The Current Status of Heterogeneous Palladium Catalysed Heck and Suzuki Cross-Coupling Reactions

In the last 30 years, C–C cross coupling reactions have become a reliable technique in organic synthesis due their versatility and efficiency. While drawbacks have been experienced on an industrial scale with the use of homogenous systems, many attempts have been made to facilitate a heterogeneous renaissance. Thus, this review gives an overview of the current status of the use of heterogeneous catalysts particularly in Suzuki and Heck reactions. Most recent developments focus on palladium immobilised or supported on various classes of supports, thus this review highlights and discuss contributions of the last decade.


Introduction
The importance of carbon-carbon bond forming reactions is well documented in literature [1][2][3]. The industrial importance of Suzuki and Heck cross-coupling reactions along with numerous others in this general field has sparked great interest in C-C bond formation reactions. In addition, the lack of a universal set of reaction conditions for every catalyst system and the drawbacks of industrially employed catalyst systems continue to fuel this interest. Heck and Suzuki reactions are typically catalysed by palladium based homogenous systems that have illustrated high turnover frequency (TOFs). These homogeneous catalysts require the use of ligands (phosphine or N-heterocyclic) to form active catalysts. As a result, separation of these catalysts (the ligand, the Pd metal or the complex) from the reaction media has presented the greatest challenge in this field. Consequently, the catalyst gets incorporated into the final product, resulting in loss of catalyst and a devalued product especially since these reactions are at the forefront of the production in the pharmaceutical industry. Successful separation of these homogeneous catalysts from the product solution requires the use of expensive nanofiltration membranes or an extensive (and usually destructive) column chromatography separation. Table 1 summarises the maximum limits of residual metal catalysts acceptable in the pharmaceutical industry.
These low allowed concentrations have prompted the development of recoverable and reuseable catalyst systems to decrease product contamination and alleviate the loss of catalyst. Many attempts have been made over the years to heterogenise homogeneous systems and to develop heterogeneous systems to displace homogenous systems and this review aims at highlighting recent contributions of the last decade at achieving this.
Alternatively, supported palladium nanoparticles as catalyst for C-C cross-coupling reactions are being developed. Most often, palladium nanoparticle become less catalytically active upon immobilisation on a solid support. Therefore, an improved catalytic system with desirable aspects of both systems (high activity and easy recovery) is needed. It is envisaged that better understanding of the reactivity of palladium nanoparticle can lead to the design of more efficient heterogeneous catalyst for C-C cross-coupling reactions.

Palladium Nanoparticles on Carbonaceous Supports
Palladium supported on carbonaceous supports has been widely applied in C-C cross-coupling reactions (Scheme 2) [16][17][18]. Amongst these, palladium on activated carbon (Pd/C) is by far the most frequently used catalyst in heterogeneous Pd-catalyzed coupling reactions. This is because of its efficiency and commercial availability [16][17][18]. It can be purchased in various quantities with a palladium content ranging from 1-20%. In addition, charcoal as a solid support ensures a higher surface area compared to the corresponding alumina and silica supported catalysts [18]. Palladium on activated carbon is reported to be stable in air and water, acids and bases, and it often doesn't require reactions to be performed under an inert atmosphere [18,19]. As a result, Pd/C catalyzed Heck and Suzuki cross-coupling reactions have been performed under a variety of reactions conditions, including in organic media [20][21][22][23], aqueous media [24][25][26], and under microwave conditions [27][28][29][30][31]. Generally, the Pd/C catalyst can convert a large variety of substrates with good to excellent yields, and it is recoverable and recyclable over many cycles [16][17][18]. Scheme 2. Schematic illustration of palladium nanoparticles supported on carbonaceous material as efficient catalyst for C-C cross-coupling reactions [32] (Adapted with permission from John Wiley and Sons). Scheme 2. Schematic illustration of palladium nanoparticles supported on carbonaceous material as efficient catalyst for C-C cross-coupling reactions [32] (Adapted with permission from John Wiley and Sons). Jadhav et al. recently (2016) developed a greener, cost effective and operationally convenient Pd/C based catalytic system for Suzuki and Heck cross-coupling reactions [33]. In their setup, the Pd/C catalyst was dispersed in an aqueous hydrotropic medium (sodium xylene sulphonate solution) that had surfactant-like properties and thus stabilised the Pd/C catalyst ( Figure 1). The hydrotropism allows for easy solubility of several types of organic functionalities in water, without the need for toxic and volatile organic co-solvent. In addition, good to excellent yields of the desired product were obtained and the catalytic system was recycled three times without any significant loss in activity. Lastly, they reported that the palladium metal was "not leached" out into the product solution. Hence, this system is highly desirable from economical as well as environmental points of view.
Molecules 2018, 23, x FOR PEER REVIEW 4 of 23 Jadhav et al. recently (2016) developed a greener, cost effective and operationally convenient Pd/C based catalytic system for Suzuki and Heck cross-coupling reactions [33]. In their setup, the Pd/C catalyst was dispersed in an aqueous hydrotropic medium (sodium xylene sulphonate solution) that had surfactant-like properties and thus stabilised the Pd/C catalyst ( Figure 1). The hydrotropism allows for easy solubility of several types of organic functionalities in water, without the need for toxic and volatile organic co-solvent. In addition, good to excellent yields of the desired product were obtained and the catalytic system was recycled three times without any significant loss in activity. Lastly, they reported that the palladium metal was "not leached" out into the product solution. Hence, this system is highly desirable from economical as well as environmental points of view. Carbon nanotubes (CNTs) have emerged as highly efficient supports for palladium and they have been extensively used as alternatives to activated carbon supports [12,32,34,35]. The CNTs can be evenly distributed in solution due to their small size, thus increasing interaction between the reactants and the catalyst. In most cases, the activity of CNTs in C-C cross-coupling reactions depends on their preparation method [35]. In a recent review, by Labulo et al., a wide range of surface functionalization techniques for carbon nanotubes that improve their properties as catalyst supports was reported [34]. The resulting CNTs catalysts displayed superior catalytic performance and better recyclability in C-C cross-coupling reactions. Scheme 3 shows the graphical representation of surface functionalization and palladium nanoparticle loading onto carbon nanotubes. Scheme 3. Graphical representation of Pd nanoparticles supported on carbon nanotubes [34] (Adapted with permission from Springer Nature). Carbon nanotubes (CNTs) have emerged as highly efficient supports for palladium and they have been extensively used as alternatives to activated carbon supports [12,32,34,35]. The CNTs can be evenly distributed in solution due to their small size, thus increasing interaction between the reactants and the catalyst. In most cases, the activity of CNTs in C-C cross-coupling reactions depends on their preparation method [35]. In a recent review, by Labulo et al., a wide range of surface functionalization techniques for carbon nanotubes that improve their properties as catalyst supports was reported [34]. The resulting CNTs catalysts displayed superior catalytic performance and better recyclability in C-C cross-coupling reactions. Scheme 3 shows the graphical representation of surface functionalization and palladium nanoparticle loading onto carbon nanotubes.  [33]. In their setup, the Pd/C catalyst was dispersed in an aqueous hydrotropic medium (sodium xylene sulphonate solution) that had surfactant-like properties and thus stabilised the Pd/C catalyst ( Figure 1). The hydrotropism allows for easy solubility of several types of organic functionalities in water, without the need for toxic and volatile organic co-solvent. In addition, good to excellent yields of the desired product were obtained and the catalytic system was recycled three times without any significant loss in activity. Lastly, they reported that the palladium metal was "not leached" out into the product solution. Hence, this system is highly desirable from economical as well as environmental points of view. Carbon nanotubes (CNTs) have emerged as highly efficient supports for palladium and they have been extensively used as alternatives to activated carbon supports [12,32,34,35]. The CNTs can be evenly distributed in solution due to their small size, thus increasing interaction between the reactants and the catalyst. In most cases, the activity of CNTs in C-C cross-coupling reactions depends on their preparation method [35]. In a recent review, by Labulo et al., a wide range of surface functionalization techniques for carbon nanotubes that improve their properties as catalyst supports was reported [34]. The resulting CNTs catalysts displayed superior catalytic performance and better recyclability in C-C cross-coupling reactions. Scheme 3 shows the graphical representation of surface functionalization and palladium nanoparticle loading onto carbon nanotubes. Scheme 3. Graphical representation of Pd nanoparticles supported on carbon nanotubes [34] (Adapted with permission from Springer Nature). Scheme 3. Graphical representation of Pd nanoparticles supported on carbon nanotubes [34] (Adapted with permission from Springer Nature).
Lakshminarayana et al. [32] extended the study by investigating the influence of various carbonaceous materials as supports in C-C cross-coupling reactions. They prepared and studied palladium oxide (PdO) nanoparticles impregnated on nanocarbon supports, such as single walled carbon nanotubes (SWCNT), multiwalled carbon nanotubes (MWCNT), carbon nanofiber (CNF), graphene oxide (GO), and reduced graphene oxide (RGO) [32]. The prepared catalysts were characterized fully and then explored in Heck cross-coupling reactions to assess their catalytic activity. Graphene oxide (GO) was found to be the best support for PdO ( Table 2). The enhanced catalytic activity of PdO/GO was believed to be due to the combined effect of a high degree of surface-bound oxygenated moieties, high surface area, electron conductivity and the mesoporous nature of graphene oxide. Lakshminarayana et al. [32] extended the study by investigating the influence of various carbonaceous materials as supports in C-C cross-coupling reactions. They prepared and studied palladium oxide (PdO) nanoparticles impregnated on nanocarbon supports, such as single walled carbon nanotubes (SWCNT), multiwalled carbon nanotubes (MWCNT), carbon nanofiber (CNF), graphene oxide (GO), and reduced graphene oxide (RGO) [32]. The prepared catalysts were characterized fully and then explored in Heck cross-coupling reactions to assess their catalytic activity. Graphene oxide (GO) was found to be the best support for PdO ( Table 2). The enhanced catalytic activity of PdO/GO was believed to be due to the combined effect of a high degree of surfacebound oxygenated moieties, high surface area, electron conductivity and the mesoporous nature of graphene oxide.

Palladium Nanoparticles on Metal Oxide Supports
A significant number of researchers have devoted themselves to finding a new and a more efficient Pd/MxOy based heterogeneous catalyst for C-C cross-coupling reactions [36,37]. In this direction, Köhler and co-workers prepared and comparatively studied the catalytic activity of Pd/Al2O3, Pd/TiO2, Pd/NaY, and Pd/CeO2 [38]. The catalysts were able to efficiently catalyse Suzuki cross-coupling reactions and the obtained TOFs for each catalyst were: Pd/Al2O3 = 9600 h −1 , Pd/TiO2 = 9700 h −1 , Pd/NaY = 4100, and Pd/CeO2 = 4100 h −1 . The Pd/Al2O3 and Pd/TiO2 catalysts showed very similar and high catalytic activities, while the activities of Pd/NaY and Pd/CeO2 were much lower. The TOF values of Pd/NaY and Pd/CeO2 catalysts are almost half of those obtained with Pd/Al2O3 and Pd/TiO2. These results clearly suggest that the nature of the support has a great influence on the activity of the catalyst.
In addition, Köhler et al. stated that the Pd concentration in solution during the reaction correlated clearly with the progress of the reaction ( Figure 2). Thus, this clearly suggests that the dissolved molecular palladium represents the "true" catalytically active species, and the prepared Pd/MxOy solids are simply precursors to the "true catalyst". Furthermore, they found that the dissolved "Pd is deposited back" onto the support at the end of the reaction. This phenomenon is commonly referred to as the palladium "release-capture or dissolution-redeposition" mechanism [39].

Palladium Nanoparticles on Metal Oxide Supports
A significant number of researchers have devoted themselves to finding a new and a more efficient Pd/M x O y based heterogeneous catalyst for C-C cross-coupling reactions [36,37]. In this direction, Köhler and co-workers prepared and comparatively studied the catalytic activity of Pd/Al 2 O 3 , Pd/TiO 2 , Pd/NaY, and Pd/CeO 2 [38]. The catalysts were able to efficiently catalyse Suzuki cross-coupling reactions and the obtained TOFs for each catalyst were: Pd/Al 2 O 3 = 9600 h −1 , Pd/TiO 2 = 9700 h −1 , Pd/NaY = 4100, and Pd/CeO 2 = 4100 h −1 . The Pd/Al 2 O 3 and Pd/TiO 2 catalysts showed very similar and high catalytic activities, while the activities of Pd/NaY and Pd/CeO 2 were much lower. The TOF values of Pd/NaY and Pd/CeO 2 catalysts are almost half of those obtained with Pd/Al 2 O 3 and Pd/TiO 2 . These results clearly suggest that the nature of the support has a great influence on the activity of the catalyst.
In addition, Köhler et al. stated that the Pd concentration in solution during the reaction correlated clearly with the progress of the reaction ( Figure 2). Thus, this clearly suggests that the dissolved molecular palladium represents the "true" catalytically active species, and the prepared Pd/M x O y solids are simply precursors to the "true catalyst". Furthermore, they found that the dissolved "Pd is deposited back" onto the support at the end of the reaction. This phenomenon is commonly referred to as the palladium "release-capture or dissolution-redeposition" mechanism [39].  In 2013, Amoroso et al. [10] extended the study of Köhler and co-workers [38] to examine the catalytic activity of Pd/MxOy type catalysts in more detail. More specifically, they wanted to investigate the influence of the "point of zero charge" on the catalytic activity of the Pd/MxOy type catalysts. The point of zero charge (PZC) is commonly used in surface science to determine the ability of a given solid material to adsorb ions. This concept describes the condition when the electrical charge density on a surface of a solid is zero [41]. Generally, the PZC is represented as the pH value at which a solid submerged in an electrolyte exhibits zero net electrical charge on the surface [41,42]. A given metal oxide (MxOy) surface will have a "positive charge" at solution pH values less than the PZC value and thus be a surface on which anions may adsorb. On the other hand, the metal oxide surface will have a negative charge at solution pH values greater than the PZC value and thus be a surface on which cations may adsorb [41,42]. The later statement is more relevant to the current discussion since the pH of the reaction mixture in most cross-coupling reactions is around 10, thus it is above the PZC values of the investigated metal oxide supports ( Figure 3). This allows one to investigate the degree of adhesion of palladium onto the support, which in turn speaks to the observed catalytic activity and the degree of palladium leaching.
Hence, Amoroso et al. [10] used the PZC values of each support to investigate the role it plays in the adhesion of the palladium nanoparticle onto the surface of the support ( Figure 3). Thus, they In 2013, Amoroso et al. [10] extended the study of Köhler and co-workers [38] to examine the catalytic activity of Pd/M x O y type catalysts in more detail. More specifically, they wanted to investigate the influence of the "point of zero charge" on the catalytic activity of the Pd/M x O y type catalysts. The point of zero charge (PZC) is commonly used in surface science to determine the ability of a given solid material to adsorb ions. This concept describes the condition when the electrical charge density on a surface of a solid is zero [40]. Generally, the PZC is represented as the pH value at which a solid submerged in an electrolyte exhibits zero net electrical charge on the surface [40,41]. A given metal oxide (M x O y ) surface will have a "positive charge" at solution pH values less than the PZC value and thus be a surface on which anions may adsorb. On the other hand, the metal oxide surface will have a negative charge at solution pH values greater than the PZC value and thus be a surface on which cations may adsorb [40,41]. The later statement is more relevant to the current discussion since the pH of the reaction mixture in most cross-coupling reactions is around 10, thus it is above the PZC values of the investigated metal oxide supports ( Figure 3). This allows one to investigate the degree of adhesion of palladium onto the support, which in turn speaks to the observed catalytic activity and the degree of palladium leaching.  In 2013, Amoroso et al. [10] extended the study of Köhler and co-workers [38] to examine the catalytic activity of Pd/MxOy type catalysts in more detail. More specifically, they wanted to investigate the influence of the "point of zero charge" on the catalytic activity of the Pd/MxOy type catalysts. The point of zero charge (PZC) is commonly used in surface science to determine the ability of a given solid material to adsorb ions. This concept describes the condition when the electrical Hence, Amoroso et al. [10] used the PZC values of each support to investigate the role it plays in the adhesion of the palladium nanoparticle onto the surface of the support ( Figure 3). Thus, they prepared and investigated the catalytic activity of Pd/La 2 O 3 , Pd/Pr 6 O 11 , Pd/Gd 2 O 3 , Pd/Sm 2 O 3 , and Pd/CeO 2 catalysts. The physicochemical properties of these materials are tabulated in Table 3. The study revealed that under basic reaction conditions (pH~10.5), the surfaces of metal oxides with high PZC values are less negatively charged than the surfaces of metal oxides with low PZC values. Thus, the La 2 O 3 support has the lowest density of negatively charged sites on the surface since it has the highest PZC value. Hence, the higher reactivity of Pd/La 2 O 3 is correlated to a marked degree of Pd 2+ leaching from the surface of La 2 O 3 , favoured by its less negatively charged surface. On the contrary, the lowest activity observed with Pd/CeO 2 is correlated to the higher degree of negatively charged surface that helps in keeping Pd 2+ ions anchored onto the support (Figure 3). prepared and investigated the catalytic activity of Pd/La2O3, Pd/Pr6O11, Pd/Gd2O3, Pd/Sm2O3, and Pd/CeO2 catalysts. The physicochemical properties of these materials are tabulated in Table 3. The study revealed that under basic reaction conditions (pH ~ 10.5), the surfaces of metal oxides with high PZC values are less negatively charged than the surfaces of metal oxides with low PZC values. Thus, the La2O3 support has the lowest density of negatively charged sites on the surface since it has the highest PZC value. Hence, the higher reactivity of Pd/La2O3 is correlated to a marked degree of Pd 2+ leaching from the surface of La2O3, favoured by its less negatively charged surface. On the contrary, the lowest activity observed with Pd/CeO2 is correlated to the higher degree of negatively charged surface that helps in keeping Pd 2+ ions anchored onto the support ( Figure 3).  Figure 4 further shows that there is a linear correlation between the yield of the coupling product and the PZC ([H3O + ] = 10 −PZC ) value of the metal oxide support. Thus, the observed reactivity trend over the MxOy support is related to the tendency of each Pd/MxOy system to deliver different amounts of palladium into the solution. In summary, the above discussion shows that there is a close relation between the "surface charge" (correlated to the PZC value) and the extent of palladium leaching at the pH of catalysis (determined by reaction conditions). Thus, the more positively charged is the surface (this is the case of Pd/La2O3), the more consistently Pd leaches, and therefore, a higher reaction rate is observed. In addition, a much slower reaction is observed when Pd/CeO2 is used as the catalyst, which shows the lowest PZC value within the series of Pd/MxOy. Consequently, prolonged recycling is possible in the case of Pd/CeO2 due to minimal palladium leaching. Thus, from the studied series of Pd/MxOy type catalysts, Pd/CeO2 was chosen as superior precatalyst based on the following considerations: (i) it can be recycled several times without a significant decrease of activity (prolonged recyclability) and (ii) the organic product is sparsely contaminated since there is minimal leaching of palladium into the solution. a Reaction conditions: 1-bromo-4-nitrobenzene (0.5 mmol), 4-methylphenylboronic acid (0.6 mmol), K 2 CO 3 (0.6 mmol), catalyst, 1.5 mL ethanol, 0.5 mL water, temperature = 25 • C. Figure 4 further shows that there is a linear correlation between the yield of the coupling product and the PZC ([H 3 O + ] = 10 −PZC ) value of the metal oxide support. Thus, the observed reactivity trend over the M x O y support is related to the tendency of each Pd/M x O y system to deliver different amounts of palladium into the solution. In summary, the above discussion shows that there is a close relation between the "surface charge" (correlated to the PZC value) and the extent of palladium leaching at the pH of catalysis (determined by reaction conditions). Thus, the more positively charged is the surface (this is the case of Pd/La 2 O 3 ), the more consistently Pd leaches, and therefore, a higher reaction rate is observed. In addition, a much slower reaction is observed when Pd/CeO 2 is used as the catalyst, which shows the lowest PZC value within the series of Pd/M x O y . Consequently, prolonged recycling is possible in the case of Pd/CeO 2 due to minimal palladium leaching. Thus, from the studied series of Pd/M x O y type catalysts, Pd/CeO 2 was chosen as superior precatalyst based on the following considerations: (i) it can be recycled several times without a significant decrease of activity (prolonged recyclability) and (ii) the organic product is sparsely contaminated since there is minimal leaching of palladium into the solution.  Alternatively, the physicochemical properties of the support can be enhanced through substitution of a foreign metal ion in its lattice structure [44]. The substitution of foreign metal ions into the lattice of reducible oxides such as CeO2 and TiO2 to form ionic solid-solution oxides (Ce1-xMxO2-δ or Ti1-xMxO2-δ) has given materials which show high activity for a variety of catalytic reactions, especially in gas-phase oxidation reactions [11,[44][45][46][47][48][49]. However, such catalysts have rarely been applied to C-C cross-coupling reactions [50][51][52]. Recently, Burange and co-workers [45] reported that the ionic ceria-zirconia (CeZrO4-δ) solid-solution oxides exhibit high redox properties Alternatively, the physicochemical properties of the support can be enhanced through substitution of a foreign metal ion in its lattice structure [43]. The substitution of foreign metal ions into the lattice of reducible oxides such as CeO 2 and TiO 2 to form ionic solid-solution oxides (Ce 1-x M x O 2-δ or Ti 1-x M x O 2-δ ) has given materials which show high activity for a variety of catalytic reactions, especially in gas-phase oxidation reactions [11,[43][44][45][46][47][48]. However, such catalysts have rarely been applied to C-C cross-coupling reactions [49][50][51]. Recently, Burange and co-workers [44] reported that the ionic ceria-zirconia (CeZrO 4-δ ) solid-solution oxides exhibit high redox properties and thermal stability that make them better catalyst supports than the pure metal oxide counterparts [52]. Hence, CeZrO 4-δ solid-solution oxide was used to immobilise palladium nanoparticles and the resultant catalyst (Pd/CeZrO 4-δ ) was explored in Suzuki cross-coupling reactions ( Figure 5). The mechanistic investigation proved that the redox couple (Ce 4+ /Ce 3+ ) in the CeZrO 4-δ support enhances the catalytic activity through creation of oxygen vacancies. Furthermore, the support displayed strong metal-support (Pd-CeZrO 4-δ ) interactions and hence, "no leaching" was observed. Thus, it was concluded that the reactions were "truly" heterogeneous [52].
Lichtenegger and co-workers [53] extended the Burange et al. [52] study by directly incorporating palladium ions into the lattice of the support (ceria), instead of modifying the properties of the support with a foreign metal. In their study, they used CeO 2 and SnO 2 as supports and  for Pd/MxOy [10] (Adapted with permission from Springer Nature).

Figure 5.
Plausible heterogeneous catalytic reaction mechanism suggested for Suzuki cross-coupling over the Pd/CeZrO4-δ redox catalyst [43] (Adapted with permission from John Wiley and Sons).
Alternatively, the physicochemical properties of the support can be enhanced through substitution of a foreign metal ion in its lattice structure [44]. The substitution of foreign metal ions into the lattice of reducible oxides such as CeO2 and TiO2 to form ionic solid-solution oxides (Ce1-xMxO2-δ or Ti1-xMxO2-δ) has given materials which show high activity for a variety of catalytic reactions, especially in gas-phase oxidation reactions [11,[44][45][46][47][48][49]. However, such catalysts have rarely been applied to C-C cross-coupling reactions [50][51][52]. Recently, Burange and co-workers [45] reported that the ionic ceria-zirconia (CeZrO4-δ) solid-solution oxides exhibit high redox properties  and thermal stability that make them better catalyst supports than the pure metal oxide counterparts [43]. Hence, CeZrO4-δ solid-solution oxide was used to immobilise palladium nanoparticles and the resultant catalyst (Pd/CeZrO4-δ) was explored in Suzuki cross-coupling reactions ( Figure 5). The mechanistic investigation proved that the redox couple (Ce 4+ /Ce 3+ ) in the CeZrO4-δ support enhances the catalytic activity through creation of oxygen vacancies. Furthermore, the support displayed strong metal-support (Pd-CeZrO4-δ) interactions and hence, "no leaching" was observed. Thus, it was concluded that the reactions were "truly" heterogeneous [43].
Lichtenegger and co-workers [53] extended the Burange et al. [43] study by directly incorporating palladium ions into the lattice of the support (ceria), instead of modifying the properties of the support with a foreign metal. In their study, they used CeO2 and SnO2 as supports and prepared a series of corresponding ionic solid-solution oxides: Ce0.99Pd0.01O2-δ, Ce0.79Sn0.2Pd0.01O2-δ, Ce0.495Sn0.495Pd0.01O2-δ, Ce0.20Sn0.79Pd0.01O2-δ, and Sn0.99Pd0.01O2-δ. These catalysts were thoroughly explored in Suzuki coupling reactions of phenylboric acid with various bromoarenes ( Figure 6). The Ce0.79Sn0.2Pd0.01O2-δ, Ce0.20Sn0.79Pd0.01O2-δ, and Sn0.99Pd0.01O2-δ catalysts showed extraordinarily high activities in Suzuki cross-coupling reactions, while the binary solid-solution oxide Ce0.99Pd0.01O2-δ proved to be the least active catalyst. Thus, the Sn containing catalysts were shown to be more active; however, the results do not show a clear correlation between Sn loadings and the catalytic activity. Furthermore, complete conversion was achieved in five subsequent reactions for all catalysts, except when Ce0.99Pd0.01O2-δ and Ce0.495Sn0.495Pd0.01O2-δ were used. Thorough catalyst leaching, recovery and recyclability studies were conducted and the results demonstrate a clear correlation between reactivity and amount of leached palladium (Figure 7). Hence, these findings also support the hypothesis that the coupling reaction is catalyzed by small amounts of leached palladium via a homogeneous reaction mechanism. Thorough catalyst leaching, recovery and recyclability studies were conducted and the results demonstrate a clear correlation between reactivity and amount of leached palladium (Figure 7). Hence, these findings also support the hypothesis that the coupling reaction is catalyzed by small amounts of leached palladium via a homogeneous reaction mechanism.  To further explore this concept of "release-capture mechanism" by ionic solid-solution oxides catalysts, we have reported on the application of a highly Pd substituted ceria catalyst (Pd0.09Ce0.91O2-δ) in the Heck cross-coupling reaction [50]. Good to excellent yields of substituted olefins were obtained with this Pd0.09Ce0.91O2-δ solid-solution catalyst. However, our findings also revealed that the catalysis occurs over the reduced two phase Pd 0 /CeO2 and not on the as prepared monophasic Pd0.09Ce0.91O2-δ (Scheme 4). In addition, it was observed that the Pd 0 /CeO2 formed in situ can be easily recovered and reused for several times without any significant loss in efficiency and only a negligible amount of palladium was detected in the product solution (0.35 ppm). The low level of palladium concentration in the product solution further highlights the excellent ability of ceria to re-adsorb leached palladium species, as discussed earlier.
We were also able to show that the reaction conditions greatly affect the catalysis. In this regard, we investigated the application of a Pd0.04Cu0.04Ce0.92O2-δ (PdCuCeO) solid-solution oxide on Suzuki cross-coupling reactions (Scheme 5) [51]. The reactions were carried-out under milder and more environmentally friendly reaction conditions, using water as sole solvent and tetrapropylammonium bromide (TPAB) as a phase transfer catalyst. The catalytic results displayed good functional group To further explore this concept of "release-capture mechanism" by ionic solid-solution oxides catalysts, we have reported on the application of a highly Pd substituted ceria catalyst (Pd 0.09 Ce 0.91 O 2-δ ) in the Heck cross-coupling reaction [49]. Good to excellent yields of substituted olefins were obtained with this Pd 0.09 Ce 0.91 O 2-δ solid-solution catalyst. However, our findings also revealed that the catalysis occurs over the reduced two phase Pd 0 /CeO 2 and not on the as prepared monophasic Pd 0.09 Ce 0.91 O 2-δ (Scheme 4). In addition, it was observed that the Pd 0 /CeO 2 formed in situ can be easily recovered and reused for several times without any significant loss in efficiency and only a negligible amount of palladium was detected in the product solution (0.35 ppm). The low level of palladium concentration in the product solution further highlights the excellent ability of ceria to re-adsorb leached palladium species, as discussed earlier. To further explore this concept of "release-capture mechanism" by ionic solid-solution oxides catalysts, we have reported on the application of a highly Pd substituted ceria catalyst (Pd0.09Ce0.91O2-δ) in the Heck cross-coupling reaction [50]. Good to excellent yields of substituted olefins were obtained with this Pd0.09Ce0.91O2-δ solid-solution catalyst. However, our findings also revealed that the catalysis occurs over the reduced two phase Pd 0 /CeO2 and not on the as prepared monophasic Pd0.09Ce0.91O2-δ (Scheme 4). In addition, it was observed that the Pd 0 /CeO2 formed in situ can be easily recovered and reused for several times without any significant loss in efficiency and only a negligible amount of palladium was detected in the product solution (0.35 ppm). The low level of palladium concentration in the product solution further highlights the excellent ability of ceria to re-adsorb leached palladium species, as discussed earlier. We were also able to show that the reaction conditions greatly affect the catalysis. In this regard, we investigated the application of a Pd0.04Cu0.04Ce0.92O2-δ (PdCuCeO) solid-solution oxide on Suzuki cross-coupling reactions (Scheme 5) [51]. The reactions were carried-out under milder and more environmentally friendly reaction conditions, using water as sole solvent and tetrapropylammonium bromide (TPAB) as a phase transfer catalyst. The catalytic results displayed good functional group We were also able to show that the reaction conditions greatly affect the catalysis. In this regard, we investigated the application of a Pd 0.04 Cu 0.04 Ce 0.92 O 2-δ (PdCuCeO) solid-solution oxide on Suzuki cross-coupling reactions (Scheme 5) [50]. The reactions were carried-out under milder and more environmentally friendly reaction conditions, using water as sole solvent and tetrapropylammonium bromide (TPAB) as a phase transfer catalyst. The catalytic results displayed good functional group tolerance and good to excellent yields of substituted biphenyls were obtained. However, catalyst leaching and recyclability studies revealed that PdCuCeO acts as a Pd reservoir and that the reactions were essentially quasi-heterogeneous, occurring over recoverable Pd/TPAB aggregates (generated in situ). The PdCuCeO solid-solution catalyst was recovered and recycled three times; however, its catalytic activity declined with each subsequent recycle. tolerance and good to excellent yields of substituted biphenyls were obtained. However, catalyst leaching and recyclability studies revealed that PdCuCeO acts as a Pd reservoir and that the reactions were essentially quasi-heterogeneous, occurring over recoverable Pd/TPAB aggregates (generated in situ). The PdCuCeO solid-solution catalyst was recovered and recycled three times; however, its catalytic activity declined with each subsequent recycle.
Hence, it can be concluded that most of the reported ionic solid-solution oxides catalysts are simply precatalysts that act as a palladium reservoir for the active palladium species. Therefore, the observed high activity and recyclability of the catalysts could be attributed to a Pd release-capture mechanism, as discussed earlier.

Palladium Nanoparticle Immobilised on Magnetic Supports
Recent studies show that magnetic nanoparticles are excellent supports for various catalysts [54][55][56][57][58][59][60][61]. Additionally, the magnetic properties can be exploited to improve catalyst separation from previous filtration and centrifugation methods. Supported palladium magnetic catalysts have the benefits of easy recovery from the reaction media by the application of an external magnetic field ( Figure 8) [57]. Magnetic nanoparticles can be categorized into four main groups, namely; metals (Fe, Co, Ni), alloys (FePt, FePd), ferrites (CoFe2O4, CuFe2O4), and most notably metal oxides (FeO, Fe2O3, Fe3O4). Iron oxides are the most widely employed in literature because they have stronger magnetic properties and can be synthesized easily by co-precipitation methods [62].  Hence, it can be concluded that most of the reported ionic solid-solution oxides catalysts are simply precatalysts that act as a palladium reservoir for the active palladium species. Therefore, the observed high activity and recyclability of the catalysts could be attributed to a Pd release-capture mechanism, as discussed earlier.

Palladium Nanoparticle Immobilised on Magnetic Supports
Recent studies show that magnetic nanoparticles are excellent supports for various catalysts [54][55][56][57][58][59][60][61]. Additionally, the magnetic properties can be exploited to improve catalyst separation from previous filtration and centrifugation methods. Supported palladium magnetic catalysts have the benefits of easy recovery from the reaction media by the application of an external magnetic field ( Figure 8) [57]. Magnetic nanoparticles can be categorized into four main groups, namely; metals (Fe, Co, Ni), alloys (FePt, FePd), ferrites (CoFe 2 O 4 , CuFe 2 O 4 ), and most notably metal oxides (FeO, Fe 2 O 3 , Fe 3 O 4 ). Iron oxides are the most widely employed in literature because they have stronger magnetic properties and can be synthesized easily by co-precipitation methods [62]. tolerance and good to excellent yields of substituted biphenyls were obtained. However, catalyst leaching and recyclability studies revealed that PdCuCeO acts as a Pd reservoir and that the reactions were essentially quasi-heterogeneous, occurring over recoverable Pd/TPAB aggregates (generated in situ). The PdCuCeO solid-solution catalyst was recovered and recycled three times; however, its catalytic activity declined with each subsequent recycle.
Hence, it can be concluded that most of the reported ionic solid-solution oxides catalysts are simply precatalysts that act as a palladium reservoir for the active palladium species. Therefore, the observed high activity and recyclability of the catalysts could be attributed to a Pd release-capture mechanism, as discussed earlier.

Palladium Nanoparticle Immobilised on Magnetic Supports
Recent studies show that magnetic nanoparticles are excellent supports for various catalysts [54][55][56][57][58][59][60][61]. Additionally, the magnetic properties can be exploited to improve catalyst separation from previous filtration and centrifugation methods. Supported palladium magnetic catalysts have the benefits of easy recovery from the reaction media by the application of an external magnetic field ( Figure 8) [57]. Magnetic nanoparticles can be categorized into four main groups, namely; metals (Fe, Co, Ni), alloys (FePt, FePd), ferrites (CoFe2O4, CuFe2O4), and most notably metal oxides (FeO, Fe2O3, Fe3O4). Iron oxides are the most widely employed in literature because they have stronger magnetic properties and can be synthesized easily by co-precipitation methods [62].  A growing number of these magnetic supports have been applied in the development of magnetically recoverable palladium based heterogeneous catalysts for C-C cross-coupling reactions. Nasrollahzadeh and co-workers have explored the application of Pd/Fe 3 O 4 magnetic nanoparticles as an efficient catalyst for C-C cross-coupling reactions [64]. The Pd/Fe 3 O 4 magnetic catalyst was found to be stable during the Suzuki coupling reaction and good yields (83-95%) were obtained over a wide range of substrates. At the end of the Suzuki reaction the catalyst was recovered by the application of an external magnet, and successively reused five times without significant loss of activity.
Although bare or "naked" iron oxide nanoparticles have been shown to be efficient supports for cross-coupling reactions, the immobilization of palladium directly on the iron oxide occasionally suffers from aggregation or surface oxidation. The oxidation and agglomeration of magnetic nanoparticles results in the loss of their magnetism; thus, it is crucial to develop efficient strategies to strengthen their chemical stability [60]. In this regard, a wide range of stabilizing or coating materials, including organic stabilizers (polymers and surfactants) and inorganic stabilizers (silica and carbon materials), have been used to protect magnetic nanoparticles [60,65]. In this direction, Kumar et al. [66] developed efficient C@Fe 3 O 4 magnetic core-shell nano-spheres by coating the Fe 3 O 4 nanoparticle with a carbon shell to protect them from being corrupted or oxidized under cross-coupling reaction conditions. Palladium was then immobilised on the synthesized C@Fe 3 O 4 magnetic core-shell to form a structurally stable Pd/C@Fe 3 O 4 catalyst (Figures 9 and 10). This catalyst was efficiently used as a heterogeneous catalyst for Heck cross-coupling reactions and good to excellent yields were obtained. The catalyst was easily separated from the reaction mixture through application of an external magnet and the catalyst was reused five times without any significant drop in activity. The authors claim that this catalytic system is "truly heterogeneous" even though some leaching was observed. They concluded that their catalytic system is advantageous because of its heterogeneity, high stability, gram scale applicability, magnetic separability, and consequent reusability [66]. A growing number of these magnetic supports have been applied in the development of magnetically recoverable palladium based heterogeneous catalysts for C-C cross-coupling reactions. Nasrollahzadeh and co-workers have explored the application of Pd/Fe3O4 magnetic nanoparticles as an efficient catalyst for C-C cross-coupling reactions [64]. The Pd/Fe3O4 magnetic catalyst was found to be stable during the Suzuki coupling reaction and good yields (83-95%) were obtained over a wide range of substrates. At the end of the Suzuki reaction the catalyst was recovered by the application of an external magnet, and successively reused five times without significant loss of activity.
Although bare or "naked" iron oxide nanoparticles have been shown to be efficient supports for cross-coupling reactions, the immobilization of palladium directly on the iron oxide occasionally suffers from aggregation or surface oxidation. The oxidation and agglomeration of magnetic nanoparticles results in the loss of their magnetism; thus, it is crucial to develop efficient strategies to strengthen their chemical stability [60]. In this regard, a wide range of stabilizing or coating materials, including organic stabilizers (polymers and surfactants) and inorganic stabilizers (silica and carbon materials), have been used to protect magnetic nanoparticles [60,65]. In this direction, Kumar et al. [66] developed efficient C@Fe3O4 magnetic core-shell nano-spheres by coating the Fe3O4 nanoparticle with a carbon shell to protect them from being corrupted or oxidized under cross-coupling reaction conditions. Palladium was then immobilised on the synthesized C@Fe3O4 magnetic core-shell to form a structurally stable Pd/C@Fe3O4 catalyst (Figures 9 and 10). This catalyst was efficiently used as a heterogeneous catalyst for Heck cross-coupling reactions and good to excellent yields were obtained. The catalyst was easily separated from the reaction mixture through application of an external magnet and the catalyst was reused five times without any significant drop in activity. The authors claim that this catalytic system is "truly heterogeneous" even though some leaching was observed. They concluded that their catalytic system is advantageous because of its heterogeneity, high stability, gram scale applicability, magnetic separability, and consequent reusability [66].   A growing number of these magnetic supports have been applied in the development of magnetically recoverable palladium based heterogeneous catalysts for C-C cross-coupling reactions. Nasrollahzadeh and co-workers have explored the application of Pd/Fe3O4 magnetic nanoparticles as an efficient catalyst for C-C cross-coupling reactions [64]. The Pd/Fe3O4 magnetic catalyst was found to be stable during the Suzuki coupling reaction and good yields (83-95%) were obtained over a wide range of substrates. At the end of the Suzuki reaction the catalyst was recovered by the application of an external magnet, and successively reused five times without significant loss of activity.
Although bare or "naked" iron oxide nanoparticles have been shown to be efficient supports for cross-coupling reactions, the immobilization of palladium directly on the iron oxide occasionally suffers from aggregation or surface oxidation. The oxidation and agglomeration of magnetic nanoparticles results in the loss of their magnetism; thus, it is crucial to develop efficient strategies to strengthen their chemical stability [60]. In this regard, a wide range of stabilizing or coating materials, including organic stabilizers (polymers and surfactants) and inorganic stabilizers (silica and carbon materials), have been used to protect magnetic nanoparticles [60,65]. In this direction, Kumar et al. [66] developed efficient C@Fe3O4 magnetic core-shell nano-spheres by coating the Fe3O4 nanoparticle with a carbon shell to protect them from being corrupted or oxidized under cross-coupling reaction conditions. Palladium was then immobilised on the synthesized C@Fe3O4 magnetic core-shell to form a structurally stable Pd/C@Fe3O4 catalyst (Figures 9 and 10). This catalyst was efficiently used as a heterogeneous catalyst for Heck cross-coupling reactions and good to excellent yields were obtained. The catalyst was easily separated from the reaction mixture through application of an external magnet and the catalyst was reused five times without any significant drop in activity. The authors claim that this catalytic system is "truly heterogeneous" even though some leaching was observed. They concluded that their catalytic system is advantageous because of its heterogeneity, high stability, gram scale applicability, magnetic separability, and consequent reusability [66].   The coated magnetic nanoparticles can be further modified by functionalising their surface with various functional groups ( Figure 11). The functionalization of the surface of magnetic nanomaterials is an important step in the design of supported catalysts, because it significantly improves the physical and chemical properties of the magnetic supports [63]. The coated magnetic nanoparticles can be further modified by functionalising their surface with various functional groups ( Figure 11). The functionalization of the surface of magnetic nanomaterials is an important step in the design of supported catalysts, because it significantly improves the physical and chemical properties of the magnetic supports [63]. Heidari and co-worker [67] used isoniazide to functionalize the surface of the Fe3O4@SiO2 magnetic nano-support ( Figure 12). The isoniazide groups were used as linkers to immobilize palladium nanoparticles and prevent agglomeration on the surface of the Fe3O4@SiO2 magnetic nanosupport [67]. The resulting material, Fe3O4@SiO2/isoniazide/Pd catalyst, was explored as a heterogeneous catalyst in Suzuki coupling reactions and good to excellent yields were obtained. The authors reported that the benefits of their heterogeneous catalyst system were its high efficiency and reuse was easily achieved through the use of a magnet. They were able to recover and recycle their catalyst six times without any noticeable loss in activity.  Heidari and co-worker [67] used isoniazide to functionalize the surface of the Fe 3 O 4 @SiO 2 magnetic nano-support ( Figure 12). The isoniazide groups were used as linkers to immobilize palladium nanoparticles and prevent agglomeration on the surface of the Fe 3 O 4 @SiO 2 magnetic nano-support [67]. The resulting material, Fe 3 O 4 @SiO 2 /isoniazide/Pd catalyst, was explored as a heterogeneous catalyst in Suzuki coupling reactions and good to excellent yields were obtained. The authors reported that the benefits of their heterogeneous catalyst system were its high efficiency and reuse was easily achieved through the use of a magnet. They were able to recover and recycle their catalyst six times without any noticeable loss in activity. The coated magnetic nanoparticles can be further modified by functionalising their surface with various functional groups ( Figure 11). The functionalization of the surface of magnetic nanomaterials is an important step in the design of supported catalysts, because it significantly improves the physical and chemical properties of the magnetic supports [63]. Heidari and co-worker [67] used isoniazide to functionalize the surface of the Fe3O4@SiO2 magnetic nano-support ( Figure 12). The isoniazide groups were used as linkers to immobilize palladium nanoparticles and prevent agglomeration on the surface of the Fe3O4@SiO2 magnetic nanosupport [67]. The resulting material, Fe3O4@SiO2/isoniazide/Pd catalyst, was explored as a heterogeneous catalyst in Suzuki coupling reactions and good to excellent yields were obtained. The authors reported that the benefits of their heterogeneous catalyst system were its high efficiency and reuse was easily achieved through the use of a magnet. They were able to recover and recycle their catalyst six times without any noticeable loss in activity.  An interesting but slightly different approach was reported by Khalili and co-workers [12], who supported palladium nanoparticles on an amino-vinyl silica functionalized magnetic carbon nanotube (CNT). Their core-shell contains CNT@Fe 3 O 4 @SiO 2 -Pd in which the functionalized SiO 2 helps with the stability of the Pd nanoparticles and is also responsible for the reduction of Pd(II) to Pd(0) without the need for adding external reducing agents ( Figure 13). This catalyst was successfully applied to Heck and Suzuki cross-coupling reactions and high yields were obtained. In addition, the catalyst exhibited good recyclability, and was used for six consecutive recycles without a significant loss in catalytic activity. An interesting but slightly different approach was reported by Khalili and co-workers [12], who supported palladium nanoparticles on an amino-vinyl silica functionalized magnetic carbon nanotube (CNT). Their core-shell contains CNT@Fe3O4@SiO2-Pd in which the functionalized SiO2 helps with the stability of the Pd nanoparticles and is also responsible for the reduction of Pd(II) to Pd(0) without the need for adding external reducing agents ( Figure 13). This catalyst was successfully applied to Heck and Suzuki cross-coupling reactions and high yields were obtained. In addition, the catalyst exhibited good recyclability, and was used for six consecutive recycles without a significant loss in catalytic activity. There are also numerous examples where organopalladium complexes, instead of palladium nanoparticles, have been immobilised on magnetic supports [68,69]. Recently, Collinson and coworkers [68] prepared a [(NHC)Pd(allyl)Cl] complex, bearing an N-heterocyclic carbene (NHC) and immobilised it on silica-coated magnetic nanoparticles (Scheme 6). Their strategy tries to combine the high activity of palladium-NHC catalysts with the facile separation and recyclability of a heterogeneous catalyst. The catalyst was then explored on Suzuki coupling reactions and good to excellent yields were obtained with a variety of substrates. However, the immobilised palladium complex could not be reused because the catalyst disassembled or decomposed under Suzuki coupling reaction conditions. The recycled catalyst only achieved 20% yield compared to the 90% yield that was obtained in its first use. The authors suspected that the base was responsible for the detachment of the magnetic silica linker.  There are also numerous examples where organopalladium complexes, instead of palladium nanoparticles, have been immobilised on magnetic supports [68,69]. Recently, Collinson and co-workers [68] prepared a [(NHC)Pd(allyl)Cl] complex, bearing an N-heterocyclic carbene (NHC) and immobilised it on silica-coated magnetic nanoparticles (Scheme 6). Their strategy tries to combine the high activity of palladium-NHC catalysts with the facile separation and recyclability of a heterogeneous catalyst. The catalyst was then explored on Suzuki coupling reactions and good to excellent yields were obtained with a variety of substrates. However, the immobilised palladium complex could not be reused because the catalyst disassembled or decomposed under Suzuki coupling reaction conditions. The recycled catalyst only achieved 20% yield compared to the 90% yield that was obtained in its first use. The authors suspected that the base was responsible for the detachment of the magnetic silica linker. An interesting but slightly different approach was reported by Khalili and co-workers [12], who supported palladium nanoparticles on an amino-vinyl silica functionalized magnetic carbon nanotube (CNT). Their core-shell contains CNT@Fe3O4@SiO2-Pd in which the functionalized SiO2 helps with the stability of the Pd nanoparticles and is also responsible for the reduction of Pd(II) to Pd(0) without the need for adding external reducing agents ( Figure 13). This catalyst was successfully applied to Heck and Suzuki cross-coupling reactions and high yields were obtained. In addition, the catalyst exhibited good recyclability, and was used for six consecutive recycles without a significant loss in catalytic activity. There are also numerous examples where organopalladium complexes, instead of palladium nanoparticles, have been immobilised on magnetic supports [68,69]. Recently, Collinson and coworkers [68] prepared a [(NHC)Pd(allyl)Cl] complex, bearing an N-heterocyclic carbene (NHC) and immobilised it on silica-coated magnetic nanoparticles (Scheme 6). Their strategy tries to combine the high activity of palladium-NHC catalysts with the facile separation and recyclability of a heterogeneous catalyst. The catalyst was then explored on Suzuki coupling reactions and good to excellent yields were obtained with a variety of substrates. However, the immobilised palladium complex could not be reused because the catalyst disassembled or decomposed under Suzuki coupling reaction conditions. The recycled catalyst only achieved 20% yield compared to the 90% yield that was obtained in its first use. The authors suspected that the base was responsible for the detachment of the magnetic silica linker. Better results were obtained by Fareghi-Alamdari et al. [69] who supported bis(N-heterocyclic carbene) palladium complex on silica coated magnetic nanoparticles ( Figure 14). Better results were obtained by Fareghi-Alamdari et al. [69] who supported bis(N-heterocyclic carbene) palladium complex on silica coated magnetic nanoparticles ( Figure 14). Their catalyst was efficiently used for Suzuki coupling reactions and at the end of the reaction, it was separated though application of an external magnet and recycled for six consecutive reactions. Hence, it not easy to identify the reaction parameters that caused the Collinson and co-workers [68] catalytic system to be unrecyclable because the reaction conditions are different (Table 4). However, the most obvious difference is the choice of the base; Collinson and co-workers used NaOH, which is very harsh, while Fareghi-Alamdari et al. used K2CO3. Hence, optimisation of reaction conditions is crucial in developing efficient and more robust catalytic systems.
In summary, the literature shows that magnetic nanoparticle-based catalytic systems have a huge potential in solving the recoverability and recyclability challenges [57].

Polymer Supported Palladium Nanoparticles
Literature reports have repeatedly shown that palladium nanoparticles can be efficiently supported on polymer frameworks [70][71][72][73][74][75][76][77]. Recently, Nemygina and co-workers immobilised various palladium precursors (PdCl2, PdCl2(CH3CN)2, and PdCl2(PhCN)2) on amino-functionalized hyper-crosslinked polystyrene [78]. They then evaluated the influence of palladium oxidation state (Pd(II) or Pd(0)) on the rate of Suzuki cross-coupling of 4-bromoanisole and phenylboronic acid. Their results revealed that the catalyst impregnated with a PdCl2(CH3CN)2 precursor resulted in better catalytic activity (Scheme 7). The PdCl2(CH3CN)2 precursor was believed to be more active because it experienced less hydrolysis and precipitation in comparison with PdCl2 and PdCl2(PhCN)2 under Suzuki coupling reaction conditions. Lastly, they observed that unreduced, Pd(II) containing precatalyst were more active [78]. Their catalyst was efficiently used for Suzuki coupling reactions and at the end of the reaction, it was separated though application of an external magnet and recycled for six consecutive reactions. Hence, it not easy to identify the reaction parameters that caused the Collinson and co-workers [68] catalytic system to be unrecyclable because the reaction conditions are different (Table 4). However, the most obvious difference is the choice of the base; Collinson and co-workers used NaOH, which is very harsh, while Fareghi-Alamdari et al. used K 2 CO 3 . Hence, optimisation of reaction conditions is crucial in developing efficient and more robust catalytic systems. In summary, the literature shows that magnetic nanoparticle-based catalytic systems have a huge potential in solving the recoverability and recyclability challenges [57].

Polymer Supported Palladium Nanoparticles
Literature reports have repeatedly shown that palladium nanoparticles can be efficiently supported on polymer frameworks [70][71][72][73][74][75][76][77]. Recently, Nemygina and co-workers immobilised various palladium precursors (PdCl 2 , PdCl 2 (CH 3 CN) 2 , and PdCl 2 (PhCN) 2 ) on amino-functionalized hyper-crosslinked polystyrene [78]. They then evaluated the influence of palladium oxidation state (Pd(II) or Pd(0)) on the rate of Suzuki cross-coupling of 4-bromoanisole and phenylboronic acid. Their results revealed that the catalyst impregnated with a PdCl 2 (CH 3 CN) 2 precursor resulted in better catalytic activity (Scheme 7). The PdCl 2 (CH 3 CN) 2 precursor was believed to be more active because it experienced less hydrolysis and precipitation in comparison with PdCl 2 and PdCl 2 (PhCN) 2 under Suzuki coupling reaction conditions. Lastly, they observed that unreduced, Pd(II) containing pre-catalyst were more active [78]. Another interesting strategy for immobilising palladium on a polymer was reported by Li and co-workers [79], who synthesised a robust and flexible cellulose sponge using a dual-cross-linking cellulose nanofiber (CNF) with γ-glycidoxypropyltrimethoxysilane (GPTMS) and polydopamine (PDA) (Scheme 8). The characterisation results revealed that the palladium nanoparticles were homogeneously dispersed on the surface of the cellulose nanofiber with a narrow size distribution. The catalyst was successfully applied to heterogeneous Suzuki and Heck cross-coupling reactions. Leaching of palladium was negligible and the catalysts could be conveniently separated from the products and reused for six times without loss of activity [79]. Scheme 8. Schematic of the stepwise formation of cellulose sponge supported palladium nanoparticles [79] (Adapted with permission from American Chemical Society).

Palladium Nanoparticles Immobilized on Hybrid Inorganic-Organic Material
The hybrid inorganic-organic materials are crystalline systems in which both inorganic and organic structural elements co-exist within a single phase [80]. These hybrid inorganic-organic materials are gaining popularity in cross-coupling reactions due to their tunable pore size, high surface areas, high crystallinity, and structural diversity [81][82][83][84][85][86][87]. Most hybrid inorganic-organic materials are synthesised through functionalization of the support surface with appropriate functional groups. This allows for a better control of the dimensions, dispersion and stability of the palladium nanoparticles. Metal-organic frameworks (MOFs) are one of the most popular classes of hybrid inorganic-organic materials and they are frequently explored as suitable palladium supports in heterogeneous cross-coupling reactions. For example, Shaikh and co-workers [88] used a zeolitic Scheme 7.
Catalysts for Suzuki cross-coupling based on functionalized hyper-crosslinked polystyrene [78] (Adapted with permission from American Chemical Society).
Another interesting strategy for immobilising palladium on a polymer was reported by Li and co-workers [79], who synthesised a robust and flexible cellulose sponge using a dual-cross-linking cellulose nanofiber (CNF) with γ-glycidoxypropyltrimethoxysilane (GPTMS) and polydopamine (PDA) (Scheme 8). The characterisation results revealed that the palladium nanoparticles were homogeneously dispersed on the surface of the cellulose nanofiber with a narrow size distribution. The catalyst was successfully applied to heterogeneous Suzuki and Heck cross-coupling reactions. Leaching of palladium was negligible and the catalysts could be conveniently separated from the products and reused for six times without loss of activity [79]. Another interesting strategy for immobilising palladium on a polymer was reported by Li and co-workers [79], who synthesised a robust and flexible cellulose sponge using a dual-cross-linking cellulose nanofiber (CNF) with γ-glycidoxypropyltrimethoxysilane (GPTMS) and polydopamine (PDA) (Scheme 8). The characterisation results revealed that the palladium nanoparticles were homogeneously dispersed on the surface of the cellulose nanofiber with a narrow size distribution. The catalyst was successfully applied to heterogeneous Suzuki and Heck cross-coupling reactions. Leaching of palladium was negligible and the catalysts could be conveniently separated from the products and reused for six times without loss of activity [79]. Scheme 8. Schematic of the stepwise formation of cellulose sponge supported palladium nanoparticles [79] (Adapted with permission from American Chemical Society).

Palladium Nanoparticles Immobilized on Hybrid Inorganic-Organic Material
The hybrid inorganic-organic materials are crystalline systems in which both inorganic and organic structural elements co-exist within a single phase [80]. These hybrid inorganic-organic materials are gaining popularity in cross-coupling reactions due to their tunable pore size, high surface areas, high crystallinity, and structural diversity [81][82][83][84][85][86][87]. Most hybrid inorganic-organic materials are synthesised through functionalization of the support surface with appropriate functional groups. This allows for a better control of the dimensions, dispersion and stability of the palladium nanoparticles. Metal-organic frameworks (MOFs) are one of the most popular classes of hybrid inorganic-organic materials and they are frequently explored as suitable palladium supports in heterogeneous cross-coupling reactions. For example, Shaikh and co-workers [88] used a zeolitic Scheme 8. Schematic of the stepwise formation of cellulose sponge supported palladium nanoparticles [79] (Adapted with permission from American Chemical Society).

Palladium Nanoparticles Immobilized on Hybrid Inorganic-Organic Material
The hybrid inorganic-organic materials are crystalline systems in which both inorganic and organic structural elements co-exist within a single phase [80]. These hybrid inorganic-organic materials are gaining popularity in cross-coupling reactions due to their tunable pore size, high surface areas, high crystallinity, and structural diversity [81][82][83][84][85][86][87]. Most hybrid inorganic-organic materials are synthesised through functionalization of the support surface with appropriate functional groups. This allows for a better control of the dimensions, dispersion and stability of the palladium nanoparticles. Metal-organic frameworks (MOFs) are one of the most popular classes of hybrid inorganic-organic materials and they are frequently explored as suitable palladium supports in heterogeneous cross-coupling reactions. For example, Shaikh and co-workers [88] used a zeolitic imidazolate framework (ZIF) to develop robust ZIF-supported palladium nanoparticles as a heterogeneous catalyst for Heck cross-coupling reactions ( Figure 15). The Pd/ZIF catalyst was found to be highly efficient and could be re-used up to four times without any loss in catalytic efficiency [88]. imidazolate framework (ZIF) to develop robust ZIF-supported palladium nanoparticles as a heterogeneous catalyst for Heck cross-coupling reactions ( Figure 15). The Pd/ZIF catalyst was found to be highly efficient and could be re-used up to four times without any loss in catalytic efficiency [88]. Another interesting example was reported by Kozell et al. [89] who immobilised palladium nanoparticles on zirconium phosphate glycine diphosphonate nanosheets (ZPGly). The immobilization of palladium nanoparticles on ZPGly nanosheets provided a stable catalyst with high palladium content (up to 22 wt. %). This Pd@ZPGly-15 catalyst proved to be highly efficient in catalysing Suzuki and Heck cross-couplings reactions. Surprisingly, the authors reported that their catalytic system operates as an efficient "heterogeneous" system, even though 3-5 ppm of palladium leached (Table 5). In addition, a "release and catch" mechanism was proposed in which soluble active palladium species are released from the solid Pd@ZPGly-15 catalyst during the oxidative addition of the halide and then re-deposited on the ZPGly support as a consequence of the reductive elimination step at the end of the reaction [89]. A magnetically separable version of a hybrid organic-inorganic support was recently reported by Omar and co-workers [90]. They supported palladium on magnetically separable organic-silica hybrid nanoparticles that were functionalized with ionic liquid groups ( Figure 16). The presence of ionic liquid groups within the framework of the hybrid nanoparticles enhanced the stability of the Another interesting example was reported by Kozell et al. [89] who immobilised palladium nanoparticles on zirconium phosphate glycine diphosphonate nanosheets (ZPGly). The immobilization of palladium nanoparticles on ZPGly nanosheets provided a stable catalyst with high palladium content (up to 22 wt. %). This Pd@ZPGly-15 catalyst proved to be highly efficient in catalysing Suzuki and Heck cross-couplings reactions. Surprisingly, the authors reported that their catalytic system operates as an efficient "heterogeneous" system, even though 3-5 ppm of palladium leached (Table 5). In addition, a "release and catch" mechanism was proposed in which soluble active palladium species are released from the solid Pd@ZPGly-15 catalyst during the oxidative addition of the halide and then re-deposited on the ZPGly support as a consequence of the reductive elimination step at the end of the reaction [89]. imidazolate framework (ZIF) to develop robust ZIF-supported palladium nanoparticles as a heterogeneous catalyst for Heck cross-coupling reactions ( Figure 15). The Pd/ZIF catalyst was found to be highly efficient and could be re-used up to four times without any loss in catalytic efficiency [88]. Another interesting example was reported by Kozell et al. [89] who immobilised palladium nanoparticles on zirconium phosphate glycine diphosphonate nanosheets (ZPGly). The immobilization of palladium nanoparticles on ZPGly nanosheets provided a stable catalyst with high palladium content (up to 22 wt. %). This Pd@ZPGly-15 catalyst proved to be highly efficient in catalysing Suzuki and Heck cross-couplings reactions. Surprisingly, the authors reported that their catalytic system operates as an efficient "heterogeneous" system, even though 3-5 ppm of palladium leached (Table 5). In addition, a "release and catch" mechanism was proposed in which soluble active palladium species are released from the solid Pd@ZPGly-15 catalyst during the oxidative addition of the halide and then re-deposited on the ZPGly support as a consequence of the reductive elimination step at the end of the reaction [89]. A magnetically separable version of a hybrid organic-inorganic support was recently reported by Omar and co-workers [90]. They supported palladium on magnetically separable organic-silica hybrid nanoparticles that were functionalized with ionic liquid groups ( Figure 16). The presence of ionic liquid groups within the framework of the hybrid nanoparticles enhanced the stability of the A magnetically separable version of a hybrid organic-inorganic support was recently reported by Omar and co-workers [90]. They supported palladium on magnetically separable organic-silica hybrid nanoparticles that were functionalized with ionic liquid groups ( Figure 16). The presence of ionic liquid groups within the framework of the hybrid nanoparticles enhanced the stability of the palladium nanoparticles on the organic-silica nanoparticles. The resultant Pd/MNP@IL-SiO 2 catalyst was efficiently applied in Heck and Suzuki cross-coupling reactions and it demonstrated high catalytic activity ( Figure 16). The Pd/MNP@IL-SiO 2 catalyst was also easily separated from the reaction mixture by application of an external magnetic field and the catalyst could be recycled over five times without a significant loss in its catalytic activity. In conclusion, the authors envisaged that this and similar catalytic systems could pave a way for the desired bridging of homogeneous and heterogeneous catalysis. palladium nanoparticles on the organic-silica nanoparticles. The resultant Pd/MNP@IL-SiO2 catalyst was efficiently applied in Heck and Suzuki cross-coupling reactions and it demonstrated high catalytic activity ( Figure 16). The Pd/MNP@IL-SiO2 catalyst was also easily separated from the reaction mixture by application of an external magnetic field and the catalyst could be recycled over five times without a significant loss in its catalytic activity. In conclusion, the authors envisaged that this and similar catalytic systems could pave a way for the desired bridging of homogeneous and heterogeneous catalysis. Another special type of hybrid organic-inorganic support was reported by Lin et al. who prepared a retrievable Pd/CMC@Ce(OH)4 catalyst for Suzuki coupling-reactions [91]. In this system, Ce(OH)4 was coated with carboxymethylcellulose (CMC) to form a hybrid organic-inorganic CMC@Ce(OH)4 support (Scheme 9). Carboxymethylcellulose has a large number of inherent carboxylate (-COO−) and hydroxyl (-OH) groups on its molecular chain. Thus, CMC is an excellent cation exchange agent, which makes it an excellent support for stabilization of palladium nanoparticles. While ceria has a unique ability of forming oxygen vacancies, that reduce Ce 4+ to Ce 3+ and create an excess of negative charge on its surface. The excess negative charge facilitates the coordination of palladium nanoparticles to the surface, as discussed earlier. These interactions have been shown to be crucial in preventing leaching and agglomeration of catalytic palladium nanoparticles. Scheme 9 shows an illustrative diagram of the conceptual mechanism of the Suzuki reaction catalyzed by Pd(0)-CMC@Ce(OH)4. The Ce 3+ on the surface of Ce(OH)4 enables the electronic charge transfer that produces a highly negatively charged palladium center. This then makes palladium a more electron rich site for facile oxidative addition of aryl halides, and thus allows catalysis to commence.
The Pd/CMC@Ce(OH)4 catalyst was applied to the Suzuki cross-coupling reaction and good to excellent yields were obtained. It is believed that the activity of the catalyst was enhanced synergistically by the unique redox properties of Ce(OH)4 (Ce 3+ /Ce 4+ ) and the coordination with the functional (carboxyl and hydroxyl) groups of the hybrid CMC@Ce(OH)4 support. Moreover, the catalyst could be easily separated by simple filtration and reused at least for five times without losing its activity. However, it was observed that Pd(0) species leached from the hybrid support during the reaction and were re-deposited onto the support at the end of the reaction. Therefore, the reaction mechanism is homogeneous in nature and the leached palladium species are the "true active species". Another special type of hybrid organic-inorganic support was reported by Lin et al. who prepared a retrievable Pd/CMC@Ce(OH) 4 catalyst for Suzuki coupling-reactions [91]. In this system, Ce(OH) 4 was coated with carboxymethylcellulose (CMC) to form a hybrid organic-inorganic CMC@Ce(OH) 4 support (Scheme 9). Carboxymethylcellulose has a large number of inherent carboxylate (-COO−) and hydroxyl (-OH) groups on its molecular chain. Thus, CMC is an excellent cation exchange agent, which makes it an excellent support for stabilization of palladium nanoparticles. While ceria has a unique ability of forming oxygen vacancies, that reduce Ce 4+ to Ce 3+ and create an excess of negative charge on its surface. The excess negative charge facilitates the coordination of palladium nanoparticles to the surface, as discussed earlier. These interactions have been shown to be crucial in preventing leaching and agglomeration of catalytic palladium nanoparticles. Scheme 9 shows an illustrative diagram of the conceptual mechanism of the Suzuki reaction catalyzed by Pd(0)-CMC@Ce(OH) 4 . The Ce 3+ on the surface of Ce(OH) 4 enables the electronic charge transfer that produces a highly negatively charged palladium center. This then makes palladium a more electron rich site for facile oxidative addition of aryl halides, and thus allows catalysis to commence.

Scheme 9.
Illustrative diagram of the conceptual mechanism of the Suzuki reaction catalyzed by Pd/CMC@Ce(OH)4 [91] (Adapted with permission from Elsevier).

Conclusions
In conclusion, the literature survey shows that remarkable progress has been made regarding the activity and recyclability of heterogeneous palladium based catalytic systems for C-C cross coupling reactions. However, thorough leaching and recyclability tests are still needed for most reported heterogeneous catalytic systems to unequivocally confirm that the palladium supported catalysts are responsible for the observed catalytic activity and not the leached material. Hence, at the moment it is hard to say what the nature of the true catalyst is, due to the dynamic nature of the reported catalytic systems for cross-coupling reactions.