Synthesis of Alkynyl Ketones by Sonogashira Cross-Coupling of Acyl Chlorides with Terminal Alkynes Mediated by Palladium Catalysts Deposited over Donor-Functionalized Silica Gel Synthesis of Alkynyl Ketones by Sonogashira Cross-Coupling of Acyl Chlorides with Terminal Alkynes Mediated by Palladium Catalysts Deposited over Donor-functionalized Silica Gel

: Palladium catalysts deposited over silica gel bearing simple amine ( ≡ Si(CH 2 ) 3 NH 2 ) and composite functional amide pendants equipped with various donor groups in the terminal position ( ≡ Si(CH 2 ) 3 NHC(O)CH 2 Y, Y = SMe, NMe 2 and PPh 2 ) were prepared and evaluated in Sonogashira-type cross-coupling of acyl chlorides with terminal alkynes to give 1,3-disubstituted prop-2-yn-1-ones. Generally, the catalysts showed good catalytic activity in the reactions of aroyl chlorides with aryl alkynes under relatively mild reaction conditions even without adding a copper co-catalyst. However, their repeated use was compromised by a signiﬁcant loss of activity after the ﬁrst catalytic run. and 1989 who reacted furoyl chlorides with alkynes in the presence of [PdCl 2 (PhCN) 2 ] / CuI and triethylamine to produce alkynyl furanyl ketones. With [PdCl 2 (PPh 3 ) 2 ] / CuI and similar catalysts, this reaction it to synthesize a number of alkynyl ketones in organic solvents in water adding dodecyl sulfate a phase transfer reagent) in a ﬂow reactor unsupported the catalyst) Abstract: Palladium catalysts deposited over silica gel bearing simple amine ( ≡ Si(CH 2 ) 3 NH 2 ) and composite functional amide pendants equipped with various donor groups in the terminal position ( ≡ Si(CH 2 ) 3 NHC(O)CH 2 Y, Y = SMe, NMe 2 and PPh 2 ) were prepared and evaluated in Sonogashira-type cross-coupling of acyl chlorides with terminal alkynes to give 1,3-disubstituted prop-2-yn-1-ones. Generally, the catalysts showed good catalytic activity in the reactions of aroyl chlorides with aryl alkynes under relatively mild reaction conditions even without adding a copper co-catalyst. However, their repeated use was compromised by a significant loss of activity after the first catalytic run.

In 2009, Tsai et al. [15] reported the application of a Pd-bipyridyl complex grafted onto the mesoporous molecular sieve MCM-41.Coupling reactions of various substrates mediated by this catalyst in neat triethylamine, in the presence of CuI and triphenylphosphine, proceeded satisfactorily at low Pd loading (0.002-0.1 mol.%). More recently, Cai et al. [16] used a related Pd catalyst prepared by depositing Pd(OAc)2 over an MCM-41 surface, modified by ≡Si(CH2)3NHCH2CH2NH2 groups. At 0.2 mol.% Pd loading, and with 0.2 mol.% CuI as a co-catalyst, this material could be reused ten times with only a marginal loss of activity (reaction in triethylamine at 50 °C). Other authors evaluated the related catalysts obtained from supports bearing phosphinedonor groups, e.g., periodic mesoporous silica with ≡CH2CH2PPh2 substituents [17] and polystyrene modified by the -CH2P + Ph2CH2CH2PPh2 Cl − moieties at the surface [18].
Alkynyl ketones are valuable synthetic building blocks, opening an access to a range of useful compounds, such as intermediates for the synthesis of various heterocycles [19][20][21][22][23], biologically active compounds [24], naturally occurring compounds [25], liquid-crystalline materials [26], and ligands for transition metal ions [27]. In particular, the promising results achieved with deposited catalysts in the cross-coupling of acyl chlorides and alkynes and the wide range of applications of coupling products led us to consider using palladium catalysts deposited over the conventional silica gel bearing donor-substituted amide pendants [28] at the surface (Scheme 2) [29], which were already evaluated in Suzuki-Miyaura biaryl coupling [30]. The results from our study of these catalysts are presented in this contribution, with a particular focus on the reaction scope and a possible influence of the donor moieties within the functional supports that were varied. . Scheme 2. Deposited catalysts used in this study.

Synthesis of the Catalysts
The deposited catalysts were prepared as reported previously (Scheme 3) [29]. In the first step, freshly calcined, commercial chromatography-grade silica gel (size fraction 63-200 µm) was mixed with (3-aminopropyl)trimethoxysilane in refluxing toluene to afford 3-aminopropylated support 1. Material 1 was subsequently treated with α-functionalized acetic acids in the presence of peptide coupling agents [31,32], yielding the corresponding amide-functionalized supports 2-4. In the final step, the resulting materials were treated with palladium(II) acetate in dichloromethane to produce the deposited Pd catalysts 5-7. As an extension of our previous work, the parent aminopropylated material 1 was also palladated to give material 8 containing only amine functional groups. Scheme 2. Deposited catalysts used in this study.

Synthesis of the Catalysts
The deposited catalysts were prepared as reported previously (Scheme 3) [29]. In the first step, freshly calcined, commercial chromatography-grade silica gel (size fraction 63-200 µm) was mixed with (3-aminopropyl)trimethoxysilane in refluxing toluene to afford 3-aminopropylated support 1. Material 1 was subsequently treated with α-functionalized acetic acids in the presence of peptide coupling agents [31,32], yielding the corresponding amide-functionalized supports 2-4. In the final step, the resulting materials were treated with palladium(II) acetate in dichloromethane to produce the deposited Pd catalysts 5-7. As an extension of our previous work, the parent aminopropylated material 1 was also palladated to give material 8 containing only amine functional groups.
Materials 1-8 were characterized by elemental analysis and infrared (IR) spectroscopy, and the data on 1-7 were compared with those on the previously studied catalysts. While the IR spectra of the newly synthesized materials were virtually identical to those previously reported (see ref. [29]), elemental analysis revealed differences, most likely reflecting the amount of residual adsorbed matter (mostly water). Full characterization data are presented in the Experimental Section.

Catalytic Assessment
Applications of deposited Pd catalysts to Sonogashira-type coupling of terminal alkynes with acyl chlorides (see Introduction) has been studied considerably less than their use in conventional Sonogashira cross-coupling between alkynes and organic halides [33]. Hence, our initial experiments with catalysts 5-8 aimed to find the optimal reaction conditions for these catalysts and to compare their performance with regard to influence of the varied functional groups modifying the support's surface. As a model reaction, we chose the coupling between equimolar amounts of ethynylbenzene (9a) and 4-methylbenzoyl chloride (10d), producing 1-(4-methylphenyl)-3-phenyl-2-propyn-1-one (11ad, see Scheme 4). The influence of the solvent and base, which are known to strongly affect these reactions (see references in the Introduction), were evaluated first. The screening experiments were performed with 0.5 mol.% of catalyst 5 and 5 mol.% of CuI in neat amines and in mixtures of triethylamine with an organic solvent as well. When using neat morpholine and pyridine, the coupling reaction did not proceed in any appreciable extent. However, when replacing these bases with N-methylmorpholine and N,N-diisopropylethylamine (Figure 1), the yields determined by gas chromatography (GC yields) of the coupling product 11ad after 8 h at 50 °C were 2% and 10%, respectively. The best (albeit still rather low) yield of 21% after 8 h was achieved in neat triethylamine.
Materials 1-8 were characterized by elemental analysis and infrared (IR) spectroscopy, and the data on 1-7 were compared with those on the previously studied catalysts. While the IR spectra of the newly synthesized materials were virtually identical to those previously reported (see ref. [29]), elemental analysis revealed differences, most likely reflecting the amount of residual adsorbed matter (mostly water). Full characterization data are presented in the Experimental Section.

Catalytic Assessment
Applications of deposited Pd catalysts to Sonogashira-type coupling of terminal alkynes with acyl chlorides (see Introduction) has been studied considerably less than their use in conventional Sonogashira cross-coupling between alkynes and organic halides [33]. Hence, our initial experiments with catalysts 5-8 aimed to find the optimal reaction conditions for these catalysts and to compare their performance with regard to influence of the varied functional groups modifying the support's surface. As a model reaction, we chose the coupling between equimolar amounts of ethynylbenzene (9a) and 4-methylbenzoyl chloride (10d), producing 1-(4-methylphenyl)-3-phenyl-2-propyn-1-one (11ad, see Scheme 4). The influence of the solvent and base, which are known to strongly affect these reactions (see references in the Introduction), were evaluated first. The screening experiments were performed with 0.5 mol.% of catalyst 5 and 5 mol.% of CuI in neat amines and in mixtures of triethylamine with an organic solvent as well. When using neat morpholine and pyridine, the coupling reaction did not proceed in any appreciable extent. However, when replacing these bases with N-methylmorpholine and N,N-diisopropylethylamine (Figure 1), the yields determined by gas chromatography (GC yields) of the coupling product 11ad after 8 h at 50 • C were 2% and 10%, respectively. The best (albeit still rather low) yield of 21% after 8 h was achieved in neat triethylamine. Materials 1-8 were characterized by elemental analysis and infrared (IR) spectroscopy, and the data on 1-7 were compared with those on the previously studied catalysts. While the IR spectra of the newly synthesized materials were virtually identical to those previously reported (see ref. [29]), elemental analysis revealed differences, most likely reflecting the amount of residual adsorbed matter (mostly water). Full characterization data are presented in the Experimental Section.

Catalytic Assessment
Applications of deposited Pd catalysts to Sonogashira-type coupling of terminal alkynes with acyl chlorides (see Introduction) has been studied considerably less than their use in conventional Sonogashira cross-coupling between alkynes and organic halides [33]. Hence, our initial experiments with catalysts 5-8 aimed to find the optimal reaction conditions for these catalysts and to compare their performance with regard to influence of the varied functional groups modifying the support's surface. As a model reaction, we chose the coupling between equimolar amounts of ethynylbenzene (9a) and 4-methylbenzoyl chloride (10d), producing 1-(4-methylphenyl)-3-phenyl-2-propyn-1-one (11ad, see Scheme 4). The influence of the solvent and base, which are known to strongly affect these reactions (see references in the Introduction), were evaluated first. The screening experiments were performed with 0.5 mol.% of catalyst 5 and 5 mol.% of CuI in neat amines and in mixtures of triethylamine with an organic solvent as well. When using neat morpholine and pyridine, the coupling reaction did not proceed in any appreciable extent. However, when replacing these bases with N-methylmorpholine and N,N-diisopropylethylamine (Figure 1), the yields determined by gas chromatography (GC yields) of the coupling product 11ad after 8 h at 50 °C were 2% and 10%, respectively. The best (albeit still rather low) yield of 21% after 8 h was achieved in neat triethylamine.
Reaction tests performed in organic solvents in the presence of 5 equiv. of triethylamine ( Figure  2) revealed a marked acceleration of the coupling reaction in acetonitrile (ca. 60% yield of 11ad within 3 h at 50 °C). In contrast, reactions in other tested solvents, viz. toluene, 1,4-dioxane, acetone and N,Ndimethylformamide, proceeded less efficiently, achieving lower yields than the aforementioned reaction in neat triethylamine (below 15% after 8 h; Figure 2); no reaction was observed in methanol. A subsequent series of experiments was designed to assess the effect of the CuI additive and relative amounts of the starting materials. Rather surprisingly, the reaction performed in neat triethylamine with 0.5 mol.% of catalyst 5 without adding CuI at 50 °C ensued in a higher yield of the coupling product than the similar reaction in the presence of the CuI co-catalyst (5 mol.%; 39% vs. 21%). Consistently, when using acetonitrile as the solvent (with added NEt3, 5 equiv.), the reaction without CuI produced 11ad in a 78% yield after 8 h, which is a higher yield than that of the reaction performed in the absence of CuI (63%). Subsequently, we determined whether the coupling reaction is affected by the amount of acyl chloride when gradually increasing the amount of 4-toulyl chloride (10d) up to 1.5 equiv. As shown in Figure 3, the yield of 11ad significantly increased with the amount Reaction tests performed in organic solvents in the presence of 5 equiv. of triethylamine ( Figure 2) revealed a marked acceleration of the coupling reaction in acetonitrile (ca. 60% yield of 11ad within 3 h at 50 • C). In contrast, reactions in other tested solvents, viz. toluene, 1,4-dioxane, acetone and N,N-dimethylformamide, proceeded less efficiently, achieving lower yields than the aforementioned reaction in neat triethylamine (below 15% after 8 h; Figure 2); no reaction was observed in methanol.  Reaction tests performed in organic solvents in the presence of 5 equiv. of triethylamine ( Figure  2) revealed a marked acceleration of the coupling reaction in acetonitrile (ca. 60% yield of 11ad within 3 h at 50 °C). In contrast, reactions in other tested solvents, viz. toluene, 1,4-dioxane, acetone and N,Ndimethylformamide, proceeded less efficiently, achieving lower yields than the aforementioned reaction in neat triethylamine (below 15% after 8 h; Figure 2); no reaction was observed in methanol. A subsequent series of experiments was designed to assess the effect of the CuI additive and relative amounts of the starting materials. Rather surprisingly, the reaction performed in neat triethylamine with 0.5 mol.% of catalyst 5 without adding CuI at 50 °C ensued in a higher yield of the coupling product than the similar reaction in the presence of the CuI co-catalyst (5 mol.%; 39% vs. 21%). Consistently, when using acetonitrile as the solvent (with added NEt3, 5 equiv.), the reaction without CuI produced 11ad in a 78% yield after 8 h, which is a higher yield than that of the reaction performed in the absence of CuI (63%). Subsequently, we determined whether the coupling reaction is affected by the amount of acyl chloride when gradually increasing the amount of 4-toulyl chloride (10d) up to 1.5 equiv. As shown in Figure 3, the yield of 11ad significantly increased with the amount A subsequent series of experiments was designed to assess the effect of the CuI additive and relative amounts of the starting materials. Rather surprisingly, the reaction performed in neat triethylamine with 0.5 mol.% of catalyst 5 without adding CuI at 50 • C ensued in a higher yield of the coupling product than the similar reaction in the presence of the CuI co-catalyst (5 mol.%; 39% vs. 21%). Consistently, when using acetonitrile as the solvent (with added NEt 3 , 5 equiv.), the reaction without CuI produced 11ad in a 78% yield after 8 h, which is a higher yield than that of the reaction performed in the absence of CuI (63%). Subsequently, we determined whether the coupling reaction is affected by the amount of acyl chloride when gradually increasing the amount of 4-toulyl chloride (10d) up to 1.5 equiv. As shown in Figure 3, the yield of 11ad significantly increased with the amount of acyl chloride. With only 1.3 equiv. of 10d, the GC yields of the coupling product were already virtually quantitative within 1 h of the reaction time.  Using 1.5 equiv. of 10d, we subsequently tried to reduce the catalyst loading. Under these conditions, the reaction proceeded satisfactorily, even in the presence of 0.1 and 0.2 mol.% of the selected model catalyst 5 and at short reaction times, as shown in Figure 4, which compares the GC yields of the coupling product 11ad achieved over different periods of time. When decreasing the reaction temperature, however, the yield of the coupling product dramatically decreased (100% at 50 °C, 67% at 40 °C and ≈14% at 30 °C after 30 min of the reaction with catalyst 5 and 0.5 mol.% Pd in the reaction mixture). Lastly, we compared all prepared catalysts and palladium(II) acetate under rather harsh reaction conditions (0.1 mol.% Pd, 30 °C reaction temperature). Regrettably, the kinetic profiles presented in Figure 5 clearly indicate that unsupported palladium(II) acetate outperforms all deposited catalysts. Among the deposited catalysts, the lowest efficiency exerted catalyst 7 bearing phosphine groups, Using 1.5 equiv. of 10d, we subsequently tried to reduce the catalyst loading. Under these conditions, the reaction proceeded satisfactorily, even in the presence of 0.1 and 0.2 mol.% of the selected model catalyst 5 and at short reaction times, as shown in Figure 4, which compares the GC yields of the coupling product 11ad achieved over different periods of time. When decreasing the reaction temperature, however, the yield of the coupling product dramatically decreased (100% at 50 • C, 67% at 40 • C and ≈14% at 30 • C after 30 min of the reaction with catalyst 5 and 0.5 mol.% Pd in the reaction mixture).  Using 1.5 equiv. of 10d, we subsequently tried to reduce the catalyst loading. Under these conditions, the reaction proceeded satisfactorily, even in the presence of 0.1 and 0.2 mol.% of the selected model catalyst 5 and at short reaction times, as shown in Figure 4, which compares the GC yields of the coupling product 11ad achieved over different periods of time. When decreasing the reaction temperature, however, the yield of the coupling product dramatically decreased (100% at 50 °C, 67% at 40 °C and ≈14% at 30 °C after 30 min of the reaction with catalyst 5 and 0.5 mol.% Pd in the reaction mixture). Lastly, we compared all prepared catalysts and palladium(II) acetate under rather harsh reaction conditions (0.1 mol.% Pd, 30 °C reaction temperature). Regrettably, the kinetic profiles presented in Figure 5 clearly indicate that unsupported palladium(II) acetate outperforms all deposited catalysts. Among the deposited catalysts, the lowest efficiency exerted catalyst 7 bearing phosphine groups, Lastly, we compared all prepared catalysts and palladium(II) acetate under rather harsh reaction conditions (0.1 mol.% Pd, 30 • C reaction temperature). Regrettably, the kinetic profiles presented in Figure 5 clearly indicate that unsupported palladium(II) acetate outperforms all deposited catalysts. Among the deposited catalysts, the lowest efficiency exerted catalyst 7 bearing phosphine groups, whereas the performance of catalysts bearing the S-and N-donor groups (5 and 6) was quite similar and slightly better than that of catalyst 8 obtained from the amine-functionalized support. whereas the performance of catalysts bearing the S-and N-donor groups (5 and 6) was quite similar and slightly better than that of catalyst 8 obtained from the amine-functionalized support. Recycled catalysts 5-8 significantly lost their activity ( Figure 6), presumably due to leaching of the deposited metal and to overall catalyst deactivation (the amount of Pd leached out during the first run was only 1%-4% of the initial amount). Notably, CuI (5 mol.%) addition to the reaction mixture increased the stability of the catalysts and even led to an activation of the phosphinefunctionalized catalyst 7, whereas the amount of leached-out Pd remained approximately the same (2%-4% during the first run; see the Supporting Information, Table S1). However, the yields of 11ad obtained with recycled deposited catalysts 5-8/CuI were still considerably lower than the yields achieved during the first runs and further decreased upon catalyst reuse. Using catalyst 5 (0.5 mol.% Pd), we also performed reaction scope tests, which are summarized in Table 1. Initially, we focused on the reactions of ethynylbenzene (9a) with substituted benzoyl chlorides. In the case of methyl-substituted acyl chlorides, the yields of the coupling products Recycled catalysts 5-8 significantly lost their activity ( Figure 6), presumably due to leaching of the deposited metal and to overall catalyst deactivation (the amount of Pd leached out during the first run was only 1%-4% of the initial amount). Notably, CuI (5 mol.%) addition to the reaction mixture increased the stability of the catalysts and even led to an activation of the phosphine-functionalized catalyst 7, whereas the amount of leached-out Pd remained approximately the same (2-4% during the first run; see the Supporting Information, Table S1). However, the yields of 11ad obtained with recycled deposited catalysts 5-8/CuI were still considerably lower than the yields achieved during the first runs and further decreased upon catalyst reuse. whereas the performance of catalysts bearing the S-and N-donor groups (5 and 6) was quite similar and slightly better than that of catalyst 8 obtained from the amine-functionalized support. Recycled catalysts 5-8 significantly lost their activity ( Figure 6), presumably due to leaching of the deposited metal and to overall catalyst deactivation (the amount of Pd leached out during the first run was only 1%-4% of the initial amount). Notably, CuI (5 mol.%) addition to the reaction mixture increased the stability of the catalysts and even led to an activation of the phosphinefunctionalized catalyst 7, whereas the amount of leached-out Pd remained approximately the same (2%-4% during the first run; see the Supporting Information, Table S1). However, the yields of 11ad obtained with recycled deposited catalysts 5-8/CuI were still considerably lower than the yields achieved during the first runs and further decreased upon catalyst reuse. Using catalyst 5 (0.5 mol.% Pd), we also performed reaction scope tests, which are summarized in Table 1. Initially, we focused on the reactions of ethynylbenzene (9a) with substituted benzoyl chlorides. In the case of methyl-substituted acyl chlorides, the yields of the coupling products Using catalyst 5 (0.5 mol.% Pd), we also performed reaction scope tests, which are summarized in Table 1. Initially, we focused on the reactions of ethynylbenzene (9a) with substituted benzoyl chlorides. In the case of methyl-substituted acyl chlorides, the yields of the coupling products increased with the decrease in steric hindrance. Similar reactions with isomeric nitrobenzoyl chlorides proceeded generally less efficiently and required longer reaction times to achieve isolated yields of the coupling products higher than 50%; the reaction of 9a with 2-nitrobenzoyl chloride, the most sterically crowded and deactivated acyl chloride, did not proceed. For the acyl chlorides, the substituents with a positive inductive (+I) or a mesomeric (+M) effect (4-Me, 4-Cl and 4-MeO) apparently facilitated the reaction (isolated yields 85% or higher), whereas the nitro group, with a strong −M effect, hampered the cross-coupling. Conversely, the outcome of the coupling reactions between benzoyl chloride (10a) and substituted phenylacetylenes (4-Me, 4-MeO and 4-CF 3 ) all proceeded with high isolated yields, in line with the long distance between the substituents in position 4 of the benzene ring and the reaction site, which inevitably minimizes their influence. The coupling of 9a with cinnamoyl chloride also proceeded satisfactorily, producing 11ak in an 87% isolated yield. In contrast, 3-phenylpropanoyl chloride (as a representative of aliphatic acyl chlorides bearing an sp 3 substituent at the acyl group) did not produce any coupling product under analogous conditions. Conversely, pivaloyl chloride was converted into 11am with an acceptable 51% isolated yield. A similar yield was obtained with 2-furoyl chloride, whereas the reaction with 2-thiophenecarbonyl chloride had a lower yield. The ethynylferrocene/benzoyl chloride pair also displayed a rather sluggish reaction, associated with side processes that were partly suppressed by lowering the amount of the acyl chloride.
In addition to spectroscopic characterization, the structure of 11af was determined by single-crystal X-ray diffraction analysis. Figure 7 shows the corresponding molecular structure along with selected interatomic distances and angles.

Methods and Materials
Infrared spectra were recorded in diffuse reflectance mode using a Fourier-transform infrared spectrometer FTIR Nicolet 6700 (Thermo Fisher Scientific, Waltham, MA, USA; (scan range 400-4000

Methods and Materials
Infrared spectra were recorded in diffuse reflectance mode using a Fourier-transform infrared spectrometer FTIR Nicolet 6700 (Thermo Fisher Scientific, Waltham, MA, USA; (scan range 400-4000 cm −1 , 64 scans, 4 cm −1 resolution). The samples analyzed in this study were diluted with KBr (grade for spectroscopy) before the measurement. Nuclear magnetic resonance (NMR) spectra were recorded at 25 • C on a Varian UNITY Inova 400 spectrometer (Palo Alto, CA, USA) operating at 399.95, 100.58 and 376.29 MHz for 1 H, 13 C and 19 F, respectively. Chemical shifts (δ in ppm) are expressed relative to internal tetramethylsilane ( 1 H and 13 C) and to external neat CFCl 3 ( 19 F). GC analyses were performed with an Agilent 6850 gas chromatograph (Santa Clara, CA, USA) equipped with a DB-5 column (0.18 mm diameter, 50 m length).
Elemental composition of the deposited catalysts was determined using the standard combustion method and a PerkinElmer PE 2400 CHN analyzer (Waltham, MA, USA). The content of palladium in solid samples and in the reaction mixtures was determined by inductively coupled plasma optical emission spectroscopy (ICP-OES) on an IRIS Interpid II instrument (Thermo Electron, Waltham, MA, USA) with axial plasma and ultrasonic CETAC nebulizer U-5000AT+. The samples were dissolved in a mixture of HF with HNO 3 (3:2, suprapure from Merck; Kenilworth, NJ, USA) at 50 • C for 15 min and evaporated. The residue was diluted with redistilled water for 105 Pd (the wavelength used for the spectrophotometric analysis was 324.270 nm).
Dichloromethane was dried over potassium carbonate and distilled under argon. Other solvents were dried over activated 3 Å molecular sieves. Triethylamine was dried over sodium metal and distilled. Other chemicals were used as obtained from commercial sources (Sigma-Aldrich, St. Louis, MO, USA). Materials 2-7 were prepared as previously described [29]. The analytical data determined for the newly prepared samples are as follows. The IR spectra were identical to those of the authentic samples.
Elemental Catalyst 8 was prepared similarly by direct palladation of material 1. Thus, palladium(II) acetate (0.449 g, 2.0 mmol) dissolved in dry dichloromethane (10 mL) was added to a suspension of support 1 (2.0 g) [29] in the same solvent (50 mL). After stirring the resulting mixture at room temperature for 1 h, the solid was filtered off and washed with dichloromethane until the washings were colorless. Then, the filter cake was washed a few more times (2-3×) and left to dry in the air. Characterization

Description of the Screening Experiments
A Schlenk tube was successively charged with the catalyst (typically 0.1-0.5 mol.% Pd), CuI (9.5 mg, 5 mol.%; if appropriate), phenylacetylene (102 mg, 1.0 mmol), 4-toluoyl chloride (230 mg, 1.5 mmol) and dodecane (internal standard; 170 mg, 1.0 mmol). The reaction vessel was flushed with nitrogen and sealed. The solvent was introduced (5 mL of pure solvent or 5 mL of a solvent with 697 µL (5 mmol) of triethylamine), and the reaction flask was transferred to a Heidolph Synthesis I parallel reactor pre-heated to the required temperature. Aliquots of the reaction mixture were periodically collected, diluted with saturated aqueous NaHCO 3 and centrifuged at 4000 rpm for 5 min. The organic phase was analyzed by gas chromatography.
During recyclation experiments, the reaction mixture obtained after 2 h at 50 • C was diluted with acetone (5 mL) and cooled on ice. A small amount of the liquid phase was separated and used to determine the conversion. The solids were filtered off, washed successively with acetone, methanol (removal of triethylammonium chloride) and acetone again. The filtrate and washings were combined and used to quantify the amount of leached-out metal. The recovered solid was used in the next catalytic experiments.

Preparative Experiments
A Schlenk tube was charged with the respective alkyne (1.0 mmol) and acyl chloride (1.5 mmol; only 1.0 mmol of the acyl chloride was used in the reaction of ethynylferrocene with benzoyl chloride to avoid decomposition). After flushing the reaction vessel with argon, catalyst 5 (0.5 mol.% Pd) was introduced, followed by dry acetonitrile (5 mL) and anhydrous triethylamine (0.7 mL, ca. 5 mmol). The reaction mixture was stirred at 50 • C for 2 h, diluted with ethyl acetate (10 mL) and cooled on ice. The cold reaction mixture was filtered, and the solid residue was washed with ethyl acetate. The combined organic washings were evaporated under vacuum, leaving a crude reaction product, which was taken up with 1,4-dioxane. Solid NaHCO 3 was added (≈0.1 g), and the resulting mixture was stirred at room temperature for 1-7 days to hydrolyze unreacted acyl chloride. The hydrolyzed reaction mixture was evaporated, and the residue was extracted with ethyl acetate. Organic washings were dried over anhydrous MgSO 4 and evaporated. Analytically pure coupling products were isolated by column chromatography over silica gel using ethyl acetate-hexane (1:10 or 1:20) as the eluent (dichloromethane was used in the case of 11ag).

Conclusions
In summary, we have described the catalytic applications of palladium catalysts deposited over silica gel bearing composite amide-donor functional moieties at the surface in the Sonogashira-type cross-coupling of acyl chlorides with terminal alkynes producing synthetically useful 1,3-disubstituted prop-2-yn-1-ones. The collected data suggest a generally good catalytic performance of these heterogeneous catalysts alone (i.e., without a co-catalyst) in the reactions of aromatic acyl chlorides with aryl alkynes under relatively mild reaction conditions. Nevertheless, a careful optimization is required for achieving good catalytic results, as the catalytic properties are significantly affected by the reaction conditions (solvent and base) and depend on the nature of the functional pendant at the support's surface. Of the tested catalysts, the poorest performance surprisingly exerted catalyst 7 bearing the phosphine moieties, which contrasts with the general notion that phosphine ligands give rise to active cross-coupling catalysts. When recycled, however, the studied catalysts lost their catalytic activity and, therefore, could not be efficiently reused. Very likely, the catalysts serve as a source of catalytically active Pd species that efficiently mediate the cross-coupling reaction but are not redeposited without deactivation.
Supplementary Materials: The following are available online at http://www.mdpi.com/2073-4344/10/10/1186/s1, Table S1: Yields of the coupling product 11ad and the amount of leached-out Pd in the recycling experiments.