Carbon–carbon bond formation reactions, especially palladium-catalyzed ones, are significant reactions in organic chemistry. Among all the palladium-catalyzed carbon–carbon bond formation reactions, Heck and Suzuki reactions have attracted interests due to their high applications in various organic syntheses (Bahadorikhalili and Mahdavi, 2018; Christoffel and Ward, 2018; Mpungose et al., 2018; González-Sebastián and Morales-Morales, 2019; Sherwood et al., 2019; Tashrifi et al., 2019; Easson et al., 2020; Fusini et al., 2020). Regarding the unique advantages of Heck and Suzuki reactions, these reactions are used for the synthesis of several compounds with complex chemical structures (Bahadorikhalili et al., 2018a; Bahadorikhalili et al., 2020; Reza Hajipour and Khorsandi, 2020; Yao et al., 2020; Zhou et al., 2020). Therefore, several efforts have been focused on the introduction of novel catalysts with improved characteristics (Hajipour et al., 2018; Ma et al., 2018).

An interesting approach for designing novel catalysts for Heck and Suzuki reactions is the immobilization of palladium onto modified nanoparticles. This method enables both the advantages of homogenous and heterogenous catalysts. Several nanoparticles including nanosilica (Isfahani et al., 2013; Khalafi-Nezhad and Panahi, 2014), nitrogen-rich polymers (Targhan et al., 2020), chitosan (Kumari et al., 2017), magnetic iron oxide (Bahadorikhalili et al., 2015; Ma’mani et al., 2014), and graphene oxide (Kumari et al., 2019) have been used as a support for palladium catalysts. Mesoporous silica nanoparticles (MSNs) are an interesting ordered porous material made of silica with high surface area and unique physical and chemical properties. MSNs are chemically and physically stable and are compatible with biological systems (Mehmood et al., 2017; Narayan et al., 2018; Jeelani et al., 2020). MSNs have been used in various applications, including catalysis (Bahadorikhalili et al., 2018b; Sayahi et al., 2020), drug delivery (Bharti et al., 2015; Moreira et al., 2016), tissue engineering (Rosenholm et al., 2016), and pollutant removal from air and solutions (Li et al., 2016; Gao et al., 2017; Zarandi, 2019). Based on the high surface area and stability and ease of functionalization of MSNs, this family of nanoporous materials has extensively been used as a support for catalysts in various reactions (Han et al., 2006; Lee et al., 2009; Sarkar et al., 2019; Xu et al., 2019).

In spite of many reports on the immobilization of the catalytic agents onto modified nanoparticles, several of them hold a number of disadvantages. It is critical to investigate this field and synthesize catalysts which are efficient and selective and, in an ideal case, have the possibility of use in green solvents. As an interesting nanoparticle for the immobilization of the catalyst, MSNs are of high interest due to their advantages, including high surface area, high thermal stability, and high surface-active groups, which could be used for surface modifications. In this study, we introduce a novel palladium catalyst supported onto dipyrido (3,2-a:2′,3′-c) phenazine–modified MSNs. The catalytic activity of Pd@Py₂PZ@MSNs is evaluated in Heck and Suzuki reactions. It should be noted that the palladium should be in the 0 oxidation state to catalyze the reaction (Zhang et al., 2006). Based on that, the catalyst should have been reduced to Pd (0) for efficiently catalyzing the reaction (Xiao et al., 2021).

In this study, SBA-15 MSNs were synthesized using the hydrothermal method and modified by (3-aminopropyl) methoxysilane. Dipyrido (3,2-a:2′,3′-c) phenazine (Py₂PZ) was synthesized in a one-step reaction between 3,4-diaminobenzoic acid and 1,10-phenanthroline-5,6-dione, which enables it to react with amine groups on the surface of MSNs via an amidation reaction to yield Py₂PZ@MSNs. The Pd@Py₂PZ@MSN catalyst was prepared using Py₂PZ@MSNs as a support, and it was fully characterized by various characterization techniques. The synthesis steps are presented in Scheme 1.

Electron microscopy images of the Pd@Py₂PZ@MSN catalyst are shown in Figure 1. The hexagonal meso-structure of the MSN backbone could be observed in the TEM image that Pd NPs appear as dark zones (Figure 1A). The rod-like structure of MSNs could clearly be observed in the SEM image (Figure 1B). In addition, the EDS analysis of the Pd@Py₂PZ@MSN catalyst confirms the presence of Pd in the structure of the catalyst (Figure 1C). As seen in Figure 1D , the XRD pattern of the Pd@Py₂PZ@MSN catalyst confirms that the crystallinity of the catalyst is similar to that of SBA-15 (Hu et al., 2016). In addition, the high-angle XRD confirms the presence of Pd NPs by two peaks of 110 and 200 planes at 2θ of 39.2 and 44.3°, respectively (Figure 1D).

To confirm the successful synthesis of Pd@Py₂PZ@MSNs, the FT-IR spectra of nanomaterials are compared (Figure 2A). In the FT-IR spectra of Pd@Py₂PZ@MSNs, a peak at 1084 cm⁻¹ represents Si–O vibrations in the structure of the catalyst. The hydroxyl groups of the catalyst could be observed at 3450 cm⁻¹, and the CH₂ stretching vibration is presented at 2927 cm⁻¹. The peak that appeared at 1739 cm⁻¹ is attributed to the C=O bond found in the ester moiety in the structure of the Pd@Py₂PZ@MSN catalyst.

The organic content in the structure of the catalyst was studied using TGA (Figure 2B). The TGA curve shows that the catalyst was thermally stable up to 250°C. A weight loss at the temperature of 250–350°C could be observed, which could be correlated to the degradation of organic moieties in the structure of the Pd@Py₂PZ@MSN catalyst. The palladium content in the structure of the catalyst was measured using ICP analysis. The results showed that the Pd content in the Pd@Py₂PZ@MSN catalyst was 0.12 mmol g⁻¹.

The surface area and pore size of the Pd@Py2PZ@MSN catalyst were characterized using BET analysis and compared with those of MSNs. Based on the BET and nitrogen adsorption–desorption results, Py₂PZ@MSNs have a high surface area of 612.37 m² g⁻¹. In addition, the pore diameter and pore volume of the synthesized nanoporous support are 5.14 nm and 0.72 cm³ g⁻¹, respectively. The surface area, pore width, and pore volumes of MSNs are 655.37 m² g⁻¹, 5.75 nm, and 0.79 cm³ g⁻¹, respectively. The decrease in surface area, pore width, and pore volume of MSNs by the synthesis of the catalyst is expected, while functionalization of the nanoparticles leads to such changes. The surface area and pore size results of MSNs and the Pd@Py₂PZ@MSN catalyst are presented in Table 1.

After the characterization of the Pd@Py₂PZ@MSN catalyst, the efficiency of the catalyst was examined in carbon–carbon bond formation reactions. For this purpose, the Pd@Py₂PZ@MSN catalyst was used in Heck and Suzuki reactions. Initially, the optimal reaction conditions were evaluated by performing the reaction under different reaction conditions including various solvents, bases, catalysts, and times of the reactions. To this end, the reaction of styrene and phenyl bromide was selected as a model reaction. The optimization results are presented in Table 2. According to the results, the best solvent for the reaction is water. However, the yields of the product were less in other solvents. Therefore, water was selected as the best solvent for the reaction. In addition, the results showed that the best yield of the product is obtained when the reaction is performed in the presence of 1 mol% of the catalyst. It should be noted that the presence of a base is critical for reaction performance. Therefore, the reaction was performed in the presence of several bases such as triethyl amine (TEA), NaOH, KOH, and pyridine. The results showed that TEA was the best base for the reaction. To study the role of the Pd@Py₂PZ@MSN catalyst in the reaction, a blank run was performed by all the reaction components, except the catalyst. No product was obtained in the absence of the catalyst. Therefore, the optimized reaction conditions were selected to be water as the solvent and 1.5 equivalent of triethyl amine base in the presence of 1.0 mol% of the Pd@Py₂PZ@MSN catalyst.

To rule out the presence of leached and homogenous Pd in the reaction mixture, in a run, the reaction was performed under the optimal reaction condition. After 60 min, the conversion of the substrates to the product measured by GC was 67%. Then, the catalyst was separated from the reaction mixture, and the filtrate was stirred under the same conditions for more 60 min. The analysis of the product by GC showed that no more product has been obtained. These observations confirmed that the Pd catalyst worked in the heterogenous phase and the presence of the Pd@Py₂PZ@MSN catalyst was necessary for reaction performance.

Based on the optimized conditions, the scope and generality of the Pd@Py₂PZ@MSN catalyst were studied in Heck and Suzuki reactions. The structures and yields of the products are presented in Table 3. The high efficiency of the Pd@Py₂PZ@MSN catalyst in the mentioned reaction could clearly be observed. The catalyst is highly active in both Heck and Suzuki reactions, and advantageous results were obtained. All substrates with electron donating and electron withdrawing functionalities have given the desired products in high isolated yields. For example, entries 4–9, 18–22, and 26–28 with phenyl halides bearing electron-donating functionalities have provided the products in 84–93% yields. In addition, phenyl halides bearing electron-withdrawing functionalities (entries 10–14, 22–23, and 29–31) have successfully participated in the reaction and have given the products in 88–99% yields. In addition, a reaction was performed with 10 mmol of the substrates to evaluate the possible scale up of the reaction. The results were advantageous, and high yields of the products were obtained in the scale-up reactions. To study the possibility of the use of the catalyst in industrial applications, scale-up reaction was performed under optimized conditions. The large-scale reaction of iodobenzene with styrene and phenylboronic acid has led to 85 and 83% yields, respectively.

A proposed mechanism was suggested, which is presented in Scheme 2. Based on the mechanism, the presence of the catalyst is critical for reaction performance. In the first step of the reaction, the catalyst activates the Ar–X bond, followed by the addition of the alkene. In the next step, the product is obtained, and the catalyst is ready for the next cycle of the reaction by the elimination of H–X.

A great advantage of the Pd@Py₂PZ@MSN catalyst is its reusability. For this purpose, the reaction was repeated for 10 sequential reactions, and the obtained results are presented in Figure 3. It could be seen that in both Heck and Suzuki reactions, no big loss in the activity of the catalyst is observed. It is significant to study the possible leaching of palladium from the catalyst. For studying the leaching of palladium from the Pd@Py₂PZ@MSN catalyst, a reaction was performed under the optimized reaction conditions. However, after 20 min when the reaction has not been completed, the catalyst was isolated from the reaction mixture. The reaction was continued without the catalyst for more than 3 h and based on GC analysis, no performance was observed in the reaction performance. In addition, the filtrate was studied by ICP, which did not show any leaching of palladium from the catalyst.

To show the structural stability of the catalyst during the reaction conditions, the catalyst was separated after the 5^th reaction run and washed with water and ethanol. Then, the dried 5^th re-used Pd@Py₂PZ@MSN catalyst was characterized by SEM and BET analysis. The results are presented in Figure 4; Table 4, respectively. Comparing the results of the recovered catalyst with the fresh ones confirm the good stability of the catalyst during the reactions.

To better study the activity of the catalyst, Pd@Py₂PZ@MSNs were compared with the previously reported catalysts. Table 5 presents the catalysts, reaction conditions, and isolated yields of the reactions for the reaction of styrene and bromobenzene. According to the results, the best performance is observed for Pd@Py₂PZ@MSNs, and the reaction is performed under mild reaction conditions.

In conclusion, Pd@Py₂PZ@MSN as a novel catalyst was designed, synthesized, and characterized based on the immobilization of palladium onto the modified SBA-15 MSNs. The catalyst was characterized using several methods, and the structure of the catalyst was confirmed. The activity of the Pd@Py₂PZ@MSN catalyst was evaluated in Heck and Suzuki reactions. The catalyst showed the best activity in water as the solvent in the presence of triethyl amine base. Several substrates with different functionalities were used and, in all cases, the desired compounds were synthesized in good isolated yields. As an advantage, the catalyst showed very good reusability after 10 sequential runs.