Polymer cocktail: A multitask supported ionic liquid-like species to facilitate multiple and consecutive C-C coupling reactions

Article abstract of DOI:10.1002/adsc.201000528

Polymer-supported ionic liquid-like species (SILLPs) with different functionalities have been combined to obtain new catalytic systems (polymer cocktail) with improved properties towards their application for C-C coupling reactions. The SILLPs have been synthesized in a simple way and combined to play different functions in the reaction medium. Those roles involve simultaneously (or consecutively) their actuation as supported reagents, precatalysts and scavengers of undesired reaction side-products. The system has been tested by employing three different C-C coupling reactions for 8 consecutive cycles. Moreover, different aryl bromides and iodides have proved to be efficiently transformed by the catalytic cocktail. Finally, the Heck reaction has been conducted in continuous-flow conditions using a packed bed of this polymer cocktail and employing supercritical carbon dioxide (scCO₂) as a solvent, which cannot be carried out with the use of soluble bases as this would lead to the clogging of the reactor. Finally, the SILLPs cocktail enables one to combine different reaction steps and purification in a single vessel, which represents a significant improvement in terms of process intensification and green chemistry. This is particularly true in the case of the use of scCO ₂ as this allows the direct production of crude materials not contaminated either by salts or by solvents. Copyright

Full text of DOI:10.1002/adsc.201000528

FULL PAPERS
DOI: 10.1002/adsc.201000528
Polymer Cocktail: A Multitask Supported Ionic Liquid-Like
À
Species to Facilitate Multiple and Consecutive C C Coupling
Reactions
Victor Sans,c Fabien Gelat,a Naima Karbass,a M. Isabel Burguete,a
Eduardo Garcꢀa-Verdugo,a,b,* and Santiago V. Luisa,*
a
Department of Inorganic and Organic Chemistry, University Jaume I, Avda. Sos Baynat s/n, E-12071 Castellꢁn, Spain
Fax : (+34)-96-472-8214; e-mail: luiss@qio.uji.es
Instituto de Catꢂlisis y Petroleoquꢀmica, CSIC, Campus de la UAM, C/Marie Curie 2, Cantoblanco, E-28049 Madrid,
b
Spain
E-mail: egverdugo@icp.csic.es
School of Engineering, University of Warwick, Coventry, CV4 7AL, U.K.
c
Received: July 5, 2010; Published online: November 17, 2010
Abstract: Polymer-supported ionic liquid-like species packed bed of this polymer cocktail and employing
(SILLPs) with different functionalities have been supercritical carbon dioxide (scCO2) as a solvent,
combined to obtain new catalytic systems (polymer which cannot be carried out with the use of soluble
cocktail) with improved properties towards their ap- bases as this would lead to the clogging of the reac-
À
plication for C C coupling reactions. The SILLPs tor. Finally, the SILLPs cocktail enables one to com-
have been synthesized in a simple way and combined bine different reaction steps and purification in a
to play different functions in the reaction medium. single vessel, which represents a significant improve-
Those roles involve simultaneously (or consecutive- ment in terms of process intensification and green
ly) their actuation as supported reagents, precatalysts chemistry. This is particularly true in the case of the
and scavengers of undesired reaction side-products. use of scCO2 as this allows the direct production of
The system has been tested by employing three dif- crude materials not contaminated either by salts or
À
ferent C C coupling reactions for 8 consecutive by solvents.
cycles. Moreover, different aryl bromides and iodides
have proved to be efficiently transformed by the cat- Keywords: C C coupling reactions; green chemistry;
alytic cocktail. Finally, the Heck reaction has been ionic liquids; polymer-supported reagents and cata-
conducted in continuous-flow conditions using a lysts; supported catalysts; supported ionic liquids
À
Introduction
achieving multiple transformations in a single pot, the
immobilization of toxic reagents and the increase in
The development of “ideal recoverable catalysts” is a selectivity of chemical processes.[8]
topic of great interest in both the academic and the
Carbon-carbon coupling reactions are very impor-
chemical industry fields.[1] One attractive approach is tant as they allow the production of a wide range of
the immobilization of the catalytic species into a dif- synthetic intermediates.[9,10] Nevertheless, achieving an
ferent phase than the reactants and products, facilitat- efficient and practical catalytic process for those reac-
ing the recovery of the catalyst and its reuse in further tions is not easy. In most cases where the use of heter-
catalytic cycles.[2] Polymer-supported reagents and ogeneous catalysts for C C coupling reactions has
À
scavengers have become widely used in organic syn- been reported,[11] a careful study of the reaction
thesis and a great variety of these materials have mechanism revealed that, under the conditions em-
been reported.[3,4] In some cases, they have distinct ad- ployed, the supported species acted as “Trojan
vantages over their solution-phase counterparts.[5] horses”,[12,13] servng as reservoirs of the “soluble”
One of them is that they allow the simultaneous use active species generated in the course of the reac-
of incompatible reagents, such as an acid and a base, tion.[14] This very often leads to a contamination of
since they can react with the solution-phase compo- the final products with metals,[15] which is a key factor
nents of the reaction mixture but not with each for the pharmaceutical industry.[16] Furthermore, the
other.[6,7] Other advantages include the possibility of reaction requires the use of a base to neutralize the
Adv. Synth. Catal. 2010, 352, 3013 – 3021
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3013

Victor Sans et al.
FULL PAPERS
acid formed giving rise to the formation of salts that of catalytic species and greatly simplifies the purifica-
complicate the downstream processing and the scale- tion and reuse protocols.
up. In this regard, it should be very convenient to de-
velop systems able to trap both the catalyst and the
salts formed during the reaction. In this way, metal- Results and Discussion
and salt-free product streams would be obtained, rep-
resenting a substantial process improvement.
The synthesis of the different SILLPs is shown in the
Ionic liquids are organic salts whose tunable physi- Scheme 1.[21,23] Polymers 2a, 3a and 7b, c were expect-
cochemical properties make them very interesting for ed to play different roles in the reaction medium.
a wide range of applications, including catalysis.[17,18] It Polymers 2a and 3a were used as the supports for the
has been found that ILs are able to stabilize nanopar- Pd catalytic species (4a and 5a),[23] while 7b, c were
[19]
À
ticles active for C C coupling reactions. Neverthe- expected to serve as the supported base needed to
less, their high cost, low biodegradability and ecotox- neutralize the acid formed during the course of the
icity concerns limit their applicability. A solution to reaction, leading to salt-free products. The supported
overcome these problems consists of supporting them base can be easily regenerated at the end of the cata-
on a insoluble material of organic or inorganic lytic cycle by simply washing the polymer with a basic
nature.[20] When the ILs are covalently anchored to solution of NaOAc in MeOH.[21b]
such a matrix they can be defined as supported ionic
liquid-like phases (SILLPs).[21] Recently we have action between iodobenzene and methyl acrylate was
demonstrated that the nature and loading of the ionic initially carried out employing 0.02 mol% Pd(OAc)2
To check the utility of this approach, the Heck re-
ACHTUNGTRENNUNG
liquid-like units in the polymer influence some impor- as the precatalyst and the polymer 7b as the support-
tant macroscopic properties of the material,[22] and ed base.[24] The system proved to be active and led to
play an important role in the catalytic activity of the a complete conversion of the substrates after 60 min
system, controlling the release of the active species (Table 1, entry 1). Reaction kinetics were monitored
and allowing their efficient recapture after the reac- by reverse-phase HPLC and no peaks corresponding
tion.[23] Furthermore, SILLPs with basic character to the salt were observed. The polymer, after filtering
proved to be efficient catalysts for the Henry reaction and washing with NaOAc solution, was active in the
in batch and in continuous-flow modes.[21b]
second and third cycles. Hence it can be concluded
In this paper the combination of both materials is that the basic SILLP 7b is able not only to act as a
reported to provide a stable polymeric cocktail for base but also to stabilize and recapture the active pal-
À
consecutive cycles of C C bond formation. This poly- ladium species from the homogeneous phase
meric cocktail is able to perform multiple tasks: It is (Scheme 2). For this to occur, it needs to be trans-
the source of the active Pd species, the base required formed through neutralization with the HI produced
in the reaction and the scavenger for the soluble Pd in the course of the reaction.[23b] Nevertheless, the ac-
present in the medium. Thus, the combination of the tivity dropped after each cycle, with TON values at
different SILLPs facilitates the delivery and recapture 60 min decreasing from 4903 to 2515 in four consecu-
tive cycles (Table 1 entries 1–4). This can be due to
Scheme 1. Polymer supported Ionic liquid-like species employed as the support for precatalysts and as the supported base.
g=gel type resin, m=macroporous resin. i) N-methylimidazole, 808C; ii) NaSbF6, MeOH; iii) Pd
DMF; iv) a) Pd(OAc)2, HCl, THF/MeOH, b) NaBH4; v) 1,2-dimethylimidazole, DMF, 808C; vi) NaOAc (aquoeus); vii)
NaOH (aqueous).
ACHTUGNTRENN(NUG OAc)2, Na2CO3, MeOH/
ACHTUNGTRENNUNG
3014
asc.wiley-vch.de
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2010, 352, 3013 – 3021

Polymer Cocktail: A Multitask Supported Ionic Liquid-Like Species
Table 1. Results for the Heck reaction employing the SILLP
7b as a supported base: recycling experiments.[a]
catalysts for the Heck reaction, we considered that a
combination of the polymer-supported precatalyst
and the supported base in the same reactor (polymer
cocktail) could provide a cooperative effect, with the
SILLPs performing the different key tasks required.[23]
Thus,Àin a second step, we selected the precatalyst 4a
(SbF6 as the counteranion). This resin was selected
due to the low coordination ability of this anion,
which can facilitate the release of the active species,
although this could also limit its capacity to subse-
quently recapture them.[23] The Heck reaction using
Et3N as the base and the precatalyst 4a was faster
than when polymer 7b was employed as the base
(TOF values at 50% of conversion: 4285 hÀ1for Et3N
vs. 691 hÀ1for the basic polymer 7b). Rather interest-
ingly, however, the polymer cocktail formed at the
end of the reaction, after filtering and washing, from
Entry Cycle Precatalyst[b] Base[b] Yield [%][c] TON[d]
1
2
3
4
1
2
3
4
Pd
–
–
(OAc)2
7b
–
–
98
78
70
50
4903
3923
3375
2515
–
–
[a]
Reaction conditions: 5 mmol PhI, 5 equiv. methyl acry-
late, 1308C, 5 mL DMF, 0.02 mol% Pd
3 equiv. of 7b.
The precatalyst and the supported base were added only
ACHTUNGRTEN(NUNG OAc)2 and
[b]
in the first reaction cycle.
[c]
[d]
Yield calculated after 1 h of reaction time.
Calculated as mmol product/mmol catalyst at 1 h of reac-
tion time.
the formation of inactive palladium species during the the precatalyst 4a and the basic SILLP 7b led, when
capture of Pd by the SILLP,[25] and revealed that fur- used for a second run, to reaction kinetics significant-
ther efforts were needed to design appropriate active ly faster than the one observed for the first cycle
catalytic systems for an efficient use on a long-term (Figure 1). Thus, for the second run, the reuse of the
scale.
polymer cocktail was clearly more efficient than the
Taking into account our previous results on the use reuse of 4a with Et3N as the base (TOF values at
of Pd species supported on SILLPs as convenient pre- 50% of conversion 750 hÀ1for Et3N vs. 2727 hÀ1 for
Scheme 2. Application of the SILLP cocktail as catalyst, supported base and supported scavenger system for consecutive
À
cycles of C C coupling reactions.
Adv. Synth. Catal. 2010, 352, 3013 – 3021
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
3015

Victor Sans et al.
FULL PAPERS
(Table 2, entry 3). Nevertheless, after an optimization
of this step, the activity was recovered (Table 2, en-
tries 4 and 5).
Once the system had proved to be active and recy-
À
clable, we focused on testing different C C coupling
reactions employing the same catalyst (Figure 2).
Thus, after five runs using the same cocktail for the
Heck reaction, in the 6th cycle (Table 2, entry 6), the
Sonogashira coupling between iodobenzene and phe-
nylacetylene proved to be very effective though small
amounts of side products, presumably dimers and ho-
mocoupling products, were observed. It is worth men-
tioning that the reaction was carried out under aero-
bic conditions and without the use of CuI as co-cata-
lyst. This is advantageous from the point of view of
the green chemistry, since it reduces the waste gener-
ated and simplifies the processing of the products. In
the 7th cycle (Table 2, entry 7), the Suzuki reaction be-
tween iodobenzene and phenylboronic acid was car-
ried out successfully. In this case, 24 h were necessary
to obtain yields over 90%. This is in agreement with
the reaction mechanism involved in this transforma-
tion. In this case the base forms initially a complex
Figure 1. Kinetic profiles of the Heck reaction between io-
dobenzene and methyl acrylate using the polymer cocktail
(4a+7b).
the cocktail). These results suggest that the polymer with the phenylboronic acid that later reacts by trans-
cocktail formed by 4a and 7b is efficiently and simul- metalation with the palladium complex to afford the
taneously acting as Pd-source, supported base and desired product. Therefore the system is controlled by
scavenger and stabilizer of the catalytically active spe- the diffusion of the species from and to the supported
cies involved in the reaction. Noteworthy, the polymer base. Finally, the control Heck reaction showed in the
formed upon initial use and regeneration of the basic 8th cycle a comparable kinetic profile to that in the 2nd
sites is a more efficient catalytic system than the one cycle showing no loss of activity after eight consecu-
À
originally synthesized. Indeed, the system was reused tive different C C bond formation reactions. Indeed,
for five consecutive cycles of the Heck reaction with- TOF50 values were very similar in both cycles (2727
out the need of prior palladium reactivation (Table 2, for 2nd cycle and 2970 for the 8th).
entries 1–5).[25] The washing step proved to be very
Once we had proved that the multifunctional poly-
important, an incomplete protocol leading to a slight meric cocktail is performing efficiently multiple tasks,
reduction in the yield as seen in the third cycle we focused on the further development of the system
by optimizing the nature of resins constituting the
cocktail. In most of the cases palladium-supported
À
Table 2. Yields obtained in succesive C C coupling reactions
complexes give rise, under the reaction conditions, to
employing a polymer cocktail from the supported precata-
the formation of soluble Pd(0) species, which are re-
lyst 4a and the base 7b: recycling experiments.[a]
sponsible for the catalytic activity.[13] Hence, the pre-
Entry Cycle Precatalyst[b] Reaction
Yield[c] [%]
catalyst previously employed (4a) could be replaced
by Pd(0) nanoparticles supported on SILLPs
(Scheme 1, 5a).[23] This simplifies the synthetic proce-
dure and reduces the waste generated. On the other
hand, the supported base is employed in excess with
regard to the substrates,[26] and large amounts of these
SILLPs are stirred in the system. Therefore, the me-
chanical stability of the backbone polymer might limit
the efficiency of the system after a high number of
reuses. Hence, we decided to explore the possibility
of using a highly cross-linked polymer (macroporous)
presenting a higher mechanical strength than the mi-
croporous ones. Thus, the same synthetic procedure
was repeated using commercial macroporous Merri-
field resin 1c with a high loading (5.5 mmol Cl/g resin,
Scheme 1).
1
2
3
4
5
6
7
8
1
2
3
4
5
6
7
8
4a
–
–
–
–
–
–
–
Heck
Heck
Heck
Heck
Heck
Sonogashira 92
Suzuki
Heck
99
99
85
96
94
91[d]
99
[a]
Reaction conditions: 5 mmol PhI, 5 equiv. of the other
substrate, 1308C, 5 mL DMF. For the first run, 0.1 mol%
of 4a and 3 equiv of 7b.
[b]
The precatalyst and the supported base were added only
in the first reaction cycle.
[c]
[d]
Yields calculated at 3 h of reaction time.
Yield calculated at 24 h of reaction time.
3016
asc.wiley-vch.de
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2010, 352, 3013 – 3021

Polymer Cocktail: A Multitask Supported Ionic Liquid-Like Species
À
Figure 2. Different C C coupling reactions catalyzed by the resin cocktail after 5 consecutive cycles of the Heck reaction.
À
The new cocktail formed by 5a and 7c proved to be Table 3. C C coupling reactions employing optimized condi-
tions: the precatalyst 5a and the supported base 7c were
active and the same catalytic sequence was repeated.
In the first place, the Heck reaction was carried out
for five consecutive cycles. Later on, the Sonogashira
and the Suzuki reactions were also conducted and, fi-
nally, the Heck reaction was repeated. The first obser-
vation was that the reaction kinetics was relatively
slow for the first cycle. 24 h were necessary to obtain
good yields. Two effects can contribute to this phe-
nomenon. First of all we must bear in mind that the
supported precatalyst needs to release the active spe-
cies to the solution. In this regard, the resin 5a has a
more complexing counteranion and a higher loading
than resin 4a, which has been proved to lead to a
slower release of the active species to the medium.[23]
Moreover, with the use of the macroporous supported
base 7c diffusion-limited processes can participate.
Upon recycling, the active Pd will be located on the
surface of both SILLPs, in close proximity to basic
sites, reducing the influence of chloride anions and
used for the resin cocktail.[a]
Entry
Cycle[b]
Reaction
Time [min]
Yield [%]
1
2
3
4
5
6
7
8
1
Heck
Heck
Heck
Heck
1440
160
160
160
160
1440
1440
160
78
82
58
99
99
99
59
99
2
3[c]
4
5
6
7
8
Heck
Sonogashira
Suzuki
Heck
[a]
Reaction conditions: 5 mmol PhI, 5 equiv. other substrate,
1308C, 5 mL DMF, 0.1 mol% 5a and 3 equiv. of 7c.
Precatalyst and supported base were added only in the
first reaction cycle.
An incomplete washing protocol was used after the
second run.
[b]
[c]
that of diffusion-limited processes on the surfaces of were found to be below 2 ppm (less than 0.2% of the
the macroporous polymer, and this leads to an in- initial palladium used). Only for the Sonogashira re-
crease of the reaction rate. After the 4th cycle, the re- action (Table 3, entry 6) were 3.8 ppm of palladium
action rate was optimized, leading to a complete con- found in solution by ICP-MS.[16]
version in less than 3 h. The Suzuki coupling (Table 3,
The scope of the catalytic system was also explored
entry 7) proved again to be slower than the other re- by testing more challenging substrates. Thus, the
actions assayed. In this reaction, the boronic acid Heck reaction was examined for different iodo- and
forms an initial complex with the base and therefore bromobenzenes and with different acrylates, as well
diffusional limitations can also be operative here, as with styrene, and the results are gathered in
which might explain the slower kinetics in this reac- Table 4. The polymer cocktail formed from 5a and 7c
tion cycle. In most cases, leaching levels of palladium was chosen as the catalytic system and N-methylpyr-
Adv. Synth. Catal. 2010, 352, 3013 – 3021
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
3017

Victor Sans et al.
FULL PAPERS
À
Table 4. Yields obtained in the C C coupling reactions employing a polymer cocktail from the supported precatalyst 5a and
the supported base 7b: recycling experiments using different substrates.[a,b]
Entry
Cycle
R1
X
R2
Z
Yield[c] [%]
1
2
3
4
5
6
7
8
9
1
2
3
4
5
6
7
8
9
C6H5
C6H5
C6H5
C6H5
4-NO2-C6H4
C6H5
4-CH3CO-C6H4
C6H5
C6H5
C6H5
I
I
I
I
Br
I
Br
I
Br
I
CH2=CH
CH2=CH
CH2=CH
CH2=CH
CH2=CH
CH2=CH
CH2=CH
CH2=CH
CH2=CH
CH2=CH
CO2Me
C6H5
CN
CO2Me
CO2Me
CO2Me
CO2Me
CO2Me
CO2Me
CO2Me
99
99
85
98
92
97
90
97
26
90
10
10
[a]
Reaction conditions: 5 mmol of RX, 5 equiv of the other substrate, 1408C, 5 mL NMP, 0.05 mol% 4a and 3 equiv. of 7c
for the first run.
The precatalyst and the supported base were added only in the first reaction cycle.
Yield calculated at 24 h of reaction time.
[b]
[c]
rolidone was used as a solvent. The different reactions filled with the resin cocktail (precatalyst 4a and sup-
were carried out by recycling the same polymer cock- ported base 7b) and introduced in a thermostatic
tail. Aryl iodides were rapidly converted to the corre- oven. The system was pressurized to 8 MPa and the
sponding Heck products with excellent yields, regard- oven heated to 2008C. Once the system was stable,
less of the nature of the alkene (Table 4, entries 1–3). the organic suspension was pumped through the reac-
The initial activity of the cocktail was maintained tor (scCO2: 0.1 mLminÀ1; substrate: 0.01 mLminÀ1).
during the whole process. This was clearly demon- The system was active giving a constant yield of ca.
strated when the benchmark model reaction between 80% of trans-methyl cinnamate for 60 min of reaction
PhI and methyl acrylate was newly assayed (entries 1 time followed by a drop in the conversion due to the
and 4). Reaction of p-bromoacetophenone and p-ni- consumption of the base. Interestingly, the system was
trobromobenzene with methyl acrylate gave the cor- kept working for more than three hours without any
responding products in good yield. Nevertheless, as overpressure. As expected, the obtained chromato-
could be expected, when the less active substrate bro- grams (Figure 3, panel C) show no signs of either salts
mobenzene was assayed, a lower yield of product was or solvent in the product stream. When the reaction
observed for the same reaction time. After each cycle was carried out using triethylamine as the base, how-
with these new substrates a cycle with the benchmark ever, a clear peak corresponding to the salts generat-
reaction was performed again to test the stability of ed during the reaction can be observed (Figure 3,
the system upon recycling.
panel A). This peak does not appear when the sup-
The resin cocktail system offers a substantial im- ported base is employed (Figure 3, panel B). Thus, by
À
provement in terms of process intensification for C C an adequate optimization of the reaction parameters,
coupling reactions, since salt-free and almost metal- including RTD and the amount of the base in the
free systems can be obtained. In this way, the down- system, it is possible to obtain completely pure prod-
stream processing of the products is highly facilitated. ucts directly from a chemical reactor.
In a further effort to improve the system in terms of
green chemistry, the polymer cocktail was employed
in a continuous-flow set-up using scCO2 as a solvent. Conclusions
In normal conditions the presence of salts would lead
to problems such as clogging of the reactor.[27] As a Summarizing, herein we have developed a multitask
proof of concept we decided to conduct the Heck re- catalytic system in a simple and inexpensive way for
À
action employing the resin cocktail in continuous-flow different C C coupling reactions. A polymeric cock-
mode using scCO2 as a solvent without any additional tail is formed through the combination of two compo-
cosolvent.[28] The low polarity of scCO2 would pro- nents based on polymeric SILLPs. The first compo-
duce the precipitation of the salts formed during the nent is a Pd-SILLP system that can efficiently act as a
reaction. This effect might eventually lead to prob- precatalyst for the generation of the active soluble Pd
lems, such as the blockage of the reactor. For this pur- species. The second component is a supported base
pose, a 1/4-inch HPLC-like stainless steel tube was obtained in a straightforward way by the use of a
3018
asc.wiley-vch.de
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2010, 352, 3013 – 3021

Polymer Cocktail: A Multitask Supported Ionic Liquid-Like Species
Figure 3. Detail of the reverse-phase HPLC chromatograms obtained under the different conditions assayed. (A) Reaction
using DMF as a solvent and Et3N as a base: presence of salts and solvent. (B) Reaction using DMF as a solvent and the sup-
ported base: absence of salts. (C) Reaction using scCO2 as a solvent and the supported base: absence of solvent and salts.
basic counteranion in the corresponding SILLP. This supported species evolve in the course of the process
polymeric cocktail is able then to simultaneously act so as to achieve a final system that is better optimized
as the catalyst precursor and as a base in the Heck for the tasks to be carried out. Thus, the supported Pd
À
and related C C coupling reactions. After the first pre-catalyst releases the active catalytic species, the
cycle(s), this cocktail is transformed into a new mate- supported base neutralizes the acid species generated
rial in which both polymeric components can possess during the reaction and, simultaneously all the pres-
simultaneously both functionalities (Pd releasing sites ent imidazolium sites act as efficient scavengers, re-
and basic sites). Rather interestingly, the resulting capturing the soluble Pd species and being able to sta-
polymeric cocktail is much more efficient than the bilize the resulting Pd nanoparticles, maintaining their
original one. Different aryl bromides and iodides activity as precatalysts for a new reaction cycle. Ac-
have proved to be efficiently transformed by this cata- cording to this multitask behaviour finally accom-
À
lytic cocktail. Moreover, three different C C coupling plished by the cocktail derived from the two different
reactions have been successively tested. The two com- SILLPs, salt-free product streams with essentially no
ponents have proved to act in a synergistic way, metallic traces are obtained. This simplifies the work-
giving as a result highly active catalytic systems, up and enables one to conduct the reaction in condi-
stable upon recycling and capable to sequentially per- tions that would be hindered by the presence of salts
À
form different C C coupling reactions. To the best of such as in scCO2. The use of scCO2 further improves
our knowledge, this is the first example in which the the environmentally friendly character of the process,
Adv. Synth. Catal. 2010, 352, 3013 – 3021
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
3019

Victor Sans et al.
FULL PAPERS
facilitating the generation of product streams not con- Acknowledgements
taminated by residual solvent traces and to work
The authors would like to thank the Spanish MICINN and
Feder funds (CTQ2008-04412/BQU), CSIC (2009-80I092),
Generalitat Valenciana (ACOMP/2010/280) and Bancaixa-
UJI (P1 1A2007-11). The collaboration of the SCIC of the
Universitat Jaume I is also acknowledged.
under continuous-flow conditions. As a result, it is
possible to perform the chemical reactions, the purifi-
cation steps and the removal of the solvent in a single
reaction vessel or reactor. This represents a significant
improvement in terms of process intensification and
green chemistry.
References
Experimental Section
[1] J. A. Gladysz, Pure Appl. Chem. 2001, 73, 1319–1324.
[2] Immobilized Catalysts: Solid Phases, Immobilization
and Applications, (Eds.: A. Kirschning, W. Bannwarth,
D. E. Bergbreiter, B. Desai), in: Topics in Current
Chemistry, Springer-Verlang, Berlin, Heidelberg, 2004.
[3] The Power of Functional Resins in Organic Synthesis,
(Eds.: F. Albericio, J. Tulla-Puche), Wiley-VCH, Wein-
heim, 2008.
The syntheses of SILLPs 2a, 3a, 4a, 5a and 7b, c were per-
formed according to the general procedures previously de-
scribed in refs.[21,23]
General Procedure for the Heterogeneous Heck
Reaction using Supported Precatalysts and Bases
[4] Reviews and specials issues: a) P. H. Toy, M. Shi, ACHTUNGTRENNUNG(Eds.
In a round-bottomed flask were introduced DMF (5 mL),
PhI (0.326 g, 1.6 mmol), methyl acrylate (0.75 g, 8 mmol),
supported base (1 g, ca. 3 mmol), mesitylene (0.372 mL,
3.09 mmol) and the corresponding palladium source
(0.1 mol% Pd). The flask was closed and heated at 1308C.
Periodically, samples of 5 mL were taken, diluted with
CH3CN (1 mL) and analyzed by HPLC. When the reaction
was over, the polymer was filtered and washed with MeOH,
CH2Cl2, MeOH, a solution of KOAc (0.2 g/mL) in metha-
nol, MeOH, CH2Cl2 and then dried under vacuum. An apro-
priate and thorough washing of the resin with a concentrat-
ed solution of NaOAc in MeOH was critical to regenerate
the supported base. Therefore 50 mL were employed, and
the suspension was left for 2–3 h to allow for a complete
anion exchange between OAc and the iodides of the resin.
The dried polymer was used for the following reaction
cycles. The Suziki and Sonogashira reactions were per-
formed using a similar protocol. The kinetics of the reac-
tions were followed by HPLC, using the methods previously
developed for the Heck, Suzuki and Sonogashira reactions.
), Polymer-Supported Reagents and Catalysts: Increas-
ingly Important Tools for Organic Synthesis, Tetrahe-
dron 2005, 61, 12013–12192; b) J. A. Gladysz, ACHTUNGTRENNUNG(Ed. ),
Recoverable Catalysts and Reagents, Chem. Rev. 2002,
102, special issue; c) T. Kehat, K. Goren, M. Portnoy,
New J. Chem. 2007, 31, 1218–1242; d) J. Lu, P. H. Toy,
Chem. Rev. 2009, 109, 815–838.
[5] a) B. Altava, M. I. Burguete, E. Garcꢀa-Verdugo, S. V.
Luis, M. J. Vicent, J. A. Mayoral, React. Funct. Polym.
2001, 48, 25–35; b) B. Altava, M. I. Burguete, J. M.
Fraile, J. I. Garcia, S. V. Luis, J. A. Mayoral, S. V. Luis,
Angew. Chem. 2000, 112, 1563–1566; Angew. Chem.
Int. Ed. 2000, 39, 1503–1506.
[6] T. Y. S. But, Y. Tashino, H. Togo, P. H. Toy, Org.
Biomol. Chem. 2005, 3, 970–971.
[7] N. T. S. Phan, C. S. Gill, J. V. Nguyen, Z. J. Zhang,
C. W. Jones, Angew. Chem. 2006, 118, 2267–2270;
Angew. Chem. Int. Ed. 2006, 45, 2209–2212.
[8] a) M. I. Burguete, A. Cornejo, E. Garcꢀa-Verdugo, J.
Garcꢀa, M. J. Gil, S. V. Luis, V. Martꢀnez-Merino, J. A.
Mayoral, M. Sokolova, Green Chem. 2007, 9, 1091–
1096; b) N. Madhavan, C. W. Jones, M. Weck, Acc.
Chem. Res. 2008, 41, 1153–1165.
[9] a) Metal-Catalyzed Cross-Coupling Reactions, Vols. 1
and 2, (Eds.: A. de Meijere, F. Diederich), Wiley-VCH,
Weinheim, 2004; b) Transition Metals for Organic Syn-
thesis; Building Blocks and Fine Chemicals, (Ed.: M.
Beller), Wiley-VCH, Weinheim, 2004.
[10] R. Chinchilla, C. Nꢂjera, Chem. Rev. 2007, 107, 874–
922.
[11] L. Yin, J. Liebscher, Chem. Rev. 2007, 107, 133–173.
[12] R. A. Sheldon, M. Wallau, I. W. C. E. Arends, U. Schu-
chardt, Acc. Chem. Res. 1998, 31, 485–493.
[13] a) J. M. Richardson, C. W. Jones, J. Catal. 2007, 251,
80–93; b) M. Weck, C. W. Jones, Inorg. Chem. 2007,
46, 1865–1875.
[14] N. T. S. Phan, M. V. D. Sluys, C. W. Jones, Adv. Synth.
Catal. 2006, 348, 609–679.
General Procedure for the Heck Reaction in
Continuous-Flow Mode in scCO2
An HPLC column (15 cmꢄ1/4 inch i.d.) was filled with a
polymeric mixture (725 mg total mass, 1:100 ratio) of poly-
mer 4a and supported base 7b. Then the column was con-
nected to a system equipped with a CO2 pump (Jasco, 304,
flow rate=0.25–1 mLminÀ1, Pmax =40 MPa), introduced in
an oven (Watlow, 240 V, 200 W, NTB25X6UA3, 47 02 DM)
and heated to 2008C. The solvent conditions were set at
708C and 80 MPa and controlled by means of a back pres-
sure regulator (Jasco Pmax =42 MPa, Cv <0.01). scCO2 was
pumped at a rate of 0.1 mLminÀ1. When stationary condi-
tions were reached, the organic phase was pumped by
means of an HPLC Jasco pump [Gilson Model 304, flow
rate=0.01–5 mLminÀ1, maximum allowed pressure (Pmax)=
40 MPa] at 0.01 mLminÀ1. The products were collected, and
dissolved in 1 mL of CH3CN. From this solution 20 mL were
collected and diluted in 1 mL of CH3CN and analyzed by
HPLC.
[15] E. J. Flahive, B. L. Ewanicki, N. W. Sach, S. A. OꢅNeill-
Slawecki, N. S. Stankovic, S. Yu, S. M. Guinness, J.
Dunn, Org. Process Res. Dev. 2008, 12 , 637–645.
3020
asc.wiley-vch.de
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2010, 352, 3013 – 3021

Polymer Cocktail: A Multitask Supported Ionic Liquid-Like Species
[16] The specifications for low metal content in the final
E. Garcia-Verdugo, S. V. Luis, V. Sans, J. Mol. Cat. A:
Chem. 2009, 309, 131–136.
product (e.g., as low as <10 ppm for palladium in an
orally administered drug substance) require efficient
separations processes for removal of dissolved metals
from reaction mixtures. European Medicines Agency,
Guideline on the Specification Limits for Residues of
Metals Catalysts, January, 2007; www.emea.europa.eu/
pdfs/human/swp/444600.pdf.
[22] M. I. Burguete, F. Galindo, E. Garcꢀa-Verdugo, N. Kar-
bass, S. V. Luis, Chem. Commun. 2007, 3086–3088.
[23] a) B. Altava, M. I. Burguete, E. Garcia-Verdugo, N.
Karbass, S. V. Luis, A. Puzary, V. Sans, Tetrahedron
Lett. 2006, 47, 2311–2314; b) N. Karbass, V. Sans, E.
Garcia-Verdugo, M. I. Burguete, S. V. Luis, Chem.
Commun. 2006, 3095–3097; c) M. I. Burguete, E.
Garcꢀa-Verdugo, F. Gelat, S. V. Luis, V. Sans, I. Garcꢀa-
Villar, P. Licence, J. Catal. 2010, 269, 150–160.
[24] a) A. H. M. de Vries, J. M. C. A. Mulders, J. H. M.
Mommers, H. J. W. Henderickx, J. G. de Vries, Org.
Lett. 2003, 5, 3285–3288; b) A. Alimardanov, L.
Schmieder-van de Vondervoort 1, A. H. M. de Vries,
J. G. de Vries, Adv. Synth. Catal. 2004, 346, 1812–1817.
[25] A. H. M. de Vries, F. J. Parlevliet, L. Schmieder-van
de Vondervoort, J. H. M. Mommers, H. J. W. Hender-
ickx, M. A. M. Walet, J. G. de Vries, Adv. Synth. Catal.
2002, 344, 996–1002.
[26] There are only few reports in the literature where cata-
lyst and base are bearing in the same material: a) A.
Corma, H. Garcia, A. Primo, J. Catal. 2006, 241, 123;
b) F. Chen, K. Toh, S. Shen, G. J. Gan, Catal. Commun.
2007, 8, 405–409; c) A. Corma, H. Garcıa, A. Leyva,
A. Primo, Appl. Catal. A: General, 2004, 257, 77–83.
[27] When the reaction is performed using 5a as the catalyst
and Et3N as the base in scCO2 the reactor is blocked,
see ref.[23b]
[17] P. Wasserscheid, T. Welton, (Eds.), Ionic Liquids in
Synthesis, Wiley-VCH, Weinheim, 2007.
[18] N. V. Plechkova, K. R. Seddon, Chem. Soc. Rev. 2008,
37, 123–150.
[19] a) D. Zhao, Z. Fei, T. J. Geldbach, R. Scopelliti, P. J.
Dyson, J. Am. Chem. Soc. 2004, 126, 15876–15882;
b) P. Migowski, J. Dupont, Chem. Eur. J. 2007, 13, 32–
39; c) X. Yang, Z. Fei, D. Zhao, W. Han Ang, Y. Li,
Paul J. Dyson, Inorg. Chem. 2008, 47, 3292–3297; d) M.
Ruta, G. Laurenczy, P. J. Dyson, L. Kiwi-Minsker, J.
Phys. Chem. C, 2008, 112, 17814–17819; e) E. Redel,
R. Thomann, C. Janiak, Chem. Commun. 2008, 1789–
1791.
[20] a) C. P. Mehnert, Chem. Eur. J. 2004, 11, 50–56; b) A.
Riisager, R. Fehrmann, S. Flicker, R. van Hal, M. Hau-
mann, P. Wasserscheid, Angew. Chem. 2005, 117, 826–
830; Angew. Chem. Int. Ed. 2005, 44, 815–819; c) Y.
Gu, G. Li, Adv. Synth. Catal. 2009, 351, 817–847.
[21] a) P. Lozano, E. Garcꢀa-Verdugo, R. Piamtongkam, N.
Karbass, T. De Diego, M. I. Burguete, S. V. Luis, J. L.
Iborra, Adv. Synth. Catal. 2007, 349, 1077–1084;
b) M. I. Burguete, H. Erythropel, E. Garcia-Verdugo,
S. V. Luis, V. Sans, Green Chem. 2008, 10, 401–407;
[28] G. A. Leeke, R. C. D. Santos, B. Al-Duri, J. P. K. Sev-
ille, C. J. Smith, C. K. Y. Lee, A. B. Holmes, I. F.
McConvey, Org. Process Res. Dev. 2007, 11, 144–148.
´
c) W. Gil. , K. Boczon, A. M. Trzeciak, J. J. Ziꢁłkowski,
Adv. Synth. Catal. 2010, 352, 3013 – 3021
ꢃ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
3021