can1st
be generation
used in catalytic
as a recyclable
phase.
and
reported
the use
of they
Grubbs
catalystreactions
1 and Grubbs
2nd generation
Buijsman
and
co-workers
reported
the
use
of
Grubbs
1st
generation
catalyst
1
and
Grubbs
2nd
catalyst in an RCM reaction of 1,5-diallyl-3-benzyl-5-isobutylimidazolidine-2,4-dione in 1-butyl-3generation catalyst hexafluorophosphate
in an RCM reaction of
1,5-diallyl-3-benzyl-5-isobutylimidazolidine-2,4-dione
methylimidazolium
(BMI
PF6) as an ionic liquid solvent. These catalysts were
in 1-butyl-3-methylimidazolium
hexafluorophosphate
PF6 ) ppm
as anin
ionic
solvent.
These
recycled
up to three times with Ru
residues of ca. 5000(BMI
and 1600
the liquid
product,
respectively
catalysts
were
recycled
up
to
three
times
with
Ru
residues
of
ca.
5000
and
1600
ppm
in
the
product,
[35]. One year later, Dixneuf reported the use of BMI PF6 as solvent in an RCM reaction of
respectively [35]. One
year later,
reported
the use salt
of BMI
solvent
in an RCM
reactionfor
of
6 asable
diallyltosylamide
catalyzed
by aDixneuf
ruthenium
benzylidine
andPF
was
to recycle
the catalyst
diallyltosylamide
catalyzed
by
a
ruthenium
benzylidine
salt
and
was
able
to
recycle
the
catalyst
for
two cycles [36]. Following this work, Guillemin and co-workers introduced the ionic liquid-tagged
two cycles [36].
Following
this
work,
Guillemin
introduced
the of
ionic
ruthenium
catalyst
33 which
was
synthesized
in and
orderco-workers
to minimize
the leaching
theliquid-tagged
catalyst from
ruthenium
catalyst
33 which
wasionic
synthesized
in order
to minimize
the leaching
of the
catalyst
the
ionic liquid
phase
[37]. This
liquid-bound
catalyst
33 completed
an RCM
reaction
of
.
from
the
ionic
liquid
phase
[37].
This
ionic
liquid-bound
catalyst
33
completed
an
RCM
reaction
diallyltosylamide with BMI PF6 as a solvent at 60 °C in 45 min using 2.5 mol % of 33. The isolation of
. PF6 as a solvent at 60 ˝ C in 45 min using 2.5 mol % of 33. The isolation
of diallyltosylamide
with BMI
the
product was achieved
by extraction
with toluene and the ionic liquid phase containing 33 was
of the product
was achieved
with toluene
theImportantly,
ionic liquid phase
containing
was
reused
for an RCM
reaction by
of extraction
diallyltosylamide
for 8 and
cycles.
this catalyst
was33
stable
reused
for
an
RCM
reaction
of
diallyltosylamide
for
8
cycles.
Importantly,
this
catalyst
was
stable
enough to catalyze the ninth cycle of this RCM reaction without any loss in activity after three
enough to
catalyze
the ninth
cycle of
this
RCM reaction
without stable
any loss
in activity
after three
months.
months.
Yao
and Sheets
reported
the
synthesis
of a similarly
ionic
liquid-tagged
catalyst
34
Yao
and
Sheets
reported
the
synthesis
of
a
similarly
stable
ionic
liquid-tagged
catalyst
34
[38].
This
[38]. This catalyst too was recycled very effectively through 17 cycles in the RCM reaction of
catalyst too was recycled
through
cycles in
the RCM
reaction analog
of diallyltosylamide
diallyltosylamide
withoutvery
anyeffectively
loss in activity.
In 17
contrast,
a similar
untagged
of 34 lost its
without in
any
in activity.
In contrast,runs.
a similar
untagged
analog
of 34 lost itsreported
activity in
second
activity
theloss
second
and subsequent
In 2007,
Dixneuf
and co-workers
thethe
synthesis
and
subsequent
runs.
In
2007,
Dixneuf
and
co-workers
reported
the
synthesis
of
35
and
36, an
of 35 and 36, an improved version of 34 [39]. However, while the activities of both catalysts, toward
improved
version ofmentioned
34 [39]. However,
whilegood
the activities
of both
toward
the same
substrate
the
same substrate
earlier, were
for the first
cycle,catalysts,
the catalyst
activities
significantly
mentioned
werecycle.
good In
forgeneral,
the firstthese
cycle,studies
the catalyst
significantly
dropped
in earlier,
the second
relied activities
on measuring
catalystdropped
reactivityinasthe
a
second
cycle.
In
general,
these
studies
relied
on
measuring
catalyst
reactivity
as
a
basis
for
catalyst
basis for catalyst recyclability. The amount of Ru leaching was typically not mentioned except in a
recyclability.
few
cases. The amount of Ru leaching was typically not mentioned except in a few cases.
Figure 6. Ionic liquid-tagged ruthenium catalysts 33–36.
Figure 6. Ionic liquid-tagged ruthenium catalysts 33–36.
An alternative scheme for liquid/liquid separation that has been used in metathesis and in other
An alternative
for liquid/liquid
separation
that has
usedwith
in metathesis
in
chemistry
is fluorousscheme
phase technology.
In this scheme,
catalysts
are been
modified
fluorinatedand
phase
other
chemistry
is
fluorous
phase
technology.
In
this
scheme,
catalysts
are
modified
with
fluorinated
tags to facilitate separation, recovery and recycling in perfluorinated solvents [40]. Examples of
phase tags to facilitate
separation,
recovery
recycling
inthe
perfluorinated
solvents
[40].[41],
Examples
of
fluorous-tagged
Ru catalysts
are shown
in and
Figure
7. Since
first report by
Horváth
fluorous
biphasic catalysis has greatly expanded and developed into a general strategy for separations [42].
This chemistry typically uses a mixture of organic and fluorous solvents, relying on the fact that
Polymers 2016, 8, 140
11 of 23
fluorous-tagged Ru catalysts are shown in Figure 7. Since the first report by Horváth [41], fluorous
biphasic catalysis has greatly expanded and developed into a general strategy for separations [42].
Polymers 2016, 8, 140
11 of 22
This chemistry typically uses a mixture of organic and fluorous solvents, relying on the fact that
fluorous
fluorous solvents
solvents are
are often
often immiscible
immiscible with
with most
most organic
organic solvents
solvents at
at room
room temperature
temperature and
and such
such
solvent
mixtures
are
often
biphasic.
Such
solvent
mixtures
can
be
used
in
two
ways.
In
one
solvent mixtures are often biphasic. Such solvent mixtures can be used in two ways. In one type
type of
of
experiment,
a
reaction
is
carried
out
under
biphasic
conditions
with
a
biphasic
liquid/liquid
separation
experiment, a reaction is carried out under biphasic conditions with a biphasic liquid/liquid
after
a reaction.
in some cases fluorous
organic
solvents
miscible
at elevated
separation
afterAlternatively,
a reaction. Alternatively,
in someand
cases
fluorous
andbecome
organic
solvents
become
temperature.
In thosetemperature.
thermomorphic
systems,
a reaction cansystems,
be effected
under monophasic
conditions
miscible at elevated
In those
thermomorphic
a reaction
can be effected
under
and
the
catalyst
and
the
fluorous
phase
can
be
separated
from
an
organic
product
in
an
organic
monophasic conditions and the catalyst and the fluorous phase can be separated from an organic
phase
at in
room
temperature
thetemperature
biphasic mixture
and the
phases
can beand
separated
one
product
an organic
phasewhen
at room
whenreforms
the biphasic
mixture
reforms
the phases
from
by aone
gravity-based
liquid/liquid
biphasicliquid/liquid
separation. In
2004, Yao
and Zhang
reported
can beanother
separated
from another
by a gravity-based
biphasic
separation.
In 2004,
Yao
the
immobilization
of
a
Grubbs-type
catalyst
on
poly(fluoroalkyl
acrylate)
37
[42].
The
catalyst
and Zhang reported the immobilization of a Grubbs-type catalyst on poly(fluoroalkyl acrylate) 37
37
was
in RCM37
reactions
of in
diallyltosylamide
in diallyltosylamide
a monophasic PhCF
v/v)3/CH
solvent
2 Cl2 (1:19
[42]. used
The catalyst
was used
RCM reactions of
in a3 /CH
monophasic
PhCF
2Cl2
system.
of the fluorous
species
perfluorohexane
and EtOAc(FC-72)
after each
(1:19 v/v)Extraction
solvent system.
Extraction
of theusing
fluorous
species using(FC-72)
perfluorohexane
andreaction
EtOAc
allowed
the
catalyst
to
be
recovered
and
reused.
The
authors
were
able
to
recycle
this
fluorous-tagged
after each reaction allowed the catalyst to be recovered and reused. The authors were able to recycle
ruthenium
catalyst for
20 cycles. catalyst
However,
was
not described.
Inspired
this work,Inspired
Curran
this fluorous-tagged
ruthenium
forleaching
20 cycles.
However,
leaching
was notbydescribed.
group
reported
the
study
of
other
recoverable
metathesis
catalysts
using
lighter
fluorous
supports,
by this work, Curran group reported the study of other recoverable metathesis catalysts using lighter
with
onlysupports,
17 fluorine
atoms
Ru in 38atoms
and 39
versus
atoms170
perfluorine
Ru in 37
[43]. per
These
fluorous
with
onlyper
17 fluorine
per
Ru in170
38 fluorine
and 39 versus
atoms
Ru
catalysts
and 39
show similar
their non-fluorous-bound
analogs in RCM reaction
in 37 [43].38These
catalysts
38 andactivity
39 showtosimilar
activity to their non-fluorous-bound
analogs of
in
diallyltosylamide.
However,
the
separations
of
catalysts
38
and
39
involved
the
use
of
fluorous
silica
RCM reaction of diallyltosylamide. However, the separations of catalysts 38 and 39 involved the use
gel
rather than
liquid/liquid
Thus,
extra solvents
including
acetonitrile
were
needed
of fluorous
silicaa gel
rather thanextraction.
a liquid/liquid
extraction.
Thus, extra
solvents
including
acetonitrile
to
obtain
the
product
and
ether
was
needed
to
recover
the
fluorous-tagged
catalyst.
The
recovered
were needed to obtain the product and ether was needed to recover the fluorous-tagged catalyst. The
catalyst
cancatalyst
be reused
least five
cycles
the average
product
yield
of 97%.yield
Laterofon,
Matsugi
recovered
canfor
be at
reused
for at
leastwith
five cycles
with the
average
product
97%.
Later
and
co-workers
two other
lighttwo
fluorous-tagged
catalysts 40 andcatalysts
41 [44]. Compared
39,
on, Matsugi
andreported
co-workers
reported
other light fluorous-tagged
40 and 41 to[44].
fluorous-tagged
catalyst
40
had
an
improvement
in
activity
with
similar
recyclability
(90%
recovery
Compared to 39, fluorous-tagged catalyst 40 had an improvement in activity with similar
of
catalyst), while
41 showed
activity
both 39higher
and 40
but was
notboth
recoverable
The
recyclability
(90% recovery
of higher
catalyst),
while than
41 showed
activity
than
39 and 40[44].
but was
fluorous-tagged
catalyst
was recycled and
reused40inwas
an RCM
reaction
of diethyl
diallylmalonate
not recoverable [44].
The40
fluorous-tagged
catalyst
recycled
and reused
in an
RCM reactionfor
of
five
times.
The
product
yield
was
95%–100%
in
each
cycle.
diethyl diallylmalonate for five times. The product yield was 95%–100% in each cycle.
Figure 7. Fluorous-tagged ruthenium catalysts 37–41.
Figure 7. Fluorous-tagged ruthenium catalysts 37–41.
Soluble polymeric supports are an alternative to ionic or fluorous tags. Poly(ethylene glycol)
Soluble
polymeric
supports
aresuch
an alternative
ionic orand
fluorous
tags. Poly(ethylene
glycol)
(PEG)
is one of
the most widely
used
supports fortoreagents
organometallic
complexes [45–47].
(PEG)
is
one
of
the
most
widely
used
such
supports
for
reagents
and
organometallic
complexes
[45–47].
PEG has been used both to bind catalysts and has been used as a solvent [48]. Examples of PEGPEG
hasRu
been
used both
to bind are
catalysts
and
been8.used
as a solvent
[48].PEG-bound
Examples ofcatalysts
PEG-bound
bound
metathesis
catalysts
shown
in has
Figure
Although
PEG and
are
Ru
metathesis
catalysts
shownincluding
in Figure 8.
Although
PEG
and PEG-bound
catalysts
are soluble
in
soluble
in many
organicare
solvents
water,
they are
insoluble
in solvents
like hexane,
diethyl
ether, and cold ethanol. Thus, these PEG derivatives can be utilized to separate catalysts and products
for catalyst recovery and recycling by either solvent precipitation or liquid/liquid extraction. An early
example of the synthesis and application of a PEG5000-bound ruthenium metathesis catalyst 42 was
reported by Yao in 2000 [49]. This soluble polymer-supported catalyst 42 was recycled 8 times in the
RCM reaction of allyl(4-pentenyl)tosylamide with more than 92% conversion in each cycle. The
Polymers 2016, 8, 140
12 of 23
many organic solvents including water, they are insoluble in solvents like hexane, diethyl ether, and
cold ethanol. Thus, these PEG derivatives can be utilized to separate catalysts and products for catalyst
recovery and recycling by either solvent precipitation or liquid/liquid extraction. An early example
of the synthesis and application of a PEG5000 -bound ruthenium metathesis catalyst 42 was reported
Polymers 2016, 8, 140
12 of 22
by Yao in 2000 [49]. This soluble polymer-supported catalyst 42 was recycled 8 times in the RCM
reaction
of allyl(4-pentenyl)tosylamide
thanether.
92% Although
conversionthe
in author
each cycle.
The catalyst
was
catalyst was
recovered by precipitationwith
withmore
diethyl
reported
the success
recovered
by
precipitation
with
diethyl
ether.
Although
the
author
reported
the
success
in
catalyst
in catalyst recycling, the analysis of Ru leaching was not reported. It also should be noted that this
recycling,
the analysis
of Ru recovery
leaching was
not reported.
It alsoof
should
be noted
that this
type of issue
process
type of process
for catalyst
requires
a large excess
the ether
solvent—a
possible
in
for
catalyst
recovery
requires
a
large
excess
of
the
ether
solvent—a
possible
issue
in
green
chemistry
green chemistry terms. In 2003, Lamaty and co-workers described the study of a PEG3400-bound
terms.
In 2003, Lamaty
and co-workers
described
the The
studysoluble
of a PEG
3400 -bound
Hoveyda-Grubbs
2nd generation
catalyst
43 [48].
support
was Hoveyda-Grubbs
attached to the
2nd
generation
catalyst
43
[48].
The
soluble
support
was
attached
to
the
benzylidene
ligand
ortho to
benzylidene ligand ortho to the metal carbene. This catalyst can catalyze RCM
reaction
of
the
metal
carbene.
This
catalyst
can
catalyze
RCM
reaction
of
diallyltosylamide
for
5
cycles
at
room
diallyltosylamide for 5 cycles at room temperature. The recovery of this catalyst from RCM reaction
temperature.
Theout
recovery
of this catalyst
fromether
RCMand
reaction
was also
carried
precipitation
was also carried
by precipitation
in diethyl
filtration.
In this
work,out
theby
authors
carried
in
diethyl
ether
and
filtration.
In
this
work,
the
authors
carried
out
some
additional
characterization
out some additional characterization of the recovered PEG-bound Ru complex. The precipitate was
1 H NMR spectroscopy,
of
the recovered
PEG-bound
Ru complex.
The precipitate
wasofanalyzed
analyzed
by 1H NMR
spectroscopy,
measuring
the intensity
the vinylbyproton
of the alkylidene
measuring
intensity
of the
protonshowed
of the alkylidene
group
in was
the catalyst
at after
16.50the
δ. first
The
group in thethe
catalyst
at 16.50
δ. vinyl
The results
that only 57%
of 43
recovered
results
showed
that
only
57%
of
43
was
recovered
after
the
first
cycle
of
RCM
reaction.
cycle of RCM reaction.
Figure 8.
8. PEG-supported
Hoveyda-Grubbs catalyst
catalyst 42
42 and
and 43.
43.
Figure
PEG-supported Hoveyda-Grubbs
The Bergbreiter and Bazzi groups have also described the use soluble polymer supports for
The Bergbreiter and Bazzi groups have also described the use soluble polymer supports for
ruthenium olefin metathesis catalysts. In their work, they focused on soluble polyolefin oligomers
ruthenium olefin metathesis catalysts. In their work, they focused on soluble polyolefin oligomers
that can be more efficiently separated as an alternative to PEG supports whose separation typically
that can be more efficiently separated as an alternative to PEG supports whose separation typically
generates large volumes of solvents waste during the polymer precipitation step. These alternative
generates large volumes of solvents waste during the polymer precipitation step. These alternative
polymers are polyethylene (PEOlig) and polyisobutylene oligomers (PIB). Several techniques for
polymers are polyethylene (PEOlig ) and polyisobutylene oligomers (PIB). Several techniques for
separation of these types of polymer-supported species including liquid/liquid separation and
separation of these types of polymer-supported species including liquid/liquid separation and
solid/liquid separation shown in Figure 9 can be used. The first synthesis and application of PIBsolid/liquid separation shown in Figure 9 can be used. The first synthesis and application of
supported Hoveyda-Grubbs type catalyst using such polymers was described in 2007 [50]. In this
PIB-supported Hoveyda-Grubbs type catalyst using such polymers was described in 2007 [50]. In this
case, the PIB (Mn = 1000) was attached to the catalyst as a benzylidene ligand following a prior
case, the PIB (Mn = 1000) was attached to the catalyst as a benzylidene ligand following a prior strategy
strategy described by Barrett [17]. This pre-catalyst 44 was then used to catalyze RCM reactions in
described by Barrett [17]. This pre-catalyst 44 was then used to catalyze RCM reactions in heptane.
heptane. The products were then separated from the catalyst solution by an acetonitrile extraction.
The products were then separated from the catalyst solution by an acetonitrile extraction. In some
In some cases, the RCM product self-separated from the catalyst. In such cases where the product
cases, the RCM product self-separated from the catalyst. In such cases where the product precipitated
precipitated from the heptane solution, a filtration was only needed to isolate the product. This PIBfrom the heptane solution, a filtration was only needed to isolate the product. This PIB-supported
supported ruthenium complex 44 was reused for at least five cycles. The Ru leaching into the product
ruthenium complex 44 was reused for at least five cycles. The Ru leaching into the product was
was measured and varied from 20 to 1000 ppm, suggesting that the “boomerang” scheme for
measured and varied from 20 to 1000 ppm, suggesting that the “boomerang” scheme for recycling the
recycling the catalyst was not as effective as desired. In order to improve the recoverability of the
catalyst was not as effective as desired. In order to improve the recoverability of the PIB-supported
PIB-supported ruthenium catalyst, an alternative type of catalyst was prepared where the PIB (Mn =
ruthenium catalyst, an alternative type of catalyst was prepared where the PIB (Mn = 1000) chains
1000) chains were attached to the non-dissociating N-heterocyclic carbene ligands of a Hoveydawere attached to the non-dissociating N-heterocyclic carbene ligands of a Hoveyda-Grubbs catalyst as
Grubbs catalyst as in complex 45 [51]. This catalyst design led to improvements in both the
in complex 45 [51]. This catalyst design led to improvements in both the consistency and effectiveness
consistency and effectiveness of Ru catalyst/product separation. Leaching levels dropped to values
of Ru catalyst/product separation. Leaching levels dropped to values as low as 73 ppm (0.37% of
as low as 73 ppm (0.37% of the charged Ru catalyst). Moreover, catalyst 45 could be reused for 20
the charged Ru catalyst). Moreover, catalyst 45 could be reused for 20 cycles. More recent work
cycles. More recent work from our group on polymer-supported olefin metathesis catalyst explored
from our group on polymer-supported olefin metathesis catalyst explored the use of Polyethylene
the use of Polyethylene oligomer (PEOlig) (Mn = 550) as catalyst supports [52]. The unique property of
PEOlig is that it is like higher molecular weight polyethylene and does not dissolve in any solvent at
room temperature. However, like high molecular weight polyethylene, it does dissolve at elevated
temperature. Moreover, polyethylene oligomers that are commercially available as 500–2000 Da
materials are soluble in toluene or THF at relatively modest temperatures like 65 °C. A PEOligsupported Hoveyda-Grubbs catalyst 46 formed from PEOlig with an Mn of 550 Da can be dissolved in
Polymers 2016, 8, 140
13 of 23
oligomer (PEOlig ) (Mn = 550) as catalyst supports [52]. The unique property of PEOlig is that it is like
higher molecular weight polyethylene and does not dissolve in any solvent at room temperature.
However, like high molecular weight polyethylene, it does dissolve at elevated temperature. Moreover,
polyethylene oligomers that are commercially available as 500–2000 Da materials are soluble in toluene
or THF at relatively modest temperatures like 65 ˝ C. A PEOlig -supported Hoveyda-Grubbs catalyst
46 formed from PEOlig with an Mn of 550 Da can be dissolved in a solvent with modest heating to
form
a monophasic
solution. The complex 46 quantitatively precipitates on cooling. Thus, 46 can
Polymers
2016, 8, 140
13 ofbe
22
added to substrate, the suspension that forms can be heated to form a monophasic reaction solution,
and
the catalyst
can be separated
from a solution
the product
simply cooling
back to room
to form
a monophasic
reaction solution,
and theof
catalyst
can bebyseparated
from aitsolution
of the
temperature.
Filtration
then
effects
phase
separation
between
the
solid
catalyst
species
and
the
product
product by simply cooling it back to room temperature. Filtration then effects phase separation
solution.
describing
this catalyst,
PEOligsolution.
-supported
catalystdescribing
46 was used
in catalyst,
a varietythe
of
between In
thereports
solid catalyst
species
and the the
product
In reports
this
RCM
forcatalyst
at least46
eight
with
the Ruof
leaching
based onfor
ICP-MS
that was
PEOligreactions
-supported
wascycles
used in
a variety
RCM reactions
at leastanalysis
eight cycles
withless
the
than
0.3% based
on the
amount analysis
of charged
Ru.was
Examples
of 0.3%
thesebased
polyolefin-supported
Rucharged
metathesis
Ru leaching
based
on ICP-MS
that
less than
on the amount of
Ru.
catalysts
areof
shown
Figure 10. The effectiveness
of the catalyst/product
these processes
Examples
theseinpolyolefin-supported
Ru metathesis
catalysts areseparation
shown inin Figure
10. The
was
further tested
sequential reactions
where the
sameprocesses
sample ofwas
46 was
usedtested
to carry
out three
effectiveness
of theincatalyst/product
separation
in these
further
in sequential
successive
RCM reactions.
This experiment
only
showed
Rusuccessive
leaching was
low
but that This
the
reactions where
the same sample
of 46 wasnot
used
to carry
outthat
three
RCM
reactions.
product
of one
was notthat
detectable
in a subsequent
cycle
the of
same
a
experiment
notreaction
only showed
Ru leaching
was low but
thatthat
the used
product
onecatalyst
reactionused
wasinnot
prior
different
reaction.
detectable in a subsequent cycle that used the same catalyst used in a prior different reaction.
a)
catalyst
heat
substrate
catalyst
product
catalyst
cool
product
1. separate
2. fresh polar solvent, substrate
b)
catalyst
substrate
solvents
rxn
catalyst perturb
product
solvents
catalyst
product
1. separate
2. fresh polar solvent, substrate
c)
substrate
heat
catalyst
product
cool
product
1. filter
2. fresh solvent, substrate
= polymer-bound catalyst
Figure 9. Schematic representation of (a) thermomorphic liquid/liquid separation where a biphasic
Figure 9. Schematic representation of (a) thermomorphic liquid/liquid separation where a biphasic
mixture of polar and nonpolar solvent is heated to form a miscible solvent mixture; (b) a latent
mixture of polar and nonpolar solvent is heated to form a miscible solvent mixture; (b) a latent
biphasic liquid/liquid separation where a room temperature mixture of miscible solvents is used to
biphasic liquid/liquid separation where a room temperature mixture of miscible solvents is used
effect a reaction and the solvent mixture is perturbed (e.g., by addition of another solvent (e.g., water)
to effect a reaction and the solvent mixture is perturbed (e.g., by addition of another solvent
or some salt) to form a separable biphasic mixture of polar and nonpolar solvents; and (c) a
(e.g., water) or some salt) to form a separable biphasic mixture of polar and nonpolar solvents; and
thermomorphic
solid/liquid
separation
systemsystem
where awhere
mixture
of substrate
and insoluble
polymer(c)
a thermomorphic
solid/liquid
separation
a mixture
of substrate
and insoluble
bound
catalyst
is
heated
to
form
a
monophasic
solution
that
is
in
turn
cooled
to
form
a
precipitate
polymer-bound catalyst is heated to form a monophasic solution that is in turn cooled to form
the polymer
catalystbound
that is then
separated
product solution
filtration.
aofprecipitate
of bound
the polymer
catalyst
that isfrom
thenthe
separated
from thebyproduct
solution
by filtration.
biphasic liquid/liquid separation where a room temperature mixture of miscible solvents is used to
effect a reaction and the solvent mixture is perturbed (e.g., by addition of another solvent (e.g., water)
or some salt) to form a separable biphasic mixture of polar and nonpolar solvents; and (c) a
thermomorphic solid/liquid separation system where a mixture of substrate and insoluble polymerbound catalyst is heated to form a monophasic solution that is in turn cooled to form a precipitate
Polymers 2016, 8, 140
14 of 23
of the polymer bound catalyst that is then separated from the product solution by filtration.
Figure 10. Polyolefin-supported Hoveyda-Grubbs catalysts 44–46.
Figure 10. Polyolefin-supported Hoveyda-Grubbs catalysts 44–46.
Up to this point, this review has focused on background examples of ring closing and cross
metathesis reactions where metathesis catalysts are separated either by liquid/solid or liquid/liquid
separations. In these applications, the need for a separation strategy depends on the reactivity of the Ru
catalyst. Catalysts with TONs that approach 105 or 106 may not even need a support and a separation
strategy. This however is not true in polymerization reactions where no chain transfer occurs. In such
cases, the living character of a ring-opening metathesis polymerization in a polymerization leads to one
metal center being used for each polymer molecule synthesized. In these cases metal or metal-ligand
separations are more critical. The balance of this review will discuss metathesis catalysts and ligand
separation from products in this more demanding context.
4. Supported Catalysts in Ring-opening Metathesis Polymerization (ROMP)
The first example of supported catalyst used in ROMP is also the first well-defined
polymer-supported olefin metathesis catalyst [16]. In this study by Grubbs and Nguyen, a series
of polystyrene-divinylbenzene (DVB-PS)-supported Cl2 (PR3 )2 Ru=CH-CH=CPh2 olefin metathesis
catalysts 6, 47, and 48 were synthesized as shown in Scheme 2. These solid-supported catalysts showed
activities that were similar to that seen for their homogeneous analogs where the catalyst reactivity
varied depending on the nature of PS-supported phosphine ligands. For example, while catalyst 48 can
only catalyze ROMP reaction of highly strained cyclic olefins such as norbornene, 6 and 47 can catalyze
ROMP of the less strained cyclic olefins such as cyclooctene. Although 6, 47, and 48 were competent
catalysts for ROMP of norbornene, the PDIs of the resulting polymer products were significantly higher
than those prepared by the homogeneous analogs. For example, polynorbornene prepared from 6 had
PDI values of 5.5, while 1 could be used to prepare similar polymers with PDI values ranging from
1.1 to 1.3. The broader molecular weight range for the polynorbornene prepared by the immobilized
catalyst was postulated to be a result of the slow initiation rate of immobilized catalyst that could
reflect its inhomogeneity or perhaps the fact that the actual catalyst is a complex that dissociates from
the support. The amount of Ru contamination in the ROMP products was not reported.
Alternative insoluble organic supports that have been used as catalyst supports for Ru ROMP
catalysts include monolithic materials. While such materials have been known since the 1970s [53]
but have received more attention [54,55] after studies by Fréchet and Svec highlighted their utility as
high-performance separation media, scavengers, and reagent supports [56,57]. Such media have been
used too as metathesis catalyst supports as noted above. That work includes studies by Buchmeiser
whose group described a synthetic route to a monolith-supported ruthenium olefin metathesis catalyst
50 (Scheme 3) [58]. The monolith was generated through ring-opening metathesis copolymerization of
norbornene (NBE) and 1,4,4a,5,8,8a-hexahydro-1,4,5,8-exo-endo-dimethanonaphthalene (DMN. H6) in
the presence of dichloromethane and 2-propanol within a borosilicate column. The functionalization
of the catalyst onto the monolith was achieved by grafting a mixture of compound 49 and norbornene
onto this solid support. After the grafting process was complete, it was terminated by addition of
ethyl vinyl ether. The resulting grafts contained imidazolium salts that were then deprotonated with
dimethylaminopyridine (DMAP) to generate an NHC ligand. The immobilization of a ruthenium
Polymers 2016, 8, 140
15 of 23
catalyst onto the monolith support using these NHC ligands was then achieved by treating this
monolith-immobilized NHC ligand with 1. As noted above, the monolith-supported ruthenium
complex 50 showed high activity toward RCM. This RCM catalyst was also used in ROMP reactions,
where the cis and trans ratio of ROMP products of norbornene and cis-cyclococtene were the same
as those obtained using the analogous homogeneous catalyst. In the ROMP chemistry, the PDI of
polymer products ranged from 1.2 to 2.6. However, while the extent of Ru leaching in RCM products
was reported as 70 ppm, Ru leaching in the ROMP products was not reported.
Schrock’s molybdenum catalysts that are used in ROMP have been also immobilized on
heterogeneous supports. However, perhaps because of their sensitivity toward air and moisture,
there are
relatively
few reports discussed about immobilizing these Mo-based ROMP catalysts15onto
Polymers
2016,
8, 140
of 22
solid supports. One example was reported by Basset group in 1996 [59]. The molybdenum complex
˝ C. The loss of neopentane in 52 was postulated to be involved
51 was immobilized onto silica at 70 °C.
in the generation of an active silica-supported catalyst 53. While the structure of 53 was inferred, the
presence of
confirmed
byby
elemental
analyses
thatthat
yield
a Mo/N/C
ratio ofratio
1/1/10
ofaacomplex
complexlike
like5353was
was
confirmed
elemental
analyses
yield
a Mo/N/C
of
(Scheme(Scheme
4). This4).
immobilized
complex
53 can53prepare
ROMP
products
of norbornene
and and
cis1/1/10
This immobilized
complex
can prepare
ROMP
products
of norbornene
˝C
cyclooctene
at 25at°C
high
yieldyield
(74%(74%
and 85%,
respectively).
These These
ROMPROMP
products
had Mnhad
values
cis-cyclooctene
25in
in high
and 85%,
respectively).
products
Mn
of
ca. 310,000
Da withDa
a PDI
1.8.
metal
analysis
of molybdenum
in ROMPinproducts
was not
values
of ca. 310,000
withofaca.
PDI
ofThe
ca. 1.8.
The
metal analysis
of molybdenum
ROMP products
described.
was
not described.
Scheme
Scheme 2.
2. DVB-PS-Supported
DVB-PS-Supported Olefin
Olefin Metathesis
Metathesis Catalysts
Catalysts 6,
6, 47,
47, and
and 48.
48.
Polymers 2016, 8, 140
Scheme 2. DVB-PS-Supported Olefin Metathesis Catalysts 6, 47, and 48.
Scheme
SynthesisofofMonolith-Supported
Monolith-Supported Ruthenium
Ruthenium Olefin
Catalyst
50.50.
Scheme
3. 3.
Synthesis
OlefinMetathesis
Metathesis
Catalyst
Polymers 2016, 8, 140
16 of 23
16 of 22
Scheme 4. Synthesis of Silica-Supported Schrock Catalyst 53.
Scheme
4. Synthesis of Silica-Supported Schrock Catalyst 53.
attemptingto
to develop
metathesis
reaction
process,
the Grubbs
group reported
In In
attempting
developa agreener
greener
metathesis
reaction
process,
the Grubbs
group several
reported
syntheses
of
highly
active
PEG-bound
ruthenium
complexes
that
can
be
used
in
aqueous
to
several syntheses of highly active PEG-bound ruthenium complexes that can be used media
in aqueous
catalyze
ROMP reactions.
One example
the PEG5000
conjugated
N-heterocyclic
carbene-containing
media
to catalyze
ROMP reactions.
Oneisexample
is the
PEG5000 conjugated
N-heterocyclic
carbeneruthenium
benzylidene
catalyst
54
[60]
(Scheme
5).
This
catalyst
initiated
the
ROMP
of 55
to give
containing ruthenium benzylidene catalyst 54 [60] (Scheme 5). This catalyst initiated the
ROMP
of 55
polynorbornene 56 in 73% conversion after 24 h, as measured by 1 H NMR spectroscopy
in D2 O. The
1
to give polynorbornene 56 in 73% conversion after 24 h, as measured by H NMR spectroscopy in
rate of ROMP reaction with a catalyst like 55 depends on the rate of phosphine dissociation. In this case
D2O. The rate of ROMP reaction with a catalyst like 55 depends on the rate of phosphine dissociation.
the dissociation of phosphine from catalyst 54 may be disfavored in water. However, the protonation
In this case the dissociation of phosphine from catalyst 54 may be disfavored in water. However, the
of free phosphine by HCl could in principle inhibit re-association of the phosphine ligand. In this case,
protonation of free phosphine by HCl could in principle inhibit re-association of the phosphine
the rate of polymerization was dramatically increased to 95% conversion in 15 min when 1 equiv of
ligand.
In this case, the rate of polymerization was dramatically increased to 95% conversion in 15 min
HCl, relative to catalyst 54, was present in the reaction. This chemistry is similar to the Phase Transfer
when
1 equiv(PTA)
of HCl,
relativeused
to catalyst
54, was and
present
ingroups
the reaction.
chemistry
is issimilar
Activation
principle
by the Gladysz
Bazzi
where This
a phosphine
that
more to
thesoluble
PhaseinTransfer
Activation
(PTA)
principle
used
by
the
Gladysz
and
Bazzi
groups
where
a fluorous or aqueous phase is used with a Grubbs catalyst under biphasic conditions where a
phosphine
that isseparation
more soluble
a fluorous
aqueous
phase
is usedre-association
with a Grubbs
catalyst
the phosphine
intoin
a fluorous
or or
aqueous
phase
precludes
[61,62].
In under
the
biphasic
conditions
where
the
phosphine
separation
into
a
fluorous
or
aqueous
phase
precludes
Grubbs’ work, the PEG-bound ruthenium catalyst 54 was shown to catalyze a ROMP reaction of the reassociation
the86%
Grubbs’
work, to
thepolymer
PEG-bound
ruthenium
was
to catalyze
monomer[61,62].
endo- 57In
with
conversion
58 within
14 h. It iscatalyst
known 54
that
thisshown
endo-monomer
a ROMP
reaction
of the ROMP
monomer
endo- 57
with
conversion
polymer
58 within
14 h.
It ROMP
is known
is a more
challenging
substrate
than
the86%
exo-monomer
55to
[60].
The extent
to which
the
products
were contaminated
by residues
of the ROMP
PEG-bound
ruthenium
54 was not 55
reported.
that
this endo-monomer
is a more
challenging
substrate
than complex
the exo-monomer
[60]. The
extent to which the ROMP products were contaminated by residues of the PEG-bound ruthenium
complex 54 was not reported.
biphasic conditions where the phosphine separation into a fluorous or aqueous phase precludes reassociation [61,62]. In the Grubbs’ work, the PEG-bound ruthenium catalyst 54 was shown to catalyze
a ROMP reaction of the monomer endo- 57 with 86% conversion to polymer 58 within 14 h. It is known
that this endo-monomer is a more challenging ROMP substrate than the exo-monomer 55 [60]. The
extent to which the ROMP products were contaminated by residues of the PEG-bound ruthenium
Polymers 2016, 8, 140
17 of 23
complex 54 was not reported.
Scheme 5. ROMP Reactions of 55 and 57 Catalyzed by 54.
The
complex 59
59 prepared
prepared with
with aa PEG
PEG2000 support
that was used in RCM and CM reactions of
The complex
of
2000 support that was used in RCM and CM reactions
17 of 22
substrates
like
2-allyl-N,N,N-trimethylpent-4-en-1-aminium
chloride
and
allyl
alcohol
in
water
has
substrates like 2-allyl-N,N,N-trimethylpent-4-en-1-aminium chloride and allyl alcohol in water has
also been
of of
57 57
(Scheme
6) [63].
TheThe
polymerization
of 57ofoccurred
in 100%
also
been used
usedininthe
theROMP
ROMPreaction
reaction
(Scheme
6) [63].
polymerization
57 occurred
in
conversion
in 5 h. in
The5 removal
process of
the PEG-bound
rutheniumruthenium
complex 59complex
from the59
products
58
100%
conversion
h. The removal
process
of the PEG-bound
from the
was not discussed
at discussed
least in terms
of the
catalyst
productsand
in aproducts
ROMP reaction.
products
58 was not
at least
in terms
ofand
the catalyst
in a ROMP reaction.
Polymers 2016, 8, 140
Scheme
Scheme 6.
6. ROMP
ROMP Reaction
Reaction of
of 57
57 Catalyzed
Catalyzed by
by PEG-Supported
PEG-Supported Hoveyda-Grubbs
Hoveyda-Grubbs Catalyst
Catalyst 59.
59.
The immobilization of a highly active Grubbs 3rd generation catalyst using a PEG350 support has
The immobilization of a highly active Grubbs 3rd generation catalyst using a PEG support has
also been reported [64]. Emrick and co-workers synthesized a PEG-supported Grubbs 350
3rd generation
also been reported [64]. Emrick and co-workers synthesized a PEG-supported Grubbs 3rd generation
by ligand exchange between pyridine ligands and PEG-bound pyridine ligands. This PEG-supported
by ligand exchange between pyridine ligands and PEG-bound pyridine ligands. This PEG-supported
catalyst 60 was used to catalyze ROMP reactions in water (Scheme 7). The results showed that this
catalyst 60 was used to catalyze ROMP reactions in water (Scheme 7). The results showed that this
catalyst works well at pH 1.5 with quantitative conversion of monomer 61 to polymer 62 in 2 h at
catalyst works well at pH 1.5 with quantitative conversion of monomer 61 to polymer 62 in 2 h at
room temperature. It was noted that the activity of 60 decreased with an increasing in pH, e.g., 23%
room temperature. It was noted that the activity of 60 decreased with an increasing in pH, e.g., 23%
conversion at pH 7 in 2 h at room temperature. However, with an addition of a Brønsted acid, i.e.,
conversion at pH 7 in 2 h at room temperature. However, with an addition of a Brønsted acid, i.e.,
CuSO4 and CuBr2, the conversion of monomer 61 was improved from 23% to 70%. The PDI of the
CuSO4 and CuBr2 , the conversion of monomer 61 was improved from 23% to 70%. The PDI of the
polymer
products prepared from 60 ranged from 1.3 to 1.5. The amounts of Ru content in the products
polymer products prepared from 60 ranged from 1.3 to 1.5. The amounts of Ru content in the products
were not reported.
were not reported.
catalyst 60 was used to catalyze ROMP reactions in water (Scheme 7). The results showed that this
catalyst works well at pH 1.5 with quantitative conversion of monomer 61 to polymer 62 in 2 h at
room temperature. It was noted that the activity of 60 decreased with an increasing in pH, e.g., 23%
conversion at pH 7 in 2 h at room temperature. However, with an addition of a Brønsted acid, i.e.,
CuSO4 and CuBr2, the conversion of monomer 61 was improved from 23% to 70%. The PDI of the
polymer
products prepared from 60 ranged from 1.3 to 1.5. The amounts of Ru content in the products 18 of 23
Polymers 2016,
8, 140
were not reported.
7. ROMP
Reaction
of 61
Catalyzed by
by PEG-Supported
Hoveyda-Grubbs
Catalyst
60.
SchemeScheme
7. ROMP
Reaction
of 61
Catalyzed
PEG-Supported
Hoveyda-Grubbs
Catalyst
60.
An early report of the use of a PIB-supported catalyst was part of a study of a
Anpolyisobutylbenzylidene-supported
early report of the use ofRu acatalyst
PIB-supported
wasmetathesis.
part of That
a study
that was used catalyst
in ring closing
work of a
showed that this “boomerang” catalyst
could bethat
usedwas
in ROMP
ca. 3%
leaching
of the charged
polyisobutylbenzylidene-supported
Ru catalyst
used with
in ring
closing
metathesis.
That work
Ru
[50].
Subsequently
in
2012,
a
more
effective
Ru
separation
was
achieved
by
the
use
of
PIBshowed that this “boomerang” catalyst could be used in ROMP with ca. 3% leaching of the charged
catalyst 45 in ROMP of norbornene derivatives as described in a report by Bazzi and
Ru [50].supported
Subsequently
in 2012, a more effective Ru separation was achieved by the use of PIB-supported
Bergbreiter [65]. The catalyst 45 can be separated from the polymer product at the end of the reaction
catalyst 45 in ROMP of norbornene derivatives as described in a report by Bazzi and Bergbreiter [65].
by dissolving the dry crude mixture of polymer and PIB-supported catalyst in an equi-volume
The catalyst 45 can be separated from the polymer product at the end of the reaction by dissolving
the dry crude mixture of polymer and PIB-supported catalyst in an equi-volume mixture of DMF
and heptane. The catalyst-containing heptane layer was then removed and the DMF layer was then
evaporated to dryness to afford the crude polymer as an off-white solid. The crude polymer was
Polymers 2016, 8, 140
18 of 22
dissolved in dichloromethane (DCM) and precipitated into an excess amount of hexane to afford
the polymer
product
for analysis
for propertiesheptane
and ruthenium
contamination
Polymers
mixture
of DMF ready
and heptane.
The catalyst-containing
layer was then
removed andlevel.
the DMF
layer
wascatalyst
then evaporated
dryness identical
to afford the
crude
polymer
as an
off-whiteby
solid.
The
crude
prepared
from
45 have to
a nearly
E/Z
ratio
to those
prepared
their
low
molecular
polymer was dissolved
in dichloromethane
and precipitated
into
an excessof
amount
of hexane
weight counterparts
with PDI
values ranging(DCM)
from 1.39
to 1.78. The
amounts
ruthenium
content in
to afford the polymer product ready for analysis for properties and ruthenium contamination level.
the resulting polymers were 111–228 ppm. In subsequent work by Bazzi and Bergbreiter, the activities
Polymers prepared from catalyst 45 have a nearly identical E/Z ratio to those prepared by their low
and recoverability
of a PIB-supported Grubbs second generation catalyst 63 formed from PIB with an
molecular weight counterparts with PDI values ranging from 1.39 to 1.78. The amounts of ruthenium
Mn of 1000
Dain(Figure
11) inpolymers
ROMP were
was studied
[66] In
insubsequent
a polymerization
of norbornene
(Scheme 8).
content
the resulting
111–228 ppm.
work by Bazzi
and Bergbreiter,
The polymerization
raterecoverability
of norbornene
63 was 93%
which
is similar
to that
of its
molecular
the activities and
of ausing
PIB-supported
Grubbs
second
generation
catalyst
63 low
formed
from PIB with64an(99%
Mn of
1000 Da (Figure
in ROMP
was studied
[66] in
polymerization
of ROMP
weight counterpart
conversion,
after 11)
60 min).
However,
complex
63aexhibited
a faster
norbornene
(Scheme
8).
The
polymerization
rate
of
norbornene
using
63
was
93%
which
is
similar
to period.
initiation with 70% conversion in 10 min than that of 64 with 30% conversion at the same time
that of its low molecular weight counterpart 64 (99% conversion, after 60 min). However, complex 63
The PDI values of the polymer products 70–74 analyzed by GPC ranged from 1.11 to 2.34. The removal
exhibited a faster ROMP initiation with 70% conversion in 10 min than that of 64 with 30% conversion
of catalyst
63 same
was examined
thePDI
ROMP
reactions
of monomer
65–69.
process
usedranged
in separation
at the
time period.inThe
values
of the polymer
products
70–74The
analyzed
by GPC
of residues
and the
polymer
involved
the of
dry
crude mixture
of
fromfrom
1.11 tocatalyst
2.34. The63
removal
of catalyst
63 product
was examined
in the dissolving
ROMP reactions
monomer
65–
The process
usedcatalyst
in separation
of residues
from catalyst
63 and
the polymer
product involved
polymer69.product
and Ru
residue
in a minimum
amount
of DCM
and precipitating
the mixture
dissolving
the the
dry homopolymer
crude mixture of as
polymer
product
RuRu
catalyst
residue in a level
minimum
amount 70–74
in hexane
affording
a white
solid.and
The
contamination
of polymers
of DCM and precipitating the mixture in hexane affording the homopolymer as a white solid. The Ru
prepared by 63 varied from 71 to 252 ppm.
contamination level of polymers 70–74 prepared by 63 varied from 71 to 252 ppm.
Figure 11. Polyisobutylene (PIB1000)-supported Grubbs second generation catalysts 63 and Grubbs
Figure 11. Polyisobutylene (PIB1000 )-supported Grubbs second generation catalysts 63 and Grubbs
catalyst 64.
catalyst 64.
Figure
Polyisobutylene (PIB1000)-supported Grubbs second generation catalysts 63 and Grubbs
Polymers
2016, 11.
8, 140
19 of 23
catalyst 64.
Scheme
Scheme 8.
8. ROMP
ROMP Reactions
Reactions of
of 65–69
65–69 Catalyzed
Catalyzed by
by 63.
63.
The PEOlig-supported ruthenium catalyst 46 has also been used in ROMP reactions of norbornene
The PE
-supported ruthenium catalyst 46 has also been used in ROMP reactions of norbornene
derivatives Olig
[67]. Examples of polymer prepared using 46 are shown in Figure 12. This PEOligderivatives [67]. Examples of polymer prepared using 46 are shown in Figure 12. This PEOlig -supported
supported catalyst 60 was used to prepare polymers 74–78 using the same solid/liquid
separation
catalyst 60 was used to prepare polymers 74–78 using the same solid/liquid separation concept as
concept as was used previously in reactions where this catalyst was used to prepare RCM products.
was used previously in reactions where this catalyst was used to prepare RCM products. In this
In this case, separation of the ROMP products from 46 was performed by filtration of reaction mixture
case, separation of the ROMP products from 46 was performed by filtration of reaction mixture
through Celite and 0.2 μm filter to yield colorless solution containing polymer products. The resulting
through Celite and 0.2 µm filter to yield colorless solution containing polymer products. The resulting
solution was concentrated and then precipitated in hexane to isolate polymers 74–78. In reactions
solution was concentrated and then precipitated in hexane to isolate polymers 74–78. In reactions
with catalyst 46, it was found that the prior scheme where polyethylene oligomers were shown to be
with catalyst 46, it was found that the prior scheme where polyethylene oligomers were shown to be
useful cosolvents facilitate Ru residue separations in ROMP chemistry [67]. Adding unfunctionalized
useful cosolvents facilitate Ru residue separations in ROMP chemistry [67]. Adding unfunctionalized
polyethylene oligomers as a cosolvent further facilitated separations of the product polymer and
polyethylene
oligomers as a cosolvent further facilitated separations of the product polymer
and
Polymers 2016,
8, 140 For example, when a commercially available unfunctionalized polyethylene
19 of 22
catalyst
residues.
catalyst residues. For example, when a commercially available unfunctionalized polyethylene
(Polywax)
(Mnn ==400)
400)[68]
[68]was
was
added
a co-solvent
toOlig
PE
catalyst,
the leaching
Olig -supported
(Polywax) (M
added
as as
a co-solvent
to PE
-supported
catalyst,
the leaching
of Ru
of
Ru residues
catalyst
60 decreased
to 19–26
ppm
range
0.5%ofofthe
thecharged
chargedRu
Ru catalyst).
catalyst). In
residues
from from
catalyst
60 decreased
to 19–26
ppm
range
(ca.(ca.
0.5%
In
addition,
control
experiments
showed
that
heating
a
suspension
of
46
and
Polywax
to
form
a
solution
addition, control experiments showed that heating a suspension of 46 and Polywax to form a solution
followed
followed by
by cooling
cooling and
and filtration
filtration led
led to
to only
only 0.04%
0.04% leaching
leaching of
of Ru
Ru leaching,
leaching, aa value
value that
that increased
increased to
to
0.42%
(0.84
ppm,
amount
of
Ru
in
the
solution)
if
the
heated
solution
of
46,
Polywax,
and
THF
was
0.42% (0.84 ppm, amount of Ru in the solution) if the heated solution of 46, Polywax, and THF was
quenched
the
ca.ca.
19–26
ppm
RuRu
residues
observed
in
quenched with
withbutyl
butylvinyl
vinylether.
ether.This
Thissuggests
suggeststhat
thatmost
mostofof
the
19–26
ppm
residues
observed
the
ROMP
products
maymay
be due
to some
side reactions
that occur
during
the quenching
of the living
in the
ROMP
products
be due
to some
side reactions
that occur
during
the quenching
of the
polymer
and
not
from
the
polymerization
process
involving
46.
The
resulting
polymers
74–78
have
living polymer and not from the polymerization process involving 46. The resulting polymers 74–78
the
same
thoseastothose
those to
prepared
by its lowby
molecular
counterpart
2, comparable
have
the E/Z
sameratio
E/Zasratio
those prepared
its low weight
molecular
weight counterpart
2,
M
that
ranged
1.23 to from
1.84. 1.23 to 1.84.
n values, and
comparable
Mn PDIs
values,
and
PDIs from
that ranged
Figure 12.
12. ROMP
ROMP products
products 74–78.
74–78.
Figure
5. Conclusions
5.
Conclusions
Althoughmany
manyexamples
examples
of newer
versions
of metathesis
olefin metathesis
with improved
Although
of newer
versions
of olefin
catalystscatalysts
with improved
reactivity,
reactivity,
air
and
moisture
stability,
functional
groups
tolerance,
and
stereoselectivity
have
been
air and moisture stability, functional groups tolerance, and stereoselectivity have been reported since
reported
since
the
discovery
of
the
first
metathesis
catalysts,
challenges
still
remain.
These
challenges
the discovery of the first metathesis catalysts, challenges still remain. These challenges include: (i) the
include:
cost of metal
the transition
metal
complexes
usedorasprecatalysts;
catalysts or (ii)
precatalysts;
(ii) the
high
cost(i)ofthe
thehigh
transition
complexes
used
as catalysts
the cost and/or
cost
and/or
tediousness
of
the
ligand
syntheses;
(iii)
the
potential
environmental
toxicological
or
tediousness of the ligand syntheses; (iii) the potential environmental toxicological or practical concerns
practical concerns that ensue when there is significant metal or ligand contamination in the product;
and (iv) the cost of and waste generated when metal or ligand contaminates have to be removed by
post reaction processing steps. Several approaches have been proposed to address these issues. One
approach is to reduce the amount of catalyst used in the reaction. Lowering the wt % of catalyst will
necessarily reduce catalyst cost, ligand cost, and product contamination. However, examples of this
Polymers 2016, 8, 140
20 of 23
that ensue when there is significant metal or ligand contamination in the product; and (iv) the cost
of and waste generated when metal or ligand contaminates have to be removed by post reaction
processing steps. Several approaches have been proposed to address these issues. One approach is
to reduce the amount of catalyst used in the reaction. Lowering the wt % of catalyst will necessarily
reduce catalyst cost, ligand cost, and product contamination. However, examples of this strategy can be
limited, especially in polymerizations where each polymer chain has one catalyst site. An alternative
strategy that has proven to be more versatile is to immobilize the catalyst on either heterogeneous or
homogeneous supports. Such scheme allow the supported catalysts to be recovered and subsequently
reused or, in other cases, effect separation of inactive catalyst residues from products. Several examples
of this chemistry have been reported in the past two decades showing the success in using this strategy
to address the issues mentioned earlier, especially those focused on the removal of metal contamination
in the final products. The benefit of using supported olefin metathesis catalysts to effectively minimize
the level of metal contamination in the products is significantly increased in the case of ROMP reactions
when reducing the amount of catalyst might not be an option. Although there is still room for further
development (such as recoverability and application in asymmetric synthesis) before the strategy
of catalyst immobilization can be widely used in both industry and academia, its potential to be
recognized as a standard tool in synthetic applications is not to be underestimated.
Acknowledgments: Funding of this work by the National Science Foundation (CHE-1362735) (DEB), the Robert
A. Welch Foundation (Grant A-0639) (DEB), the Qatar National Research Fund (project numbers NPRP-28-6-7-7
and NPRP-4-081-1-016) (HSB and DEB) are gratefully acknowledged.
Author Contributions: Jakkrit Suriboot, Hassan S. Bazzi, and David E. Bergbreiter wrote and edited the paper.
Conflicts of Interest: The authors declare no conflict of interest.
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
Grubbs, R.H. Handbook of Metathesis, 1st ed.; Wiley-VCH Verlag GmbH: Weinheim, Germany, 2003;
Volume 1, p. 204.
Grela, K. Olefin Metathesis: Theory and Practice, 1st ed.; John Wiley & Sons, Inc.: Hoboken, NJ,
USA, 2014; p. 608.
Grubbs, R.H. Olefin-metathesis catalysts for the preparation of molecules and materials. Angew. Chem. Int.
Ed. 2006, 45, 3760–3765. [CrossRef] [PubMed]
Schrock, R.R. Multiple metal–carbon bonds for catalytic metathesis reactions. Angew. Chem. Int. Ed. 2006, 45,
3748–3759. [CrossRef] [PubMed]
Vougioukalakis, G.C. Removing ruthenium residues from olefin metathesis reaction products. Chem. Eur. J.
2012, 18, 8868–8880. [CrossRef] [PubMed]
Chu, D.S.H.; Schellinger, J.G.; Shi, J.; Convertine, A.J.; Stayton, P.S.; Pun, S.H. Application of living free
radical polymerization for nucleic acid delivery. Acc. Chem. Res. 2012, 45, 1089–1099. [CrossRef] [PubMed]
Kuhn, K.M.; Bourg, J.-B.; Chung, C.K.; Virgil, S.C.; Grubbs, R.H. Effects of NHC-backbone substitution
on efficiency in ruthenium-based olefin metathesis. J. Am. Chem. Soc. 2009, 131, 5313–5320. [CrossRef]
[PubMed]
Hong, S.H.; Day, M.W.; Grubbs, R.H. Decomposition of a key intermediate in ruthenium-catalyzed olefin
metathesis reactions. J. Am. Chem. Soc. 2004, 126, 7414–7415. [CrossRef] [PubMed]
Courchay, F.C.; Sworen, J.C.; Ghiviriga, I.; Abboud, K.A.; Wagener, K.B. Understanding structural
isomerization during ruthenium-catalyzed olefin metathesis: A deuterium labeling study. Organometallics
2006, 25, 6074–6086. [CrossRef]
Dinger, M.B.; Mol, J.C. High turnover numbers with ruthenium-based metathesis catalysts. Adv. Synth. Catal.
2002, 344, 671–677. [CrossRef]
Merrifield, R.B. Solid phase peptide synthesis. I. The Synthesis of a tetrapeptide. J. Am. Chem. Soc. 1963, 85,
2149–2154. [CrossRef]
Collman, J.P.; Belmont, J.A.; Brauman, J.I. A silica-supported rhodium hydroformylation catalyst: Evidence
for dinuclear elimination. J. Am. Chem. Soc. 1983, 105, 7288–7294. [CrossRef]
Polymers 2016, 8, 140
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
21 of 23
Drago, R.S.; Pribich, D.C. A new method for enhancing site isolation on silica gel and for improving the
lifetime of site-isolated catalysts. Inorg. Chem. 1985, 24, 1983–1985. [CrossRef]
Töllner, K.; Popovitz-Biro, R.; Lahav, M.; Milstein, D. Impact of molecular order in Langmuir-Blodgett films
on catalysis. Science 1997, 278, 2100–2102. [CrossRef] [PubMed]
Annis, D.A.; Jacobsen, E.N. Polymer-supported chiral Co(salen) complexes: Synthetic applications and
mechanistic investigations in the hydrolytic kinetic resolution of terminal epoxides. J. Am. Chem. Soc. 1999,
121, 4147–4154. [CrossRef]
Nguyen, S.T.; Grubbs, R.H. The syntheses and activities of polystyrene-supported olefin metathesis catalysts
based on Cl2 (PR3 )2 Ru=CH–CH=CPh2 . J. Organomet. Chem. 1995, 497, 195–200. [CrossRef]
Ahmed, M.; Barrett, A.G.M.; Braddock, D.C.; Cramp, S.M.; Procopiou, P.A. A recyclable “boomerang”
polymer-supported ruthenium catalyst for olefin metathesis. Tetrahedron Lett. 1999, 40, 8657–8662. [CrossRef]
Schürer, S.C.; Gessler, S.; Buschmann, N.; Blechert, S. Synthesis and application of a permanently immobilized
olefin-metathesis catalyst. Angew. Chem. Int. Ed. 2000, 39, 3898–3901. [CrossRef]
Weskamp, T.; Kohl, F.J.; Hieringer, W.; Gleich, D.; Herrmann, W.A. Highly active ruthenium catalysts for
olefin metathesis: The synergy of N-heterocyclic carbenes and coordinatively labile ligands. Angew. Chem.
Int. Ed. 1999, 38, 2416–2419. [CrossRef]
Randl, S.; Buschmann, N.; Connon, S.J.; Blechert, S. Highly efficient and recyclable polymer-bound catalyst
for olefin metathesis reactions. Synlett 2001, 2001, 1547–1550. [CrossRef]
Mennecke, K.; Grela, K.; Kunz, U.; Kirschning, A. Immobilisation of the grubbs III olefin metathesis catalyst
with polyvinyl pyridine (PVP). Synlett 2005, 2005, 2948–2952. [CrossRef]
Prühs, S.; Lehmann, C.W.; Fürstner, A. Preparation, reactivity, and structural peculiarities of
hydroxyalkyl-functionalized “second-generation” ruthenium carbene complexes. Organometallics 2004,
23, 280–287. [CrossRef]
Allen, D.P.; van Wingerden, M.M.; Grubbs, R.H. Well-defined silica-supported olefin metathesis catalysts.
Org. Lett. 2009, 11, 1261–1264. [CrossRef] [PubMed]
Lim, J.; Seong Lee, S.; Ying, J.Y. Mesoporous silica-supported catalysts for metathesis: Application to a
circulating flow reactor. Chem. Commun. 2010, 46, 806–808. [CrossRef] [PubMed]
Bek, D.; Gawin, R.; Grela, K.; Balcar, H. Ruthenium metathesis catalyst bearing chelating carboxylate ligand
immobilized on mesoporous molecular sieve SBA-15. Catal. Commun. 2012, 21, 42–45. [CrossRef]
Monge-Marcet, A.; Pleixats, R.; Cattoën, X.; Man, M.W.C. Catalytic applications of recyclable silica
immobilized NHC–ruthenium complexes. Tetrahedron 2013, 69, 341–348. [CrossRef]
Blanc, F.; Copéret, C.; Thivolle-Cazat, J.; Basset, J.-M.; Lesage, A.; Emsley, L.; Sinha, A.; Schrock, R.R. Surface versus
molecular siloxy ligands in well-defined olefin metathesis catalysts: [{(RO)3 SiO}Mo(=NAr)(=CHtBu)(CH2 tBu)].
Angew. Chem. Int. Ed. 2006, 45, 1216–1220. [CrossRef] [PubMed]
Lopez, L.P.H.; Schrock, R.R. Formation of dimers that contain unbridged W(IV)/W(IV) double bonds. J. Am.
Chem. Soc. 2004, 126, 9526–9527. [CrossRef] [PubMed]
Blanc, F.; Thivolle-Cazat, J.; Basset, J.-M.; Copéret, C.; Hock, A.S.; Tonzetich, Z.J.; Schrock, R.R. Highly active,
stable, and selective well-defined silica supported Mo imido olefin metathesis catalysts. J. Am. Chem. Soc.
2007, 129, 1044–1045. [CrossRef] [PubMed]
Blanc, F.; Rendon, N.; Berthoud, R.; Basset, J.-M.; Coperet, C.; Tonzetich, Z.J.; Schrock, R.R. Dramatic
enhancement of the alkene metathesis activity of Mo imido alkylidene complexes upon replacement of one
tBuO by a surface siloxy ligand. Dalton Trans. 2008, 3156–3158. [CrossRef] [PubMed]
Hultzsch, K.C.; Jernelius, J.A.; Hoveyda, A.H.; Schrock, R.R. The first polymer-supported and recyclable
chiral catalyst for enantioselective olefin metathesis. Angew. Chem. Int. Ed. 2002, 41, 589–593. [CrossRef]
Dolman, S.J.; Hultzsch, K.C.; Pezet, F.; Teng, X.; Hoveyda, A.H.; Schrock, R.R. Supported chiral Mo-based
complexes as efficient catalysts for enantioselective olefin metathesis. J. Am. Chem. Soc. 2004, 126,
10945–10953. [CrossRef] [PubMed]
Mayr, M.; Wang, D.; Kröll, R.; Schuler, N.; Prühs, S.; Fürstner, A.; Buchmeiser, M.R. Monolithic disk-supported
metathesis catalysts for use in combinatorial chemistry. Adv. Synth. Catal. 2005, 347, 484–492. [CrossRef]
Dupont, J.; de Souza, R.F.; Suarez, P.A.Z. Ionic liquid (molten salt) phase organometallic catalysis. Chem. Rev.
2002, 102, 3667–3692. [CrossRef] [PubMed]
Buijsman, R.C.; van Vuuren, E.; Sterrenburg, J.G. Ruthenium-catalyzed olefin metathesis in ionic liquids.
Org. Lett. 2001, 3, 3785–3787. [CrossRef] [PubMed]
Polymers 2016, 8, 140
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
54.
55.
56.
57.
58.
22 of 23
Semeril, D.; Olivier-Bourbigou, H.; Bruneau, C.; Dixneuf, P.H. Alkene metathesis catalysis in ionic liquids
with ruthenium allenylidene salts. Chem. Commun. 2002, 146–147. [CrossRef]
Audic, N.; Clavier, H.; Mauduit, M.; Guillemin, J.-C. An Ionic liquid-supported ruthenium carbene complex:
A robust and recyclable catalyst for ring-closing olefin metathesis in ionic liquids. J. Am. Chem. Soc. 2003,
125, 9248–9249. [CrossRef] [PubMed]
Yao, Q.; Sheets, M. An ionic liquid-tagged second generation Hoveyda–Grubbs ruthenium carbene complex
as highly reactive and recyclable catalyst for ring-closing metathesis of di-, tri- and tetrasubstituted dienes.
J. Organomet. Chem. 2005, 690, 3577–3584. [CrossRef]
Thurier, C.; Fischmeister, C.; Bruneau, C.; Olivier-Bourbigou, H.; Dixneuf, P.H. Ionic imidazolium containing
ruthenium complexes and olefin metathesis in ionic liquids. J. Mol. Catal. A Chem. 2007, 268, 127–133.
[CrossRef]
Fustero, S.; Simón-Fuentes, A.; Barrio, P.; Haufe, G. Olefin metathesis reactions with fluorinated substrates,
catalysts, and solvents. Chem. Rev. 2015, 115, 871–930. [CrossRef] [PubMed]
Horváth, I.T.; Rábai, J. Facile catalyst separation without water: Fluorous biphase hydroformylation of
olefins. Science 1994, 266, 72–75. [CrossRef] [PubMed]
Yao, Q.; Zhang, Y. Poly(fluoroalkyl acrylate)-bound ruthenium carbene complex: A fluorous and recyclable
catalyst for ring-closing olefin metathesis. J. Am. Chem. Soc. 2004, 126, 74–75. [CrossRef] [PubMed]
Matsugi, M.; Curran, D.P. Synthesis, Reaction, and recycle of light fluorous grubbs´hoveyda catalysts for
alkene metathesis. J. Org. Chem. 2005, 70, 1636–1642. [CrossRef] [PubMed]
Matsugi, M.; Kobayashi, Y.; Suzumura, N.; Tsuchiya, Y.; Shioiri, T. Synthesis and RCM reactions using a
recyclable grubbs´hoveyda metathesis catalyst activated by a light fluorous tag. J. Org. Chem. 2010, 75,
7905–7908. [CrossRef] [PubMed]
Bergbreiter, D.E. Using soluble polymers to recover catalysts and ligands. Chem. Rev. 2002, 102, 3345–3384.
[CrossRef] [PubMed]
Bergbreiter, D.E.; Tian, J.; Hongfa, C. Using soluble polymer supports to facilitate homogeneous catalysis.
Chem. Rev. 2009, 109, 530–582. [CrossRef] [PubMed]
Gravert, D.J.; Janda, K.D. Organic synthesis on soluble polymer supports: Liquid-phase methodologies.
Chem. Rev. 1997, 97, 489–510. [CrossRef] [PubMed]
Varray, S.; Lazaro, R.; Martinez, J.; Lamaty, F. New soluble-polymer bound ruthenium carbene catalysts:
Synthesis, characterization, and application to ring-closing metathesis. Organometallics 2003, 22, 2426–2435.
[CrossRef]
Yao, Q. A Soluble polymer-bound ruthenium carbene complex: A robust and reusable catalyst for ring-closing
olefin metathesis. Angew. Chem. Int. Ed. 2000, 39, 3896–3898. [CrossRef]
Hongfa, C.; Tian, J.; Bazzi, H.S.; Bergbreiter, D.E. Heptane-soluble ring-closing metathesis catalysts.
Org. Lett. 2007, 9, 3259–3261. [CrossRef] [PubMed]
Hongfa, C.; Su, H.-L.; Bazzi, H.S.; Bergbreiter, D.E. Polyisobutylene-anchored N-heterocyclic carbene ligands.
Org. Lett. 2009, 11, 665–667. [CrossRef] [PubMed]
Hobbs, C.; Yang, Y.-C.; Ling, J.; Nicola, S.; Su, H.-L.; Bazzi, H.S.; Bergbreiter, D.E. Thermomorphic
polyethylene-supported olefin metathesis catalysts. Org. Lett. 2011, 13, 3904–3907. [CrossRef] [PubMed]
Hansen, L.C.; Sievers, R.E. Highly permeable open-pore polyurethane columns for liquid chromatography.
J. Chromatogr. A 1974, 99, 123–133. [CrossRef]
Hjertén, S.; Liao, J.-L.; Zhang, R. High-performance liquid chromatography on continuous polymer beds.
J. Chromatogr. A 1989, 473, 273–275. [CrossRef]
Hjerten, S.; Yi-Ming Li, Y.; Liao, J.-L.; Mohammad, J.; Nakazato, K.I.; Pettersson, G. Continuous beds:
High-resolving, cost-effective chromatographic matrices. Nature 1992, 356, 810–811. [CrossRef]
Tripp, J.A.; Stein, J.A.; Svec, F.; Fréchet, J.M.J. “Reactive filtration”: Use of functionalized porous polymer
monoliths as scavengers in solution-phase synthesis. Org. Lett. 2000, 2, 195–198. [CrossRef] [PubMed]
Tripp, J.A.; Svec, F.; Fréchet, J.M.J. Grafted macroporous polymer monolithic disks: A new format of
scavengers for solution-phase combinatorial chemistry. J. Comb. Chem. 2001, 3, 216–223. [CrossRef]
[PubMed]
Mayr, M.; Mayr, B.; Buchmeiser, M.R. Monolithic materials: New high-performance supports for permanently
immobilized metathesis catalysts. Angew. Chem. Int. Ed. 2001, 40, 3839–3842. [CrossRef]
Polymers 2016, 8, 140
59.
60.
61.
62.
63.
64.
65.
66.
67.
68.
23 of 23
Herrmann, W.A.; Stumpf, A.W.; Priermeier, T.; Bogdanović, S.; Dufaud, V.; Basset, J.-M. A Molecularly
defined, grafted olefin metathesis catalyst from tris(neopentyl)-nitridomolybdenum(VI). Angew. Chem. Int.
Ed. 1996, 35, 2803–2805. [CrossRef]
Gallivan, J.P.; Jordan, J.P.; Grubbs, R.H. A neutral, water-soluble olefin metathesis catalyst based on an
N-heterocyclic carbene ligand. Tetrahedron Lett. 2005, 46, 2577–2580. [CrossRef]
Da Costa, R.C.; Gladysz, J.A. Syntheses and reactivity of analogues of grubbs’ second generation metathesis
catalyst with fluorous phosphines: A new phase-transfer strategy for catalyst activation. Adv. Synth. Catal.
2007, 349, 243–254. [CrossRef]
Tuba, R.; Corrêa da Costa, R.; Bazzi, H.S.; Gladysz, J.A. Phase transfer activation of fluorous analogs of
grubbs’ second-generation catalyst: Ring-opening metathesis polymerization. ACS Catal. 2012, 2, 155–162.
[CrossRef]
Hong, S.H.; Grubbs, R.H. Highly active water-soluble olefin metathesis catalyst. J. Am. Chem. Soc. 2006, 128,
3508–3509. [CrossRef] [PubMed]
Samanta, D.; Kratz, K.; Zhang, X.; Emrick, T. A synthesis of PEG- and phosphorylcholine-substituted
pyridines to afford water-soluble ruthenium benzylidene metathesis catalysts. Macromolecules 2008, 41,
530–532. [CrossRef]
Al-Hashimi, M.; Hongfa, C.; George, B.; Bazzi, H.S.; Bergbreiter, D.E. A phase-separable second-generation
hoveyda-grubbs catalyst for ring-opening metathesis polymerization. J. Polym. Sci. A Polym. Chem. 2012, 50,
3954–3959. [CrossRef]
Al-Hashimi, M.; Bakar, M.D.A.; Elsaid, K.; Bergbreiter, D.E.; Bazzi, H.S. Ring-opening metathesis
polymerization using polyisobutylene supported Grubbs second-generation catalyst. RSC Adv. 2014,
4, 43766–43771. [CrossRef]
Suriboot, J.; Hobbs, C.E.; Guzman, W.; Bazzi, H.S.; Bergbreiter, D.E. Polyethylene as a cosolvent and catalyst
support in ring-opening metathesis polymerization. Macromolecules 2015, 48, 5511–5516. [CrossRef]
Baker Hughes. Polywax. Available online: http://www.bakerhughes.com/news-and-media/resources/
technical-data-sheet/polywax-polyethylenes (accessed on 9 March 2016).
© 2016 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC-BY) license (http://creativecommons.org/licenses/by/4.0/).