Mohsen Golestanzadehab and
Hossein Naeimi*a
aDepartetment of Organic Chemistry, Faculty of Chemistry, University of Kashan, Kashan, 8731751167, Iran. E-mail: naeimi@kashanu.ac.ir; golestanzadeh@grad.kashanu.ac.ir; Fax: +98-31-55912397; Tel: +98-31-55912388
bChild Growth and Development Research Center, Research Institute for Primordial Prevention of Non-Communicable Disease, Isfahan University of Medical Sciences, Isfahan, 8174673461, Iran
First published on 2nd September 2019
Immobilized Pd nanoparticles on a new ligand, namely, tris(pentaethylene-pentamine)triazine supported on graphene oxide (Pdnp-TPEPTA(L)-GO) was introduced as a novel and robust heterogeneous catalyst for use in C–C bond formation reaction. The Pdnp-TPEPTA(L)-GO catalyst was synthesized by complexation of Pd with TPEPTA as a ligand with high N-ligation sites that were supported on graphene oxide through 3-chloropropyltrimethoxysilane. The prepared catalyst was characterized using some microscopic and spectroscopic techniques. The TPEPTA(L)-GO substrate is a 2D heterogeneous catalyst with a high specific surface area and a large amount of N-ligation sites. The Pdnp-TPEPTA(L)-GO catalyst used in the C–C bond formation reaction between aryl or heteroaryl and phenylboronic acid derivatives was applied towards the synthesis of biaryl units in high isolated yields. Notably, a series of competing experiments were performed to establish the selectivity trends of the presented method. Also, this catalyst system was reusable at least six times without a significant decrease in its catalytic activity.
For example, the Suzuki–Miyaura cross coupling reaction is a highly efficient and robust organic reaction that includes cross coupling between a sp2-hybridized halide and sp2-hybridized boronic acid to form a Csp2–Csp2 bond. The biaryl or biheteroaryl motifs are ubiquitous among the wide range of industrially and pharmaceutically important compounds (Fig. 1).
Generally, the homogeneous catalysts of Pd complexes catalyze the cross coupling Suzuki–Miyaura reaction.4–6 However, using homogeneous Pd complexes has several disadvantages including low efficiency in the separation of the catalyst and high cost of the Pd element at the industrial scale. Therefore, the immobilization of Pd complexes or Pd nanoparticles on different organic, inorganic, and hybrid supports has been made to provide heterogeneous Pd catalysts. Unfortunately, the supported Pd complex catalysts often suffer from low specific surface area (SSA), space confinement in mesoporous supports, less reactivity of support, reduced availability of the metal complex, and low thermal and chemical stability with respect to leaching of metals.7–9
Recently, two-dimensional supports such as graphene,10 graphene oxide (GO),11 and reduced graphene oxide (RGO)12 have attracted extensive attention in heterogeneous catalysis due to their high SSA,13 more surface active sites,14 excellent photocatalyst support,15 superior electron mobility,16 the availability of functional groups from all side of the sheets,17,18 and other excellent properties.19,20 As examples of the utilization of two-dimensional graphene in coupling reactions, we have considered previously reported papers. Very recently, Huang and co-workers reported a robust 3D ionic liquid supported on RGO/Pd nanocomposites for the Suzuki cross coupling reaction.21 Sengupta et al. reported an efficient aminobis(phosphine)-PdII complex on GO in different coupling reactions.22 Also, the effect of GO support on the catalytic performance of Pd–Fe3O4/GO, Pd–Co3O4/GO and Pd–Ni(OH)2/GO in the Suzuki cross coupling reaction was considered by Elazab and co-workers.23,24 In accordance with the various, constantly evolving fields, the development of efficient two-dimensional graphene/Pd nanocomposites is of enormous importance.25–43 In addition, Li et al. reported a new catalyst (palladium decorated on nitrogen-doped graphene nanoshells) for N-allylation reaction.44 Herein, it is notable that most of the reported catalysts applied in coupling reactions were unstable in air, less reusable or totally non-reusable, and had low loading of active sites. Undoubtedly, the design, preparation, and application of superior heterogeneous catalysts with high SSA and more reactive sites as well as highly coordinating donor atoms for the stabilization of Pd(0) in cross coupling reactions is still appealing. In continuation of our previous studies on two-dimensional heterogeneous catalysts,45–48 this research highlights the application of graphene oxide modified with a novel ligand with high N-ligation sites for the stabilization of palladium nanoparticles with oxidation state (0) in the Suzuki–Miyaura cross coupling reaction. The schematic representation of Pdnp-TPEPTA(L)-GO is shown in Fig. 2.
In the second step of the preparation of the Pdnp-TPEPTA(L)-GO catalyst, 3-chloropropyltrimethoxysilane was anchored on graphene oxide, according to the previous literature with some modification.54,55 Similar to the first step, the second step is also very important because a large amount of 3-chloropropyl on GO results in the stabilization of the highest amount of ligand on the surface of GO. We determined the amount of chloropropylsilyl groups on the surface of GO using Mohr's method (see ESI†).56,57 The total density of chloropropylsilyl groups on GO was estimated to be about 0.95 mmol g−1. The chloropropylsilyl groups not only cover the GO sheets but also make them ready for further surface functionalization.54 Finally, in the third step, through two sequential steps, the stabilization of TPEPTA on the surface of GO was performed at 80 °C for 12 h. Then, TPEPTA(L)-GO was reacted with Pd(OAc)2, leading to the immobilization of palladium nanoparticles in the presence of sodium dodecyl sulfate (SDS) as the surfactant for controlling the size of palladium particles58,59 in order to produce Pdnp-TPEPTA(L)-GO catalyst.
To confirm the functionalization of GO, the Pdnp-TPEPTA(L)-GO catalyst was characterized by the following techniques: Scanning Electron Microscopy (SEM), Energy Dispersive X-ray spectroscopy (EDX), Transmission Electron Microscopy (TEM), Fourier Transform Infrared Spectroscopy (FT-IR), Powder X-ray Diffraction (XRD), and Thermogravimetric Analysis (TGA).
In order to study the surface morphologies of GO and the Pdnp-TPEPTA(L)-GO catalyst, SEM images were recorded. As shown in Fig. 3a, the plate-like forms including the flaky layered texture of GO was observed. The EDX analysis of GO is shown in Fig. 3b. The results of EDX analysis show the presence of carbon, nitrogen, and oxygen. The absence of metal elements such as potassium and sodium clearly showed that the washing step was successful. The surface morphology of the Pdnp-TPEPTA(L)-GO catalyst was also studied by SEM (Fig. 3c). As shown in this figure, the sheets of GO were covered by the foreign matter Pdnp-TPEPTA(L). In this image, both the plate-like sheets of GO and Pd nanoparticles were observed. Moreover, the Pdnp-TPEPTA(L)-GO catalyst was investigated by EDX (Fig. 3d). This analysis clearly shows the presence of Pd and N in the concentration of 30.2 wt% and 46.3 wt%, respectively. According to the Pd wt%, the total mmol of Pd nanoparticles on TPEPTA(L)-GO was calculated to be 2.58 mmol g−1. Notably, the amount of Pd element in the Pdnp-TPEPTA(L)-GO catalyst measured using atomic absorption spectrometry (AAS) was 28 wt% (see ESI†). To the best of our knowledge, this amount is the highest amount of Pd immobilized on GO or 2D supports. In addition, the amount of TPEPTA ligand supported on the GO sheets was calculated using EDX analysis.60 Based on this method, the total amount of TPEPTA ligand detected from the nitrogen content was 1.43 mmol g−1 (see ESI†). In order to quantify the amount of Pd in the catalyst, ICP analysis was carried out after its treatment with HCl (37%) and HNO3 (65%). Based on the ICP results, the Pd quantity was measured to be 286 mg L−1 (286 ppm). This amount is equal to 28.6 wt% (∼2.57 mmol g−1). The quantitative analysis is in good agreement with the EDX data.
Then, the Pdnp-TPEPTA(L)-GO catalyst was investigated by TEM in order to investigate the size of Pd nanoparticles and 2D layers of the GO support (Fig. 3e–g). From the TEM images in Fig. 3f and g, the layers of GO were found to be covered by functional groups such as the Pdnp-TPEPTA complexes. Also, the particle size distribution of the Pd nanoparticles was estimated using Image J software and the average size was about 7.63 ± 0.5 nm.61 This data indicates that the Pd nanoparticles did not aggregate on TPEPTA(L)-GO and the SDS surfactant assisted this event as well as delete the formation of smaller sized Pd particles.62
The FT-IR spectrum of the Pdnp-TPEPTA(L)-GO catalyst and its comparison with GO and TPEPTA ligand shows the peaks that confirm the successful synthesis of the abovementioned catalysts (Fig. 4). Investigating the FT-IR spectrum of GO, the bands appearing at 3394, 1719, 1573, and 1198 cm−1 may be attributed to the stretching vibrations of O–H, CO, CC, and C–O, respectively.63 According to the FT-IR spectrum of the TPEPTA ligand, the absorption peaks observed at 3412, 2925, 1575, and 1091 cm−1 are associated with the vibrations of N–H, C(sp3)–H, CN, and C–N, respectively.64,65 The presence of vibration bands of GO and the TPEPTA ligand in the FT-IR spectrum of Pdnp-TPEPTA(L)-GO confirms the successful synthesis of the target catalyst. The observation of a slight shift (∼4 cm−1) in the N–H bands and other peaks is due to the chemical environment of Pd–N coordination.66–68
In addition, the crystalline structure of Pdnp-TPEPTA(L)-GO was studied by XRD analysis and the obtained data are displayed in Fig. 5. In the XRD pattern of GO, the peak at 26.4° (002),69 which is related to the crystalline structure of graphite, is moved to a lower Bragg angle of 12.1° in the crystalline structure of GO corresponding to the (001) plane with d-spacing of 0.92 nm.70
Fig. 5 The XRD patterns of GO (blue) and the Pdnp-TPEPTA(L)-GO catalyst (green); the vertical axes at left and right are the intensity values of GO and the Pdnp-TPEPTA(L)-GO catalyst, respectively. |
The increase in d-spacing value is due to the presence of different functional groups including oxygen motif between the graphite layers during the oxidation process using KMnO4 and HNO3.69 However, in the XRD patterns of Pdnp-TPEPTA(L)-GO, the peaks are indexed as the (111) 38.4°, (200) 44.1°, and (220) 67.8° plane, which can be attributed to the face-centered-cubic (fcc) structure of the palladium nanoparticles on TPEPTA(L)-GO (JCPDS card 26-1081 and 01-0646).71,72 Also, a peak appeared at 21.2° (002), whereas the reflection corresponding to the (001) plane of GO at 12.1° disappeared. This observation confirms the functionalization of GO sheets with TPEPTA ligand.73 Furthermore, using Scherrer's equation, the size of the palladium nanoparticles 10.5 nm was calculated.74
TGA analysis was used to investigate the thermal stability of the Pdnp-TPEPTA(L)-GO catalyst (Fig. 6). The TGA of the catalyst shows two main weight losses. The first weight loss was obtained at 88–112 °C, which is attributed to the trapped water and organic solvents.75 The second weight loss of the catalyst occurred at 197–239 °C, which can be related to the TPEPTA ligand.76 Notably, the remaining non-decomposable sample was approximately 29 wt%, which released the amount of Pd immobilized on the structure of the catalyst. In addition, in accordance with the weight loss amount, the concentration of the TPEPTA ligand per C atoms in the graphene layers was measured, which was found to be one ligand per ninety six carbon atoms (see ESI†).
Entry | Base | Ratio (1:2:base) | Catalyst (mg) | Solvent | T (°C) | Time (min) | Yieldb (%) |
---|---|---|---|---|---|---|---|
a Reaction conditions: phenylboronic acid (1) (x mmol), 4-iodoanisole (2) (y mmol), base (z mmol), solvent (6 mL).b Isolated yields.c N-Methylpyrrolidone.d The Pd-TPEPTA catalyst without graphene oxide as the support was used. | |||||||
1 | Et3N | 1.2:1:2 | 30 | DMF:H2O (2:1) | 80 | 60 | 20 |
2 | NaOH | 1.2:1:2 | 30 | DMF:H2O (2:1) | 80 | 40 | 45 |
3 | Na3PO4 | 1.2:1:2 | 30 | DMF:H2O (2:1) | 80 | 24 | 50 |
4 | Na2CO3 | 1.2:1:2 | 30 | DMF:H2O (2:1) | 80 | 15 | 75 |
5 | K2CO3 | 1.2:1:2 | 30 | DMF:H2O (2:1) | 80 | 10 | 95 |
6 | K2CO3 | 1.2:1:2 | 30 | DMF:H2O (1:1) | 80 | 15 | 79 |
7 | K2CO3 | 1.2:1:2 | 30 | H2O | Reflux | 20 | 75 |
8 | K2CO3 | 1.2:1:2 | 30 | DMF | 80 | 15 | 75 |
9 | K2CO3 | 1.2:1:2 | 30 | EtOH:H2O (1:1) | Reflux | 25 | 65 |
10 | K2CO3 | 1.2:1:2 | 30 | NMPc:H2O (1:1) | 80 | 20 | 78 |
11 | K2CO3 | 1.2:1:2 | 30 | NMPc:H2O (2:1) | 80 | 15 | 87 |
12 | K2CO3 | 1.2:1:1.5 | 30 | DMF:H2O (2:1) | 80 | 20 | 55 |
13 | K2CO3 | 1:1:2 | 30 | DMF:H2O (2:1) | 80 | 20 | 70 |
14 | K2CO3 | 1.2:1:2 | 25 | DMF:H2O (2:1) | 80 | 15 | 72 |
15 | K2CO3 | 1.2:1:2 | 35 | DMF:H2O (2:1) | 80 | 10 | 93 |
16 | K2CO3 | 1.2:1:2 | 30 | DMF:H2O (2:1) | 100 | 10 | 95 |
17 | K2CO3 | 1.2:1:2 | 30 | DMF:H2O (2:1) | 60 | 15 | 60 |
18b | K2CO3 | 1.2:1:2 | 80 | DMF:H2O (2:1) | 80 | 25 | 85 |
With optimization results in hand, the generality and scope of this methodology was evaluated in the synthesis of diverse biaryl units using reactions of different substrates containing I, Br, and/or Cl with phenylboronic acid derivatives (Scheme 3). As shown in Scheme 3, the target products (P1-P21) were obtained in 85–97% and 82–96% yields from the reaction of aryl iodides and bromides involving electron-donating and electron-withdrawing substituents, and phenylboronic acid and 4-methoxyboronic acid as the starting materials, respectively. As we know from the previously reported literature, the presence of electron-withdrawing groups in aryl halide or electron-donating groups in aryl boronic acid enhances the reaction rate and yield of the Suzuki cross coupling reaction due to the facilitation of the rate limiting oxidative addition step.81–83 It is notable in this research that the catalyst can improve the Suzuki reaction irrespective of the nature of the substituent (electron-withdrawing or electron-donating) on the aryl halides and aryl boronic acids, and the type of halogen in the aryl halide, compared with the previously reported methods.21,37,40
In addition, it is important to note that the aryl chlorides are less reactive than aryl iodides and bromides in the Suzuki cross coupling reaction due to the bond strength. However, many reported papers have focused on the utilization of aryl chlorides in the Suzuki cross coupling reaction.84–90 All of the papers report interesting results but research is still appealing in this field. Due to the less reactivity of aryl chlorides, some researchers used a higher amount of the catalyst.91–94 In order to study the applicability and efficiency of the Pdnp-TPEPTA(L)-GO catalyst, the Suzuki cross coupling reaction using aryl chlorides to produce products P4, P5, P8, P9, P12, P18, P20, and P21 was studied using the same amount of catalyst as that for the Suzuki reaction using aryl iodides and aryl bromides. The obtained results were very promising. The products were obtained in 82–91% isolated yields in 20–30 minutes. When the reactions were carried out at 120 °C, the isolated yields of the mentioned products increased slightly but the time of the reactions decreased similar to the time of the reactions for aryl iodides and bromides (see ESI-Table 2S†).
Interestingly, the preparation of compound P21 was studied directly from the reaction of 2,4-dimethylphenol (3) and 4-phenylbenzaldehyde (4) by a condensation reaction in the presence of RGO-SO3H as the catalyst under solvent free conditions, according to a previous work71 and the Suzuki cross coupling reaction through two synthetic pathways (Scheme 4). As can be seen, the approach (b) is more efficient and interesting when the researchers do not use 4-phenylbenzaldehyde. Also, the approach (a) is less unfavorable due to the formation of O-alkylated product during the condensation reaction95,96 and higher reaction time.
Scheme 4 Synthesis of compound 21 using two synthetic pathways, reaction conditions for the approach (a): 2,4-dimethylphenol (3) (6 mmol), 4-phenylbenzaldehyde (4) (2 mmol), solvent-free, T: 100 °C, RGO-SO3H (sulfonated reduced graphene oxide): 40 mg, time: 2.5 h, isolated yield: 81%. Reaction conditions for approach (b): the condensation step was similar to the approach (a) and X: Br or Cl, the reaction conditions for the Suzuki reaction are the same as the footnote in Scheme 3. Time: 90 min for X: Br and 110 min for X: Cl, and for X: Br and X: Cl the final yields obtained were 91% and 86%, respectively (see ESI†). |
As shown in Scheme 5, it is possible to obtain the compound P22 (as an important chemical feedstock, CAS-RN: 118727-34-7 and Reaxys-RN: 7822565) using different approaches (Scheme 5, approaches: a–f). Despite much attempts for the synthesis of compound P22 using different starting materials, the successful synthesis with high yields and lower reaction time have been largely overlooked. For example, Schwab and co-workers reported the synthesis of compound P22 using [Pd(PPh3)4] catalyst in 62% isolated yield (Scheme 5, approach b).97 Also, for different approaches with two sequential steps, the total yields were lower than those for the one step approaches (Scheme 5, approaches c and e).98,99 Therefore, some experiments were conducted to obtain an efficient method for the preparation of the desired product (Scheme 5, approaches g–j). The cyclotrimerization of three equivalents of 4-aminoacetophenone was carried out using RGO-SO3H as the heterogeneous acid catalyst and the target product was produced in 62% yield within 16 h (approach g vs. approach d100,101 in Scheme 5). However, the presence of NH2 group on the starting material caused a higher amount of the used catalyst to be used and the amount of by-product produced was higher during the process. To overcome this problem, we protected the NH2 group using tert-butyloxycarbonyl ((Boc)2O), and then applied RGO-SO3H and de-protection of N-Boc to obtain the compound P22 (Scheme 5, approach h). Due to the three reaction steps, the total yield of the desired product was lower than that in our previous approach. These experiments show the significant role of Pdnp-TPEPTA(L)-GO catalyst in the preparation of compound P22 (Scheme 5, approaches i and j). The yield of the desired product increased to 89% isolated yield when the reaction conditions for approach (i) were used. Furthermore, using diverse starting materials, the desired product was obtained in good yield (80%) (Scheme 5, approach j).
Scheme 5 Different approaches for the synthesis of compound P22 from various starting materials (A comparison study between our work and previously reported works). Reaction conditions for (a): 10% palladium on activated charcoal, N2H4 in ethanol, time: 10.5 h, T: 70 °C, yield: 85%.102 (b) tetrakis(triphenylphosphine)palladium(0), Aliquat 336, K2CO3 in H2O, toluene, time: 24 h, T: 100 °C, inert atmosphere, yield: 62%, TBB is 1,3,5-tribromobenzene.97 (c) Two reaction steps, 1: CF3SO3H, toluene, heating, 2:32%, hydrazine hydrate, RANEY® nickel, THF, time: 6 h, T: 85 °C.98 (d) Toluene-4-sulfonic acid, H2O, neat (no solvent), time: 16 h, T: 142 °C, yield: 71% (ref. 100) and toluene-4-sulfonic acid, time: 16 h, T: 145 °C, yield: 25%.101 (e) Two reaction steps, 1: Na2CO3 in H2O, tetrakis(triphenylphosphine)palladium(0) [Pd(PPh3)4], 1,2-dimethoxy-ethane, time: 18 h, heating, 2: 76%, CF3SO3H, CH2CL2, time: 2 h, T: 20 °C.99 (f) Arachno-[(η5-C5Me5RuCO)2B2H6] in toluene, time: 40 h, T: 75 °C, inert atmosphere.103 (g) Cyclotrimerization reaction, RGO-SO3H (80 mg), toluene, reflux, time: 16 h, yield: 62% (see ESI†) [this work]. (h) Three reaction steps, 1: protection of amino groups using (Boc)2O (tert-butyloxycarbonyl), 2: cyclotrimerization reaction, RGO-SO3H (80 mg), toluene, time: 12 h, reflux, 3: de-protection of N-Boc, total yield: 45%, (see ESI†) [this work]. (i) Pdnp-TPEPTA(L)-GO (30 mg), DMF:H2O (2:1) (6 mL), K2CO3, T: 80 °C, time: 1 h, yield: 89% (see ESI†) [this work]. (j): Pdnp-TPEPTA(L)-GO (60 mg), DMF:H2O (2:1) (6 mL), K2CO3, T: 100 °C, time: 1.5 h, yield: 80% (see ESI†) [this work]. |
For extended study, a series of competing experiments were also performed to establish the selectivity trends of the presented method (Scheme 6). First, the competing reactions were carried out between two boronic acids including phenylboronic acid (PBA) and 4-methoxyphenylboronic acid (4-OMe-PBA) with 1.2 mmol, 1.0 mmol of bromobenzene, and different amounts of K2CO3 as the base (Scheme 6, experiments 1–3). The typical procedure for each experiment is illustrated in ESI.† The researchers studying the Pd catalyst have found that a relationship exists between the Suzuki–Miyaura reaction catalytic cycle and acid–base chemistry.78 Notably, in the Suzuki cross coupling reaction, the main role of the base such as K2CO3 is to increase the reactivity of PBA towards the Ph-halide complex by changing it into the corresponding phenylborate (PBO). Hence, the pKa values of different PBA were reported in the literature.104,105 Therefore, a lower pKa value of PBA directly captures the greater tendency to obtain the OH− anion. Among PBA and 4-OMe-PBA, PBA has a lower pKa value.105 In the competing experiment 1 (CE-1), a limited amount of K2CO3 was used and good selectivity was observed for the two expected products 5a and 5b. When 1.0 mmol of K2CO3 was used, PBA with stronger acidity (lower pKa) reacts more extensively with the OH− anion and the compound 5a was obtained in 75% yield. For further investigation of the amount of base in the selectivity of the present method, we performed CE-2 and CE-3 with 2.0 and 4.0 mmol of K2CO3, respectively. As shown in these experiments, the selectivity decreases with increase in the stoichiometry of K2CO3. Additionally, we performed CE-4 and CE-5 for the study of para and meta isomers of arylbromides including electron-withdrawing groups (EWG) in the presence of PBA and 4-OMe-PBA in isolated experiments. In CE-4, 4-bromobenzaldehyde was more reactive relative to 3-bromobenzaldehyde in the presence of PBA. According to the reported literature,106 arylhalides with EWG accelerate the oxidative addition step in the mechanism of the Suzuki reaction. Also, this differentiation between 8a and 8b is related to the influence of inductive and resonance effects on the electronic charge stabilization in the Suzuki cross coupling reaction pathway. As shown in Scheme 6, CE-5, 4-OMe-PBA is less reactive towards the base (OH− anion) and a very good selectivity was observed for the two compounds 9a and 9b. Finally, in CE-6, the competition between aryl bromides including EWG and electron-donating groups (EDG) at the para positions were performed and we exclusively observed that aryl bromide including EDG cannot react with 4-OMe-PBO in the presence of aryl bromide including EWG. This observation refers to the mechanistic aspect of the Suzuki reaction progress, thus representing the existence of EDG and EWG on the aryl halides.106
Scheme 6 Competing experiments in the Suzuki–Miyaura cross coupling reaction; PBA is phenylboronic acid, 4-OMe-PBA is 4 methoxyphenylboronic acid. Experiments 1–3 for competition between PBA and PBA including electron-donating groups such as OMe at different mmol of K2CO3 as the base in the presence of the same arylbromide. Experiment 4 for studying the effect of electron-withdrawing group at meta or para positions in the presence of the same phenylboronic acid. Experiment 5 for the investigation of the effect of electron-withdrawing group at meta or para positions in the presence of the same phenylboronic acid including electron-donating groups such as OMe. Experiment 6 for studying the effect of the electron-donating and electron-withdrawing groups on the aryl halide in the Suzuki–Miyaura cross coupling reaction in the presence of the same phenylboronic acid with methoxy group at the para position. General reaction conditions for all the competing experiments: solvent (6 mL), Pdnp-TPEPTA(L)-GO catalyst (30 mg). All the yields refer to the isolated yield. For the typical procedures and separation of the desired compound from the reaction mixture, please see ESI.† |
Usually, there are two competing reactions with the main coupling reaction (Scheme 7, part A). In this study, all the possible products including the homocoupling, deboronation, and Suzuki–Miyaura reactions were monitored during the processes using gas chromatography (GC). These analyses revealed that in all the experiments, the conversion of PBA and aryl halides was 100% and the reaction occurred until completion. Also, in all the experiments, the main product of Suzuki reaction was obtained in large yields and two other byproducts from the homocoupling and deboronation reactions were detected at 0.1–0.5% and 2.3–3.0%, respectively. Interestingly, the present method is highly appealing and selective in limiting the production of by-products relative to the other methods.
It seems that the influence of the substituent plays a dual role in the mechanism of the Suzuki–Miyaura cross coupling reaction (Scheme 7 part B).78 As can be seen, PBA with EWG at different positions, especially the para position, increases the acidity of PBA due to delocalization of the negative charge on PBO as a conjugate base by the resonance effect. The EWG substitution on PBA decreased the nucleophilicity of PBO towards the electrophilic Pd complex, which can decrease the rate of the transmetalation step.107 The aryl halide with EDG substitution has a slower rate of transmetalation than that for the aryl halide with EWG. The presence of EDG on PBA has the opposite effects. However, according to some previous studies,78 a higher rate of transmetalation can be obtained when the EDG and EWG exist on PBA and the aryl halide, respectively.
We have proposed a plausible reaction mechanism for this protocol, according to the literature,108–110 as shown in Scheme 8. Depending on the reaction conditions such as the concentration and type of the base, the stability of Pd nanoparticles immobilized on the ligand and reduced reaction conditions, the reaction can be performed through different pathways. First, the free Pd(0) was released in the reaction medium near the high SSA of the Pdnp-TPEPTA(L)-GO catalyst. Then, the oxidative addition of Pd(0) to the aryl halide to produce the organopalladium complex intermediate (I) occurred. The two main pathways (a) and (b) can be proposed according to the concentration and power of the base in the reaction medium. When the base has a significant nucleophilicity to replace X as the leaving group from the complex (I), the pathway (a) was followed. However, in the opposite conditions, the base was attached to PBA and formed PBO (III) (pathway b). Subsequently, the migration of the Ph group to intermediates (I) or (II) took place in pathways (a) or (b) to form the intermediate (IV) through the transmetalation step. The Suzuki product was obtained through the reductive elimination of the Pd(II) complex (IV). At this point, the Pd(II) complex (V) was reduced and free Pd(0) can follow two main pathways including rebounding to the catalyst via pathway (c) or the initiation of the next catalytic cycle via pathway (d).
The greatest advantages of Pdnp-TPEPTA(L)-GO catalyst are the ease of separation and compatibility with the organic compounds as reactants due to the large organic ligand supported on GO as the inorganic support. Also, the reusability and recyclability of Pdnp-TPEPTA(L)-GO catalyst is very important from the economic, environmental, and industrial points of view. Therefore, the reusability of the Pdnp-TPEPTA(L)-GO catalyst was tested in the Suzuki–Miyaura cross coupling reaction for the preparation of compound P1 under the optimized conditions within 10 minutes. The Pdnp-TPEPTA(L)-GO catalyst showed a reusability of at least six runs without a remarkable decrease in its catalytic activity performance (Table 2). As shown in this table, the first catalytic run resulted in the production of 95% of the desired product within 10 minutes. In the 5th catalytic run, approximately 87% of the final product was obtained. According to the recovered catalyst values in each run, about 97% of the Pdnp-TPEPTA(L)-GO catalyst remained until the 6th run. However, the decreasing catalytic activity in the 6th run can be related to the lost catalyst during catalyst recovery and not to palladium leaching. During the reusability experiments, the Pd content of the catalyst decreased by only 6 ppm (about 2.3%) of the total Pd content immobilized on TPEPTA(L)-GO. Therefore, this observation showed that the Pdnp-TPEPTA(L)-GO catalyst is a robust and satisfactory heterogeneous catalyst. For the completion of our observation related to the Pd leaching, the hot filtration study for this catalyst was performed and the catalyst immediately following the end of the reaction was isolated. According to the ICP analysis obtained for the hot filtration test, the Pd detected in the 1st, 3rd, and 6th runs were 0.08, 0.1, and 0.06 ppm, respectively. After the isolation of the catalyst, the hot filtration reaction mixture was charged with iodobenzene and PBA. The GC yields for the 1st, 3rd, and 6th runs were only < 2%. This experiment strongly revealed that the leaching of Pd nanoparticles during the reaction progress is very low and the Pdnp-TPEPTA(L)-GO catalyst is a reusable and air-stable heterogeneous catalyst in practice. For further investigation of the used catalyst after the 6th run, the morphology of Pdnp-TPEPTA(L)-GO was studied and the catalyst remained mostly unchanged after use in the Suzuki–Miyaura cross coupling reaction, according to the FT-IR and TEM studies (see ESI†).
Run | First | Second | Third | Fourth | Fifth | Sixth |
---|---|---|---|---|---|---|
a 4-Iodoanisole (1 mmol), phenylboronic acids (1.2 mmol), K2CO3 (2 mmol), solvent: DMF:H2O (2:1) (6 mL), Pdnp-TPEPTA(L)-GO (30 mg), T: 80 °C.b Isolated yields.c Not determined.d GC yield. | ||||||
Time (min) | 10 | 10 | 10 | 10 | 10 | 10 |
Yield (%)b | 95 | 95 | 93 | 92 | 90 | 87 |
TON | 12.3 | 12.3 | 12.1 | 11.9 | 11.7 | 11.3 |
TOF (h−1) | 74.1 | 74.1 | 72.9 | 71.7 | 70.5 | 68.1 |
Recovered catalyst (%) | >99 | >99 | >98 | >98 | >97 | >97 |
Pd amount (ppm) | 258 | 257 | ND | ND | 253 | 252 |
Pd leached after hot filtration test (ppm) | 0.08 | NDc | 0.1 | ND | ND | 0.06 |
Yield of hot filtration test (%)d | <2 | ND | <2 | ND | ND | <2 |
TPEPTA | Tris(pentaethylene-pentamine)triazine |
GO | Graphene oxide |
1D | One-Dimensional |
2D | Two-Dimensional |
3D | Three-Dimensional |
SSA | Specific Surface Area |
TCT | 2,4,6-Trichloro-1,3,5-triazine |
PEHA | Pentaethylenehexamine |
SDS | Sodium dodecyl sulfate |
AAS | Atomic absorption spectrometry |
PBA | Phenylboronic acid |
4-OMe-PBA | 4-Methoxyphenylboronic acid |
PBO | Phenylborate |
EWG | Electron-withdrawing group |
EDG | Electron-donating group |
Footnote |
† Electronic supplementary information (ESI) available: Detailed experimental procedures for the preparation of the catalyst, synthesized compounds, spectroscopic data, and hard copies of the NMR spectra. See DOI: 10.1039/c9ra04511b |
This journal is © The Royal Society of Chemistry 2019 |