ssil

8/13/2019 SSIL

1/14

PERSPECTIVE www.rsc.org/dalton | Dalton Transactions

Immobilization of molecular catalysts in supported ionic liquid phases

Charlie Van Doorslaer, a Joos Wahlen, a Pascal Mertens, a Koen Binnemans b and Dirk De Vos* a

Received 19th January 2010, Accepted 18th March 2010DOI: 10.1039/c001285h

In a supported ionic liquid phase (SILP) catalyst system, an ionic liquid (IL) lm is immobilized on ahigh-surface area porous solid and a homogeneous catalyst is dissolved in this supported IL layer,thereby combining the attractive features of homogeneous catalysts with the benets of heterogeneouscatalysts. In this review reliable strategies for the immobilization of molecular catalysts in SILPs aresurveyed. In the rst part, general aspects concerning the application of SILP catalysts are presented,focusing on the type of catalyst, support, ionic liquid and reaction conditions. Secondly, organicreactions in which SILP technology is applied to improve the performance of homogeneoustransition-metal catalysts are presented: hydroformylation, metathesis reactions, carbonylation,hydrogenation, hydroamination, coupling reactions and asymmetric reactions.

Introduction

Ionic liquids (ILs) are organic salts with low melting points andvery lowvapour pressures. 1-3 Their non-volatile nature is oneof themain motives to explore them as alternatives for volatile organicsolvents. 4 Depending on their composition, ILs can dissolve orreject organic compounds. Since ILs are ionic and polar, theyare excellent media to hold charged or polar catalyst species liketransition-metal complexes. 3-11 In analogy to the well-establishedsupported aqueous-phase catalysis, 12-13 ILs have recently beenintroduced for the immobilization of homogeneous catalysts. 14-16

In a supported ionic liquid phase (SILP) catalyst system, an ILlmis immobilized on a high-surfacearea porous solid ( e.g. , silica)and the homogeneous catalyst is dissolved in the IL layer. Theresultingcatalyst is a solid,with theactive species being solubilized

in the IL phaseand behavingas a homogeneous catalyst. Typically,

aCentre for Surface Chemistry and Catalysis, Department of Microbial and Molecular Systems, K.U.Leuven, Kasteelpark Arenberg 23, box 2461, 3001,Leuven, Belgium. E-mail: [email protected] for Molecular Design and Synthesis, Department of Chemistry,K.U.Leuven, Celestijnenlaan 200F, box 2404, 3001, Leuven, Belgium

Charlie Van Doorslaer

Charlie Van Doorslaer (born1983) obtained his Masters degree in Applied Bioscience and Engineering at the Catholic Uni-versity of Leuven (K.U.Leuven,Belgium) in 2006 with a the-sis on carbamate synthesis fromamines, alcohols and CO 2 at theCentre of Surface Chemistry and Catalysis. SinceSeptember 2006,he has been preparing a PhDthesis on ionic liquids as media for catalytic reactions.

Joos Wahlen

Joos Wahlen (born 1977) completed his doctoral research atthe Centre for Surface Chem-istry and Catalysis in 2004 witha dissertation on heterogeneouscatalysis for the generation of singlet oxygen and performed a post-doc in the same group. Hisexpertise is in thebroad domainof organic catalysis with solid mate-rials and immobilized molecularcatalysts, selective oxidation and organocatalysis.

there is no direct interaction between the supports surface and thehomogeneous catalyst.

SILPs allowa more efcientutilization of theIL andthe catalystsince theILs surface area is increased relative to itsvolume andthesubstrate can readily diffuse to the catalyst. In contrast, operationunder classical liquid/liquid biphasic conditions requires largeramounts of ILs and the ILs high viscosity may cause mass-transfer limitations. The latter is particularly important for gas-phase reactions, since ILs generally display a limited solubility forgases such as H 2 , O 2 and CO. SILP catalysis combines the mostattractive features of homogeneous catalysis like high activity andselectivity with benets of heterogeneous catalysts such as largeinterfacial reaction areas and ease of product separation. Indeed,the reaction products can be recovered from the organic phase,while the catalyst remains immobilized in the SILP. Therefore,

SILP catalysis has potential for efcient catalyst recycling andmight enable the application of xed-bed reactor technologyto homogeneous catalysis. Some reviews about this interestingtopic have already appeared. 17-20 Wasserscheid et al. focused ina microreview on the hydroformylation of propene. 17 In anotherreview they tackle the use of SILP technology in catalytic andseparation technologies. 19 Gua and Li deal with solid-based

This journal is The Royal Society of Chemistry 2010 Dalton Trans. , 2010, 39 , 83778390 | 8377

A

H

8/13/2019 SSIL

2/14

heterogeneous catalytic systems that have been developed withthe aid of ILs. 18 Thus, topics like the use of solid catalysts in ILs,metal nanocatalysts in ILs and the preparation of heterogeneouscatalysts in ILs, can be found in this reference. In previousreviews, various aspects of the immobilization of transition-metalcomplexes in a SILP have only partially been covered. Cleartrends can be discerned now in the preparation procedures of therespective SILP catalysts with regard to choice of catalyst, IL andsupport. Moreover, as the body of experimental work grows, it isnow possible to make quantitative comparisons between homoge-nous, heterogeneous and biphasic systems. Therefore the presentreview focuses on recent developments in the immobilization of molecularly dened catalysts in SILPs.

Firstly, general aspects of SILP catalysis are discussed. Theseinclude the preparation of SILP catalysts and the choice of thehomogeneous catalyst, ionic liquid, support, and reaction condi-tions. Next, the use of SILP catalysis is illustrated with examplesfrom hydroformylation, metathesis, carbonylation, hydrogena-tion, hydroamination, CC coupling and various enantioselectivereactions.

Pascal Mertens

Pascal Mertens (born 1980) ob-tained his Masters degree in Ap- plied Bioscience and Engineeringat theCatholic University of Leu-ven (K.U.Leuven, Belgium) witha thesison colloidalgoldcatalystsin alcohol oxidation at the Centreof Surface Chemistry and Catal- ysis. He also completed his doc-toral research with a dissertationon colloidal metal catalysts inliquid phase redox reactions. Hisresearch interests involve selectivetransformations in the synthesis

of ne chemicals and pharmaceuticals.

General considerations

Catalysts, ILsandsolidsupports canbe combined in various ways.An overview is presented in Scheme 1. In this review, catalystsare denoted SILP catalysts (SILPCs) if the support is loadedwith a separate IL phase (Type 1). The IL is deposited on thesupport as a multilayer and acts as an immobilizing mediumfor the catalyst, behaving as a homogeneous catalyst. The moststraightforward procedure for the preparation of a Type 1 SILPC(Type 1a, catalyst/IL/support) is via simple impregnation of theIL on a supports surface. Prior to the impregnation step, themetal and the ligand are dissolved in the IL phase. In a differentapproach (Type 1b, catalyst/IL/IL-support), the surface of thesupport is rst modied with a monolayer of covalently anchoredIL fragments.Next, additional IL and thecatalystare impregnatedon the modied support.

In Type 2 catalysts, the IL is directly anchored to the supportvia covalent anchoring, forming a monolayer of IL fragments.This can be achieved by grafting a functionalized IL fragmenton a preformed support or by solgel synthesis. 21 Accordingto our denition, the resulting solid catalysts are in fact noimmobilized ionic liquid phase catalysts, since there is noseparate IL phase supported on the solid. The IL in this caseshould rather be considered as a covalently anchored ligand.In this approach (Type 2, catalyst-IL-support + IL-support),IL fragments are rst covalently anchored to a support via theorganic cation, and part of the anion is then ion-exchangedwith anionic catalyst species ( e.g. , [WO 4]2- , [RuO 4]- , [PdCl 4]2- ,[NiCl 4]2- , [CuCl 4]2- , [Al 2Cl 7]- , [SnCl 5]- ).22-25 In this context, itshould be noted that earlier reported catalytic systems withimmobilized quaternary ammonium moieties and ion-exchangedmetalcatalysts arenowadays more popularly called supported ionicliquid catalysts (SILCs). Anchoring via the anionic part of the IL isalso possible. Alternatively, the catalyst or ligand is incorporatedin the IL fragment via covalent binding. 26 In a third strategy(Type 3, IL/catalyst-support), an IL is impregnatedon the surfaceof classical heterogeneous catalysts such as covalently anchored

Koen Binnemans

Koen Binnemans (born 1970)obtained his M.Sc. (1992) and Ph.D. (1996) in Chemistry at theCatholic University of Leuven. Inthe period 19992005, he was a postdoctoral fellow of the Fund for Scientic Research Flanders(Belgium). He did postdoctoral work with Prof. Jacques Lucas(Rennes, France) and Prof. Dun-can W. Bruce (Exeter, U.K.). In2000, he received the rst ERES (European Rare-Earth and Ac-tinide Society) Junior Award .

From 2002 until 2005, he was (part time) associate professor.Presently, he is professor of chemistry at the Catholic Universityof Leuven. His current research interests are metal-containing liquid crystals (metallomesogens), lanthanide-mediatedorganicreactions,lanthanide spectroscopy, luminescent molecular materials, and ionicliquids.

Dirk De Vos

Dirk De Vos (born 1967) isa full professor at the CatholicUniversity of Leuven. After doc-toral research in Leuvenand post-doctoral work at Purdue Univer-sity, he focuses on selective cat-alytic oxidation, metalorganic frameworks, ionic liquids and insitu spectroscopy and microscopyof working catalysts. His awardsinclude the D. W. Breck Award in Zeolite Science and Technol-ogy (1998, together with P. Ja-cobs and co-workers), the BASF

catalysis award (2001), the award of the Flemish scientic founda-tion (2006), and the IPMI Student Advisor Award (2008).

8378 | Dalton Trans. , 2010, 39 , 83778390 This journal is The Royal Society of Chemistry 2010

A

H

8/13/2019 SSIL

3/14

Scheme 1 Various catalyst/ionic liquid/support combinations.

catalysts, 27 or supported transition-metal nanoparticles. 28-30 Thistype of catalyst was recently named solid catalyst with ionic liquid layer (SCIL). 31 The present review is limited to Type 1 catalysts,for which a SILP is used to immobilize homogeneous catalystspecies. For the immobilization of transition-metal nanoparticles,we refer to other reviews, which elaborate this topic moreexhaustively. 18-19

Homogeneous catalyst

A prerequisite for the industrial application of SILP catalysis is

that the homogeneous catalyst has an inherently stable catalyticperformance, guaranteeing a long lifetime. In addition, the loss of catalyst into the organic layer containing the products should benegligible. Since ILs are charged and have polar properties, theyare well suited to hold ionic or highly polar catalyst species, likeionic metal species ( e.g. , [PdCl 4]2- , [WO 4]2- ). On the other hand,the preference of the catalyst for the IL phase can be increasedby incorporation of ionic groups in the ligand of the catalystcomplex. For instance, phosphine ligands can be modied withanionic sulfonate groups or cationicgroupsbasedon guanidinium,imidazolium or pyridinium moieties. Thisreviewfocuseson molec-ularly dened catalysts; therefore, the combination of ILs withsupported metal catalysts, 31 metal nanoparticles 32 or supported

enzymes33

is not discussed. Regarding the eld of applications, theuse of SILPs is being studied extensively for the immobilization of rhodium and palladium complexes, since their heterogenizationvia classical routes ( e.g. , impregnation or covalent anchoring)has only been developed with limited success. Although lessfrequently studied in this context, other metals such as Ru, Cu,Mn, Co, Zn or Cr have also been considered for immobilizationin SILPs.

Ionic liquid

ILs are often called designer solvents, since their properties suchas density, melting point, viscosity, or miscibility with water or

organic solvents can be ne-tuned by an appropriate choice of anion and cation. 1-3 The cations in ILs are mostly large organiccations, like N ,N -dialkyl-imidazolium, N -alkylpyridinium, N ,N -dialkylpyrrolidinium, tetraalkylammoniumor tetraalkylphospho-nium ions. Several cation types can be used for SILP catalysis.Especially the 1-ethyl-3-methylimidazolium (emim), 1-butyl-3-methylimidazo-lium (bmim) and 1-hexyl-3-methylimidazolium(hmim) ions are often used as the ILs cationic part. Theanions can vary widely, e.g. , Cl - , Br - , PF 6 - , BF 4 - , CF 3SO 3 - and(CF 3SO 2)2N - . Whereas Cl - and Br - yield hydrophilic ILs that aremiscible with water, uorinated anions like BF 4 - or PF 6 - allowthe preparation of hydrophobic ILs immiscible with H 2O. Thehydrophobicity of ILs with BF 4 - depends on the cations alkylchain length. The most widely used IL in SILP applications is 1-butyl-3-methylimidazoliumtetrauoroborate, [bmim][BF 4],or thehexauorophosphate salt, [bmim][PF 6]. These ILs combine a highsolvation power for polar catalyst complexes with a weak coordi-nation strength.This specic combinationmakes themsuitable forreactionswith electrophilic catalysts. Themajor problemis that theBF 4 - and PF 6 - anions are sensitive to hydrolysis. The resulting HFmight cause side reactions or corrosion problems, or the stronglycoordinating uoride ions might poison the transition-metal cat-alyst by irreversible complexation. Consequently, the applicationof these ILs is restricted to anhydrous conditions. In this context,it is not surprising that the ILs water content is an importantissue, as the presence of small amounts of water might inducehydrolysis of the metal catalyst or decomposition of the BF 4 - orPF 6 - anions. Moreover, impurities such as residual halide ionsmight bind tightly to the transition-metal complex and impedeits catalytic activity. Recently, bis(triuoro-methylsulfonyl)imideanion (CF 3SO 2)2N - has emerged as a popular anion for synthe-sizing hydrophobic ILs, because the resulting ILs are chemicallyand thermally more robust. However, disposal of such ILs, e.g. , bycombustion, is more complicated due to potential HF liberation.General concerns about environmental impact and toxicologyassociated with ILs involving halogen-containing anions recentlyled to the introduction of halogen-free ILs as environmentallyfriendly solvents e.g. , imidazolium-based ILs with weakly coordi-nating anions like alkylsulfates. 34 In summary, such low-melting,halogen-free IL systems that combine weak coordination toelectrophilic metal centres and low viscosity with high stability tohydrolysis are highly desirable for applications in transition-metalcatalysis.

Support

The majority of the supports used in SILP catalysis are poroussilica gels with a high surface area of 300500 m 2 g- 1. Thermalpretreatment of the silica gel reduces the number of acidic silanolgroups. This can be important for a stable catalyst performancesince the silanol groups may react with ligands or catalysts, herebylowering the actual concentration of the ligand or the catalystin the IL phase. Mesoporous silica materials ( e.g. , MCM-41 andMCM-48) and crystalline silicaaluminamaterials suchas zeolitesand clays ( e.g. , montmorillonite) have also been used. 35-39 Otherinorganic materials (Al 2O 3 , TiO 2, ZrO 2 etc. ) are less frequentlyused due to their low surface areas and pore volumes. However,under certain conditions alumina might be preferred, since thissupport shows higher stability at high pH values as compared to


A

H

8/13/2019 SSIL

4/14

SiO 2 . A few studies focusedon theuse oforganicpolymer materialsincluding poly(diallyldimethylammonium chloride), polystyrenebased polymers and chitosan. 40-42 More advanced supports suchas membranes and carbonnanobers supportedon sintered metalbres have also been used. 43-44

Catalyst preparation

The general method for the preparation of SILPCs is impregna-tion. The catalyst precursor and the ligand are rst dissolved in asmall amountof a volatile organic solvent ( e.g. , methanol, acetone,acetonitrile, tetrahydrofuran, dichloromethane and chloroform)that dissolves both the catalyst complex and the IL, and then theIL is added.Next, this solution is mixed with the support material,and upon stirring, the volatile solvent is removed by evaporationunder reduced pressure. Alternatively, the homogeneous catalystand the IL are impregnated on a support that already containsa monolayer of covalently anchored IL fragments. In both cases,the homogeneous catalyst is dissolved in the IL lm. The resultingSILPCs are characterized by the nature of the support, thecatalytic metal loading, the molar ligand/metal ratio (L/M),the IL loading and the pore lling degree ( a ). The IL loadingis usually dened as the amount of IL relative to the mass of the support (weight percentage), while the pore lling degree ais dened as the volume of the IL divided by the pore volumeof the support. Ideally, the IL should form a thin layer onthe surface, without completely lling the pores. In general,nitrogen adsorption measurements show that the BET surfacearea, the total pore volume and the mean pore size of thesupport decrease upon adsorption of the IL. Most of the SILPcatalyst systems described in the literature have IL loadings inthe range of 1025 wt.%, while the metal loadings vary between0.010.2 mmol g - 1 .

Reaction conditions

SILP catalysis is ideally suited for gas-phase reactions since noproblems regarding catalyst leaching or washing out of the ILare encountered if the reaction products are not too polar. Inaddition, several ILs show high thermal stability and low vapourpressure, and are therefore excellent media for reactionsat elevatedtemperatures at which conventional solvents would evaporate ordecompose. On the other hand, liquid-phase reactions requirecareful tuning of the polarities of all reaction constituents inorder to prevent leaching of the catalyst or the IL. Applicationof a SILPC in liquid-phase conditions supposes extremely lowsolubility of theIL in the liquid reactant/productmixture. Physicalremoval of the IL lm from the support by mechanical forcesshould also be prevented. The SILPC can be used in solvent-freeconditions, with the reagents and products forming a separateorganic layer. Alternatively, an additional solvent may be used.This solvent contains the reagents and the products and shouldbe immiscible with the IL phase to prevent leaching of the ILfrom the surface of the support. Obviously, the solvent should notdissolve the catalyst to any extent. For this purpose, non-polarsolvents such as alkanes ( e.g. , n-hexane, n-heptane, n-dodecane)are commonly used. Slightly more polar solvents are tolueneand dichloromethane. On the other hand, water, ethanol, or 2-propanol should be used in combination with hydrophobic ILs

(e.g. , containing BF 4 - or (CF 3SO 2)2N - anions), although evenin this case, limited extraction of the IL and the catalyst inthe solvent is difcult to avoid. Supercritical carbon dioxide(scCO 2) has recently been introduced as an organic solvent.Although scCO 2 shows good solubility in some ILs, the reverseis not the case, with no detectable IL solubilization in thescCO 2 phase. scCO 2 reduces the viscosity of the IL and thusimproves the solubility and mass transport of permanent gases.Disadvantagesare the highinvestment and operating costsand thelimited solvating ability of scCO 2 compared with classical organicsolvents.

Selected examples

1. Olen hydroformylation

Rhodium complexes are the most effective catalysts for olenhydroformylation under mild conditions (100 C, 15 bar CO/H 2).Typically, Rh(CO) 2(acac) (acac = acetylacetonate) is used asthe catalyst precursor and a phosphine ligand is added tofavour the formation of linear over branched aldehydes ( n/ iso-aldehyde ratio). Since the heterogenization of Rh catalystsvia classical routes ( e.g. , impregnation or covalent anchoring)has only been achieved with limited success, the Rh com-plex immobilization in supported liquid phases emerged as analternative. 12-13

Liquid-phase hydroformylation in batch reactors. The hydro-formylation of 1-hexene was one of the rst reactions carried outwith a homogeneous catalyst immobilized in a SILP (Scheme 2). 32

Initially, the hydroformylation was performed with Rh-monophosphine complexes (Scheme 3, 1). As support a [1-butyl-3-(3-triethoxysilyl-propyl)-4,5-dihydroimidazolium][BF 4] or [PF 6]modied dehydrated silica gel (474 m 2 g- 1) was used

(Scheme 4).45

Scheme 2 Hydroformylation of 1-hexene.

The resulting SILP catalyst ([Rh]/IL/IL-SiO 2] activity of theSILPC (TOF = 3900 h - 1) was almost three times higher comparedto the biphasic IL system (Table 1). This improvement wasattributed to a higher concentration of the active Rh species at theinterfaceandthegenerally larger interfacearea of thesolidsupportin comparison with the biphasic system. The selectivity ( n/ isoaldehyde ratio = 2.4) was comparable for both systems. Note thatthe homogeneous system is still signicantly more active (Table 1),which is probably due to mass transfer limitations, caused by thelowersolubility of thegases in theIL. Howeverthe IL introductionenables the recycling of the catalytic system. Although the activeRh species are insoluble in the organic medium, at high aldehydeconcentrations [bmim][BF 4] partially dissolved in the organicphase which led to rhodium complex leaching (Rh loss 2 mol%).This was solved by keeping the aldehyde concentration below 50wt.% and increasing the phosphine concentration in the IL. Notethat the Rh leaching could still be minimized.


A

H

8/13/2019 SSIL

5/14

Scheme 3 Classical ligands and ionic liquid tagged ligands in SILP catalysis ( 1 sulfonated monophosphine ligands; 2 ionic liquidtagged Pd complex; 3 monophosphine ligands containing ionic guanidinium or sulfonate groups; 4 [1,1,3,3-tetramethylguanidinium][lactate](TMGL); 5 imidazolium-tagged bis(oxazoline) ligand; 6 (S )-BINAP ligand; 7 triphenylphosphine; 8 the bisphosphine ligand sulfoxantphos; 95,5-isopropylidenebis[(4 R)-4-phenyl-4,5-dihydro-1,3-oxazole]; 10 (S ,S )-Mn( III ) salen chloride; 11 dimeric ( R,R)-Cr( III ) salen complex).

Scheme 4 Covalently anchored imidazolium fragments on silica gel.

Table 1 Comparison of thecatalyticperformance of SILP, homogeneous,

and biphasic catalytic systems in the liquid-phase hydroformylation inbatch reactors

Sa Catalytic system L/Rh b n/ i c TOF d /h - 1 Ref.

C6 SILP/[bmim][BF 4]-tppti-Rh 10 2.4 3900 41C6 Biphasic/[bmim][BF 4]-tppti-Rh 10 2.2 1380 41C6 Homog/PPh 3-Rh 10 2.6 24 000 41C6 Biphasic/TMGL-TPPTS-Rh 5 3.8 82 42C6 SILP/TMGL-TPPTS-Rh/MCM-41 5 3.0 389 42C6 SILP/TMGL-TPPTS-Rh/SiO 2 5 2.1 87 42

a S = substrate, C6 = 1-hexene. b L/Rh = ligand to Rh molar ratio. c n/ i =normal/iso aldehyde ratio. d TOF is dened as mol (aldehyde) per mol(Rh) per hour.

A related Rh-monophosphine based system was pre-pared using a mesoporous MCM-41 silica (1041 m 2 g- 1)as the support and [bmim][BF 4], [bmim][PF 6] or [1,1,3,3-

tetramethylguanidinium][lactate] (TMGL) as ILs (Scheme 3, 4).46

In the hydroformylation of 1-hexene, the MCM-41 catalyst(tppts/Rh = 5) wasmore active than a similarly prepared silica gelcatalyst (Table 1). The highest activity (TOF = 389 h - 1 , n/ iso = 3)was obtained with a 15 wt.% TMGL loading. The n/ iso aldehyderatio increased with increasing TMGL loading and tppts/Rhmolar ratio. This system is nearly 5 times more active than thecorresponding biphasic system (Table 1). The catalyst was reused5 times without a drop in activity or selectivity, and theRh contentin the organic phase was below the AAS detection limit. In thehydroformylation of higher olens, theactivityof theSILPcatalystdecreased with increasing olen chain length.

Gas- and liquid-phase hydroformylation in xed-bed reactors.In a further development, the hydroformylation was studied incontinuous-ow xed-bed reactors. Rh complexes with mono- orbisphosphine ligands were used as the catalysts. The gas-phasepropene and liquid-phase 1-octene hydroformylation have rstbeen carried out using a SILPC comprised of silica (298 m 2 g- 1),[bmim][PF 6] and ionic Rh-monophosphine complexes (0.2 wt.%Rh loading) (Scheme 3, 3). 47 In the continuous-owhydroformyla-tion of propene at120 C, it was observed that catalysts containingthe NORBOS-Cs 3 ligand were signicantly more active than theguanidinium-based analogues (Table 2, entries 3 and 4). However,the SILPc were less active compared to the heterogeneous catalyst(Table 2, entry 1). With respect to the ligand/Rh ratio, the


A

H

8/13/2019 SSIL

6/14

Table 2 Comparison of the catalytic performance of SILP and heterogeneous catalysts in the hydroformylation of propene in xed-bed reactors

Catalytic system a a b L/Rh c n/i d TOF, h - 1 e Rf f

1 Heterog/NORBOSC3-Rh-SiO 2 0 2.9 1.0 114.8 432 SILP/[bmim][PF 6]- NORBOSCs 3-Rh-SiO 2 0.05 11.3 2.0 88.4 433 SILP/[bmim][PF 6]- NORBOSCs 3-Rh-SiO 2 0.78 2.9 0.9 38.0 434 SILP/[bmim][PF 6]-guanidinium-Rh-SiO 2 0.78 2.9 0.9 20.6 435 Heterog/sulfox.-Rh-SiO 2 2.5 1.7 37.4 44

6 SILP/[bmim][PF 6]-sulfox.-Rh-SiO 2 0.49 2.5 2.0 5.1 447 Heterog/sulfox.-Rh-SiO 2 10.2 16.9 40.8 448 SILP/[bmim][PF 6]-sulfox.-Rh-SiO 2 0.08 10.0 23.3 37.0 449 SILP/[bmim][PF 6]-sulfox.-Rh-SiO 2 0.20 20.0 23.7 16.7 4410 SILP/[bmim][PF 6]-sulfox.-Rh-SiO 2 0.1 10.0 18.2 308 48

a Heterog = heterogeneous reaction system (without IL), sulfox = sulfoxantphos ligand. b Pore lling degree of support as the ratio IL volume/supportpore volume. c L/Rh = ligandto Rh molar ratio. d n/ i = normal/iso aldehyde ratio. e TOFis denedas mol(aldehyde) permol (Rh) perhour. f References.

catalyst activity of the SILPs showed an optimum at L/Rh = 11.3(Table 2, entry 2). Furthermore, the selectivity obtained usingboth SILP catalysts was in the usual range obtained for Rh-monophosphine catalysts, i.e. , n/ iso < 2.8 and independent of a .Very low n/ iso ratios and activities were obtained for SILPCs with

L/Rh ratios of 2.9 (entries 3 and 4). This indicates that underthese conditions, the catalytically active complexes are ligand-free complexes interacting with the support. The poor solubilityof CO/H 2 in [bmim][PF 6] suggested that the catalyst sufferedfrom gas mass-transfer limitations. ICP-AES analysis of outletsamples taken at steady state conversion demonstrated negligibleRh leaching ( < 0.7%, below detection limit).

Similarly, Rh-bisphosphine complexes were applied in thecontinuous-ow gas-phasehydroformylationof propeneat 100 C.The IL [bmim][PF 6] or the halogen-free [bmim][ n-octylsulfate]were impregnated on silica gel (298 m 2 g- 1).48 The L/Rh ratioand the pore lling degree of the IL had a strong inuence on thecatalytic performance. Ligand/Rh ratios of 2.5 resulted in rather

low activities (TOF = 5.1 h - 1) and unusually low n/ iso ratiosaround 2 (entry 6). This is in contrast with analogous reactionsperformed with the IL free catalyst where high activities wereobserved even at L/Rh ratios of 2,5 (entry 5). On the other hand,L/Rh ratios of 10 and 20 resulted in much higher activities andn/ iso ratios up to 23.7 (entries 89) ( 96% linearity ). The highestinitial activity was obtained in the absence of IL (entry 7), whichmight be attributed to the formation of surface-bound ligand-free Rh species exhibiting high activities and to the absence of any mass-transport limiting factors. As for the reactions withcatalysts having a low L/Rh ratio, increased IL loading resulted indecreasing activity. This supports the conclusion that the catalystsare operating under mass-transfer limitation due to the poor gas

dissolution rate in ILs.A breakthrough for SILPCs with improved long-term stability

was achieved by thermal pre-treatment of the silica support.These studies contributed signicantly to a better understandingof various aspects of SILP catalysis. The improved catalyst wasprepared by dispersing Rh(CO) 2(acac), sulfoxantphos (Scheme 3,8) and [bmim][ n-C 8H 17OSO 3] on a silica gel (304 m 2 g- 1).

49-51 Theresulting partial dehydroxylation of the surface silanol groupsproved to be crucial for obtaining SILP catalysts with highstability. This was demonstrated by temperature-programmeddesorption of ammonia (NH 3-TPD) and MAS 29Si NMR spec-troscopy. The SILPC was very active and highly selective for the

hydroformylation of propene at 100 C. In situ high-pressure IRspectroscopy showed that the rhodium catalyst indeed behaves asa homogeneous catalyst, i.e. , a complex dissolved in an IL lmon a support. Experiments in the absence of the IL revealed adeactivation of the catalyst within 10 h time-on-stream. Catalysts

with low ligand/Rh loadings (L/Rh = 3 and 5) were initiallyvery active but deactivated rapidly while the selectivity droppedto very low n/ iso ratios. In contrast, with higher ligand loadingsthe initial catalytic performance was retained for at least 60 h(TOF = 44 h- 1 , TON = 2600, n/ iso = 20). FTIR spectroscopyindicated that the deactivation is related to a gradual degradationand removal of the ligand from the catalyst complex. Evidencefor the ligand-support interaction was provided by solid stateMAS 31P NMR, which showed that the support effect is relatedto the irreversible reaction of the ligand with the acidic silanolsurface groups. Consequently, successful catalyst systems requirea relatively large excess of ligand to compensate for the losscaused by surface reactions. The catalyst lifetime exceeded 200 h

time-on-stream, with only a slight decrease in activity. Since theselectivity towards n-butanal remained high at around 97%, it wasconcluded that possibly high-boiling side products ( e.g. , 2-ethyl-hexanal and 2-ethyl-hexanol from consecutive aldol condensationof the formed product) accumulate in the IL lm, thus diluting theeffective Rh concentration. Upon exposure of the SILP catalystto vacuum, these condensation products could be removed andthe initial activity was restored, allowing total lifetimes exceeding700 h. ICP analyses of the exit gas streams showed Rh contentsbelow the 3 ppm detection limit.

The same Rh-sulfoxantphos complex in [bmim][ n-C 8H 17OSO 3]on pre-treated silica gel ( a = 0.1, L/Rh = 10) was used forthe continuous gas-phase hydroformylation of 1-butene. 52 As

previously described for propene hydroformylation, the SILP cat-alyst displayed the typical characteristics of a truly homogeneouscatalyst. At 120 C, the SILPC showed a 2.1-fold higher activitywhen 1-butene (TOF = 647 h - 1) was used instead of propene(TOF = 308 h - 1) (Table 2, entry 10). The selectivity to n-aldehydeswas also higher for 1-butene (98%) compared to propene (95%).Solubility measurements showed that the difference in solubilitybetween propene and 1-butene in silica-supported [bmim][ n-C 8H 17OSO 3] parallels theobserved differences in activity well. Thesame group also carried out the SILP catalyzed hydroformylationof 1-butene in a gradient-free gas-phase reactor. Such Bertytype reactors allowed kinetic studies without artefacts caused by


A

H

8/13/2019 SSIL

7/14

concentration or temperature gradients. The results showed thatat least in this case, there were no mass transport limitations fromthe gas into the IL phase. Furthermore, the limitation by lm orpore diffusion seemed to be negligible. 53

Hydroformylation in supercritical CO 2 . A further technicaladvancement was introduced with the use of scCO 2 as themobile phase. Compared to conventional organic solvents, the

hydroformylation reaction in ILs generally suffers from low gassolubility and slow diffusion of the reactive CO/H 2 gases in theionic medium. A key aspect in this context is to enhance thesolubility of the reactants in the IL and to enhance the masstransfer between the gas phase/organic liquid and the IL phase.Swelling with supercritical CO 2 reduces the ILs viscosity and thusimproves the solubility and mass transport of permanent gases.Importantly, although scCO 2 shows good solubility in some ILs,the reverse is not the case, with no detectable IL solubilization inthe CO 2 phase.

For the hydroformylation of 1-octene, a SILP catalyst wasprepared by supporting Rh(CO) 2(acac), 3-(diphenylphosphino)-benzenesulfonate 1-propyl-3-methyl-imidazolium salt (L/Rh =

10) and [1-octyl-3-methyl-imidazolium][NTf 2] onsilica gel (322m 2g- 1).54 High activities were obtained (TOF up to 800 h - 1) andproved stable for at least 40 h. The corresponding aldehydes wereproduced with an n/ iso ratio of 3. The IL loading had little effecton the reaction rate. This observation suggests that using thesupercritical ow, diffusion is not rate limiting. An additionaladvantage of the supercritical uid is that it prevents catalystfouling by extracting aldol condensation products from the ILlayer. Leaching levels measured by ICP-MS tended to start high(24 ppm) in the rst fractions, but dropped to 0.5 ppm later on.

2. Olen metathesis

A solution of the rst generation Grubbs Ru catalyst and[hmim][PF 6] was impregnated on amorphous alumina. The Ru-SILP catalyst (0.03 mmol Ru g - 1) was effective for multiple olenmetathesis reactions like macrocyclizations and dimerizations. 55

The activity of the SILPC was dependent on the solvent choice:whereas in polar solvents no reaction occurred, the reaction pro-ceeded well in aromatic hydrocarbons. For diethyl diallylmalonatemetathesis in benzene the TON number reached 154 (82% yield).ICP analysis of the hydrocarbon layer revealed no Ru leachingfrom the SILP catalyst. However, upon recycling, a moderateactivity decrease was observed: 82% yield in 1 h upon rst usevs. 70% yield in 2 h after the 5th recycle.

3. Methanol carbonylation

The [bmim][Rh(CO) 2I2] IL was prepared by stirring the dimericcatalyst precursor [Rh(CO) 2I]2 with an excess of [bmim][I] inmethanol. 37 Subsequently, silica gel (304 m 2 g- 1, dried at 500 C)was added to this solution and next the volatile solvent wasevaporated. The SILP catalyst was used for the continuous gas-phase carbonylation of a mixture of methanol and methyl iodide(75 : 25 wt.%) with CO at 180 C. At low gas space velocity,complete conversion of methanol was obtained with a TOFfor thedesired acetyl products (acetic acid andmethyl acetate) of 76.5 h - 1 .The selectivity towards the ester was high (Scheme 5). Li et al.remarked that theprocess designof thexed bedSILPC requires a

Scheme 5 Carbonylation of methanol.

smaller reactor size than the current technology in order to obtain

the same productivity, which makes the SILPC carbonylationconcept potentially interesting for technical applications.

4. Carbonylation of amines and nitrobenzene to ureas

A modied solgel synthesis procedure was used for the in-corporation of [1-alkyl-3-methylimida-zolium][BF 4] and a metal-phosphine complex (Rh, Pd, Ru, or Co) into the pores or cavitiesof a silicamaterial. 56 The silica precursor (tetraethyl orthosilicate),the IL, and the metal complex were mixed with a HCl solutionin aqueous ethanol. After ageing at 60 C, the solid was dried at150 C.N 2 adsorption measurementsafterremoval of [bmim][BF 4]and TEM and XRD analysis showed that the silica gel matrix is

amorphous and non-uniformly mesoporous with pore sizes of 511 nm. Whereas the ethyl- and butyl-substituted ILs could bewashed out in reuxing acetone, ILs with larger molecular sizessuch as decyl- and hexadecyl-substituted ILs were retained morermly. It was proposed that these ILs are retained in the cavities of the silica gel by physical connement or encapsulation. FT-Ramanspectroscopy showed that the connement of the IL resulted inunusual changes in the symmetry and coordination geometry of the ILs. The resulting SILPCs (0.11 wt.% Rh, 35 wt.% IL) weretestedin thecarbonylation of aminesand nitrobenzene to ureas. Inthe carbonylation of aniline andnitrobenzene, thehighest catalyticactivity was observed for the Rh(PPh 3)3Cl catalyst immobilized in[1-decyl-3-methylimidazolium][BF 4] on silica (Scheme 6).

Scheme 6 Synthesis of diphenyl urea.

The enhanced catalytic activity compared to homogeneous ILcarbonylation systems was ascribed to the effect of the high

local concentration of IL containing the metal complex due toits connement in the silica gel matrix. Upon reuse, the SILPCshowed reduced activity and the Rh content of the catalyst wasslightly lower than in the rst reaction. The Rh and Pd SILPCswere also active in the oxidative carbonylation of aniline to methylN -phenyl-carbamate in methanol at 135 C and 50 bar CO/O 2 .

5. Hydrogenation

The catalytic reduction of unsaturated bonds with molecular hy-drogen is an importanttransformation in the industrial productionof ne and bulk chemicals. Typical catalysts are transition-metalcomplexes based on Ru, Rh, Ir, Pd or Pt. Most applications of


A

H

8/13/2019 SSIL

8/14

Table 3 Comparison of the catalytic performance of SILP, homogeneousand biphasic catalytic systems in the hydrogenation of 1-hexene

Entry Reaction system T / C Y (%) a TOF ave /min - 1 b kh c

1 SILP 25 99 447 11.22d Homogeneous 50 29 46 0.43 Biphasic 30 20 4

a Y (%) = yield of the hydrogenated product byproducts are isomerizedolens. b TOF ave is denedas mol(hydrogenatedproduct) permol (Rh) perminute. c kh is the initialrate constantfor the formation of thehydrogenatedproduct. d Homogeneous reaction: acetone is used as solvent.

SILP catalysis in this eld concern the immobilization of Rh-phosphinecomplexes forthe hydrogenationof C = C double bonds.In this respect, the tunability of the solvation properties of ILsmight offer interesting opportunities for selective hydrogenationreactions, e.g. , in the partial hydrogenation of dienes.

The Rh complex [Rh(norbornadiene)(PPh 3)2][PF 6] and[bmim][PF 6] were impregnated on silica gel (289 m 2 g- 1).57 TheIL loading was 25 wt.% and the average IL layer thickness wasestimated at 0.6 nm. The resulting catalyst was used for the hydro-genation of alkenes like 1-hexene, cyclohexene and 2,3-dimethyl-2-butene at room temperature and41 bar of H 2 in n-heptane as thesolvent. In the hydrogenation of 1-hexene, the SILPC showed anenhanced activity in comparison with homogeneous and biphasicIL/heptane reaction systems (Table 3). The enhanced activity of the Rh complex in an IL phase compared to the homogeneoussystem has been observed and might be explained due to theabsence of any coordinating solvent. Therefore, if the catalysisis carried out in an IL medium no coordinating species shouldbe present, in that way leading to easy accessible metal centersand consequently higher reactivities. The improved performance(higher TOF) of the SILP catalytic system with respect to thebiphasic IL system was ascribed to a concentration effect: by usinga high-surface area support containing only a small amount of IL the concentration of the active rhodium species is signicantlyincreased on the interface in comparison with thebiphasic system.The catalyst could be reused 18 times without activity loss. TheRh leaching remained below the detection limit of 33 ppb. SEMand TEM analysis of fresh and spent catalysts indicated no Rhaggregation.

Serp et al. used multiwalled carbon nanotubes (MWCNTs)instead of the commonly used oxide supports as support for theheterogenization of ILs and homogeneous Rh catalysts for thehydrogenation of 1-hexene. 58 CNTs are attractive supports becauseof inherent advantageous properties such as good mechanicalstrength, high chemical stability, and large surface area-to-volumeratio. They discovered that the covalent modication of the CNTsurface with imidazolium based ionic functionalities is necessaryto prevent leaching of the IL lm. Ionophilic ligands are indeedknown to be better retained in the IL phase compared to theconventional ligands. 59 ILs are after all excellent solvents for otherionic species. One method to prepare a highly ionophilic catalystis to confer a charge to the catalyst. However, there are concernsthat charged species react more slowly in ILs than equivalentneutral species, dueto a toostrong binding between the IL and themetal centre. 59 Therefore, both in SILP and biphasic catalysis, theionophilicity of the catalyst has been increased by using an ionic

ligand. This has been achieved by introduction of IL (ammonium,imidazolium) ponytails in the ligands. 60

Theloading of freeowingIL, [bmim][PF 6], on thesefunctional-ized CNTs canbe increasedto 55 wt.% without detectable leachingof the IL, compared to 2030 wt.% on the traditionally employedoxide supports. TheRh/CNT SILPcatalyst exhibitsa signicantlyhigher activity (TOF 2850 h - 1) than SILPCs supported on oxidesupports, including silica (TOF 1400 h - 1) or activated carbon(TOF 950 h - 1), with full retention of the Rh/IL active phaseupon recycling. This higher activity might be partially attributedto a more efcient use of Rh catalyst in the hydrogenation becauseof the good dispersibility of the CNT support in the reactionmixture which minimizes the mass transport effects. 18

The Wilkinson catalyst [Rh(PPh 3)3][Cl] and [bmim][PF 6] weremixed with the organic polymer poly(diallyl-dimethylammoniumchloride). 40 The resulting SILPC (5 mmol Rh g - 1) was used for theselective hydrogenation of 2-cyclohexen-1-one to cyclohexanoneand 1,3-cyclooctadiene to cyclooctene at 30 C and 5 bar of H 2 in diethyl ether as the solvent. The authors claim that theSILPC showed similar activity and selectivity compared to thehomogeneous reaction system in CH 2Cl 2 and the activity washigher compared to a conventional biphasic IL/ether system,whichis probably dueto improved mass transfer between theSILPand the substrate containing solvent phase. However it should benoted that upon water addition to the homogeneous reaction thereaction rate increased signicantly (TOF = 15 h - 1), compared toa TOF of 9 h - 1 for the SILP system. The introduction of the IL isnevertheless advantageous because of the reusability of the SILPCwithout loss of activity. AASspectroscopy showedno tracesof Rhin the organic layer.

Sintered metal bers (SMFs) were used as structured supportsfor the [bmim][BF 4] IL containing the homogeneous catalyst pre-cursor [Rh(norbornadiene)Cl] 2, triphenylphosphine (Scheme 3, 7)and HBF 4 .44 The SMFs were rst coated with a layer of carbonnanobers (CNF) or zeolite ZSM-5 to increase the specic surfacearea of the support and to favour a homogeneous coverage by theIL. The coating of a thin lm of IL on this support ensured a fulluse of the catalyst without mass transfer limitations. The resultingSILP system(0.06 wt.% Rh,12 wt.% IL)was used for thegas-phasehydrogenation of 1,3-cyclohexadiene to cyclohexene at 60 C ina xed-bed tubular reactor. The catalysts were only active in thepresence of acid ( e.g. , HBF 4 or H 3PO 4) and the optimal molarratio acid/IL was 0.5. In the absence of acid, the catalyst rapidlydecomposed, thereby yielding metallic Rh. Based on 1H NMRspectroscopy, Rh(H) 2Cl(PPh 3)3 was proposed as the catalyticallyactive species. Theamount of PPh 3 wasalso important; a PPh 3/Rhratio of 8 resulted in a high activity (TOF up to 285 h - 1), a veryhigh selectivity ( > 96%) towards cyclohexene and an acceptablestability during 6 h on stream.

6. Hydroamination

Supported hydroamination catalysts were prepared byimpregnation of the metal complexes [Rh(1,1 -bis(diphenyl-phosphino)ferrocene)(2,5-norbornadiene)][ClO 4], [Pd(dppf)][(CF 3SO 3)2], [Cu 2(C 6H 5CH 3)][(CF 3SO 3)2] or Zn(CF 3SO 3)2 and theIL [1-ethyl-3-methylimidazolium] [CF 3SO 3] on a diatomaceousearth support. 61 Before, the diatomaceous earth (10 m 2 g- 1) waswashed with acid and silylated with Me 2SiCl 2 . The resulting


A

H

8/13/2019 SSIL

9/14

catalysts were tested in the addition of 4-isopropylaniline tophenylacetylene in n-heptane as the co-solvent. The reactionyields the enamine which in situ isomerizes to the correspondingimine (Scheme 7). At 98 C, the highest initial activity wasobserved for the supported palladium catalyst (TOF = 4.3 h - 1 , 0.1wt.% Pd, 18.8 wt.% IL). The other supported metal complexesshowed much lower activities. Details on the extent of the metalleaching or the recyclability of the SILPCs were not reported.

Scheme 7 Addition of 4-isopropylaniline to phenylacetylene.

The Lewis acid [Pd(dppf)][(CF 3CO 2)2] and the Brnstedacid CF 3SO 3H were immobilized in a lm of [1-alkyl-3-methylimidazolium][CF 3SO 3] (alkyl = ethyl, butyl, hexyl) on silicagel (150 m 2 g- 1, dried in vacuo at 200 C). 62 MAS 1H NMRmeasurements showed a large increase of the line width whenthe Pd complex and CF 3SO 3H were immobilized in the supportedILs. This suggests a substantial decrease in the mobility of theimidazolium cations. It was proposed that the IL forms a solvent

cage around the Pd complexes, thereby establishing a long-rangeordered system. The solvent cage around one metal complexwould consist of up to 2533 IL ion pairs. The resulting aggre-gates arrange into a regular packing with a glass-like structure.Temperature-resolved MAS 1H NMR measurements showed aphase transition from glassy to liquid state at ca. 75 C. Thisindicates that the solvent cages break down and the moleculesacquire similar mobility as in the parent supported IL, whichdoes not contain the dissolved complex. This bifunctional catalystsystem (H + /Pd 2+ = 10, 0.042 mmol Pd g - 1) was used for thehydroamination of styrene with aniline in n-octane as the solvent(Scheme 8). The addition reaction can yield the Markownikoff product N -(1-phenylethyl)aniline or the anti -Markownikoff prod-

uct N -(2-phenylethyl)aniline.

Scheme 8 Addition of aniline to styrene.

In batch experiments performed at 125 C, the Markownikoff addition product was the only product apart from some styrene

oligomerization products. The highest activity was observed forthe SILPCs containing more polar 1-alkyl-3-methylimidazoliumbased ILs (ethyl > butyl > hexyl). The initial TOF for the mostactive material was 33.3 h - 1. The same activity was observedunder biphasic IL/heptane conditions, suggesting the absenceof diffusion limitation across the liquidliquid phase-boundary.The catalytic activity increased with increasing IL amount. Theamount of Pd leached into the reaction solution was below1 mmol L - 1 . The temperature dependence was explored in axed-bed reactor. Aniline conversion started at 150 C and thecatalytic activity increased exponentially with temperature toreach a maximum at approximately 240 C. At this temperature,the steady-state conversion of 65% corresponds with a TOF of 200 h - 1 . In the kinetic regime ( T < 240 C), the selectivity to N -(1-phenylethyl)aniline was 100%below 180 C anddecreased steadilyat higher temperatures. Above 250 C, the conversion decreasedas the thermodynamic limit of the reaction was encountered. Inthe thermodynamic regime ( T > 240 C), the selectivity to N -(1-phenylethyl)aniline (44%) was the same for all SILP Cs and nearlytemperature independent.

7. Pd-catalyzed coupling reactions

Palladium is an excellent catalyst for various selective CCcoupling reactions. It is not always clear whether the reactionis catalyzed by molecularly dened palladium species or by in situformed palladium nanoparticles. In some cases, these nanoparti-cles are stabilized in ILs and the resulting catalyst systems haveshown excellent performance. 28 On the other hand, if palladium-catalyzed reactions are carried out in 1,3-dialkylimidazoliumbased ILs as the solvent, palladium might react with the imi-dazolium cation forming a metal carbene complex. 26 This is dueto the relatively high acidity of the H atom in the 2-position of a1,3-dialkylimidazolium ion. The latter can be deprotonated ( e.g. ,

by basic ligands or by basic reactants) to form a metal-carbenecomplex. In these cases, the IL acts a ligand precursor.

CC coupling reactions. Pd(OAc) 2 and [bmim][PF 6] wereimpregnated on amorphous silica (600700 m 2 g- 1). The SILPcatalyst (0.25 mmol Pd g - 1) was used for the MizorokiHeckreaction between iodo-benzene and cyclohexyl acrylate with tri-n-butylamine (2 equiv.) as the base. 63 In order to prevent removalof the IL layer from the silica surface, the reaction was carriedout in n-dodecane as the co-solvent. At 150 C, the TON andTOF numbers reached 68 400 and 8000 h - 1 respectively (96%yield). Upon reuse of the catalyst, the catalytic activity slightlydecreased after two recycles. This might arise from fouling of thesilica surface with the formed n-Bu 3NHI, which is insoluble inn-dodecane. This problem could be solved by washing the catalystwith dilute aqueous NaOH after each run. ICP analysis of thehydrocarbon layer showed that 0.24% of the Pd dissolved fromthe SILPC.

In a related study, Pd(OAc) 2 and [bmim][PF 6] were immo-bilized on reversed-phase silica gels such as aminopropylatedor N ,N -diethylaminopropylated silica. 64 Using the N ,N -diethyl-aminopropylated silica gel, the MizorokiHeck reaction betweeniodobenzene and cyclohexyl acrylate was carried out in water asthesolvent andtri- n-butylamineas thebase (Scheme 9).At 100 C,the reaction was faster than in n-dodecane as the solvent and theuse of water eliminated the need for a washing step to remove the


A

H

8/13/2019 SSIL

10/14

Scheme 9 MizorokiHeck reaction of iodobenzene and cyclohexylacrylate.

ammonium iodide from the surface. The catalyst could be reused

5 times without loss of catalytic activity. Pd leaching into theaqueous layer was 1.1 ppm after the rst run (ICP). Normal phasesilica resulted in the removal of the IL from the silica surface andleaching of palladium. At low Pd loadings (0.043 mmol g - 1), TONand TOF numbers reached 1 600 000 and 71000 h - 1 respectively.

A similar catalyst system was used for the SuzukiMiyauracoupling of aryl halides with arylboronic acids. Pd(OAc) 2 and[bmim][PF 6] were impregnated on N ,N -diethylaminopro-pylatedalumina. 65 Alumina was preferred over silica owing to its higherstability in aqueous basic reaction conditions.

The coupling reaction between p-bromoacetophenone andphenylboronic acidproceeded veryfast in thepresenceofK 2CO 3 (2equiv.) at room temperature in 50% aqueous ethanol as thesolvent

(Scheme 10). The catalyst was reused up to ve times with 95%average yield, though the catalytic activity gradually decreasedupon recycling. At low Pd loadings (0.020.025 mmol g - 1),TON and TOF numbers amounted to 2 000000 and 30 000 h - 1

respectively. At higher loadings (0.250.3 mmol g - 1), Pd leachingwas observed.

Scheme 10 SuzukiMiyaura coupling of p-bromoacetophenone andphenylboronic acid.

An IL-tagged Pd( II ) complex (Scheme 3, 2) and [bmim][PF 6]were immobilized on silica or on a polymeric imidazolium salt. Asdescribed earlier for the hydrogenation reactions, the introductionof a more ionophilic ligand is expected to induce here aswell a positive effect on the Pd leaching from the SILPC. 60

Alternatively, the polyvinyl imidazolium salt was prepared insidethe megapores of Raschig rings. 66 The immobilized catalysts (26mmol Pd g - 1 silica or 5 mmol Pd g - 1 ring) were used for theMizorokiHeck coupling reaction between p-iodoacetophenoneand tert -butylacrylate in toluene as the co-solvent. The reactionwas carried out at150 C using triethylamine (2 equiv.) as the base.Although Pd leaching was about 5% after the rst run (ICP-MSanalysis), the catalysts could be reused up to 10 times withoutloss of activity. STEM measurements showed the presence of Pdnanoparticles that remain on the polymeric surface of the spentcatalysts. Strictly speaking this catalyst is not a pure SILP catalyst,since Pd nanoparticles are deposited on the support during thereaction.

The Pd( II ) complexes in the above mentioned Heck and Suzukicoupling reactions are expected to serve as a dormant species

(pre-catalysts) that are not involved in the real catalytic cycle butare a source of colloidal Pd(0) which also can show catalytic ac-tivity at low concentrations. It was suggested that the leached andsoluble palladium species do not followthe proposedPd( II )/Pd( IV)catalytic cycle.This has severe consequences regarding the claimedrecyclability of these systems: the immobilized Pd( II ) complexesare reusable catalyst if, and only if, the in situ produced Pd(0)species remain wholly with the support and do not freely leachinto solution. 67

Allylic substitution reactions. The biopolymer chitosan wasused as a support for Pd(OAc) 2 , tppts and [bmim][BF 4] or[bmim][PF 6].42 Chitosan consists of D-glucosamine units witha small amount of N -acetyl- D -glucosamine units. The resultingcatalyst was used for the allylic substitution between morpholineand ( E )-1,3-diphenyl-3-acetoxyprop-1-ene at room temperature(Scheme 11). After reaction, the product was separated from theIL phase by extraction with diethyl ether.

Scheme 11 Allylic substitution of ( E )-1,3-diphenyl-3-acetoxyprop-1-eneby morpholine.

TheSILPCsystemcouldbe recycled andreused at least 10 timeswithout any decrease in activity. Except for the rst cycle, in whichthe amount of palladium leached was 9%, the detected Pd levelsby ICP-MS in the ether phase after product extraction were verylow.

8. Asymmetric reactions

Asymmetricmetal complexes are oftenusedas molecularcatalyststo introduce chirality in the synthesis of enantiopure organiccompounds like pharmaceuticals or agrochemicals. In particular,transition-metal complexes with chiral P- or N-donor ligandsprovide high enantioselectivities. Classical strategies for the im-mobilization of chiral metal complexes include cross-linking with(in)organic groups and covalent attachment to solid supports.However, these methods require chemical modication of theligand and/or the support and the obtained enantioselectivityand activity values are not seldom inferior compared to theirhomogeneous analogues. These limitations might be overcome

by using a supported IL as the solvent phase for dissolving thechiral catalyst. In that case, even commercially available ligandsand catalysts can be used.

Asymmetric hydrogenation. For the asymmetric b-ketoester hydrogenation a SILPC was prepared by mixingthe chiral [Ru( S -BINAP)][Cl] ( S -BINAP = (S )-(-)-2,2 -bis(diphenylphosphino)-1,1 -binaphthalene) (Scheme 3, 6),[bmim][PF 6] and poly(diallyldimethylammonium chloride).The resulting catalyst (10 mmol Ru g - 1) was applied in theasymmetric hydrogenation of methyl acetoacetate to methylhydroxybutyrate in 2-propanol as the co-solvent (Scheme 12). TheSILPC was less active compared to the homogeneous catalyst in


A

H

8/13/2019 SSIL

11/14

Scheme 12 Asymmetric hydrogenation of methyl acetoacetate.

2-propanol but showed higher activity compared to homogeneousbiphasic conditions in IL/2-propanol. Importantly, the highenantioselectivity of the homogeneous catalyst was maintainedin the SILP system. The supported Ru catalyst was reused oncewithout loss of activity and no traces of Ru were observed in theorganic layer.

The asymmetric acetophenone hydrogenation was studiedusing SILPCs containing Ru- or Rh-BINAP complexes andphosphonium-based ILs supported on a modied silica gel. 68 Asthereactionrequires basic pHconditions, thesilica beads were rstimpregnatedwithK 2CO 3 or K 3PO 4 . Phosphonium-based ILs wereused because of their stability at high pH. In a typical catalyst syn-thesis, the Rh precursor [Rh(1,5-cyclooctadiene) 2][CF 3SO 3] wasrst stirred with ( R)-BINAP in CH 2Cl 2. Next, tetradecyl(trihexyl)phosphonium decanoate and the base-promoted silica gel (141 m 2

g- 1) were added and the volatile solvent was removed in vacuo.The asymmetric hydrogenation of acetophenone was carried

out at 50 C and 50 bar of H 2 in n-hexane as the solvent. Theactivity of the supported Rh catalysts was comparable to thehomogeneous analogues. Ru catalysts showed high initial activityand moderate enantioselectivity, but deactivated rapidly. At 90%conversion, the above-mentioned supported Rh catalyst (11 mmolRh g - 1, 12 wt.% ionic liquid) showed a selectivity towards 1-phenylethanol of 70% and an enantiomeric excess of 30%. Themain side reaction was the phenyl ring saturation, leading to1-cyclohexylethanone and 1-cyclohexylethanol. A ltration testproved that no catalytically active Rh species leached to the hexanephase. In methanol as the solvent, the homogeneous Rh catalystswere highly selective for the reduction of the carbonyl group butshowed no stereoselectivity. Apparently, the interactions betweenacetophenone and the chiral phosphine ligand are enhanced inthe SILP. As already proposed in recent reports on imidazoliumsalts, 62 this is attributed to the formation of solvent cages of ILmolecules around the substrate-catalyst adducts. The formationof these solvent cages was indicated by a strong broadening of theMAS 1H and 31P NMR signals of the supported ILs. Within therestricted space of the solvent cage, the preferred binding modeof acetophenone to the metal centre would be different from thatin classical solvents such as methanol. A model was proposed inwhich acetophenone preferentially binds to the metal complex viasimultaneous h2/ h 2-coordination of both the carbonyl and thephenyl group. Improved stereoselection is obtained, as the h 2/ h 2-coordinatedacetophenone has to t tightly into thepocket formedby the 4 phenyl groups of the chiral BINAP ligand.

Recently, the chiral Ru complex RuCl 2(PPh 3)2(S,S)-1,2-diphenylethylenediamine) was synthesized and immobilizedin [bmim][BF 4] conned on the surface of 1-methyl-3-(3-triethoxysilylpropyl)imidazolium tetrauoroborate modiedmesoporous materials MCM-48, MCM-41, SBA-15 and amor-phous SiO 2 (0.6 mmol IL g - 1;0.2 mmol Ru g - 1). In 2-propanol, theprepared heterogeneous catalysts showed high activity and enan-tioselectivity in the asymmetric hydrogenation of acetophenone. 69

All 4 catalysts could be used 4 times without observable decreasein conversions and ee values, while the SiO 2-based catalyst still

maintained high conversion and ee values even in the fthrun. This might be due to the larger pore size and complexstructure in SiO 2 particles, which is moreeffective to avoid channelblockage in the recycling experiments. The total turnover number(TON) of catalyst IL/1-SiO 2/a (in which the chiral Ru complexa and the [bmim][BF 4] IL were absorbed onto 1-methyl-3-(3-triethoxysilylpropyl)imidazolium tetrauoroborate (1) modiedSiO 2) after ve catalytic runs ( ca. 1000) was much higher thanthat of the homogeneous catalyst (200). To investigate the possiblesynergetic interaction between the two kinds of ILs, the catalystIL/M48/a (inwhich thechiral Ru complex a and the [bmim][BF4]IL were directly adsorbed onto unmodied MCM-48) and 1-M48/a (without adsorbed ionic liquid [bmim][BF4]; imidazoliumbased IL 1 modied M48) were also evaluated. Both catalystsshowed comparable catalytic activity ( > 99% conversion) to thatof IL/1-M48/a, however, they presented a much lower conversion(< 10%) when reused for the second run. This indicated thatthe synergistic interaction between the two kinds of ILs in theheterogeneous catalyst can signicantly enhance the stability of active species.

Asymmetric aldol reactions. The organocatalyst L-prolinehas been supported on silica gels modied with a monolayerof covalently anchored ILs, with or without additionally ad-sorbed IL. 70 An active and selective SILP catalyst was ob-tained by the impregnation of L-proline and [bmim][BF 4] or[1-butyl-4-methylpyridinium] [BF 4] on silica gel (453 m 2 g- 1)which was previously modied with [1-(3-trimethoxysilyl-propyl)-2,3-dimethylimidazolium][BF 4] or [1-(3-trimethoxy-silylpropyl)-4-methylpyridinium][BF 4] respectively. The BET surface area, theBJH average pore diameter and the cumulative pore volume of the materials decreased slightly upon covalent attachment of IL fragments and a further steep decrease of these parameterswas observed upon impregnation with additional IL. In theasymmetric aldol reaction between acetone and benzaldehyde, theyield and enantioselectivity obtained with the modied silica gelwith additional IL were comparable with those obtained underhomogeneous conditions in DMSO as the solvent. The physicallyadsorbed IL didnot haveany effecton thee.e.values. Impregnationof L-proline and IL on silica without covalently anchored ILfragments resulted in decreased yields and lowenantioselectivities.The imidazolium-based SILPCs were recycled three times in thealdol condensation between p-nitrobenzaldehyde and acetonewithlimitedactivityor enantioselectivity losses.However, starting fromthe fth run, lower yields were observed. According to NMRmeasurements, no proline leaching was observed. Regeneration of the catalytic materials was possible by methanol washing, drying,and recharging with L-proline.

For the asymmetric Mukaiyama aldol condensation, a SILPCwas prepared by impregnation of Cu(OTf) 2 , an imidazolium-tagged bis(oxazoline)ligand (Scheme3, 5) and [emim][NTf 2]onsil-ica(20m 2 g- 1).71 Alternatively, Cu, ligand and IL were impregnatedon a support that already contained a monolayer of covalentlyanchored imidazolium fragments. The resulting catalysts wereapplied in the asymmetric Mukaiyama aldol condensation of 1-phenyl-1-trimethylsiloxyethene and methyl pyruvate in diethylether as the co-solvent (Scheme 13).

The SILPC systems were less active than the homogeneouscatalyst in IL but at least as active as the homogeneous analogue


A

H

8/13/2019 SSIL

12/14

Scheme 13 Asymmetric Mukaiyama aldol condensation between1-phenyl-1-trimethylsiloxyethene and methyl pyruvate.

in CH 2Cl 2 . The unmodied silica impregnated with [emim][NTf 2]was efciently recycled and gave stable conversions and enan-tioselectivities over 5 reaction runs. Ligand leaching was belowthe detection limit. Analysis of the solvent phase after extractionshowed that IL leaching into the ether layer was below the 1 ppmdetection limit of suppressed ion chromatography. The activitydrop after the fth run was ascribed to aggregation of silicaparticles.

Asymmetric ring opening of epoxides. A dimeric ( R,R)-Cr( III )salen complex (Scheme 3, 11 ) and [bmim][PF 6] were impregnatedon silica and theresulting SILPC wasused for theasymmetric ringopening of epoxides, in batch and xed-bed reactors. 72 The SILPCshowed enhanced activity and selectivity compared to a related

supported catalyst prepared by conventional impregnation in theabsence of an IL.

Under continuous-ow operation, the asymmetric ring openingof 1,2-epoxyhexane with TMSN 3 in n-hexane resulted in 51%conversion and an e.e. valueof theepoxideof 94%. ThecumulativeTON over three runs amounted to 314 and the overall leachingwas only 0.7% (Scheme 14).

Scheme 14 Asymmetric ring opening of epoxides.

Asymmetric epoxidation of olens. The IL fragment [1-(3-trimethoxysilylpropyl)-3-methyl-imidazolium][PF 6] was cova-lently anchored to mesoporous MCM-48 silica and the IL-modied material was impregnated with additional [bmim][PF 6]and a chiral substituted ( S ,S )-Mn( III ) salen chloride complex(Scheme 3, 10).73 The immobilized catalyst (0.2 mmol Mn g - 1) wasused for the asymmetric epoxidation of unfunctionalized olensin CH 2Cl 2 as the solvent using a mixture of m-chloroperbenzoicacid and N -methylmorpholine- N -oxide as the oxidant at0 C. For

olens like 1-phenyl-1-cyclohexene, the SILPC showed higher e.e.values compared to the homogeneous catalyst. The immobilizedcatalyst could be reused 3 times without loss of activity orselectivity. NMRand ICP-AES measurements showed no leachingof IL or Mn-salen complex.

Asymmetric cyclopropanation of olens. The asymmetriccyclopropanation of olens with ethyl diazoacetate wascarried out using a SILPC prepared by impregnationof CuCl, 5,5 -isopropylidenebis[(4 R)-4-phenyl-4,5-dihydro-1,3-oxazole] (Scheme 3, 9) and [bmim][PF 6] on a laponite clay(370 m 2 g- 1) (Scheme 15). 74 Compared to homogeneous reactionconditions, the SILPCs showed much lower yields. However, the

Scheme 15 Asymmetric cyclopropanation of styrene with ethyldiazoacetate.

supported catalyst showed interesting variations in the cis/ trans -and enantioselectivities. As an example, the cyclopropanation of styrene changed from a preference for the (1 S ,2S )-trans isomer inbulk solution to the (1 R,2S )-cis isomer in the supported IL lm.These variations were strongly dependent on the thickness of theIL lm and the supports nature. A decrease of the IL loading ledto a decrease in yield and cis/ trans selectivity. Remarkably, uponreuse the yield increased and the enantioselectivity increased fortrans -isomers, whereas for the cis-isomers, the enantioselectivitytended to zero or was even reversed, albeit with low values. Only

layered solids with negative layer charges ( e.g. , clays) displayedthis behaviour. It was proposed that the close proximity to thesurface of the support results in a restricted rotational mobility of the bis(oxazoline)-copper complexes.

Conclusions

Supported ionic liquid phase (SILP) catalysis combines the ad-vantages of homogeneous and heterogeneous catalysis. Amongstthe most attractive features are the high activity and selectivity,the ease of product separation, and the efcient catalyst recy-cling. Compared to biphasic reaction conditions, SILP catalysisallows a more efcient utilization of the ionic liquid and thecatalyst. Summarizing, these properties make this technologyvery attractive for future applications. However, as the eldof SILP catalysis has only developed relatively recently, somelimitations remain and signicant progress is still to be expected.For instance, whereas application of SILP catalysis in gas-phaseprocesses is quite straightforward, liquid-phase reactions requiretuning of all reaction constituents polarities in order to preventleaching of the catalyst or the ionic liquid. It is generally acceptedthat ionophilic ligands, typically prepared by introduction of IL(ammonium, imidazolium) ponytails in the ligands, are superiorto the conventional ligands, concerning catalyst leaching. This isparticularly relevant for reactions in which highly polar reactionproducts are formed. In that case, limited extraction of the ionicliquid and the catalyst into the polar product phase seems difcultto avoid. Furthermore, a more fundamental understanding of various aspects of SILP catalysis is lacking, for example thesupports inuence on activity and selectivity of the catalyst,the precise interaction of the ionic liquid with the support, andthe issue of mass transfer in supported ionic liquid phases. Inaddition, the nature of the active species in SILP catalysis shouldbe studied in more detail. Therefore, detailed mechanistic studiesneed to be performed to investigate SILP catalysis for specictransformations. Additionally, the basic kinetics of SILP catalysisshould be studied more thoroughly. Finally, prior to industrial use,the performance of SILP catalysts must be superior to that of the


A

H

8/13/2019 SSIL

13/14

existing catalytic systems, both with respect to activity, selectivityand recyclability.

Acknowledgements

Financial support from the K.U. Leuven (project IDO/05/005CECAT grant and CASAS Methusalem grant) and from the IWT(SBO-project Materials processing in ionic liquids and PhDfellowship of CVD) is acknowledged.

Notes and references

1 Ionic liquids in synthesis , ed. P. Wasserscheid and T. Welton, Wiley-VCH, Weinheim, 2008, vol. 1.

2 H. Weing artner, Angew. Chem., Int. Ed. , 2008, 47, 654.3 P. Wasserscheid and W. Keim, Angew. Chem., Int. Ed. , 2000, 39, 3772.4 V. Plechkova and K. R. Seddon, Chem. Soc. Rev. , 2008, 37, 123.5 C. M. Gordon, Appl. Catal. A: General , 2001, 222 , 101.6 H. Olivier-Bourbigou and L. Magna, J. Mol. Catal. A , 2002, 182 , 419.7 D. Zhao, M. Wu, Y. Kou and E. Min, Catal. Today , 2002, 74, 157.8 J. Dupont, R. F. de Souza and P. A. Z. Suarez, Chem. Rev. , 2002, 102 ,

3667.9 T. Welton, Coord. Chem. Rev. , 2004, 248 , 2459.

10 V. I. P arvulescu and C. Hardacre, Chem. Rev. , 2007, 107 , 2615.11 P. Wasserscheid and P. Schulz, in Ionic liquids in synthesis , ed.

P. Wasserscheid and T. Welton, Wiley-VCH, Weinheim, 2008, vol. 2, p.369.

12 J. P. Arhancet, M. E. Davis, J. S. Merola and B. E. Hanson, Nature ,1989, 339 , 454.

13 P. Tundo and A. Perosa, Chem. Soc. Rev. , 2007, 36, 532.14 C. P. Mehnert, Chem.Eur. J. , 2005, 11 , 50.15 A. Riisager, R. Fehrmann, M. Haumann and P. Wasserscheid, Top.

Catal. , 2006, 40, 91.16 A. Riisager and R. Fehrmann, in Ionic liquids in synthesis , ed.

P. Wasserscheid and T. Welton, Wiley-VCH, Weinheim, 2008, vol. 2, p.527.

17 A.Riisager, R.Fehrmann,M.Haumannand P. Wasserscheid, Eur. J. In-org. Chem. , 2006, 695.

18 Y. Gua and G. Li, Adv. Synth. Catal. , 2009, 351 , 817.19 A. Riisager, R. Fehrmann, M. Haumann and P. Wasserscheid, Top.

Catal. , 2006, 40, 91.20 C. P. Mehnert, Chem.Eur. J. , 2005, 11 , 50.21 M. H. Valkenberg, C. de Castro and W. F. H olderich, Green Chem. ,

2002, 4, 88.22 (a) K. Yamaguchi, C. Yoshida, S. Uchida and N. Mizuno, J. Am.

Chem. Soc. , 2005, 127, 530; (b) J. Kasai, Y. Nakagawa, S. Uchida,K. Yamaguchi and N. Mizuno, Chem.Eur. J. , 2006, 12, 4176.

23 R. Ciriminna,P. Hesemann,J. J. E.Moreau,M. Carraro, S.Campestriniand M. Pagliaro, Chem.Eur. J. , 2006, 12, 5220.

24 (a) T. Sasaki, C. Zhong, M. Tada and Y. Iwasawa, Chem. Commun. ,2005, 2506; ( b) C. Zhong, T. Sasaki, M. Tada and Y. Iwasawa, J. Catal. ,2006, 242 , 357.

25 W. Chen, Y. Zhang, L. Zhu, J. Lan, R. Xie and J. You, J. Am. Chem.Soc. , 2007, 129 , 13879.

26 B. Karimi and D. Enders, Org. Lett. , 2006, 8, 1237.27 Y. Gu, C. Ogawa, J. Kobayashi, Y. Mori and S. Kobayashi, Angew.Chem., Int. Ed. , 2006, 45, 7217.

28 J. Huang, T. Jiang, H. Gao, B. Han, Z. Liu, W. Wu, Y. Chang and G.Zhao, Angew. Chem., Int. Ed. , 2004, 43, 1397.

29 S.Mao, Z. Liu, B.Han,J.Huang, Z. Sun, J. Zhangand T. Jiang, Angew.Chem., Int. Ed. , 2006, 45, 266.

30 (a) J.-P. Mikkola, P. Virtanen, H. Karhu, T. Salmi and D. Y. Murzin,Green Chem. , 2006, 8, 197; (b) J.-P. T. Mikkola, P. P. Virtanen, K.Kord as, H. Karhu and T. O. Salmi, Appl. Catal. A , 2007, 328 , 68; (c) P.Virtanen, H. Karhu, K. Kordas and J.-P. Mikkola, Chem. Eng. Sci. ,2007, 62, 3660.

31 U. Kernchen, B. Etzold, W. Korth and A. Jess, Chem. Eng. Technol. ,2007, 30, 985.

32 P. Migowski and J. Dupont, Chem.Eur. J. , 2007, 13, 32.33 F. van Rantwijk and R. A. Sheldon, Chem. Rev. , 2007, 107 , 2757.

34 P. Wasserscheid, R. V. Hal, A. B osmann, J. Eer and A. Jess, in IonicLiquids as Green Solvents:Progress andProspects , ed. R.D. Rogers andK. R. Seddon, American Chemical Society, 2003, ch. 5, pp. 57.

35 Y. Yang, H. Lin, C. Deng, J. She and Y. Yuan, Chem. Lett. , 2005, 34,220.

36 A. Corma, H. Garc a and A. Leyva, Tetrahedron , 2004, 60,8553.

37 A. Riisager, B. Jrgensen, P. Wasserscheid and R. Fehrmann, Chem.Commun. , 2006, 994.

38 L.-L. Lou, X. Peng, K. Yu and S. Liu, Catal. Commun. , 2008, 9, 18911893.

39 G. Roger, Chem. Eng. Techn. , 2006, 7, 869.40 A. Wolfson, I. F. J. Vankelecom and P. A. Jacobs, Tetrahedron Lett. ,

2003, 44, 1195.41 D. W. Kim and D. Y. Chi, Angew. Chem., Int. Ed. , 2004, 43,

483.42 J. Baudoux, K. Perrigaud, P.-J. Madec, A.-C. Gaumont and I. Dez,

Green Chem. , 2007, 9, 1346.43 E. Miyako, T. Maryyama, N. Kamiya and M. Goto, Biotechnol. Lett. ,

2003, 25, 805.44 M. Ruta, I. Yuranov, P. J. Dyson, G. Laurenczy and L. Kiwi-Minsker,

J. Catal. , 2007, 247 , 269.45 C. P. Mehnert, R. A. Cook, N. C. Dispenziere and M. Afeworki, J. Am.

Chem. Soc. , 2002, 124 , 12932.46 (a) Y. Yang, H. Q. Lin, C. X. Deng, J. R. She and Y. Z. Yuan, Chem.

Lett. , 2005, 34, 220; (b) Y. Yang, C. Deng and Y. Yuan, J. Catal. , 2005,

232 , 108.47 A. Riisager, K. M. Eriksen, P. Wasserscheid and R. Fehrmann, Catal.Lett. , 2003, 90, 149.

48 A. Riisager, P. Wasserscheid, R. van Hal and R. Fehrmann, J. Catal. ,2003, 219 , 452.

49 A. Riisager, R. Fehrmann, S. Flicker, R. van Hal, M. Haumann and P.Wasserscheid, Angew. Chem., Int. Ed. , 2005, 44, 815.

50 A. Riisager, R. Fehrmann, M. Haumann, B. S. K. Gorle and P.Wasserscheid, Ind. Eng. Chem. Res. , 2005, 44, 9853.

51 A.Riisager, R.Fehrmann, M.Haumannand P. Wasserscheid, Eur. J. In-org. Chem. , 2006, 695.

52 M. Haumann, K. Dentler, J. Joni, A. Riisager and P. Wasserscheid,Adv. Synth. Catal. , 2007, 349 , 425.

53 M. Haumann, M. Jakuttis, S. Werner and P. Wasserscheid, J. Catal. ,2009, 263 , 321.

54 U. Hintermair, G. Zhao, C. C. Santini, M. J. Muldoon and D. J. Cole-Hamilton, Chem. Commun. , 2007, 1462.

55 H.Hagiwara,N. Okunaka, T. Hoshiand T. Suzuki, Synlett , 2008, 1813.56 (a) F. Shi, Q. Zhang, Y. Gu and Y. Deng, Adv. Synth. Catal. , 2005, 347 ,

225; (b) F. Shi, Q. Zhang, D. Li and Y. Deng, Chem.Eur. J. , 2005, 11 ,5279.

57 C. P. Mehnert, E. J. Mozeleski and R. A. Cook, Chem. Commun. , 2002,3010.

58 L. Rodrguez-Perez, E. Teuma, A. Falqui, M. Gomez and P. Serp,Chem. Commun. , 2008, 4201.

59 T. Welton, Coord. Chem. Rev. , 2004, 248 , 2459.60 (a) A. Corma, H. Garcia and A. Leyva, Tetrahedron , 2004, 60 , 8553;

(b) C. S. Consorti, G. L. P. Aydos, G. Ebeling and J. Dupont, Org.Lett. , 2008, 10, 237; ( c) F. Favre, H. Olivier-Bourbigou, D. Commereucand L. Saussine, Chem. Commun. , 2001, 1360; ( d ) K. W. Kottsieper, O.Stelzer and P. Wasserscheid, J. Mol. Catal. A: Chem. , 2001, 175 , 285;(e) N.Audic, H. Clavier, M. Mauduit andJ. C. Guillemin, J. Am. Chem.Soc. , 2003, 125 , 9248; ( f ) J. J. Tindale and P. J. Ragogna, Can. J. Chem. ,

2010, 88, 27; ( g) R. Sebesta, I. Kmentova and S. Toma, Green Chem. ,2008, 10, 484.61 S. Breitenlechner, M. Fleck,T. E. M uller and A. Suppan, J. Mol. Catal.

A, 2004, 214 , 175.62 (a) O. Jimenez, T. E. M uller, C. Sievers, A. Spirkl and J. A. Lercher,

Chem. Commun. , 2006, 2974; ( b) C. Sievers, O. Jimenez, T. E. M uller,S. Steuernagel and J. A. Lercher, J. Am. Chem. Soc. , 2006, 128,13990; ( c) C. Sievers, O. Jim enez, R. Knapp, X. Lin, T. E. M uller,A. T urler, B. Wierczinski and J. A. Lercher, J. Mol. Catal. A , 2008, 279,187.

63 H. Hagiwara, Y. Sugawara, K. Isobe, T. Hoshi and T. Suzuki, Org.Lett. , 2004, 6, 2325.

64 H. Hagiwara, Y. Sugawara, T. Hoshi and T. Suzuki, Chem. Commun. ,2005, 2942.

65 H. Hagiwara, K. H. Ko, T. Hoshi and T. Suzuki, Chem. Commun. ,2007, 2838.


A

H

8/13/2019 SSIL

14/14

66 C. Brochwitz, A. Feldhoff, U. Kunz, M. Vaultier and A. Kirschning,Lett. Org. Chem. , 2006, 3, 442.

67 K. Yu, W. Sommer, J. M. Richardson, M. Weck and C. W. Jones, Adv.Synth. Cat. , 2005, 347 , 161.

68 K. L. Fow, S. Jaenicke, T. E. M uller and C. Sievers, J. Mol. Catal. A ,2008, 279 , 239.

69 LL Lou, X. Peng, K. Yu and S. Liu, Catal. Commun. , 2008, 9, 1891.70 (a) M. Gruttadauria, S. Riela, P. Lo Meo, F. DAnna and R. Noto,

Tetrahedron Lett. , 2004, 45, 6113; ( b) M. Gruttadauria, S. Riela, C.

Aprile, P. Lo Meo, F. DAnna and R. Noto, Adv. Synth. Catal. , 2006,348 , 82.

71 S. Doherty, P. Goodrich, C. Hardacre, V. P arvulescu and C. Paun, Adv.Synth. Catal. , 2008, 350 , 295.

72 B. M. L. Dioos and P. A. Jacobs, J. Catal. , 2006, 243 , 217.73 L.-L. Lou, K. Yu, F. Ding, W. Zhou, X. Peng and S. Liu, Tetrahedron

Lett. , 2006, 47, 6513.74 M. R. Castillo, L. Fousse, J. M. Fraile, J. I. Garc a and J. A. Mayoral,

Chem.Eur. J. , 2007, 13, 287.

8390 | Dalton Trans 2010 39 8377 8390 This journal is The Royal Society of Chemistry 2010

A

H

ssil

Documents