Introduction

In the context of green chemistry, the search for sustainable catalytic systems for cross-coupling reactions is always desired (Wang et al. 2020). High yields (above 90%) of the desired reaction product are good indicators of the efficiency of a catalytic system; however, the working conditions can be harmful to the environment. Typical catalytic systems for cross-coupling reactions employ a metallic catalyst (Pd normally), auxiliary ligand to stabilize the metal, Lewis base to neutralize the system, solvent, inert atmosphere, and heating (Diederich and Meijere 2004; Negishi 2002). Efforts have been reported in order to reduce the amount of catalytically active species in reactions (Roy and Uozumi 2018) or replace it with a more eco-friendly and less expensive metal (Ma et al. 2021); reduce the use of auxiliary ligands (Serrano et al. 2022); employ less toxic solvents (Su et al. 2022) and bases (Long et al. 2019); and make the reaction conditions milder through the use of lower temperatures (close to room temperature) (Ma et al. 2022), microwave heating (Salih 2022), sonication (Wilson et al. 2018), or elimination of the inert atmosphere (Tian et al. 2022). In addition, the search for new, greener catalytic systems employing biomass sources appears frequently in literature, such as in the use of biomass extracts to prepare metal nanoparticles (NPs) via biogenic routes (Schmitt et al. 2021) or green solvents applied in carbon–carbon and carbon–heteroatom couplings (Peixoto et al. 2022). On the other hand, biomass material is also used as a catalytic support of active metals, giving rise to efficient catalysts which employ, for example, biochar (Gholinejad et al. 2019), calcined materials (Rosa et al. 2019), or polymeric films (Faria et al. 2014) in cross-coupling reactions. In terms of novel catalytic systems, the Suzuki-Miyaura reaction is commonly preferred because it is one of the most widely studied couplings and because of its importance as a key step in the synthesis of bioactive molecules (Miyaura and Suzuki 1995).

As previously shown, the search for new eco-friendly catalytic systems based on palladium is current and of interest to the chemical industry of organic synthesis; thus, this line of investigation is the main objective of the present manuscript. Focusing on the search for a biomass source capable of being used as a catalytic support in coupling reactions, materials such as chitosan, cellulose, and corn stem have high potential for application given their wide availability and low cost (Kumari and Kishor 2020; Heinze 2016; CONAB 2022). Thus, chitosan (Fig. 1, top), a polysaccharide derived from the deacetylation of chitin, is extremely abundant as it composes the exoskeletons of marine crustaceans, for example (Kumari and Kishor 2020). In this way, the investigation of new uses for chitosan corroborates the fishing industry’s need to value its generated waste. Likewise, cellulose (Fig. 1, bottom), an extremely abundant polysaccharide present in plants, stands out for its low cost and wide range of applications (Heinze 2016). And, finally, the corn stem biomass residue includes the prerequisites abundance and low cost, since only Brazil recently had a cereal planted area of 72.9 million hectares in the 2020/2021 harvest (CONAB 2022).

Fig. 1
figure 1

Chemical structure of chitosan (top) and cellulose (bottom)

Full size image

Given this scenario, in this article we report the production of catalytic supports for Pd NPs using raw materials from biomass (chitosan and cellulose) widely available commercially, in addition to an agro-industry residue (corn stem). Emphasizing the use of these low value-added materials will contribute to more sustainable processes since such materials were obtained via simpler routes in terms of unit operations and processing, normally requiring lower energy consumption (Arancon et al. 2013; Maroušek and Trakal 2022). Also, the valorization of biomass corroborates for a better use of residues within its production chain, allowing a greater fixation of income in small producers (Arancon et al. 2013; Maroušek and Trakal 2022). Thus, chitosan/cellulose film (CCF) and corn stem biochar (CSB) were produced and evaluated as catalytic supports for Pd NPs, generating catalysts named Pd/CCF and Pd@CSB, respectively. The solids were tested in the Suzuki-Miyaura cross-coupling, showing efficiency and robustness.

Experimental section

The reagents used in the assays are of analytical purity as the reported work was conducted and tested only at the laboratory bench level.

Preparation of Pd/CCF catalyst

Initially Pd NPs were obtained by dissolution of 24 mg of palladium (II) chloride (PdCl2; Sigma-Aldrich, 99%) in 10 mL of isopropyl alcohol (C3H8O; Proquimios, 99.5%) stirring by 30 min on a stir plate (Ika, model C-MAG HS 7 D). During this process, 1 mL of 0.1 M hydrochloric acid (HCl; Tedia, 37%) was added to dissolve the salt. After fully dissolution, sodium borohydride (NaBH4; Sigma-Aldrich, 99%) reducer solution (53.2 mg in 20 mL of 0.01 M of potassium hydroxide [KOH; Vetec, 85%]) was added and stirred for 1 h until the formation of Pd NPs. Centrifugation step (centrifuge manufactured by Nova Técnica, model NT-825) with isopropyl alcohol washes (3×, 5 min, 3500 rpm) was necessary to Pd NP purification, and finally the solid was dried over laboratory oven (Odontobrás, model EL-1.2 [80 °C, 30 min]). Yield: 11 mg.

Subsequently, the chitosan/cellulose film (CCF) was produced by preparing a solution containing 0.2 g of commercial chitosan (Sigma-Aldrich, deacetylation degree of 75%), 1.7 mL of 1 M acetic acid (C2H4O2; Synth, 99.5%), and 10 mL of distilled water, being stirred for 2 h at 60 °C. After the formation of a viscous liquid, a second solution containing microgranular cellulose (Synth, 90% [0.2 g in 10 mL of distilled water]) was added together with 10 mg of previously prepared Pd NPs. Stirring was continued until a homogeneous phase was obtained. The film was produced in a glass Petri plate having a thickness of 20 μm, being left to rest for solvent evaporation (12 h, room temperature). Yield: 789 mg.

Preparation of Pd@CSB catalyst

Corn stem samples were collected in the city of Santo Antônio da Patrulha-RS (Brazil). After manual cutting, washing with deionized water, oven drying (105 °C, 24 h), grounding, and sieving to obtain particle size in the range of 1.68 to 2.00 mm, the samples were carbonized in a muffle furnace (Quimis, model Q318M21 [7 g, 300 °C, 20 min]) yielding 2.59 g of corn stem biochar (CSB). To prepare the catalyst containing Pd NPs, 135 mg of CSB was weighed and 20 mL of a solution containing 25 mg of PdCl2 diluted in isopropanol/water (1:1) was added. The resulting system stayed in an ultrasonic bath (Cole-Parmer, model 08895-31) for 5 min and then the reducing agent was added (26.6 mg of NaBH4 in 10 mL of 0.01 M KOH). Stirring continued for 2 h (room temperature) on a magnetic stirrer (Ika, model C-MAG HS 7 D) and, at the end, the Pd@CSB solid was filtered and oven dried (80 °C, 2 h). Yield: 148 mg of Pd@CSB.

Transmission electron microscopy analysis (TEM)

TEM analyses were performed using a JEOL JEM 1400 operated at 120 kV. The samples were prepared by deposition of the solid from an isopropanol suspension onto a carbon-coated copper grid at room temperature.

Scanning electron microscopy (SEM)

The morphology of the solids was analyzed using a JEOL model JSM 6610 at 20 kV. The samples were gold coated for analysis.

Fourier transform infrared spectroscopy (FT-IR)

FT-IR analyses were conducted using a Shimadzu spectrophotometer, model IR Prestige 21, combining 32 scans at a 2-cm−1 resolution. The Pd@CSB sample was analyzed as pellet with an approximate thickness and diameter of 2 mm and 5 mm, respectively. The sample mass dilution was performed with KBr. The Pd/CCF sample was analyzed as a film using direct reading with attenuated total reflectance (ATR) mode after 45 scans.

Porosimetric evaluation (BET isotherms)

N2 adsorption isotherms were performed on Micrometrics Tristar II 3020 equipment. Initially, samples were preheated at 70 °C for 8 h under vacuum and the surface area was determined using the Brunauer–Emmett–Teller (BET) method at −196 °C with P/Patm in the range of 0.01 to 0.35. The average diameter of mesopores and their distribution were calculated using the Barrett–Joyner–Halenda (BJH) method with Halsey standards for the desorption isotherm. The micropore volume was calculated using the t-plot method and the standard isotherm of Harkins and Jura.

Elemental analysis (CHN)

The carbon, hydrogen, and nitrogen compositions (CHN analysis) were performed on a PerkinElmer M-CHNS/O analyzer, model PE 2400 series II.

Inductively coupled plasma optical emission spectrometry (ICP-OES)

A 50 mg sample of the solid Pd@CSB was decomposed using 4 mL of 14.4 mol L−1 nitric acid (HNO3; Qhemis, 65%) and heating to 100 °C (thermostatic bath in Ika apparatus, model C-MAG HS 7 D) and added to 10 mL with ultrapure water. The palladium level was determined (emission line at 340.458 nm) using an inductively coupled plasma optical emission spectrometer (PerkinElmer, model Optima 4300 DV) with an axial view configuration.

Suzuki-Miyaura cross-couplings

Experimental

All reactions were conducted under a nitrogen atmosphere in a Schlenk reactor. Bases and solvents were purchased from Synth (Brazil). Phenylboronic acid (95%) and aryl halides were purchased from Sigma-Aldrich (Brazil). All chemicals were used directly from the original container without prior purification. NMR spectra were recorded on a Bruker Ascend 400-MHz spectrometer. Mass spectra were obtained on a GC/MS PerkinElmer Clarus 600 T (EI, 70 eV) equipped with a 30-m capillary DB-5 column with a dimethylpolysiloxane stationary phase. Gas chromatography was performed on a PerkinElmer Clarus 400 GC equipped with a flame ionization detector (FID) and a 30-m capillary column with a dimethylpolysiloxane stationary phase.

Typical procedure for the Suzuki-Miyaura cross-coupling reaction

A Schlenk reactor was charged with base (2 mmol), phenylboronic acid (187 mg, 1.5 mmol), aryl halide (1 mmol), 0.5 mol% of Pd using the catalysts Pd@CSB (5 mg) or Pd/CCF (42 mg), undecane (Sigma-Aldrich, 99% [internal standard, 10 μL]), and solvent (3 mL). The reaction mixture was stirred at 100 °C in a thermostatic oil bath with the aid of a magnetic stir plate (Ika apparatus, model C-MAG HS 7 D). The reaction crude was cooled to room temperature, extracted with 20 mL of diethyl ether (C4H10O; Qhemis, 98%), and washed with 5 mL of sodium hydroxide (NaOH; Synth, 99%) and brine (2 × 5 mL). The organic phase was dried over magnesium sulfate (MgSO4; Synth, 98%), filtered, and concentrated in vacuum; afterward, the crude material was purified by flash chromatography on silica gel (Supelco, pore size 60 Å, 230–400 mesh particle size). The corresponding biaryl products were characterized by 1H and 13C NMR, and by GC-MS.

  • 4-Cyanobiphenyl (Faria et al. 2014). White solid. 1H NMR (400 MHz, CDCl3) δ: 7.72–7.66 (m, 4H), 7.59–7.56 (m, 2H), 7.50–7.39 (m, 3H) ppm. 13C NMR (100 MHz, CDCl3) δ: 145.6, 139.1, 132.5, 129.0, 128.6, 127.7, 127.2, 118.9, 110.8 ppm. GC–MS (IE, 70 eV) m/z (%): 179 (100, M+), 178 (25), 151 (17), 180 (14), 76 (14), 89 (11), 177 (8), 152 (8).

  • Biphenyl (Faria et al. 2014). White solid. 1H NMR (400 MHz, CDCl3) δ: 7.59–7.57 (m, 4H), 7.44–7.40 (m, 4H), 7.34–7.31 (m, 2H) ppm. 13C NMR (100 MHz, CDCl3) δ: 141.2, 128.7, 127.2, 127.1 ppm. GC–MS (IE, 70 eV) m/z (%): 154 (100, M+), 153 (43), 152 (29), 76 (19), 155 (12), 151 (9), 77 (9), 115 (7).

  • 4-Methylbiphenyl (Faria et al. 2014). White solid. 1H NMR (400 MHz, CDCl3) δ: 7.58–7.56 (m, 2H), 7.48 (d, J = 8.1 Hz, 2H), 7.42–7.39 (m, 2H), 7.32–7.28 (m, 1H), 7.23 (d, J = 7.8 Hz, 2H), 2.38 (s, 3H) ppm. 13C NMR (100 MHz, CDCl3) δ: 141.1, 138.3, 137.0, 129.4, 128.7, 127.0, 126.9, 21.1 ppm. GC–MS (IE, 70 eV) m/z (%): 168 (100, M+), 167 (75), 165 (30), 152 (27), 153 (19), 83 (19), 82 (17), 169 (12).

  • 4-Methoxybiphenyl (Faria et al. 2014). White solid. 1H NMR (400 MHz, CDCl3) δ: 7.56–7.51 (m, 4H), 7.43–7.39 (m, 2H), 7.31–7.28 (m, 1H), 6.97 (d, J = 8.8 Hz, 2H), 3.84 (s, 3H) ppm. 13C NMR (100 MHz, CDCl3) δ: 159.1, 140.8, 133.8, 128.7, 128.1, 126.7, 126.6, 114.2, 55.3 ppm. GC–MS (IE, 70 eV) m/z (%): 184 (100, M+), 141 (69), 169 (68), 115 (53), 139 (19), 185 (17), 76 (13), 63 (12).

Results and discussion

Chitosan/cellulose film (CCF)

The preparation of palladium nanoparticles (Pd NPs) was carried out as described in the “Experimental section” prior to the production of CCF. Thus, a solution of PdCl2 in isopropyl alcohol was subjected to the NaBH4 reducer until a visible change of color (orange to black) by the formation of Pd NPs. Afterward, the system was washed with isopropyl alcohol, centrifuged, and the Pd NPs were isolated and dried. The size of the nanoparticles was assessed by transmission electron microscopy (Fig. 2).

Fig. 2
figure 2

TEM micrograph of the Pd NPs (left) and a histogram (right) illustrating the particle size distribution

Full size image

Figure 2 indicates that the Pd NPs produced had a diameter compatible with other similar reports (Faria et al. 2014). Thus, the preparation of the catalyst continued with the addition of a mixture of Pd NPs, microgranular cellulose, and water in a viscous chitosan solution previously prepared (see “Experimental section”). After homogenization, the liquid was poured into a glass plate for evaporation of the solvent and formation of the film, which was named as Pd/CCF.

A SEM micrograph of Pd/CCF is shown in Fig. 3 (left). The film appeared to have a fibrous structure. EDS analysis of the Pd/CCF (Fig. 3, right) indicated the presence of palladium metal.

Fig. 3
figure 3

SEM micrograph illustrating the fibrous surface of Pd/CCF (left) and an EDS pattern showing the detection of Pd metal (right)

Full size image

Following the investigation, the FT-IR analysis was conducted to assess the presence of organic groups on the Pd/CCF surface (Fig. 4).

Fig. 4
figure 4

FT-IR spectrum of Pd/CCF catalyst showing the presence of organic groups on its surface

Full size image

In Fig. 4 it is possible to observe characteristic bands of chitosan (Dotto et al. 2013) and cellulose (Fonseca et al. 2020) existing in the Pd/CCF. The existing broad band at 3257 cm−1 is associated with O–H and N–H groups. In the range of 2925–2849 cm−1, elongation interactions of the C–H bond can be observed. The signal corresponding to the stretching of the carbonyl is evidenced at 1643 cm−1. The bands located at 1555 cm−1 and 1067 cm−1 can be associated with elongation of the C–N bond. The signals observed in the region of 1370 cm−1 refer to CH2. And, finally, the band in the region of 1011 cm−1 can be associated with the C–O elongation.

Corn stem biochar (CSB)

Initially, corn stalks were collected from rural properties in the city of Santo Antônio da Patrulha-RS (Brazil). After manual cutting, washing, drying in an oven, and sieving, the samples were carbonized as described in the “Experimental section.” Porosimetric (BET isotherms) and CHN analyses were performed on the solids corn stem biochar (CSB) and commercial activated carbon (CAC; Sigma-Aldrich activated carbon #242276) to compare surface properties (Table 1).

Table 1 Surface areas and CHN analysis of CSB and CAC
Full size table

Table 1 shows that the surface area of the CAC was about 126× greater than of CSB (entries 1 and 2). This result was expected since the CSB did not undergo an activation process. By analyzing the pore volume, the same tendency was observed; that is, CAC had a pore volume about 411× greater than CSB (entries 1 and 2). In the CHN analysis, it was seen that CSB had a lower percentage of C than CAC due to the heterogeneous composition (see EDS analysis).

For Pd NPs impregnation into the CSB was mixed and stirred (sonication) with PdCl2 solution in isopropanol/water (1:1) for 5 min. After, addition of reducer (NaBH4 in basic solution) was performed to form the Pd(0) NPs. Stirring was carried out for 120 min at room temperature. Finally, the resulting solid was filtered, oven dried, and named Pd@CSB. This material was characterized by SEM-EDS, TEM, ICP-OES, and FT-IR analyses.

A SEM micrograph of Pd@CSB solid is shown in Fig. 5 (left). The Pd@CSB solid appeared to have a fibrous structure. EDS analysis of the Pd@CSB (Fig. 5, right) indicated the presence of palladium metal.

Fig. 5
figure 5

SEM micrograph illustrating the fibrous structure of Pd@CSB (left) and an EDS pattern showing the detection of Pd metal (right)

Full size image

TEM analysis was also performed on the prepared Pd@CSB solid; thus, Pd(0) NPs can be encountered as irregular-shaped structures, with a predominant size of 10 nm (Fig. 6).

Fig. 6
figure 6

TEM micrograph of the Pd@CSB indicating Pd(0) NPs (left) and a histogram (right) illustrating the particle size distribution

Full size image

ICP-OES analysis was performed on the prepared Pd@CSB sample. The Pd loading encountered was 106,476 ± 264 mg kg−1 (~0.01% Pd by weight) and this data served as a parameter for determining the Pd@CSB weight used in the catalytic tests.

FT-IR analysis was conducted on the Pd@CSB solid in order to obtain information about the existence of functional groups. The result found is shown in Fig. 7.

Fig. 7
figure 7

FT-IR spectrum of Pd@CSB catalyst showing the presence of organic groups on its surface

Full size image

In Fig. 7, a characteristic band corresponding to the stretching vibration of −OH groups is evidenced by the broad band near 3340 cm−1. This stretching is attributed to the hydroxyl groups of cellulose, hemicellulose, and lignin present in the CSB, or by the presence of water in the material. Evidence of C–H stretching of alkyl groups of cellulose, hemicellulose, and lignin can also be observed by the peak at around 2905 cm−1. The carbonyl band observed at 1730 cm−1 is likely derived from ester groups of hemicellulose or lignin monomers. The presence of a band at approximately 1604 cm−1 may be due to the C–C bond stretching of the aromatic ring of the lignin or to the C=O group, corresponding to the carbonyl of aldehydes and ketones originating from lignin (Farinella et al. 2008). The band at 1239 cm−1 corresponds to =C–O–C, which is commonly observed when =C–O− in ether, ester, and phenol groups are present. Finally, the band near 1032 cm−1 may be assigned to the (C–O)–C bond stretching of cellulose present in the template structure (Fonseca et al. 2020).

Catalytic tests

Having both solids (Pd/CCF and Pd@CSB) characterized, the Suzuki-Miyaura reaction between phenylboronic acid and 4-bromobenzonitrile was chosen to optimize the catalytic systems (Scheme 1; Table 2). For this purpose, bases and solvents were varied based on previous works of our research group and the use of the eco-friendly reagents was prioritized (Faria et al. 2014; Rosa et al. 2019; Schmitt et al. 2021).

Scheme 1
scheme 1

The Suzuki-Miyaura cross-coupling reaction investigated in this study

Full size image
Table 2 Optimization of reaction conditions for the Suzuki-Miyaura cross-coupling of Pd/CCF and Pd@CSB with phenylboronic acid and 4-bromobenzonitrilea
Full size table

The results presented in Table 2 show that both developed catalysts were successful in the studied reaction (entries 7 and 8), characterizing sustainable catalytic systems as they do not require phosphine ligands and/or quaternary ammonium salts, in addition to employing eco-friendly reagents (ethanol and K2CO3). In this sense, the pair K2CO3/ethanol (base/solvent) proved to be the most efficient in terms of reaction yield, corroborating previous works by our research group (Rosa et al. 2019; Schmitt et al. 2021). On the other hand, the nonpolar solvent dioxane and the organic base NEt3 did not provide the maximum catalytic activity of the catalysts tested (entries 1–6). The effect of the catalytic supports (CCF and CSB) on the reaction yield is evident when the reaction under study is repeated only with the Pd NPs (entry 9), indicating the superiority of the Pd/CCF catalyst (entry 7). And obviously, the Suzuki-Miyaura cross-coupling does not occur without the presence of Pd, confirming the real function of the catalysts reported here (entry 10).

With the actual knowledge about the composition of the catalysts, the scope of its use in the Suzuki-Miyaura reaction was increased by varying aryl halides and/or functional groups on the aromatic ring (Table 3).

Table 3 Suzuki-Miyaura cross-coupling of phenylboronic acid with different aryl halides catalyzed by Pd/CCF and Pd@CSBa
Full size table

Table 3 shows the efficiency of the catalysts in the Suzuki-Miyaura cross-coupling without auxiliary ligands. Excellent yields for the coupling of aryl iodides were observed using both catalysts (entries 1–6); however, Pd@CSB needed shorter reaction times than Pd/CCF for the same reactions. Even using electron-donating ArI, the catalysts showed high efficiency in the cross-coupling (entries 3–6). However, using ArBr as cross-coupling partner, Pd@CSB presents intolerance in aryl bromides containing electron donating groups in para position (entries 9–12). In general, reaction times increased significantly (to 24 h) due to the difficulty of promoting the first stage of the catalytic cycle (oxidative addition) when ArBr is used. Both catalysts did not tolerate ArCl as cross-coupling partner in the Suzuki-Miyaura reaction (entries 13–18): even employing activated aryl chloride (containing NO2 group) the reaction yield stayed in order of traces (entries 13 and 14). The other aryl chlorides tested showed no reaction (entries 16–18) but, surprisingly, Pd/CCF promoted the reaction of PhCl in low yield (entry 15). Although the result is minimal, biphenyl was obtained by direct reaction and no evidence of homocoupling of PhB(OH)2 was encountered. In this study, TON and TOF values were calculated and the Pd@CSB catalyst proved to be superior in terms of TOF values (see entries 4 and 6) to other catalytic systems that employ similar conditions (conventional heating and the same amount of catalyst) (Faria et al. 2014; Rosa et al. 2019; Schmitt et al. 2021).

In order to compare the results obtained in Table 3 with other studies using Pd NPs as a catalyst, the Suzuki-Miyaura reaction of 4-bromoanisole with phenylboronic acid was chosen. Thus, Table 4 shows different catalytic systems developed with their particularities.

Table 4 Suzuki-Miyaura cross-couplings between phenylboronic acid and 4-bromoanisole by different catalysts containing Pd NPs
Full size table

The best result shown in Table 4 (entry 1) combines a low amount of Pd (0.1 mol%) with a high yield of 4-methoxylbiphenyl (96%) through the use of nanostructured hafnium (IV) oxide as a catalytic support (Wu et al. 2022). The same cross-coupling product yield was obtained by the Pd/CCF catalyst described in the present manuscript employing a higher amount of Pd (entry 7). Identical results in terms of 4-methoxybiphenyl yield were found using cellulose acetate thin film (Faria et al. 2014), soil-derived humic acid-coated iron oxide NPs (Chinchole et al. 2019), and dendritic structure (Sheikh et al. 2022) as catalytic supports of Pd NPs (entries 4, 8, 9). On the other hand, very interesting results are reported employing Pd NP biosynthesis through plant extracts (entries 2, 3, 5). In these reports, it is observed that organic residues from the plant extract remain together with the Pd NPs after their isolation, providing stability to the catalyst in the same way as the catalytic support (Lebaschi et al. 2017; Hekmati et al. 2017; Schmitt et al. 2021). However, the recycling of catalysts containing Pd NPs from biosynthesis is normally less (3–6 runs) than when using a more robust catalytic support (Lebaschi et al. 2017; Hekmati et al. 2017; Schmitt et al. 2021). Obviously, each catalytic system has its pros and cons due to the large number of items that must be observed. Thus, focusing on the results here presented, it can be said that the catalysts reported in this manuscript are among those with potential technological use due to their simplified, eco-friendly, and low-cost production routes.

The term low cost used in the designation of the catalysts produced is justified by the low acquisition value of the materials that were used in the preparation of the Pd NP supports, as well as by the ease of obtaining them. In order to exemplify this fact, the acquisition cost of the main reagents (chitosan, cellulose, and PdCl2) for the production of 1 g of the Pd/CCF catalyst, based on the prices stipulated in Brazil via the suppliers’ websites, was USD 5.58. This value is quite low compared to the most popular Pd catalyst, tetrakis (triphenylphosphine)palladium(0) (Negishi 2002), used in Suzuki-Miyaura cross-coupling, which costs about USD 89.40 per gram (Sigma-Aldrich 2022).

In this same approach, the effect of the COVID-19 pandemic on the prices of less noble metals such as copper, aluminum, zinc, and silver is observed (Vochozka et al. 2021; Rowland et al. 2022; Novakova et al. 2022; Bartos et al. 2022). Obviously, nobler metals like palladium also experienced an increase in their commercial value due to this same cause. Thus, it is interesting to think about the economic reality when developing technological products derived from these metals. And, in this sense, the catalytic systems reported in the present manuscript are in line with this trend since they employ an adequate palladium load (0.5 mol%) and provide a very good to high yield of product (78–96%) according to Table 4 (entries 6 and 7).

The recycle of the catalysts under study were carry out by choosing the reaction of 4-iodoanisole with phenylboronic acid to analyze this parameter (Table 3, entries 5 and 6). For Pd/CCF was used reaction time of 5 h and, for Pd@CSB, the reaction time was of 1 h as is shown in Table 3 (entries 5 and 6). In this way, in the first run, 98% yield was obtained within Pd/CCF and 95% yield with Pd@CSB. After first run, the reaction crude was filtered and the solid (catalyst and K2CO3) was washed with ethanol and returned to Schlenk reactor to another run. The same procedure was repeated in the other runs. The performance of the catalysts in the recycling study can be seen in Fig. 8.

Fig. 8
figure 8

Recycle of Pd/CCF and Pd@CSB catalysts in repeated runs of the Suzuki-Miyaura cross-coupling of phenylboronic acid with 4-iodoanisole

Full size image

As shown in Fig. 8, Pd/CCF showed excellent catalytic activity during ten successive runs. The tests stopped in tenth run because the film was disintegrating. On the contrary, Pd@CSB showed to lost catalytic activity during recycle, showing very good yield (~80%) until the fifth run. Anyway, both catalysts were actives during 10 runs and this fact represents the robustness of the catalytic precursors developed.

As Pd/CCF catalyst have shown interesting performance in recycling study, further investigations were conducted to clarify the catalytically active Pd species involved. In this sense, an important tool to assess Pd leaching is the hot filtration test (Phan et al. 2006; Rosa et al. 2019). In general, the test consists of filtering the reaction mixture while it is still hot in order to separate the added catalyst (solid) from the reaction medium, which is subsequently heated again. Finally, it is evaluated whether there was an increase in the product yield in the absence of the catalyst. If this occurs, there is evidence of Pd soluble and active in the system. In the investigation conducted with Pd/CCF, the reaction chosen was that of 4-iodoanisole with phenylboronic acid (Table 3, entry 5). This reaction generated 70% of the respective biphenyl (monitored via GC) in a time of 1 h (100 °C). Thus, after 1 h of reaction, the reaction crude was filtered through a heated filtration system, and the filtrate was again taken to a new reactor. The system was heated and stirred for another 4 h under the same usual reaction conditions. Finally, the reaction crude was read again via GC and no significant increase in the amount of the respective biphenyl was evidenced. The comparison of the results obtained with and without Pd/CCF catalyst proved that the evolution of the reaction yield is totally inhibited after removing the catalytic precursor from the reaction medium. These data indicate that the active Pd remains in the Pd/CCF and, thus, it can be inferred that the reaction is promoted on the surface of the CCF in the case of heterogeneous catalysis.

Assessment of the homogeneous or heterogeneous nature of the active Pd species continued with the metallic mercury poisoning test (Phan et al. 2006; Baran et al. 2015; Rosa et al. 2019). Such an assay, when effective, inhibits the catalytic activity of the metal under analysis due to the formation of an alloy (amalgam) with Hg, resulting in a catalytically inactive material. This fact is an indication that catalytically active Pd(0) is present in the catalytic cycle. On the other hand, if the addition of metallic mercury does not influence the catalytic activity, it may be possible to have a catalyst that does not include unprotected Pd(0) supported or in solution in the reaction medium (Phan et al. 2006). Based on this information, the reaction between phenylboronic acid and 4-iodoanisole was chosen (Table 3, entry 5), with 0.2 mmol of Hg(0) being added to the reactor concomitantly with the reagents. After 5 h of reaction, GC analysis showed that only 16% of the product (4-methoxybiphenyl) was formed, also indicating traces of biphenyl, which is the by-product from the homocoupling of phenylboronic acid. It is important to remember that in the absence of Hg(0) the same reaction had a yield of 98%.

In a second Hg(0) poisoning study, the same reaction started like as usual (Table 3, entry 5) but, after 1 h, 0.2 mmol de Hg(0) was added and the stirring continued until complete 5 h. Again, the appearance of reaction by-products was observed. An aliquot for GC analysis was removed before Hg(0) was added (reaction time of 1 h) and indicated a 55% yield of 4-methoxybiphenyl. This value did not change after the addition of Hg(0).

In summary, it can be said that the results of hot filtration and Hg(0) poisoning tests together indicate that the real catalytic species is unprotected Pd(0) and supported in CCF.

Conclusion

In conclusion, in this work two catalysts were produced containing Pd NPs and using easy-to-prepare catalytic supports from chitosan, cellulose, and corn stem. Such solids were properly characterized and tested in the Suzuki-Miyaura reaction, showing interesting catalytic activity. The catalytic systems reported can be characterized as sustainable due to the fact that they use eco-friendly reagents, do not use toxic auxiliary ligands, in addition to demonstrating excellent catalyst recyclability. It is believed that both catalysts have technological appeal given the simplicity of obtaining them, robustness, and low cost in the acquisition of their raw materials. Focusing on cost and comparing the mass of 1 g, the reagents needed for the production of the Pd/CCF catalyst cost about 6% of the value of the best-selling Pd-based catalyst, tetrakis (triphenylphosphine)palladium(0). Thus, our group believes in the potential of the catalysts reported here and continues their study seeking the development of new applications in cross-coupling reactions (mainly Heck-Mizoroki and Sonogashira). Also, with the results presented in the manuscript, it is expected to add value to biomass and agro-industrial residues, contributing to the state of the art in the development of new catalysts for the chemical industry.