The water used in the experiments was deionized. Aryl bromides and arylboronic acids were purchased from Alfa Aesar. Other chemicals were obtained commercially and used without purification. ¹H nuclear magnetic resonance (¹H NMR) spectra were recorded with a Bruker Avance II 400 MHz spectrometer, using tetramethylsilane (TMS) as an internal standard. All the products were isolated by short chromatography on a silica-gel (200-300 mesh) column, using petroleum ether (60-90 °C) as the eluent, unless otherwise stated. Compounds that had previously been described in the literature were characterized by comparison of their ¹H NMR spectra with the reported data.

An aryl halide (0.5 mmol), arylboronic acid (0.75 mmol), (i-Pr)₂NH (1.0 mmol), Pd/C (1.5 mol%, 16 mg), and water (1.0 mL) were allowed to react at 100 °C. After a certain period, the reaction mixture was added to brine (10 mL) and extracted with ethyl acetate (3 × 10 mL). The solvent was concentrated under vacuum, and the product was isolated by short chromatography on a silica-gel (200-300 mesh) column.

For the first run, a mixture of 4-bromobenzonitrile (0.5 mmol), 4-methylphenylboronic acid (0.75 mmol), (i-Pr)₂NH (1.0 mmol), Pd/C (1.5 mol%, 16 mg), and water (1.0 mL) was allowed to react at 100 °C for the indicated time. The system was cooled to room temperature and the reaction mixture was passed through a membrane filter (0.45 µm), and washed sequentially with ethanol (15 mL) and water (5 mL). The filter residue was naturally air dried for 24 h for the next reaction cycle. The filtrate was concentrated under vacuum, and then added to brine (10 mL) and extracted with ethyl acetate (3 × 10 mL). The residue was purified by flash column chromatography on silica gel (200-300 mesh) to give the corresponding biaryl compound.

The cross-coupling of 4-bromoanisole and phenylboronic acid in the presence of 1.5 mol% Pd/C at 100 °C in water was chosen as the model reaction. The influence of various bases on the cross-coupling reaction was investigated; the results are summarized in Table 1. Initially, a series of typical organic bases was examined (Table 1, entries 1-8). Triethylamine or diisopropylethylamine, which are often used as bases in Pd-catalyzed cross-coupling reactions ^{[26, 27]}, showed similar activity in the present protocol (Table 1, entries 1 and 2). Diisopropylamine and isopropylamine gave very different results (Table 1, entries 3 and 4). It is clear that diisopropylamine is the most efficient base in this catalytic system, providing 93% isolated yield of the cross-coupled product in 20 min. Tripropylamine and dipropylamine were inefficient (Table 1, entries 5 and 6), even though their structures are similar to that of diisopropylamine. The cyclic amine piperazine and 1,4-diazabicyclo[2.2.2]octane (DABCO) were also tested in this catalytic system (Table 1, entries 7 and 8). Lower yields were obtained with the organic alkali sodium methoxide (Table 1, entry 9). Common inorganic bases gave disappointing results in the present protocol (Table 1, entries 10-12), even though they are soluble in water and have been successfully used in efficient ligand-free catalytic systems ^[28]. In this Pd/C-catalyzed system, the reason for the high efficiency is probably that diisopropylamine acts not only as a base but also as a ligand, and has a high tendency to coordinate to Pd to form active species, as reported by Boykin et al. ^[26] and as suggested in our previous work ^[27].

With the optimized conditions in hand, we further explored the scope and limitations of this catalytic system; the results are listed in Table 2. Various aryl bromides bearing electron-withdrawing groups, such as nitro, nitrile, and acetyl moieties, and electron-donating groups, such as hydroxyl, methyl, and methoxy moieties, coupled with arylboronic acid with almost complete conversion within 45 min (Table 2, entries 1-7). Arylboronic acids with electron-withdrawing groups such as fluoro and nitrile coupled with 4-bromoanisole to give good to excellent yields on extension of the reaction time (Table 2, entries 8 and 9). This catalytic system followed the basic rule of the Suzuki reaction, i.e., that electron-withdrawing groups on the aryl bromides and electron-donating groups on the arylboronic acids accelerate the rates of the cross-coupling reactions (Table 2, entries 3 and 9). When the steric effect of the aryl bromide or arylboronic acid increased, the cross-coupling reaction proceeded smoothly within 2 h (Table 2, entries 10-14). Increasing the steric effect of both the aryl bromide and arylboronic acid decreased the reactivity of the cross-coupling, but 98% of isolated yield was obtained in 3 h for the reaction between 2-bromobenzonitrile and o-tolylboronic acid (Table 2, entry 15). Overall, this catalytic system tolerated a broad range of aryl bromides and arylboronic acids - not only aryl bromides with hydrophilic groups but also with hydrophobic groups.

Table 2 Pd/C-catalyzed cross-coupling of arylbromides with arylboronic acids. Reaction conditions: aryl bromide 0.5 mmol, arylboronic acid 0.75 mmol, Pd/C 1.5 mol%, (i-Pr)2NH 1.0 mmol, H2O 1.0 mL, 100 °C, in air. a The time was not optimized.

Table 2 Pd/C-catalyzed cross-coupling of arylbromides with arylboronic acids.

Aryl-substituted pyridines are the most common N-heteroaryl units present in pharmaceutically active compounds. To extend the scope of this methodology, we further investigated the Suzuki reactions of pyridyl bromides with arylboronic acids. 3-Pyridyl bromides were successfully coupled with phenylboronic acid in excellent yields (Table 2, entries 16 and 17). 2-Pyridyl bromide was less active in coupling with arylboronic acids using this catalytic system, providing 50%-86% isolated yields within 24 h (Table 2, entries 18-20). However, the present method is more efficient than our previously result ^[27] for the cross-coupling of 2-pyridyl bromide with phenylboronic acid, which did not activate 2-pyridyl bromide at all. It should be noted that this catalytic system was also suitable for sulfur-containing heteroaryl bromide. The reaction between 2-bromothiophene and phenylboronic acid gave 83% yield after 4 h (Table 2, entry 21).

The reusability of Pd/C is a great advantage in process chemistry, both economically and environmentally ^{[29, 30, 31, 32, 33, 34]}. We examined the recyclability of Pd/C in the cross-coupling reaction, using 4-bromobenzonitrile (0.5 mmol) and 4-methylphenylboronic acid (0.75 mmol) as substrates, in the presence of Pd/C (1.5 mol%) and diisopropylamine (two equiv.) at 100 °C in water (1 mL). As shown in Table 3, the recovered Pd/C gave complete conversion within a slightly prolonged period (60 min) for the second run. The catalyst was reused three times without significant loss of catalytic efficiency. From the fourth to the sixth runs, the catalyst still provided 85%, 81%, and 79% isolated yields, respectively, but the efficiency decreased in the seventh run. To identify the actual active catalyst in the Suzuki reaction, Pd/C was pretreated with (i-Pr)₂NH in neat water for 20 min under the conditions described in Table 3, and passed through a 0.45 μm membrane filter. The filtrate was then used for the Suzuki reaction of 4-bromoanisole with phenylboronic acid. However, the reaction was sluggish, and only a trace amount of cross-coupled product was observed after 20 min. These results suggest that the Pd/C-catalyzed Suzuki reaction proceeds heterogeneously, and the active component is Pd located on the carbon; this is consistent with Sajiki’s report ^[29]. The cause of the decrease in catalytic activity is under investigation.