Polyethyleneimine-Modified Polymer as an Efficient Palladium Scavenger and Effective Catalyst Support for a Functional Heterogeneous Palladium Catalyst.

The polyethyleneimine-modified polymers, polystyrene-divinylbenzene-based (TAs) and polymethacrylate-based polymers (TAm), were used as palladium scavengers to eliminate residual palladium species after palladium on carbon-catalyzed Sonogashira-type coupling reaction. Since both TAs and TAm indicated relatively favorable elimination abilities toward residual palladium species in the reaction mixture, the affinities of TAs and TAm for palladium species were used as supports for palladium catalysts. The TAm-supported palladium catalyst (Pd/TAm) indicated better catalyst properties for the chemoselective hydrogenation compared to those of the corresponding TAs-supported palladium catalyst (Pd/TAs). Aromatic benzyl ethers; aromatic and aliphatic N-Cbzs; and aromatic carbonyl groups were smoothly hydrogenated in the presence of 1-5 mol % of Pd/TAm in MeOH or 2-PrOH. In contrast, the hydrogenation of aromatic ketones was selectively suppressed in morpholine which act as appropriate catalyst poison and solvent. Furthermore, Pd/TAm-catalyzed chemoselective hydrogenation was applicable to continuous-flow reaction.

Introduction

Chemoselective hydrogenation as an alternative reductive transformation of reducible functionalities in the presence of multiple reducible functionalities is a useful synthetic methodology for the novel short-step synthetic route.[1,2] Heterogeneously catalyzed hydrogenation methods have been widely used as environmentally friendly reactions due to their easy removability from reaction mixtures by the simple filtration, reusability, and reduction of contaminants requiring separation after the reaction.[2d,2e,3,4]

Two main strategies to establish heterogenous palladium (Pd)-catalyzed chemoselective hydrogenation are available. The first is the utilization of appropriate catalyst poisons that partially block the catalytic active site by the moderate coordinating property toward the Pd species[5−7] such as Lindlar’s catalyst.[5] We have developed various Pd/C-catalyzed chemoselective hydrogenation methods using catalyst poisons, such as diphenyl sulfide (Ph2S),[6a,6b] amines,[6c] and ammonium acetate (NH4OAc).[6d] The use of the complexes of Pd/C and coordinable catalyst poisons, such as Pd/C-diphenyl sulfide [Pd/C(Ph2S)],[7a] and Pd/C-ethylenediamine (en) complexes [Pd/C(en)][7b−7e] was effective for chemoselective hydrogenation as heterogeneous catalysts.[7] In contrast, it is well known that the development of chemoselective heterogeneous catalysts is largely dependent on the unique characteristics of catalyst supports.[8−10] We have also previously reported many heterogeneous and chemoselective Pd catalysts immobilized on polyethyleneimine (PEI) (Pd/PEI),[9j,9k] fibroin silk protein (Pd/Fib),[9h,9l−9n] molecular sieves 3A (Pd/MS3A),[9c,9f,9i] boron nitride (Pd/BN),[9c,9e,9g] ceramics (Pd/ceramic),[9b] and monolithic ion-exchange resins (Pd/monolith)[9a] each exhibiting different catalytic activities. Chelate resins bearing metal-coordinating chelate functionalities on a polymer backbone, such as iminodiacetate[10b] and tertiary amine moieties,[10a,10c] are well-known effective supports of Pd for the production of chemoselective hydrogenation catalysts. Chelate resins have been used as scavengers to eliminate residual metals in reaction mixtures to prevent the contamination of the products such as medicines and fine chemicals.[11] The maximum permissible amount of residual Pd in medicines is strictly defined in the guideline ICH Q3D due to its toxicity and function depression properties on functional materials.[12] The information of ICH harmonization for better health is available online: http://www.ich.org/products/guidelines.html. Therefore, the elimination of residual Pd from products and/or reaction solutions in each reaction step is quite important.

Herein, we have evaluated the Pd scavenging abilities of polystyrene–divinylbenzene-based (TAs) and polymethacrylate-based polymers (TAm) modified with polyethyleneimine (PEI, range of MW ca. 200–10 000) as a chelate for the elimination of leached Pd species during the Pd/C-catalyzed Sonogashira-type reaction.[13] Because the TAs and TAm indicated appropriate elimination efficiencies of leached Pd species based on the chelating ability of the PEI moiety, we have also applied TAm as a sufficiently capable support for Pd (Pd/TAm) for chemoselective hydrogenation.

Results and Discussion

Elimination of Leached Pd Species Using TAm and TAs

We have recently developed effective scavenging methods for the elimination of leached Pd species during the Pd/C-catalyzed Sonogashira-type reaction[13] using thiol-modified dual-pore silica beads as a Pd scavenger.[14] Because non-negligible amounts of both two and zero valent Pd species were leached from the heterogeneous Pd/C catalyst during the Pd/C-catalyzed Sonogashira-type reaction using 4′-iodoacetophenone and 3-butyn-1-ol due to the strong coordination of the alkyne moiety of 3-butyn-1-ol,[13] the reaction was suitable for evaluating the Pd scavenging abilities of chelate resins. The Pd scavenging abilities of the PEI-modified resins (TAs and TAm) from the diluted filtrate (23.8 μM of the leached Pd species) during the Sonogashira-type reaction were compared with those of commercially available Pd scavengers based on silica beads modified with propylamine or diethylenetriamine functionalities. The filtrate of 0.4 mol % of 10% Pd/C (85.2 mg, 80.0 μmol)-catalyzed Sonogashira-type coupling reaction of 4′-iodoacetophenone (4.92 g, 20.0 mmol), 1.5 equiv of 3-butyn-1-ol (2.27 mL, 30.0 mmol), and Na3PO4·12H2O (15.2 g, 40.0 mmol) in 2-PrOH/H2O (40/40 mL) at 80 °C for 2 h was extracted with EtOAc (100 mL) and H2O (100 mL), and the organic layer was diluted to 200 mL with EtOAc. The Pd concentration in the diluted EtOAc solution was 23.8 μM and was divided into four equal portions (50 mL each), and 1.00 g of Pd scavengers (TAm and TAs) was added to each portion. The scavenger in each EtOAc solution was removed by simple filtration after stirring at 25 °C for 3 h, and the Pd concentrations in the filtrates were measured via atomic absorption spectrophotometry (AAS) (the Pd detection limit of the AAS analysis: 1 ppm) (Table ). Although the propylamine-modified silica beads were only modestly beneficial for Pd scavenging in the Pd/C-catalyzed Sonogashira-type reaction (23%, entry 1), diethylenetriamine-modified silica beads and TAm and TAs scavengers showed relatively efficient scavenging abilities (41–47%, entries 2–4) due to the appropriate bidentate coordinable activity of the alkyldiamine moieties toward Pd(0) species. The highest Pd scavenging efficiency (47%) was achieved using polyethyleneimine-modified polystyrene–divinylbenzene-based scavenger (TAs, entry 4). The higher scavenging ability is likely because of the higher alkyldiamine moiety content per unit weight of TAs and the auxiliary π-interaction effect toward Pd species of the aromatic rings constituting the polystyrene–divinylbenzene backbone of the TAs. The use of Pd scavengers with high Pd capture capacity per unit weight is essential from the practical and industrial point of view. The diethylenetriamine-modified silica beads indicated the highest Pd scavenging efficiency per amine functional group substituting on the polymer backbone (%/mmol/g) as compared to the other scavengers (entry 2 vs 1, 3, and 4).

Table 1

Comparison of the Pd Scavenging Abilities of TAm, TAs, and Commercially Available Scavengersa

entry	scavenger (mmol/g)b	removed Pd (%)	Pd capture efficiency per amine functional group substituting on the polymer backbone (%/mmol/g)
1	propylamine-modified silica beadsc (1.5–2.0)	23	12–15
2	diethylenetriamine-modified silica beadsc (0.5–1.5)	42	28–84
3	TAm (2.87)	41	14
4	TAs (6.66)	47	7

The Sonogashira-type coupling reaction was performed using a solution of 4′-iodoacetophenone (20.0 mmol, 4.9 g), 1.5 equiv of 3-butyn-1-ol (30.0 mmol, 2.3 mL), 0.4 mol % of 10% Pd/C (80.0 μmol, 85.2 mg), and NaPO3·12H2O (40.0 mmol, 15.2 g) in 2-PrOH/H2O (40 mL/40 mL) stirred at 80 °C for 2 h. The concentration of Pd species in the diluted filtrate of the reaction mixture was measured by atomic absorption spectrometry (AAS, AA-7000, SHIMADZU, Japan), and the Pd detection limit of the AAS analysis is 1 ppm.

The amine functionalities per gram of each scavenger.

Propylamine-modified silica beads and diethylenetriamine-modified silica beads are commercially available from Fujifilm Wako Pure Chemical Corporation, catalog number; http://www.wako-chem.co.jp/english/labchem/product/Org/QuadraSil/index.htm, contact; http://ffwk.fujifilm.co.jp/en/contact/index.html#.

Development of the Pd/TAm Catalyst and Its Application for Chemoselective Hydrogenation

We have previously developed a heterogeneous Pd(0)–polyethyleneimine (Pd/PEI) catalyst for the chemoselective hydrogenation of alkyne functionalities by exploiting the appropriate catalyst poison effect and strong chelating ability of the polyethyleneimine moiety toward Pd(0).[9j,9k] Because the polyethyleneimine-modified TAm and TAs exhibit comparatively efficient Pd-adsorption abilities as Pd scavengers and proportionally lower ratios of polyethylenediamine moieties within the molecule compared to the support (polyethyleneimine) of Pd/PEI,[9j,9k] they were used as the Pd catalyst supports to develop active hydrogenation catalysts due to their moderate coordinable properties toward Pd. Pd species were embedded on TAm and TAs in a manner analogous to our previously established preparation method for the chelate resin-supported heterogeneous Pd catalysts[10] (Scheme ). TAm and TAs were immersed in an EtOAc solution of Pd(OAc)2 at 25 °C and stirred for 24 h under an argon atmosphere. Colorless TAm (A, mean particle diameter; 144 μm) and TAs (A′, mean particle diameter; 710 μm) gradually turned to yellow (B and B′) with decoloring of the dark-orangish EtOAc solution of Pd(OAc)2. The filtered yellowish [Pd(OAc)2-adsorbed] TAm (B) and TAs (B′) were washed using MeOH and H2O, dried under reduced pressure, and subsequently stirred in the aqueous solution of hydrazine monohydrate at 25 °C under argon for 3 h. The resulting pale gray resins were collected by filtration, washed with MeOH and H2O, and dried under reduced pressure to afford Pd/TAm (C) and Pd/TAs (C′). The Pd contents of the products were both 3.3 wt % based on the concentration of the residual Pd(OAc)2 in the EtOAc filtrate determined by AAS (the Pd detection limit of the AAS analysis: 1 ppm).

Scheme 1

Preparation of 3.3% Pd/TAm and 3.3% Pd/TAs

We first evaluated the catalytic activity for the hydrogenation of diphenylacetylene (1) using the 3.3% Pd/TAm and 3.3% Pd/TAs (Table ). Although 1 was absolutely hydrogenated to afford corresponding diphenylethane (2) in the presence of 3.3% Pd/TAm (1 mol %) under atmospheric H2 conditions in MeOH at 25 °C for 2.5 h (entry 1), the 3.3% Pd/TAs-catalyzed hydrogenation of 1 was incomplete even after 24 h under the same conditions (entry 2). Furthermore, 5% Pd/PEI catalyst as a chemoselective semihydrogenation catalyst of alkynes[9j,9k] exhibited the low catalyst activity for the hydrogenation of 1 under nearly the same conditions as compared to the 3.3% Pd/TAm catalyst (entry 3 vs 1). Therefore, the 3.3% Pd/TAm catalyst was selected as the most promising heterogeneous Pd catalyst for further screening.

Table 2

Comparison of the Catalyst Activities of 3.3% Pd/TAm, Pd/TAs, and Pd/PEI

entry	catalyst	time (h)	ratio (1:2:3:4)a
1	3.3% PdTAmb	2.5	0:100:0:0
2	3.3% Pd/TAsb	24	2:92:2:4
3	5% Pd/PEIc	24	0:30:2:68 (ref (9j, 9k))

The products ratio was determined by 1H NMR.

Diphenylacetylene (0.25 mmol, 44.6 mg) and 1 mol % of 3.3% Pd/TAm or 3.3% Pd/TAs (2.5 μmol, 8.1 mg) in MeOH (1.0 mL) were stirred at 25 °C under hydrogen atmosphere.

Diphenylacetylene (1.0 mmol, 178.2 mg) and 0.83 mol % of 5% Pd/PEI (8.3 μmol, 17.8 mg) in MeOH (2.0 mL) were stirred at 25 °C under hydrogen atmosphere.

Next, the 3.3% Pd/TAm catalyst was analyzed in detail using high-angle annular dark-field scanning transmission electron microscopy (HAADF-STEM), electron probe microanalysis (EPMA), and X-ray photoelectron spectroscopy (XPS). The mean particle size of the Pd clusters was estimated to be approximately 2.7 nm with a standard deviation of 1.5 nm based on the HAADF-STEM image (Figure a). EPMA indicated that the Pd clusters were uniformly distributed in all regions of the 3.3% Pd/TAm catalyst (Figure b,c). Regions with relatively high Pd concentration in 3.3% Pd/TAm catalyst are shown in red in Figure c. Furthermore, it was clear that the 3.3% Pd/TAm catalyst was composed of a combination of approximately equal amounts of Pd(0) metal (characteristic peaks at ca. 340.2 and 335.1 eV corresponding to Pd 3d3/2 and Pd 3d5/2) and Pd(II) (characteristic peaks at ca. 342.9 and 337.2 eV corresponding to Pd 3d3/2 and Pd 3d5/2) by XPS analysis, despite the hydrazine reduction during catalyst preparation (Scheme ). The strong coordination ability of the polyethyleneimine moiety on TAm toward Pd(II) partially suppressed the reduction of Pd(II) to Pd(0) due to the enhanced electron density of Pd(II).[10c] Furthermore, XPS analysis of the 3.3% Pd/TAm after hydrogenation revealed that the content ratio of Pd(II) and Pd(0) was almost entirely maintained before and after use (Figure d,e). These results strongly supported that no reduction of Pd(II) to Pd(0) species by PEI and/or hydrogen gas has occurred during hydrogenation. The characteristic peaks of Pd 3d3/2 and Pd 3d5/2 and the area% of Pd(II) ions and Pd(0) species are indicated in the Supporting Information (Figure S1).

Figure 1

(a) HAADF-STEM image, (b) EPMA line profile, (c) EPMA element mapping, and (d, e) XPS spectra of the 3.3% Pd/TAm catalyst before and after use, respectively.

The hydrogenation catalytic activity of 3.3% Pd/TAm was evaluated at 25 °C under atmospheric H2 conditions in MeOH (Table ). Alkyne (entry 1), azide (entry 2), nitro (entry 3), N-Cbz (entries 4 and 5), alkene (entries 6 and 12), as well as aromatic and aliphatic benzyl ester functionalities (entries 6 and 7, respectively) were smoothly hydrogenated. p-Anisaldehyde was selectively hydrogenated to the corresponding benzyl alcohol without hydrogenolysis of the benzylic hydroxyl group (entry 8). In contrast, 4-hydroxyphenyl benzyl ether as an aromatic benzyl ether was debenzylated to afford the corresponding hydroquinone (entry 9), and 2-(benzyloxy)ethanol as an aliphatic benzyl ether was tolerated under the same hydrogenation conditions (entry 10). The reductive ring opening reaction of trans-stilbene oxide was observed to quantitatively afford 1,2-diphenylethanol without hydrogenolysis of the benzylic hydroxyl group (entry 11). tert-Butyldimethylsilyl ether was tolerated under the same hydrogenation conditions, whereas a co-existing olefin was selectively hydrogenated (entry 12).

Table 3

3.3% Pd/TAm-Catalyzed Hydrogenation of Various Reducible Functionalitiesab

Substrate (0.25 mmol) and 1 mol % of the 3.3% Pd/TAm catalyst (2.5 μmol, 8.1 mg) in MeOH (1 mL) stirred at 25 °C under a hydrogen atmosphere.

Recovery of the substrate.

Furthermore, the 10 mmol scale hydrogenation of 1 (1.8 g; 10 mmol) was successful, and Pd leaching was not detected in the reaction mixture by AAS (below the Pd detection limit of 1 ppm) (eq ).The catalytic activity of 3.3% Pd/TAm for the hydrogenation of aromatic ketones was relatively low, and the hydrogenation of isobutyrophenone (5a) was incomplete in MeOH at 25 °C for 24 h (Table , entry 1; standard conditions). The hydrogenation of 5a was achieved in 2-PrOH solvent at 50 °C to afford the corresponding benzyl alcohol (6a) without further hydrogenolysis of the benzylic hydroxyl group by increasing the loading of 3.3% Pd/TAm to 5 mol % (entry 2). The total suppression of hydrogenation of the aromatic ketone functionality of 5a was achieved using morpholine solvent in the presence of 1 mol % of 3.3% Pd/TAm due to the moderate poisonous effect of morpholine (entry 3). In contrast, the hydrogenation of 5a catalyzed by 5% Pd/C or 7% Pd/WA30[10a,10c] with tertiary amine functionalities on the polystyrene–divinylbenzene skeleton of WA30 was incompletely suppressed in morpholine (entries 3 vs 4 and 5). Acetophenone (5b) was selectively hydrogenated to the corresponding benzyl alcohol (6b) in the presence of 5 mol % of the 3.3% Pd/TAm catalyst (entry 6), whereas the reduction of 5b was completely suppressed in the morpholine solvent (entry 7).

Table 4

Catalytic Activity Control of 3.3% Pd/TAm for the Hydrogenation of Aromatic Ketone Derivativesab

Reaction conditions: substrate (0.25 mmol) and Pd catalyst in MeOH, 2-PrOH, or morpholine (1 mL) stirred at 25 °C for 24 h under a hydrogen atmosphere.

The ratio was determined by 1H NMR and isolated yields indicated in parentheses.

Furthermore, the 3.3% Pd/TAm-catalyzed hydrogenation of the compound (8) containing alkyne, benzyl ether and aromatic ketone functionalities was investigated (Scheme ). Although the all reducible functionalities of 8 were completely hydrogenated to afford the corresponding ethylbenzene derivative (9) in a quantitative yield using a 10% Pd/C catalyst (top of Scheme ), the 3.3% Pd/TAm-catalyzed hydrogenation of 8 in MeOH at 50 °C chemoselectively formed 10 in excellent yield (93%) without hydrogenolysis of the benzylic hydroxy group or benzyl ether (middle of Scheme ). Furthermore, only the alkyne functionality was chemoselectively hydrogenated, and the corresponding aromatic ketone and benzyl ether survived in morpholine solvent (11, the bottom of Scheme ).

Scheme 2

Chemoselective Hydrogenation of an Acetophenone Derivative Possessing Various Reducible Functionalities

Reuse test of the 3.3% Pd/TAm catalyst was tested under the same hydrogenation conditions using benzyl benzoate as a substrate (Table ). The catalyst was quantitatively recovered by simple filtration and reused at least 5 times. Although the turnover frequency (TOF) of 3.3% Pd/TAm catalyst in each reaction was gradually decreased with the prolongation of the reaction time due to physical abrasion of the catalyst by stirring, the hydrogenation of benzyl benzoate was completed to give the corresponding benzoic acid in moderate turnover numbers (TON) of 3.3% Pd/TAm catalyst (TON; 98–100).

Table 5

Reuse Testab

entry	yield of the product (%)	time (h)	recovered yield of the Pd catalyst (%)	TON	TOF (h–1)
1st	98	4.5	98	98	21.8
2nd	quant.	5.5	99	100	18.2
3rd	quant.	9	quant.	100	11.1
4th	quant.	24	97	100	4.2
5th	quant.	24	98	100	4.2

Benzyl benzoate (2.0 mmol, 425 mg) and 1 mol % of the 3.3% Pd/TAm catalyst (20.0 μmol, 64.8 mg) in MeOH (8 mL) stirred at 25 °C under a hydrogen atmosphere.

Isolated yields were indicated in parentheses.

A comparison of the catalytic activities of 3.3% Pd/TAm, Pd/C, and several heterogeneous Pd catalysts previously developed for chemoselective hydrogenations in our laboratory[1a−1e,6,7,9,10] is shown in Figure . Each catalyst surrounded by a square catalyzed the hydrogenation of the reducible functionalities below each line, and various reducible functionalities surrounded by a thick-frame were hydrogenated using 3.3% Pd/TAm. It should be noted that the 3.3% Pd/TAm-catalyzed hydrogenation of aromatic ketones could be controlled using a 2-PrOH or morpholine solvent.

Figure 2

Comparison of the hydrogenation catalytic activity of 3.3% Pd/TAm, Pd/C, and several previously developed chemoselective heterogeneous Pd catalysts.

Next, we have applied 3.3% Pd/TAm as a solid in the form of beads (ca. 144 μm diameter) to continuous-flow hydrogenation reactions utilizing the avoidable shape of the unfavorable back pressure. A 0.05 M MeOH solution of benzyl benzoate (13.5 mmol in 270 mL) was flowed through a catalyst cartridge (φ 4.6 × 50 mm) filled with 3.3% Pd/TAm (100 mg) once at 0.4 mL/min and 50 °C under an atmosphere of hydrogen gas 10 mL/min: (eq ). The continuous-flow system using 3.3% Pd/TAm was successfully applied for the gram scale reaction without the deterioration of catalytic activity due to the physical damage- and attrition-free conditions resulting from the stirring (eq vs Table ). Higher TON and TOF could be achieved in a continuous-flow hydrogenation reaction using a 3.3% Pd/TAm cartridge (TON; 409.4 and TOF; 36.4 h–1, eq vs Table ). Furthermore, the 3.3% Pd/TAm-catalyzed flow hydrogenation of 8 was investigated (eq 3). Although product 10 was obtained under batch conditions in MeOH at 50 °C (Scheme , middle), only the alkyne functionality was chemoselectively hydrogenated to afford 11 in nearly quantitative yield during the single pass through the cartridge filled with the 3.3% Pd/TAm catalyst (TON; 15.8 and TOF; 38.0 h–1, eq 3).

Conclusions

We have demonstrated that polyethylene-modified TAm and TAs show beneficial Pd scavenging activities compared to those of commercially available Pd scavengers (diethylenetriamine-modified silica beads) for the elimination of leached Pd in the filtrate of the ligand- and copper-free Pd/C-catalyzed Sonogashira-type reaction.[13] Because the TAm and TAs indicated comparable removal (acquisition) capabilities for Pd species due to the adequate coordinable property of the polyethyleneimine moiety, TAm was used as a support of the heterogeneous Pd catalyst (3.3% Pd/TAm) for the development of a novel chemoselective hydrogenation catalyst. The optimized catalyst, 3.3% Pd/TAm, was developed as a recoverable, reusable and Pd-leaching free catalyst for chemoselective hydrogenation. Furthermore, 3.3% Pd/TAm was amenable to the continuous-flow hydrogenation reactions due to its high durability and unique chemoselectivity.

Experimental Section

Material

Pd(OAc)2 and 10% Pd/C were obtained from N.E. Chemcat Corporation, Japan. Polyethyleneimine (PEI)-functionalized resins (TAm and TAs) were obtained from Mitsubishi Chemical Corporation, Japan. Other reagents were obtained from commercial sources and used without further purification unless otherwise noted. 1H and 13C NMR spectra were recorded by a JEOL ECZ-400 (1H: 400 MHz, 13C: 100 MHz), AL-400 (1H: 400 MHz, 13C: 100 MHz), or ECA-500 spectrometer (1H: 500 MHz, 13C: 125 MHz). Chemical shifts (δ) are indicated in ppm and are internally referenced (0.00 ppm for tetramethylsilane for 1H NMR, 7.26 ppm for CDCl3 for 1H NMR, and 77.0 ppm for CDCl3 for 13C NMR, 2.05 ppm for acetone-d6 for 1H NMR, 206.4 ppm for acetone-d6 for 13C NMR). IR spectra were recorded by a Brucker FT-IR ALPHA. ESI high-resolution mass spectra were measured by a Shimadzu hybrid IT-TOF mass spectrometer. High-angle annular dark-field scanning transmission electron microscopy analyses (HAADF-STEM) were measured by a FEI TECNAI G2 F20. Electron probe microanalyses (EPMA) were measured by a JEOL JXA-8100. X-ray photoelectron spectroscopy (XPS) of 3.3% Pd/TAm was measured by ULVAC Quantera-SXM-GC. The Pd content of the AcOEt filtrates was measured by atomic absorption spectrometry (SHIMADZU AA-7000).

Experimental

Elimination Study of the Pd Species Leached during the Pd/C-Catalyzed Sonogashira-Type Reaction Using Four Kinds of Pd Scavengers (Table )

To a solution of 4′-iodoacetophenone (4.92 g, 20.0 mmol), dry-type 10% Pd/C (85.2 mg, 80.0 μmol) and Na3PO4·12H2O (15.2 g, 40.0 mmol) in 2-PrOH (40 mL), and H2O (40 mL) was added 3-butyn-1-ol (2.27 mL, 30.0 mmol) at 25 °C under an argon atmosphere. The reaction mixture was stirred at 80 °C (external oil bath temperature), and the reaction progress was monitored by thin-layer chromatography. After completion of the reaction (2 h), the reaction mixture was cooled to 25 °C and filtered through celite to remove Pd/C and insoluble salts. The filtrate was extracted with additional EtOAc (100 mL) and water (100 mL), and the organic layer was diluted to 200 mL with EtOAc (Pd concentration: 23.8 μM) and divided to four equal portions (each 50 mL). Subsequently, each 1.00 g of propylamine-modified silica beads, triamine-modified silica beads, TAm and TAs were added to each divided EtOAc solution (50 mL) as Pd species scavengers, respectively, and stirred for 3 h at 25 °C. Each reaction mixture was filtered to remove the scavenger. Pd concentration in each filtrate was measured by atomic absorption spectrometry. Elimination (23, 42, 41, and 47%) of Pd species could be achieved.

Preparation of Pd/TAm and Pd/TAs (Scheme )

A suspension of dry TAm (1.00 g, colorless powder) in EtOAc solution (40 mL) of Pd(OAc)2 [72.0 mg, 321 μmol (34.1 mg, palladium quantity)] was stirred under an argon atmosphere at 25 °C for 1 day. The resulting yellow solid was collected by filtration (1 mm filter paper), washed with H2O (5 mL × 5) and MeOH (5 mL × 5), and dried in vacuo for 12 h. The filtrate was transferred to a 100 mL volumetric flask and diluted to 100 mL with H2O, and 8.14 ppm (814 μg) of palladium species was observed in the diluted filtrate by the use of Atomic Absorption Spectrometry (SHIMADZU AA-7000). The collected solid was then stirred with NH2NH2·H2O (24.0 μL, 481 μmol) in H2O (40 mL) under an argon atmosphere at 25 °C for 4 h. The pale gray solid was collected by filtration (1 mm filter paper), washed with H2O (5 mL × 5) and MeOH (5 mL × 5), and dried in vacuo for 12 h to give Pd/TAm (999.0 mg). The filtrate was transferred to a 100 mL volumetric flask and diluted to 100 mL with H2O, and 0.35 ppm (35.0 μg) of palladium species was observed in the diluted filtrate. Since the total palladium species which was not absorbed on TAm were 849 μg, the palladium content of Pd/TAm was estimated to be 3.3% [(34.1 – 0.849)/999 × 100].

Pd/TAs were also prepared using TAs (166.6 mg, white solid), and Pd(OAc)2 [11.2 mg, 50.0 μmol (5.3 mg, palladium quantity)] in a method similar to prepare 3.3% Pd/TAm and 156.1 mg of Pd/TAs was obtained. Since the total free palladium species in the filtrates after the palladium absorption and subsequent reduction process were 21.5 μg, the palladium content of Pd/TAs was estimated to be 3.3% [(5.3 – 0.022)/156 × 100].

General Procedure for Chemoselective Hydrogenation (Tables 2–4)

A mixture of a substrate (250 μmol) and 3.3% Pd/TAm or Pd/TAs (8.1 mg, 2.5 μmol) in MeOH, 2-PrOH, or morpholine (1 mL) was stirred at 25 or 50 °C using a test tube equipped with H2 balloon. The reaction was continuously monitored by thin-layer chromatography. After the specific time as indicated in Tables –4, the mixture was filtered by membrane filter (pore size: 0.45 μm). The catalyst on the filter was washed with EtOAc (5 mL × 3). The combined filtrates were concentrated in vacuo to afford the corresponding analytically pure product (1H and 13C spectra are described in the Supporting Information).

Reuse Test of 3.3% Pd/TAm for Hydrogenation (Table 5)

The first run shown in Table : a mixture of benzyl benzoate (425 mg, 2.0 mmol) and 3.3% Pd/TAm (64.8 mg, 2.0 μmol) in MeOH (8 mL) was stirred at 25 °C under atmospheric hydrogen atmosphere (balloon) conditions. The mixture was filtered using a funnel (1 mm filter paper) after 4.5 h reaction. The catalyst on the filter was washed with EtOAc (3 mL × 5), and the combined filtrates were concentrated in vacuo to afford 240 mg of benzoic acid (1.97 mmol, 98%). The recovered catalyst was dried at 25 °C in vacuo overnight (63.7 mg, 98% recovery). The second run reaction was carried out in analogy with the first run except for the usage of benzyl benzoate (425 mg, 2.0 mmol) and 3.3% Pd/TAm (63.7 mg, 2.0 μmol) for 5.5 h, and benzoic acid was obtained in quantitative yield (252 mg, 2.0 mmol). The 3–5 run reactions were also carried out in a manner similar to the first run except for the usage of the substrate and catalyst in association with a slight loss of the catalyst during the recovery by filtration, and the reaction progress was monitored by thin-layer chromatography. The detailed results are described in the Supporting Information.

Gram Scale Hydrogenation in Continuous-Flow Conditions (eq )

A solution of benzyl benzoate (2.87 g, 13.5 mmol) in MeOH (270 mL, 0.05 M) was pumped at 0.4 mL/min into the catalyst-packed cartridge [3.3% Pd/TAm (100.0 mg); φ 4.6 × 50 mm] together with a flow rate of 10 mL/min hydrogen gas at 50 °C after the flow of MeOH and hydrogen gas to the cartridge under the same conditions for ca. 10 min. The whole reaction mixture was collected and concentrated in vacuo to give the corresponding benzoic acid (1.55 g, 12.7 mmol) in 94% yield as a colorless solid.

General Procedure for Chemoselective Hydrogenation in Continuous-Flow Conditions (eq 3)

A solution of the substrate (250 μmol) in MeOH (10 mL, 0.05 M) was pumped at 0.4 mL/min into the catalyst-packed cartridge [3.3% Pd/TAm (100.0 mg); φ 4.6 × 50 mm] together with a flow rate of 10 mL/min hydrogen gas at 50 °C after the flow of MeOH and hydrogen gas to the cartridge under the same conditions for ca. 10 min. The whole reaction mixture was collected and concentrated in vacuo to give the corresponding hydrogenated product. If necessary, the product was further purified by silica gel-column chromatography (hexane/EtOAc).