Palladium Catalysts Supported on Mesoporous Molecular Sieves Bearing Nitrogen Donor Groups: Preparation and Use in Heck and Suzuki CC Bond-Forming Reactions

DOI: 10.1002/cssc.200900006

Palladium Catalysts Supported on Mesoporous MolecularSieves Bearing Nitrogen Donor Groups: Preparation andUse in Heck and Suzuki C�C Bond-Forming ReactionsJan Demel,[a, b] Martin Lamac,[a] Jir� Cejka,*[b] and Petr Stepnicka*[a]

Introduction

The Heck[1] and Suzuki[2] reactions are well-established and effi-cient tools for the construction of new carbon–carbon bonds.Although many different palladium catalysts have been de-signed for these reactions since their discovery in the 1970s,the development of new catalytic systems remains an activearea. A particular attention is currently being paid to heteroge-neous (or heterogenized) catalysts that are highly active and,simultaneously, fulfill the requirements for green catalytic sys-tems (e.g. , they are highly efficient and recyclable).[3]

In many catalytic systems the palladium precatalyst hasbeen reported to be converted into fine metal particles duringthe course of these coupling reactions.[4, 5] Indeed, palladiumnanoparticles themselves exert high catalytic activities. It is be-lieved that their “mode of action” involves chipping of reactiveatoms or small clusters from the metal particle and their re-ad-sorption.[6] This is also the likely reason by which unprotectedpalladium nanoparticles grow during the reaction and, conse-quently, lose their catalytic activity. In order to increase the life-time of nanoparticle catalysts and to prepare catalysts that canbe easily separated from the reaction mixture and (potentially)reused, palladium nanoparticles have been combined with var-ious stabilizing agents (e.g. , tetraalkylammonium salts,[7] den-drimers,[8] polymers,[9] or polar solvents[10]) and also immobi-lized onto siliceous solid supports.[3] Mesoporous molecularsieves proved to be very useful for the preparation of support-ed catalysts as they offer high surface areas, regular structures,and also allow for various structural modifications.[11] In at-tempts to stabilize supported catalysts, the mesoporous mo-lecular sieves were surface-modified with phosphine,[12] sulfan-yl,[13] or nitrogen donor groups[14, 15] that can provide additionalstabilization to the metal particle by means of coordination. Inthis regard, sieves modified with nitrogen groups are particu-

larly attractive because they not only form very active catalystsbut also are environmentally benign. Many different nitrogendonor groups and mesoporous supports have been studied.However, the influence of palladium loading and of additional(innocent) surface modifications in such catalysts has oftenbeen overlooked.

Previously, we have prepared several supported palladiumcatalysts from MCM-48- and SBA-15-type mesoporous sievesbearing various nitrogen groups at their surface, and com-pared their catalytic performance in a model Heck reaction.[16]

Among these catalysts, the best properties were exertedby those possessing�SiCH2CH2CH2NHCH2CH2NEt2 and �SiCH2CH2CH2NH2 groups.The former material also showed the best results when themodifying group served simultaneously for anchoring of palla-dium species and as scavenger for acid byproducts (bifunction-al catalyst). These results stimulated us to further study the cat-alytic properties of such catalysts, particularly the one basedon the mesoporous molecular sieve SBA-15 bearing the �SiCH2CH2CH2NHCH2CH2NEt2 groups at the wall surface. In ourwork, we focused on materials that contain different amounts

A series of supported catalysts is prepared by treatment ofSBA-15-type mesoporous molecular sieve bearing�SiCH2CH2CH2NHCH2CH2NEt2 groups with palladium(II) acetate.These catalysts are studied in Suzuki biaryl couplings and inHeck reactions to establish the influence of metal loading andinnocent surface modifications (trimethylsilylation). The Suzukireaction proceeded efficiently with model and practically rele-vant substrates; the catalyst performance increasing with anincreasing degree of metalation (decreasing N/Pd ratio). Cata-lyst poisoning tests revealed that the reaction takes place in

the liquid phase with the catalyst serving as a reservoir ofactive metal species and also as a stabilizing support once thereaction is performed. In the Heck reactions, on the otherhand, the catalyst performance strongly changed with the re-action temperature and with the N/Pd ratio. The material withthe lowest metal loading (0.01 mmol palladium per gram ofmaterial, N/Pd ratio ca. 100:1) proved particularly attractive inthe Heck coupling, being highly active at elevated tempera-tures, recyclable, and capable of acting as a bifunctional cata-lyst (i.e. , functioning without any external base.

[a] J. Demel, Dr. M. Lamac, Prof. P. StepnickaDepartment of Inorganic Chemistry, Faculty of ScienceCharles University in PragueHlavova 2030, 12840 Prague 2 (Czech Republic)Fax: (+ 42) 0221 951 253E-mail : [email protected]

[b] J. Demel, Prof. J. CejkaJ. Heyrovsky Institute of Physical ChemistryAcademy of Sciences of the Czech Republicv.v.i, Dolejskova 3, 18223 Prague 8 (Czech Republic)Fax: (+ 42) 0286 582 307E-mail : [email protected]

442 � 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim ChemSusChem 2009, 2, 442 – 451

of palladium and are (optionally) surface-modified with trime-thylsilyl groups. In this contribution we report the preparationand characterization of such materials, and the results of theirtesting as catalysts for model and practically relevant Suzukiand Heck reactions.

Results and Discussion

Preparation and Characterization of the Catalysts

The preparation of the catalysts is depicted in Scheme 1. Theparent support (material 1) bearing N-{2-(N,N-diethylamino)eth-yl}-3-aminopropyl groups at the wall surface was obtained by

reacting 2-(N,N-diethylamino)ethylamine with 3-chloropropyl-modified, SBA-15-type mesoporous molecular sieve.[16] Thelatter material was prepared by direct solvothermal synthesisfrom sodium silicate, (3-chloropropyl)triethoxysilane, and astructure-directing agent.[17] Another support possessing simul-taneously Me3Si- and N-{2-(N,N-diethylamino)ethyl}aminopropylgroups (material 2) was prepared analogously from the 3-chloropropyl-modified SBA-15-type mesoporous molecularsieve, which was first trimethylsilylated with Me3SiCl and(Me3Si)2O[11a] and then reacted with the amine.

Materials 1 and 2 were subsequently treated with palladi-um(II) acetate in dichloromethane to afford a series of catalysts

with varying surface properties and metal loadings (Scheme 1).Thus, catalyst 3 was obtained from material 1 at maximum pal-ladium loading (reagent ratio: 0.65 mmol Pd ACHTUNGTRENNUNG(O2CCH3)2 pergram of support, achieved degree of metalation:0.46 mmol(Pd) g�1). Catalyst 4 was prepared similarly from ma-terial 2, but was shown to contain only 0.15 mmol(Pd) g�1,probably because of a lower metal-binding affinity of the tri-methylsilylated support. To allow for a comparison of supports1 and 2, another catalyst having a palladium loading similar tocatalyst 4 was prepared from material 1 (catalyst 5 :0.14 mmol(Pd) g�1). In all cases, the molar amount of the nitro-gen groups was higher than the amount of anchored palladi-um (Table 1). In particular, catalyst 6 was designed to have aN/Pd molar ratio of around 100:1 so that the solid supportcould not only anchor the palladium but also bind acidic by-products.

The resulting materials were characterized by elementalanalysis (conventional and optical emission spectroscopy), IRspectroscopy, nitrogen adsorption isotherm measurements,and powder X-ray diffraction. The absence of chlorine in mate-rial 1 indicated complete conversion of the 3-chloropropyl to�ACHTUNGTRENNUNG(CH2)3NH ACHTUNGTRENNUNG(CH2)2NEt2 groups. By contrast, the relatively high re-sidual chlorine content in material 2 (ca. 75 % of the initialamount) implied that not all 3-chloropropyl groups of the sup-port reacted with the amine, likely owing to a blocking effectof the bulky SiMe3 moieties. A rather low amount of the nitro-gen substituents and the presence of the SiMe3 group obvi-ously account for the relatively lower metal-binding affinity ofsupport 2 as compared with support 1.

The palladium content was determined by inductively cou-pled plasma–optical emission spectroscopy (ICP–OES) analysis.In the case of catalysts 5 and 6, for which the metal loadingwas well below the maximum binding capacity of the support,the determined metal content was in good agreement withthe amount of palladium(II) acetate added to the reaction mix-ture. It is also worth noting that the maximum palladium load-ing for materials 1 and 2 (i.e. , in catalysts 3 and 4) was foundat the same molar ratio of amine groups to palladium.

Powder X-ray diffraction patterns of 1–6 consisted of threepeaks in the range 2q<108, as is typical for SBA-15-type hex-agonal mesoporous molecular sieves. Although catalysts 3–6exerted slightly lower diffraction intensities compared to theirparent materials, the diffraction data still confirmed that thesupports remain largely unchanged upon the deposition of

Scheme 1. Preparation of catalysts 1–6.

Table 1. Selected analytical data for catalysts 3–6.

Catalyst Pd loading [mmol g�1] N/Pd atomic ratio[a]

3 0.46 2.5:14 0.15 2.5:15 0.14 8.9:16 0.01 124:1[b]

[a] The nitrogen-group/palladium molar ratios are half of the nitrogen-atom/palladium ratios. [b] This value may be somewhat uncertain owingto low Pd loading.

ChemSusChem 2009, 2, 442 – 451 � 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.chemsuschem.org 443

Palladium Catalysts Supported on Nitrogen-Donor Mesoporous Molecular Sieves

palladium. Nitrogen adsorption isotherms recorded for material1 and catalysts 3–6 displayed a sharp increase in the adsorbedamount at p/p0 = 0.6–0.8. Such behavior is characteristic formesoporous sieves with narrow pore size distributions andpore dimensions around 6–7 nm. After metalation, the overallsurface area slightly decreased while the pore volume did notchange significantly (Table 2). This can be attributed to thepresence of some ‘surface roughness’ on the SBA-15 walls, asrecently described.[18]

IR spectra of the parent chloropropylated mesoporous sieveand of materials 1–6 (Figure 1) displayed bands resulting fromaliphatic pendants at ca. 2975–2890 cm�1 (nC�H), at ca. 1450–1465, and around 1385 cm�1 (C�H bending). Trimethylsilylationof the support surface was reflected by an increased intensityof the nC-HACHTUNGTRENNUNG(CH3) band at 2962 cm�1. Besides, catalysts 3–6showed two strong bands attributable to acetate nas and ns vi-brations at ca. 1570 and 1395 cm�1, the relative intensity ofwhich (expressed as the intensity ratio nasACHTUNGTRENNUNG(acetate)/nC�H) in-creased with the metal loading. Notably, the acetate bandswere shifted towards lower energies versus those of [Pd-ACHTUNGTRENNUNG(O2CCH3)2]3 (1605 and 1435 cm�1 in Nujol) whilst the energydifference Dn=nasACHTUNGTRENNUNG(COO)�ns ACHTUNGTRENNUNG(COO)[19] remained roughly thesame (ca. 175 cm�1 in 3, and 170 cm�1 in [Pd ACHTUNGTRENNUNG(O2CCH3)2]3). Thisindicates that the acetate ion in catalysts 3–6 coordinates as abridging donor albeit with support from some additional inter-actions (N!Pd). It should be noted, however, that the carbox-

ylate bands are rather broad, suggesting some chemical none-quivalence of the acetate ions.

Catalytic Tests

The catalytic properties of catalysts 3–6 were initially tested inmodel Suzuki cross-coupling reactions between phenylboronicacid and 4-substituted halobenzenes (7 a–d and 8 a–d), yield-ing the respective substituted biphenyls 9 a–d (Scheme 2), andin a model Heck reaction between bromobenzene and n-butylacrylate to give n-butyl cinnamate (10, Scheme 3). Next, the

catalysts were also probed in several practically relevant reac-tions. The selected examples included the preparation of ge-neric nonsteroidal anti-inflammatory drugs: (biphenyl-4-yl)ace-tic acid (Felbinac) and 5-(2,4-difluorophenyl)-2-hydroxy-benzoicacid (Diflunisal),[20] both in the form of methyl esters (11 and12 in Scheme 4); and intramolecular Heck cyclization of N-(2-bromophenyl)-(E)-cinnamamide to give 3-benzylideneoxindole(13, Scheme 5).[21]

The Suzuki Reaction

Biaryl coupling reactions were carried out with a 1.2 molarexcess of phenylboronic acid with respect to the aryl halide,with potassium carbonate as the base, and in the presence ofan amount of catalyst corresponding to 0.5 mol % Pd (vs. thearyl halide). The reaction time was 24 h. Reactions performedin different solvents with catalyst 5 and bromide 7 a as the

Table 2. Textural properties and analytical data for catalysts 3–6.

Entry BET surface area[a] [m2 g�1] Pore diameter [nm] Pore volume [cm3 g�1] Pd loading [mmol g�1]

Material 1 671 6.9 0.83 0Catalyst 3 494 6.8 0.75 0.46Catalyst 4 556 6.5 0.70 0.15Catalyst 5 660 6.8 0.85 0.14Catalyst 6 499 6.8 0.80 0.01Catalyst 3, reused[b] 450 6.8 0.73 n.a.Catalyst 3, reused[c] 380 6.8 0.71 n.a.

[a] BET: Brunauer–Emmett–Teller. [b] Catalyst recovered after Suzuki reaction in dioxane. [c] Catalyst recovered after Suzuki reaction in dry EtOH.

Figure 1. Representative IR spectra of A) the parent 3-chloropropylated mes-oporous sieve, B) material 1, and C) catalyst 3.

Scheme 2. Suzuki coupling of substituted aryl halides with phenylboronicacid to give biaryls 9 a-d (R = Me (a), OMe (b), C(O)Me (c), and NO2 (d)).

Scheme 3. Heck reaction of n-butyl acrylate with PhBr to give n-butyl cinna-mate (10).

444 www.chemsuschem.org � 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim ChemSusChem 2009, 2, 442 – 451

J. Cejka, P. Stepnicka et al.

substrate revealed that reactions carried out in anhydrous eth-anol at 75 8C and in dioxane at 100 8C gave the best yields(Table 3). Subsequent tests with catalysts 3–6 showed that thecatalyst activity strongly depended on the N/Pd ratio: Thehigher the N/Pd ratio, the lower the yield of the coupling prod-uct in both solvents (see entries 1, 3, and 6–10 in Table 3).

Because catalyst 3 displayed slightly higher turn-over num-bers (TONs) than its silylated analogue 4 under identical condi-

tions, the former material was subsequently tested in reactionswith a series of bromobenzenes bearing both deactivating(R = Me (a) and OMe (b)) and activating (R = C(O)Me (c) andNO2 (d)) substituents (entries 6, 8, and 11–16 in Table 3). Again,the coupling reactions provided quantitative or at least satis-factory yields of the corresponding biaryls 9. Lowering of thepalladium amount to 0.1 and 0.05 mol % in the reaction ofPhB(OH)2 with 7 a led to a decrease in the yield of 9 a whileTONs grew from 200 (0.5 mol %) to 630 and 1140, respectively.

The recyclability of catalyst 3 was studied in the standard re-action of 7 a with PhB(OH)2 (Table 4). After each catalytic run,

the catalyst was recovered by centrifugation, washed with eth-anol, and dried in air at room temperature. The activity of therecycled catalysts strongly decreased after each catalytic run;the deactivation being more pronounced in dioxane than inanhydrous ethanol. The observed activity-lowering upon recy-cling can be attributed to a combination of the loss of palladi-um (Table 5) and growth of palladium particles as was estab-lished by powder X-ray diffraction.[22] Textural data for the recy-cled catalysts (Table 2) indicate that mesoporous structure re-

mains largely intact during thecatalyzed reaction, there beingobserved only a decrease in BETsurface areas.

Catalyst 3 was also studied inreactions of aryl chlorides 8 withPhB(OH)2. In addition to K2CO3,KF and K3PO4 were employed asbase to possibly optimize the re-action course. The results aresummarized in Table 6. Non-acti-vated substrates 8 a and 8 b pro-duced the corresponding bi-phenyls in yields less than 2 % ordid not react at all. For activatedsubstrates 8 c and 8 d, the high-est yields of biphenyls 9 c and9 d were 47 % (EtOH/K3PO4/75 8C) and 29 % (dioxane/KF/100 8C), respectively, after 24 h.Biaryl couplings leading to esters11 and 12 were also carried outwith catalyst 3 (0.5 mol %) using20 % molar excess of the respec-tive arylboronic acid and potassi-

Scheme 4. Suzuki coupling affording the analogues of nonsteroidal anti-in-flammatory drugs 11 and 12.

Scheme 5. Intramolecular Heck reaction of N-(3-bromophenyl)-(E)-cinnama-mide to afford isomeric oxindoles (E)- and (Z)-13.

Table 3. Catalytic results for the Suzuki reaction.[a]

Entry Catalyst Substrate Solvent Yield of 8 [%][b] TON[c]

1 5 7 a EtOH 56 1122 5 7 a EtOH/H2O[d] 32 643 5 7 a dioxane 62 1244 5 7 a DMF[e] 0 05 5 7 a MeCN 0 06 3 7 a dioxane 95 1907 4 7 a dioxane 83 1668 3 7 a EtOH quant. 2009 4 7 a EtOH quant. 20010 6 7 a EtOH 0 011 3 7 b dioxane 87 17412 3 7 c dioxane quant. 20013 3 7 d dioxane quant. 20014 3 7 b EtOH 79 17815 3 7 c EtOH quant. 20016 3 7 d EtOH quant. 20017 3[f] 7 a EtOH 63 63018 3[g] 7 a EtOH 57 1140

[a] Reactions were performed under an argon atmosphere at 75 8C in EtOH or 100 8C in dioxane. The reactionmixture consisted of aryl halide (1.0 mmol), phenylboronic acid (1.2 mmol), base (K2CO3, 1.1 mmol), catalyst(0.5 mol % Pd with respect to aryl halide), and 5 mL of dry solvent. Reaction time was 24 h. [b] Determined by1H NMR. [c] TON: turn-over number, defined as mmol ACHTUNGTRENNUNG(product)/mmol(Pd). [d] 1:1 (v/v) mixture. [e] DMF: N,N-di-methylformamide. [f] 0.1 mol % of Pd. [g] 0.05 mol % of Pd.

Table 4. Recycling experiments of catalyst 3.[a]

Catalytic run Yield of 9 a [%] ACHTUNGTRENNUNG(TON)In dioxane In EtOH

1 (fresh) 87 (174) 94 (188)2 23 (46) 51 (102)3 3 (6) 10 (20)

[a] For conditions, see Table 3.

ChemSusChem 2009, 2, 442 – 451 � 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.chemsuschem.org 445

Palladium Catalysts Supported on Nitrogen-Donor Mesoporous Molecular Sieves

um carbonate as the base. The reactions were performed inanhydrous ethanol at 75 8C. Both reactions proceeded cleanly,affording the esters in approximately 70 % yields (determinedby 1H NMR) after 24 h.

The Heck Reaction

As model Heck reaction we chose the coupling between bro-mobenzene and n-butyl acrylate to give n-butyl cinnamate(10). Because our previous work showed that sodium acetateand N,N-dimethylacetamide (DMA) as the solvents give thebest results,[16] this base–solvent combination was also usedthroughout the current study. Initial testing of catalysts 3–6 at160 8C (Figure 2) revealed that the silylated catalyst 4 affordedthe lowest yield of the coupling product. Among the nonsily-lated catalysts (3, 5, and 6), those with a lower palladium load-ing performed better (i.e. , the higher the N/Pd ratio, the betterthe yield of 10)—unlike the Suzuki coupling. The decrease inthe yield of the coupling product observed in the reaction cat-alyzed with 6 at later stages can be attributed to double aryla-tion, giving rise to n-butyl 2,2-diphenylacrylate, which wasidentified in the reaction mixture by GC-MS measurements.When the reaction with catalyst 6 showing the highest N/Pdratio was performed without sodium acetate, it proceeded atroughly half the rate of the reaction conducted in the presenceof the “external” base (Figure 3).

Upon lowering the reaction temperature from 160 8C to 140and 120 8C, the yields of n-butyl cinnamate decreased with allcatalysts. However, the decrease was markedly more pro-nounced for the catalysts having higher N/Pd ratios (Figure 4).Such a trend can be accounted for by a lower accessibility ofthe palladium for the organic reagents when the donor nitro-

gen groups are present in large excess. Nonetheless, when thetemperature reaches a point beyond which the palladium canliberate from the solid support, the reaction is forced to pro-ceed even with such catalysts. After the reaction, the palladiumis readily recaptured by the highly functionalized support andthus preserves its activity (also because the palladium particlescan disperse further). These properties result in a lower degreeof aggregation and, consequently, increase the catalyst life-time. As evidenced by the results in Table 7, catalyst 6 main-tained its activity during four subsequent runs. The somewhathigher yields of 10 during the third and fourth catalytic runsprobably result from suppressed consecutive reactions of theprimary coupling product (twofold arylation). When the recy-

Table 5. Fraction of leached-out palladium from the total palladium pres-ent in the reaction mixture

Catalyst In dioxane[a] [%] In EtOH[b] [%]

3 1 84 11 25 5 5

[a] Reaction of 7 a with phenylboronic acid at 100 8C in dioxane for 24 h.[b] Reaction of 7 a with phenylboronic acid at 75 8C in EtOH for 24 h.

Table 6. Yields of the coupling products 9 attained in the Suzuki reac-tions with chloroarenes 8.[a]

Halide (R) Solvent K2CO3 [%] KF [%] K3PO4 [%]

8 a (Me) EtOH 2 n.a. n.a.8 b (OMe) EtOH 0 n.a. n.a.8 c (C(O)Me) EtOH 19 7 58 d (NO2) EtOH 14 24 478 c (C(O)Me) dioxane 19 29 208 d (NO2) dioxane 43 39 21

[a] The reactions were performed at 75 8C in EtOH or at 100 8C in dioxanewith 0.5 mol % Pd of catalyst 3 with respect to aryl halide. Yields of therespective biaryls are given.

Figure 2. Time-dependence of the yield of the coupling product 10 for themodel Heck reaction performed in the presence of catalysts 3–6 at 160 8C(in DMA and at 1 mol % Pd loading).

Figure 3. Time-dependence of the yield of 10 for the model Heck reactionperformed with catalyst 6 in the presence and the absence of sodium ace-tate at 160 8C (in DMA and at 1 mol % Pd loading).

446 www.chemsuschem.org � 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim ChemSusChem 2009, 2, 442 – 451

J. Cejka, P. Stepnicka et al.

cling experiments were carried out without any external basethe catalyst activity decreased right after the first catalytic rundespite the catalyst being regenerated after each run by stir-ring overnight with triethyla-mine.

The amounts of leached-outpalladium presented in Table 8imply that the metal is lostsomewhat easier from the palla-dium-rich materials (i.e. , at lowerN/Pd ratios). During five runs,the content of palladium in cata-lyst 6 decreased by 34 %. Thiscorresponds to the loss of 8 % ofpalladium per catalytic run onaverage, which in turn indicatesthat the liberation of palladiummust have accelerated duringthe later cycles (Table 8).[23] Inter-

estingly, after four consecutive reactions performed withoutany external base, catalyst 6 lost less than 1 % of the anchoredpalladium.

Catalytic results for intramolecular Heck reaction (Scheme 4)are summarized in Table 9. The reaction catalyzed with catalyst5 (1 mol % Pd) at 160 8C was complete within 2 h, cleanly af-fording oxindole 13 as a mixture of E and Z isomers in ca. 5:2ratio. Upon lowering the reaction temperature, the yield of thecyclization product decreased while the E/Z ratio remainedroughly constant. At 140 8C, catalyst 5 exerted higher activitythan catalysts 3 and 6 (see entries 5 and 6). When the amountof catalyst 5 was lowered to 0.5 mol %, the coupling reactionstill proceeded with a very good yield (87 %) whereas a furtherlowering of the catalyst amount to 0.1 mol % Pd ensued inonly 7 % yield.

Catalyst Poisoning Tests

In order to study the nature of the prepared catalysts, severalpoisoning experiments were carried out. In addition to thestandard mercury poisoning test,[24] we also utilized an ap-proach recently introduced by Jones et al. , who added of anexcess of �ACHTUNGTRENNUNG(CH2)3SH modified molecular sieve to palladium cat-alyst supported with the same material as a scavenger of theactive metal.[13e] In our case, the parent material 1 was utilizedanalogously.

Because higher reaction temperatures were required for theHeck reaction, the poisoning tests were performed only for theSuzuki reaction. The testing reactions of 7 a with phenylboron-ic acid to give 9 a were performed in the presence of catalyst 3at a reduced 0.1 mol % Pd loading in ethanol at 75 8C. After15 min of reaction, the reaction mixture was “poisoned” by ad-

Figure 4. The yields of the coupling product 10 after 6 h of reaction for cata-lysts 3, 5 and 6 at different temperatures (in DMA and at 1 mol % Pd load-ing).

Table 7. Recycling experiments for the model Heck reaction with catalyst6 performed in the presence and in the absence of sodium acetate as anexternal base.

Catalytic run Yield of 10 [%][a]

With base[b] Without base[c]

1 (new catalyst) 73 472 71 173 84 154 84 85 33 n.a.

[a] Determined by GC. [b] After reaction the catalyst was recovered by fil-tration, and washed with ethanol and acetone. [c] After the reaction thecatalyst was recovered by filtration, washed with ethanol and acetone,and regenerated by overnight stirring with excess triethylamine.

Table 8. Percentages of leached-out palladium.[a]

Entry Leached Pd [%]

Catalyst 3 6Catalyst 5 5Catalyst 6 4

[a] After 6 h of reaction at 160 8C.

Table 9. Catalytic results from intramolecular Heck reaction.[a]

Entry Catalyst Temperature [8C] Yield of 13 [%][b] TOF [h�1][c]

1 5 160 100 502 5 140 43 223 5 120 7 44 5 100 0 05 3 140 27 146 6 140 12 67 5[d] 160 83 428 5[e] 160 7 4

[a] The reaction mixture consisted of N-(2-bromophenyl)-(E)-3-phenylacrylamide (0.5 mmol), sodium acetate(0.75 mmol), DMA (2.5 mL) and 1 mol % of catalyst with respect to the substrate. The reaction time was 2 h.[b] Determined by 1H NMR. [c] Turnover frequency: TOF= mmol of product/(mmol of Pd � reaction time)[d] 0.5 mol % Pd. [e] 0.1 mol % Pd.

ChemSusChem 2009, 2, 442 – 451 � 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.chemsuschem.org 447

Palladium Catalysts Supported on Nitrogen-Donor Mesoporous Molecular Sieves

dition of either mercury or material 1 in such an amount thatthe overall N/Pd ratio in the reaction mixture was ca. 120 (i.e. ,roughly the same as in catalyst 6). A comparison of kinetic pro-files for the standard and poisoned catalytic systems is pre-sented in Figure 5. The reaction stopped immediately after the

addition of the poison and the yields of the coupling productsremained unchanged during the following 27 h. These resultsindicate that the catalytic activity comes from leached-out pal-ladium and also that the reaction does not occur inside thepores of mesoporous molecular sieve (particles of material 1are too large to enter the ca. 7 nm wide channels). The likelyreason for the reaction not to occur inside the pores is proba-bly a too high local concentration of amino groups, whichstrongly bind the metal component.

Conclusions

The catalysts prepared by treatment of SBA-15-type mesopo-rous molecular sieves bearing �SiCH2CH2CH2NHCH2CH2NEt2

groups at the surface with palladium(II) acetate efficiently pro-mote Heck and Suzuki coupling reactions of model substratesas well as practically useful educts to afford 3-benzylideneoxin-dole and analogues (esters) of Felbinac ((biphenyl-4-yl)aceticacid) and Diflunisal (5-(2,4-difluorophenyl)-2-hydroxybenzoicacid). The catalytic activity of such materials varies with thedegree of metalation and is also controlled by the propertiesof the support surface.

In the Suzuki reactions, the catalyst performance increasedwith increasing amount of the palladium deposited on thesupport. The catalyst containing the highest amount of palladi-um (lowest N/Pd ratio) performed best in the series. Such re-sults are in agreement with the results reported by Cruddenet al.[13d] and by Jones et al. ,[13e] who studied palladium cata-

lysts prepared from mesoporous molecular sieves possessingsulfanyl groups. The higher the S/Pd ratio, the lower the yieldof the coupling products in the Suzuki reaction of 4-bromoace-tophenone with phenylboronic acid pinacol ester at 80 8C andin the Heck reaction of 4-bromoacetophenone with n-butyl ac-rylate at 120 8C. Furthermore, we demonstrated that nonfunc-tional surface modifications also play an important role in de-termining the catalyst efficiency. Even simple silylation of thesupport strongly changed the properties of the support (theextent of amination reaction and the metal binding ability)and affected catalytic reactivity of a catalyst prepared thereof(n.b. , the yields achieved with the trimethylsilylated catalyst 4were somewhat higher than with its nonsilylated counterpart5 in both ethanol and dioxane).

Both the mercury poisoning test and a test similar to thatperformed by Jones et al.[13e] indicate that the Suzuki reactiontakes place in the liquid phase. The catalysts thus very likelyserve as the source of active palladium. However, for the pres-ervation of their catalytic activity it is vital that the support effi-ciently recaptures and stabilizes the active palladium speciesso that it can be used again for the reaction.

The trends observed in Heck reactions were strikingly differ-ent from those noted in Suzuki couplings. The catalysts withrelatively higher N/Pd ratios (5 and 6) were not only active butalso, at 160 8C, outperformed catalyst 3 showing the lowest N/Pd ratio (i.e. , the highest metal loading). Moreover, catalyst 6could be recycled three times without any significant loss ofactivity. Such properties can enable one to tune the catalystperformance for a particular reaction system and also enableto boost the bifunctional character of a catalysts having N/Pdratios. Unfortunately, the yields achieved in the reactionswhere the nitrogen groups are used to bind both the palladi-um and the acidic byproducts still remained lower than thosewith the same catalyst in the presence of an “external” base.Nevertheless, our results were by far better than the ones pre-viously achieved with Pd nanoparticles deposited onto ion-ex-changed MCM-41[25] and with Pd-containing ion-exchangedzeolite X and sepiolites.[26]

Experimental Section

Materials and Methods

Palladium(II) acetate (Aldrich), mesitylene (Fluka), bromobenzene(Aldrich), n-butyl acrylate (Aldrich), n-butyl cinnamate (Alfa Aesar),dry N,N-dimethylacetamide (Aldrich), poly(ethylene glycol)-block-poly(propylene glycol)-block-poly(ethylene glycol) (Pluronic P123,(ethylene oxide)20(propylene oxide)70(ethylene oxide)20 ; Aldrich), (3-chloropropyl)triethoxysilane (Aldrich), 2-(N,N-diethylamino)ethyl-ACHTUNGTRENNUNGamine (Aldrich), hexamethyldisiloxane (Fluka), trimethylchlorosilane(Fluka), phenylboronic acid (Fluka), 4-bromotoluene (7 a, Fluka), 4-bromoanisole (7 b, Fluka), 4-bromoacetophenone (7 c, Fluka), 1-bromo-4-nitrobenzene (7 d, Fluka), 4-chlorotoluene (8 a, Fluka), 4-chloroanisole (8 b, Fluka), 4-chloroacetophenone (8 c, Fluka), 1-chloro-4-nitrobenzene (8 d, Fluka), o-bromoaniline, and absoluteethanol (Riedel de Haen) were used without further purification.1,4-Dioxane and toluene were distilled from sodium. Dichlorome-thane was dried over K2CO3 and distilled. Sodium acetate was

Figure 5. Kinetic profiles for the coupling of phenylboronic acid to 4-bromo-toluene in ethanol at 75 8C: A) control reaction, B) with Hg as poison, andC) with material 1 as poison. The arrow indicates the time at which the poi-sons were added. The yields remained unchanged during the following 6–27 h.

448 www.chemsuschem.org � 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim ChemSusChem 2009, 2, 442 – 451

J. Cejka, P. Stepnicka et al.

freshly melted before use in order to remove traces of water. Allcatalysts were handled in air.

5-Bromosalicylic acid methyl ester was prepared as previously de-scribed[27] by refluxing a mixture of 5-bromosalicylic acid (4.4 g,20 mmol), concentrated sulfuric acid (2.2 mL, 12 mmol), and meth-anol (30 mL) for 46 h. The resulting white crystals were filtered off,washed with a cold mixture of methanol and water (1:1), and driedin air. Yield: 3.5 g, 76 %. 1H NMR (CDCl3): d= 3.96 (s, 3 H; CH3), 6.89,7.53, 7.95 (3 � m, 1 H, C6H3) ; 10.69 (s, 1 H, OH) ppm.[23] ESI-MS:m/z = 229 ([M�H]�).

4-Bromophenylacetic acid methyl ester: excess diazomethane (pre-pared from 50 mmol of 4-tolylsulfomethylnitrosamide) in diethylether was added to a solution 4-bromophenylacetic acid (5.4 g,25 mmol) in diethyl ether (50 mL). The mixture was stirred for30 min at room temperature and evaporated under reduced pres-sure to give the analytically pure product as a colorless oil. Yield:5.5 g, 96 %. 1H NMR (CDCl3): d= 3.58 (s, 2 H, CH2), 3.69 (s, 3 H, CH3),7.13–7.17 and 7.42–7.47 (2 � m, 2 H, C6H4) ppm.[28] ESI+MS: m/z =250ACHTUNGTRENNUNG([M+Na]+).

N-(2-bromophenyl)-(E)-3-phenylacrylamide: a mixture of 2-bromoa-niline (16.7 g, 0.10 mol) in acetone (40 mL) and aqueous K2CO3

(20 g, 0.15 mol in 50 mL) was cooled to 0 8C. Cinnamoyl chloride(17.5 g, 0.10 mol) was added and the resulting mixture was stirredat 0 8C for 2 h. Subsequently, it was poured into icecold water andthe separated solid was filtered off, washed with water, and driedover P2O5. The crude product was subsequently recrystallized fromhexane/diethyl ether (1:1). Yield: 16.8 g, 56 %. 1H NMR (CDCl3): d=6.59 (d, 3JHH = 15.5 Hz, 1 H, CH=CH), 6.99 (m, 1 H, aromatics), 7.25–7.45 (m, 4 H, aromatics), 7.53–7.58 (m, 3 H, aromatics), 7.77 (d,3JHH = 15.5 Hz, 1 H, CH=CH), 7.81 (br s, 1 H, NH), 8.50 (br d, JHH =8.2 Hz, 1 H, aromatics) ppm.[29] ESI+MS: m/z = 302 ([M+H]+), 324([M+Na]+).

Nitrogen adsorption isotherms were determined with a Micromerit-ics ASAP 2020 volumetric instrument at �196 8C. Prior to the meas-urements the samples were degassed at 250 8C until a pressure of10�3 Pa was attained (at least for 24 h). X-ray powder patternswere recorded on a Bruker D8 X-ray powder diffractometerequipped with a graphite monochromator and position-sensitivedetector (V�ntec-1) using Cu Ka radiation (l= 1.5412 �) andBragg–Brentano arrangement (data acquisition details : low 2q

region: 2q= 0.58–108, step 0.0085368 at 0.5 s per step; high 2q

region: 2q= 36.58–41.88, step 0.0085368 at 25 s per step). IR spec-tra were recorded with an FTIR Nicolet Magna-IR AEM spectrome-ter (64 scans, 4 cm�1 resolution) using a sealed cell connected to avacuum apparatus. The samples were pressed into self-supportingwafers with a density of ca. 10 mg cm�2 and evacuated for ca. 6 hin order to remove adsorbed water.

Palladium content was determined with an ICP-OES spectrometer(IRIS Intrepid II ; Thermo Electron) equipped with axial plasma andultrasonic CETAC nebulizer, model U-5000AT+ (conditions: plasmapower 1150 W, nebulizer pressure 25.0 psi (1 psi = 6.895 � 103 Pa),auxiliary gas flux 1.0 mL min�1, sample uptake 2.40 mL min�1). Thesamples were first dried at 100 8C for 3 h and then mineralized in amixture of concentrated HF and HNO3 (2:3 v/v; Suprapur Merck) at50 8C for 15 min and diluted with redistilled water. The analyticalline of 324.270 nm was used for Pd determination. The content ofcarbon, hydrogen, nitrogen, and chlorine was determined by stan-dard combustion analysis.

NMR spectra were measured on a Varian UNITY Inova 400 spec-trometer at 25 8C. Chemical shifts (d/ppm) are given relative to in-

ternal tetramethylsilane (1H and 13C). Gas chromatography analyseswere performed on a GC Agilent 6850 chromatograph equippedwith a flame ionization detector and DB-5 column. The identity ofthe products was confirmed by GC-MS analysis (Thermo FocusDSQ equipped with a Thermo TR-5MS capillary column).

Preparation of the Chloropropylated Mesoporous MolecularSieve

The SBA-15-type chloropropylated mesoporous molecular sievewas prepared by direct synthesis as follows: template triblock co-polymer P123 (10.0 g) and sodium silicate nonahydrate (27.35 g,0.0962 mol) were dissolved in water (256 mL) by stirring at roomtemperature for 18 h. Then, 35 % HCl (67.5 mL, 0.78 mol) wasadded and the reaction vessel was immersed into a water bathkept at 45 8C. After 30 min, (3-chloropropyl)triethoxysilane (1.88 g,7.8 mmol; corresponding to 7.5 % of all Si atoms in the reaction so-lution) was added, and the resulting mixture was stirred at 45 8Cfor 1 h and then put into a preheated oven and kept at 90 8C for7 days. The separated white powder was filtered off, washed withwater and ethanol and, finally, extracted with ethanol in a Soxhletextractor for 24 h. Powder X-ray diffraction: 2q= 0.88 (s), 1.48 (m),1.58 (m). IR: nmax/cm�1 2980 (s), 2938 (s), 1864 (w), 1641 (w), 1448(m), 1381 (w), 1351 (w). Textural properties from N2 adsorption iso-therms: BET surface area 668 m2 g�1, pore diameter 6.7 nm, porevolume 0.80 cm3 g�1. Elemental analysis: C 7.6 mmol g�1, Cl0.84 mmol g�1.

Preparation of Material 1

A mixture of chloropropylated mesoporous molecular sieve (1.0 g)and 2-(N,N-diethylamino)ethylamine (2.4 mL, 17 mmol) in dry tolu-ene (30 mL) was stirred under argon at 100 8C for at least 30 h. Thesolid was filtered off, washed with ethanol, and dried in air. Theyellowish powder was added to a mixture of 25 % aqueous ammo-nia and ethanol (1:3, ca. 100 mL) and the suspension was stirredfor 3 h. The solid material was recovered by filtration, washed withethanol and dried in air at room temperature. The resulting solid,material 1, was directly used in the next step. Powder X-ray diffrac-tion: 2q= 0.88 (s), 1.48 (m), 1.58 (m). IR: nmax/cm�1 2978 (s), 2941 (s),2887 (m), 1857 (w), 1631 (w), 1462 (m), 1393 (w), 1383 (w), 1351(w). Textural properties from N2 adsorption isotherms: BET surfacearea 671 m2 g�1, pore diameter 6.9 nm, pore volume 0.83 cm3 g�1.Elemental analysis: C 7.6 mmol g�1, Cl 0.84 mmol g�1.

Preparation of Material 2

The silylation was performed as described elsewhere.[11a] To chloro-propylated molecular sieve (1.0 g) were added trimethylchlorsilane(20 g) and hexamethyldisiloxane (30 g) and the mixture was re-fluxed under argon atmosphere for 20 h. It was then evaporatedunder vacuum and the residue was suspended in acetone. Thesolid material was filtered off, washed three times with acetone,and dried in air to give a silylated molecular sieve, which was con-verted to amine-modified material as follows: A mixture of silylatedmolecular sieve (1.0 g) and 2-(N,N-diethylamino)ethylamine (2.4 mL,17 mmol) in dry toluene (30 mL) was stirred at 100 8C for at least30 h under argon atmosphere. The resulting solid was filtered off,washed with ethanol, and dried. The resulting yellowish powderwas added to a mixture of 25 % aqueous ammonia and ethanol(1:3, ca. 100 mL) and the suspension was stirred for 3 h. The solid

ChemSusChem 2009, 2, 442 – 451 � 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.chemsuschem.org 449

Palladium Catalysts Supported on Nitrogen-Donor Mesoporous Molecular Sieves

was filtered off, washed with ethanol and dried in air at room tem-perature to give material 2.

Preparation of Catalysts 3–6

Material 1 or material 2 (1.0 g) was added to a solution of palladiu-m(II) acetate (0.65 mmol for catalysts 3 and 4, 0.15 mmol for cata-lyst 5, and 0.0115 mmol for catalyst 6) in dry dichloromethane(30 mL). The mixture was stirred for 1 h at room temperature andthen filtered (in the case of catalysts 3 and 4 the filtrate was col-ored by unanchored palladium(II) acetate). The solid was thorough-ly washed with dichloromethane and dried in air at room tempera-ture.

Characterization data for catalyst 3 : powder X-ray diffraction: 2q=0.98 (s), 1.48 (m), 1.68 (m). IR: nmax/cm�1 2977 (s), 2937 (s), 2887 (m),1864 (w), 1573 (s), 1448 (w), 1396 (s), 1332 (w). Textural propertiesfrom N2 adsorption isotherms: BET surface area 494 m2 g�1, pore di-ameter 6.8 nm, pore volume 0.75 cm3 g�1. Elemental analysis: C9.74 mmol g�1, N 1.15 mmol g�1, Cl traces, Pd 0.46 mmol g�1.

Characterization data for catalyst 4 : powder X-ray diffraction: 2q=0.88 (s), 1.48 (m), 1.68 (m). IR: nmax/cm�1 2962 (s), ca. 2935 (shoulderm), 2903 (m), 1946 (w), 1862 (m), 1710 (w), 1573 (s), 1446 (m), 1408(s). Textural properties from N2 adsorption isotherms: BET surfacearea 556 m2 g�1, pore diameter 6.5 nm, pore volume 0.70 cm3 g�1.Elemental analysis: C 10.48 mmol g�1, N 0.38 mmol g�1, Cl0.57 mmol g�1, Pd 0.15 mmol g�1.

Characterization data for catalyst 5 : powder X-ray diffraction: 2q=0.88 (s), 1.48 (m), 1.68 (m). IR: nmax/cm�1 2977 (s), 2939 (s), 2888 (m),1860 (w), 1564 (s), 1453 (m), 1396 (s). Textural properties from N2

adsorption isotherms: BET surface area 660 m2 g�1, pore diameter6.8 nm, pore volume 0.85 cm3 g�1. Elemental analysis: C9.58 mmol g�1, N 1.25 mmol g�1, Cl traces, Pd 0.14 mmol g�1.

Characterization data for catalyst 6 : powder X-ray diffraction: 2q=

0.88 (s), 1.48 (m), 1.68 (m). IR: nmax/cm�1 2977 (s), 2941 (s), 2887 (m),1856 (w), 1462 (s), 1393 (m), 1350 (w). Textural properties from N2-adsorption isotherms: BET surface area 499 m2 g�1, pore diameter6.8 nm, pore volume 0.80 cm3 g�1. Elemental analysis: C8.83 mmol g�1, N 1.37 mmol g�1, Cl 0.09 mmol g�1,Pd 0.01 mmol g�1.

Catalytic Testing in Suzuki Reactions

The coupling reactions were carried out in a 25 mL flask equippedwith a rubber septum. The reaction mixture was prepared bymixing aryl halide (1.0 mmol), phenylboronic acid (1.2 mmol),K2CO3 (1.1 mmol), catalyst (0.5 mol % Pd with respect to the arylhalide), and dry solvent (5 mL). The reaction vessel was flushedwith argon and transferred into a preheated oil bath. Sampleswere withdrawn with a syringe, filtered through a 0.45 mm PTFE sy-ringe filter, and diluted with CDCl3. The composition of the reactionmixture was determined from 1H NMR spectra. In recycling experi-ments the catalysts were recovered by filtration, washed with etha-nol and acetone, and dried in air.Poisoning tests were performed all at once (in one oil bath) toensure identical conditions. Three identical reaction mixtures (A–C)consisting of 4-bromotoluene (1.0 mmol), phenylboronic acid(1.2 mmol), K2CO3 (1.1 mmol), catalyst 3 (0.1 mol % Pd with respectto aryl halide), and absolute ethanol (5 mL) were heated to 75 8C.Reaction mixture A was a control reaction, while reaction mixtur-es B and C were, after 15 min, poisoned by 0.5 mL of Hg and by107 mg of material 1, respectively. The composition of the reaction

mixtures was established by 1H NMR measurements. The couplingproducts 9,[30] 11,[31] and 12[32] were unequivocally identified bytheir 1H NMR spectra. ESI+MS for 11: m/z = 249 ([M+Na]+). ESI+MSfor 12 : m/z = 265 ([M+H]+), 287 ([M+Na]+).

Catalytic Testing in Heck Reactions

Catalytic experiments were performed with a Heidolph Synthesis Iallowing for 16 parallel runs. The reaction mixture consisted ofbutyl acrylate (64 mg, 0.5 mmol), bromobenzene (118 mg,0.75 mmol), sodium acetate (63 mg, 0.75 mmol; if appropriate),mesitylene (50 mg, 0.42 mmol; internal standard), DMA (2.5 mL),and catalyst (1 mol % Pd). Because of the large amount of catalyst6 corresponding to 1 mol % of palladium, the reactions were per-formed in 5 mL of solvent.The reaction mixture was thoroughly flushed with nitrogen priorto the reaction. Samples were centrifuged at 4000 rpm and ana-lyzed by high-resolution gas chromatography. The identity of thereaction product 10 was confirmed by a comparison with an au-thentic sample (GC-MS and NMR spectra). In the recycling experi-ments the catalysts were recovered by filtration, thoroughlywashed with ethanol and acetone, and dried in air. In the case ofcatalyst 6 used in bifunctional mode (i.e. , without sodium acetate),the recovered catalyst was stirred at room temperature overnightin neat triethylamine (25 mL). Subsequently, it was filtered off,washed with ethanol and acetone, and dried in air. Samples for de-termination of leached-out palladium were taken after cooling ofthe reaction mixture and were filtered through 0.2 mm PTFE micro-filter.Intramolecular Heck couplings to give 13 were also performedwith Heidolph Synthesis I. The reaction mixture consisted of N-(2-bromophenyl)-(E)-3-phenylacrylamide (151 mg, 0.5 mmol), sodiumacetate (63 mg, 0.75 mmol), DMA (2.5 mL), and catalyst (1 mol %Pd with respect to the substrate). The reaction mixture was thor-oughly flushed with nitrogen prior to the reaction. After the reac-tion time, the mixture was quenched by adding saturated aqueousNaCl solution and then extracted with diethyl ether. The composi-tion of the extract was established from 1H NMR spectra.[33]

Acknowledgements

This work was financially supported by the Grant Agency of theCzech Republic (project no. 104/09/0561) and forms part of long-term research projects of the Faculty of Science, Charles Universi-ty supported by the Ministry of Education, Youth and Sports ofthe Czech Republic (project nos. MS M0021620857 and LC06070).

Keywords: supported catalysts · Heck reaction · mesoporousmaterials · palladium · Suzuki reaction

[1] a) S. Br�se, A. de Meijere, Metal-Catalyzed Cross-Coupling Reactions, 2ndEdition (Eds. : A. de Meijere, F. Diederich), Wiley-VCH, Weinheim, 2004,pp. 217–315; b) I. P. Beletskaya, A. V. Cheprakov, Chem. Rev. 2000, 100,3009–3066.

[2] a) N. Miyaura, in Metal-Catalyzed Cross-Coupling Reactions, 2nd Edition(Eds. : A. de Meijere, F. Diederich), Wiley-VCH, Weinheim, 2004, pp. 41–123; b) N. Miyaura, A. Suzuki, Chem. Rev. 1995, 95, 2457–2483; c) N.Miyaura, Top. Curr. Chem. 2002, 219, 11–59.

[3] N. T. S. Phan, M. Van Der Sluys, C. W. Jones, Adv. Synth. Catal. 2006, 348,609–679.

[4] J. G. de Vries, Dalton Trans. 2006, 421–429.[5] D. Astruc, Inorg. Chem. 2007, 46, 1884–1894.

450 www.chemsuschem.org � 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim ChemSusChem 2009, 2, 442 – 451

J. Cejka, P. Stepnicka et al.

[6] M. B. Thathagar, J. E. ten Elshof, G. Rothenberg, Angew. Chem. 2006,118, 2952–2956; Angew. Chem. Int. Ed. 2006, 45, 2886–2890.

[7] a) T. Jeffery, Chem. Commun. 1984, 1287–1289; b) I. Pryjomska-Ray, A. M.Trzeciak, J. J. Zi�łkowski, J. Mol. Catal. A 2006, 257, 3–8; c) M. T. Reetz,J. G. de Vries, Chem. Commun. 2004, 1559–1563.

[8] R. W. J. Scott, O. M. Wilson, R. M. Crooks, J. Phys. Chem. B 2005, 109,692–704.

[9] a) S. Chandrasekhar, Ch. Narsihmulu, S. S. Sultana, N. R. Reddy, Org. Lett.2002, 4, 4399–4401; b) A. Gniewek, A. M. Trzeciak, J. J. Zi�łkowski, L. Ke-pinski, J. Wrzyszcz, W. Tylus, J. Catal. 2005, 229, 332–343; c) A. Corma,H. Garc�a, A. Leyva, J. Catal. 2006, 240, 87–99; d) M. A. R. Meier, M. Filali,J.-F. Gohy, U. S. Schubert, J. Mater. Chem. 2006, 16, 3001–3006; e) I. P.Beletskaya, A. N. Kashin, A. E. Litvinov, V. S. Tyurin, P. M. Valetsky, G.van Koten, Organometallics 2006, 25, 154–158; f) P. Ribire, V. Declerck,Y. Ndellec, N. Yadav-Bhatnagar, J. Martinez, F. Lamaty, Tetrahedron2006, 62, 10456–10466.

[10] M. T. Reetz, G. Lohmer, Chem. Commun. 1996, 1921–1922.[11] a) J. S. Beck, J. C. Vartuli, W. J. Roth, M. E. Leonowicz, C. T. Kresge, K. D.

Schmitt, C. T-W. Chu, D. H. Olson, E. W. Sheppard, S. B. McCullen, J. B.Higgins, J. L. Schlenker, J. Am. Chem. Soc. 1992, 114, 10834–10843; b) D.Zhao, Q. Huo, J. Feng, B. F. Chmelka, G. D. Stucky, J. Am. Chem. Soc.1998, 120, 6024–6036.

[12] Q. Hu, J. E. Hampsey, N. Jiang, C. Li, Y. Lu, Chem. Mater. 2005, 17, 1561–1569.

[13] a) Q. Xu, W. Hao, M. Cai, Catal. Lett. 2007, 118, 98–102; b) B. Nohair, S.MacQuarrie, C. M. Crudden, S. Kaliaguine, J. Phys. Chem. C 2008, 112,6065–6072; c) C. M. Crudden, M. Sateesh, R. Lewis, J. Am. Chem. Soc.2005, 127, 10045–10050; d) J. D. Webb, S. MacQuarrie, K. McEleney,C. M. Crudden, J. Catal. 2007, 252, 97–109; e) J. M. Richardson, C. W.Jones, J. Catal. 2007, 251, 80–93.

[14] a) J. Horniakova, T. Raja, Y. Kubota, Y. Sugi, J. Mol. Catal. A 2004, 217,73–80; b) J. H. Clark, D. J. Macquarrie, E. B. Mobofu, Green Chem. 2000,2, 53–55; c) E. B. Mubofu, J. H. Clark, D. J. Macquarrie, Green Chem.2001, 3, 23–25; d) S. Paul, J. H. Clark, J. Mol. Catal. A 2004, 215, 107–111;e) M. Lagasi, P. Moggi, J. Mol. Catal. A 2002, 182–183, 61–72; f) C. Gon-z�lez-Arellano, A. Corma, M. Iglesias, F. S�nchez, Adv. Synth. Catal. 2004,346, 1758–1764; g) D. J. Macquarrie, B. Gotov, S. Toma, Platinum Met.Rev. 2001, 45, 102–110.

[15] a) P. Yi, Z. Zhuangyu, H. Hongwen, J. Mol. Catal. 1990, 62, 297–306; b) J.Zhou, R. Zhou, L. Mo, S. Zhao, X. Zheng, J. Mol. Catal. A 2002, 178, 289–292; c) S. Mandal, D. Roy, R. V. Chaudhari, M. Sastry, Chem. Mater. 2004,16, 3714–3724; d) S. F. Zhao, R. X. Zhou, X. M. Zheng, J. Mol. Catal. A2004, 211, 139–142; e) O. Vassylyev, J. Chen, A. P. Panarello, J. G. Khinast,Tetrahedron Lett. 2005, 46, 6865–6869; f) V. Polshettiwar, P. Hesemann,J. J. E. Moreau, Tetrahedron 2007, 63, 6784–6790; g) H. Li, L. Wang, P. Li,Synthesis 2007, 1635–1642.

[16] J. Demel, Sujandi, S.-E. Park, J. Cejka, P. Stepnicka, J. Mol. Catal. A 2009,302, 28–35.

[17] Sujandi, S.-E. Park, Stud. Surf. Sci. Catal. 2007, 170, 1446–1453.

[18] A. Zukal, H. Siklov�, J. Cejka, Langmuir 2008, 24, 9837–9842.[19] The separation of the antisymmetric and symmetric carboxylate stretch-

ing bands, Dn= nas ACHTUNGTRENNUNG(COO)�nsACHTUNGTRENNUNG(COO), reflects the nature of acetate coordi-nation. For a Review, see: G. B. Deacon, R. J. Phillips, Coord. Chem. Rev.1980, 33, 227–250.

[20] a) A. O. King, N. Yasuda, Top. Organomet. Chem. 2004, 6, 205–245; b) R.Franzn, Y. Xu, Can. J. Chem. 2005, 83, 266–272.

[21] M. O. Terpko, R. F. Heck, J. Am. Chem. Soc. 1979, 101, 5281–5283.[22] The fresh catalyst did not show any diffraction peak in the region of Pd

(111) diffraction whilst the recovered materials displayed broad halosdue to increased diffraction volume. PDF card 5–0681: palladium, cubic,space group Fm-3 m (no. 225), a = 3.8898 � at 26 8C; hkl 111, d =

2.246 �, relative intensity = 100, 2q�408 at Cu Ka.[23] Kçhler et al. have shown (for the Heck coupling of PhBr with styrene)

that the amout of leached-out palladium from a Pd/C catalyst changeswith the progress in the catalyzed reaction. In their case the palladiumamount in the liquid phase decreased with the amount of the unreact-ed aryl halide. Hence, one can assume the metal leaching to be morepronounced when the reaction does not reach completion. For refer-ence, see: K. Kçhler, R. G. Heidenreich, J. G. E. Krauter, J. Pietsch, Chem.Eur. J. 2002, 8, 622–631.

[24] a) G. M. Whitesides, M. Hackett, R. L. Brainard, J. P. P. M. Lavalleye, A. F.Sowinski, A. N. Izumi, S. S. Moore, D. W. Brown, E. M. Staudt, Organome-tallics 1985, 4, 1819–1830; b) J. A. Widegren, R. G. Finke, J. Mol. Catal. A2003, 198, 317–341.

[25] a) J. Demel, J. Cejka, P. Stepnicka, J. Mol. Catal. A 2007, 274, 127–132;b) J. Demel, S.-E. Park, J. Cejka, P. Stepnicka, Catal. Today 2008, 132, 63–67.

[26] a) A. Corma, H. Garc�a, A. Leyva, A. Primo, Appl. Catal. A 2003, 247, 41–49; b) A. Corma, H. Garc�a, A. Leyva, A. Primo, Appl. Catal. A 2004, 257,77–83.

[27] J. A. Mikroyannidis, I. K. Spiliopoulos, J. Polym. Sci. Polym. Chem. Ed.2004, 42, 1768–1778.

[28] N. Sakai, T. Moriya, T. Konakahara, J. Org. Chem. 2007, 72, 5920–5922.[29] G. Evindar, R. A. Batey, J. Org. Chem. 2006, 71, 1802–1808.[30] http://riodb01.ibase.aist.go.jp/sdbs/cgi-bin/cre_index.cgi?lang = eng

(accessed April 2009).[31] S. Crosignani, J. Gonzalez, D. Swinnen, Org. Lett. 2004, 6, 4579–4582.[32] S. L. Adamski-Werner, S. K. Palaninathan, J. C. Sacchettini, J. W. Kelly, J.

Med. Chem. 2004, 47, 355–374.[33] a) T. Miura, Y. Takahashi, M. Murakami, Org. Lett. 2007, 9, 5075–5077;

b) A. C. Coda, A. G. Invernizzi, P. P. Righetti, G. Tacconi, G. Gatti, J. Chem.Soc. Perkin Trans. 2 1984, 615–619; c) G. E. Lathourakis, K. E. Litinas, J.Chem. Soc. Perkin Trans. 1 1996, 491–493.

Received: January 7, 2009

Published online on May 5, 2009

ChemSusChem 2009, 2, 442 – 451 � 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim www.chemsuschem.org 451

Palladium Catalysts Supported on Nitrogen-Donor Mesoporous Molecular Sieves