Solventless synthesis of Ru(0) composites stabilized with polyphosphorhydrazone (PPH) dendrons and their use in catalysis

Abstract

Ruthenium nanoparticles (NPs) are prepared by milling under air ruthenium chloride (RuCl₃), sodium borohydride (NaBH₄) and a polyphosphorhydrazone (PPH) dendron (generation 0 to 2) having an alkyl chain at the focal point and triarylphosphines on the surface. The resulting NPs have a diameter in the 2 to 3 nm range and they are stable upon storage in solution or as powders. They can efficiently catalyze hydrogenation of styrene. The interaction between the dendrons and the NPs is studied, and the influence of the alkyl chain length and dendron generation is also discussed.

---

Introduction

Since the early reports on nanoparticle-dendrimer composites from Crooks,¹ Tomalia² and Esumi³ groups in 1998, these hybrid systems have paved the way for innovative dendrimer- or dendron-based nanoparticle (NP) stabilizing strategies,⁴ which could possibly find valuable applications in material science, biomedical applications or catalysis.⁵ To the best of our knowledge, most of the work done so far has been devoted to obtain copper^6–8 or iron^9,10 NPs as well as noble metal nanocomposites, including mostly palladium,^11–14 platinum,^15–18 gold^19–22 or silver.^23–26 This finding could be related to the easiness on the synthesis of these nanoparticles following the well-established Crooks' method.¹ The latter is based on the pre-complexation of metallic ions within the dendrimer scaffold and the subsequent reduction of these complexes to afford the so-called dendrimer-encapsulated nanoparticles (DEN). Despite the fact that the formation of such well-defined objects has been discussed and unambiguously evidenced, for instance in the case of Pd-encapsulated NPs,²⁷ it should be noted that in many cases rather undefined composite materials are obtained, mostly because the surface functions of the dendrimers are also involved in the stabilization of the NPs. The alternative use of non-commercially available dendrons, which are dendrimers having different functional groups on their surface and at their focal point, can offer significant advantages to design well-defined, ligand coated NPs.^4,28

As stated above, among the variety of dendrimer- or dendron- stabilized NPs involving noble metals, ruthenium NP-dendrimer composites are significantly less represented despite their relative low price when compared with other noble metals and the multiple applications this metal has in catalysis. Ruthenium-containing catalytic systems are actually described in many examples^29,30 ranging from metathesis,^31,32 to hydrogenation,³³ ammonia synthesis,³⁴ or methanol electro-oxidation,³⁵ to name a few. To the best of our knowledge, there are only a few investigations concerning ruthenium NPs-dendrimers composites, and most of them involve well-known PAMAM dendrimers, and unsurprisingly they are mostly dedicated to catalytic purposes,³⁶ independently from the oxidation state of the ruthenium atoms.³⁷ For instance, early examples involving Ru and Ni/Ru oxides obtained by electrodeposition of these metals coordinated within PAMAM dendrimers³⁸ have been tested for the electro-oxidation of ethanol. In the same scope, platinum and ruthenium NPs encapsulated in PAMAM dendrimers have also been used for the electro-oxidation of methanol.³⁹ Other bimetallic systems incorporating ruthenium and rhodium have also been described.^40,41 The chemistry of dendrimeric nano-complexes that are further reduced to afford DENs has reached a degree of sophistication that allows elegant experiments in which atom displacements and core–shell structures can be achieved.⁴² The question of the structural definition of the dendrimeric stabilizing agent is not a major concern, and as exposed in the previous examples PAMAM dendrimers⁴³ are perfectly adapted for this purpose. Conversely, the thermal stability of the stabilizing dendrimer is an important feature as thermal decomposition could affect the stability of the nano-composite.⁴⁴ In this regard, it can be noted that PPI dendrimers⁴⁵ and cross-linked⁴⁰ dendrimeric systems have been explored to stabilize Ru-based nano-composites that have been assayed for hydrogenation of phenols in aqueous media. The stability or operating conditions of such dendrimer-based systems can also be controlled by immobilization on solid supports like alumina.^46,47

Two decades ago, Caminade and coworkers developed a new synthetic approach using thiophosphoryl hydrazide building block to prepare polyphosphorhydrazone (PPH) dendrimers.⁴⁸ These macromolecules have proved to be easy to prepare and to purify on large laboratory scales, and the presence of phosphorus atoms within their structure allows a precise control on their synthesis and purity.⁴⁹ In addition, the backbone of PPH dendrimers is stable under multiple conditions,⁵⁰ and a large diversity of PPH-containing architectures can be designed by means of regioselective modifications that can be accurately performed at the core,⁵¹ on the periphery⁵² and within the backbone.⁵³ Consequently, PPH dendrimers have proved to be suitable for a wide range of applications,⁵⁴ from biology,⁵⁵ optics,⁵⁶ materials,⁵⁷ to catalysis.⁵⁸ These PPH dendrimers have also been used in the past to synthesize dendrimer–nanoparticle composites.^13,57,59,60

In this report we present a new series of dendrimer–nanoparticle composites using a new series of PPH dendrons as stabilizing agents for zero-valent ruthenium nanoparticles, following a solventless mechanochemical approach. This strategy circumvents the issues related to purification techniques that generally include tedious processes, and high amounts of solvents in the case of dialysis or reverse osmosis,⁶¹ or the use of other extraction agents, such as thiols.⁶²

Mechanochemistry⁶³ comprises physicochemical and chemical transformations driven by a mechanical force that changes the crystal structure of solids. The applied force generates fresh surfaces, which are rich in active catalytic sites prone to enhance the mass transfer required to initiate a chemical reaction.⁶⁴ The use of milling to produce new materials from solids has been applied and studied for many centuries,^65–67 yet its potential has been underestimated. Scientists are now rediscovering milling as an environmentally friendly, economical and fast process.^63,68 Mechanochemical synthesis can also be used to prepare products that may not be stable under atmospheric conditions like some organic compounds,^69,70 coordination complexes,⁷⁰ macromolecular and organometallic compounds.⁶⁷ Moreover, different authors have proven the utility of this technique when obtaining catalytically interesting materials. Although the use of mechanochemical approaches has already been reported for the preparation of nano-catalysts and catalytic nanomaterials, it should be noticed that all examples, to the best of our knowledge, involved mostly oxidized species.^71,72

There are some examples about previous efforts to synthesize zero-valent noble metal nanoparticles from mechanochemistry.^73,74 For instance, some of us have obtained pure, small, and cheap zero-valent ruthenium nanoparticles employing a mechanochemical approach.⁷⁵ However, the resulting nano-objects were found to agglomerate rapidly, and their instability was also associated to hazard issues as these nano-objects were prone to react violently when they were exposed to air.

Herein, we present a new strategy to avoid agglomeration and sudden explosions, keeping the advantages of using a solventless technique when affording zero-valent ruthenium nanoparticles. Contrarily to straightforward strategies involving commercially available small stabilizing agents, we have designed a series of PPH dendrons having a long alkyl tail of variable length at the focal point to facilitate the solubility of the composites in weakly polar media and triphenyl phosphine-derived surface functions to enable stabilizing dendron–nanoparticle interactions.

Interestingly, the as-obtained systems were found to be air-stable and they were successfully tested in a model catalytic styrene hydrogenation reaction.^76–78

Results and discussion

Dendron chemistry

The chemistry of PPH dendrimers is highly compatible with the design of core-functionalized dendrons. The design of PPH dendrons can be achieved by modification of an activated vinyl core, which offers the possibility to modify the focal point of a PPH dendron after its outgrowth.⁷⁹ Alternatively, the functionalized core can be implemented before the dendrimer outgrowth by selective substitution of a defined number of chlorine atoms on cyclotriphosphazene core,⁵¹ or by the use of any arylaldehyde-containing building block which can be used as a starting point for the dendrimer outgrowth.⁸⁰ This last approach is the most straightforward strategy provided that the core function is compatible with the dendrimer synthesis pathway. The dendrons used in the present investigation were designed with a long alkyl chain (dodecyl or hexadecyl) at the focal point and triphenylphosphine moieties on the surface. The aldehyde-derived methodology was then implemented starting from non-commercial para-alkoxy benzaldehydes (Scheme 1).

Scheme 1 Synthesis of alkoxy-cored dendrons with triaryl phosphine surface.

Small size ligands 3a and 3b have been obtained by Pd-catalyzed P–C cross-coupling reactions⁸¹ involving para-alkoxy iodobenzenes 1a⁸² and 1b⁸³ which were readily prepared according to routine procedures. The length of the dodecyl and hexadecyl alkyl chains was elicited in order to provide a good solubility in organic media. Interestingly, these triaryl phosphine 3a and 3b are sufficiently air-stable to authorize silica gel chromatography purification, and both compounds were obtained with good yields (40–55% over two steps). The presence of the phosphine is easily identified by the presence of a singlet at ca. −7 ppm on ³¹P NMR spectra.

Para-alkoxy-benzaldehydes 4a and 4b were obtained by a procedure analogous to the one used for 1a and 1b, the aromatic aldehyde function was used as a starting point for the dendron outgrowth synthesis. The synthesis involves the two-step substitution–condensation procedure routinely used for PPH dendrimeric structures. The resulting dendrons equipped either with aldehyde 7a–b or P–Cl surface groups 5a–b and 8a–b were easily purified by silica-gel chromatography and obtained with good yields (above 85%) on multigram scale. P–Cl terminated dendrons 5a–b and 8a–b were used to prepare phosphine-terminated stabilizing dendrons 6a–b and 9a–b by nucleophilic substitution of the chlorine atoms with phenol 2 in the presence of cesium carbonate. All compounds were characterized by means of multinucleus NMR and by mass spectrometry, and in all cases, ³¹P NMR was found to be a convenient tool to monitor the reactions (Fig. 1). Typically, a new singlet at ca. 63 ppm attributed to the dichlorothiophosphorhydrazone end groups characterized the condensation step leading to dendrons 5a–b and 8a–b. The reaction completion was also confirmed on proton NMR spectra by complete disappearance of the aldehyde singlet at ca. 10 ppm. Nucleophilic substitutions of the chlorine atoms of the dichlorothiophosphorhydrazone surface with 4-hydroxy-benzaldehyde and phosphine-bearing phenol 2 was accompanied by a limited shielding of the chemical shifts of the corresponding phosphorus atoms from ca. 63 ppm to 60.6 ppm and 61.5 ppm, respectively.

Fig. 1 ³¹P NMR monitoring of the dendron outgrowth and functionalization.

The solubility of these dendrons in non-polar solvents was found to be governed by the generation number. Indeed, dendrons 3a–b and 6a–b were found to be rather soluble in many solvents ranging from heptane to THF. On the contrary, dendrons 9a–b were found to be soluble in THF, but insoluble in alkanes, as a result of the presence of several heteroelements and polar groups in their structure.

Mechanochemical synthesis of Ru nanocomposites

All nanocomposites were obtained by solvent free mechanochemical reactions, involving hand milling of a small quantity of RuCl₃ precursor (100 mg), a stabilizing dendron and sodium borohydride (NaBH₄) as a reducing agent (Fig. 2) in an agate mortar. Typically, the quantity of dendron was adjusted to have 1 equivalent of terminal phosphine per ruthenium atom and the reducing agent was used in large excess (8 equivalents). According to previous studies,⁶³ this grinding and fine milling step allows the intimate mixing of the reactants and results at first in a size reduction of the reactant particles and subsequent increasing of their specific surface area and surface energy. This step is generally regarded as a mechanochemical activation leading to modifications of the structure of the materials, of their chemical compositions, and their chemical reactivity.⁸⁴

Fig. 2 Mechano-chemical preparation of dendron–Ru nanocomposites.

Washings and centrifugation cycles managed the removal of excess of sodium borohydride. In a first attempt, a series of washing cycles with ethanol afforded stable colloidal solutions that were found to present large signals at ca. 6 and 30 ppm on ¹¹B NMR spectra. The presence of boron and residual salts (NaCl mainly) was also confirmed by XRD experiments which revealed the presence of sodium chloride in halite phase, as compared to the Powder Diffraction File (PDF) #75-0306 provided by the International Center for Diffraction Data (ICDD). The presence of this phase pointed out that the ethanol washings were not efficient enough to eliminate the unwanted by-products, probably because the dendron molecules can retain the latter within their structure.

The alternate use of degassed methanol and water in between each centrifugation cycle allowed the complete disappearance of signals corresponding to boron species on proton and ¹¹B NMR spectra, and the absence of NaCl was also confirmed by XRD. After these washing cycles, the composites were suspended in freshly distilled THF, affording black, stable, colloidal solutions. Noteworthy, these solutions could be kept under atmospheric conditions for more than 4 months without agglomeration, precipitation or macroscopic change, and evaporation/dissolving cycles did not affect their properties. As expected, the solubility of the nanocomposites was found to be highly related to the solubility of the capping dendrons.

TEM imaging

Exhaustive TEM and HRTEM imaging of nanocomposites confirmed the presence of well dispersed nanocomposites in all cases, with small size ranges (2 to 3 nm in diameter) and relatively narrow size distributions. As detailed hereunder, some differences could be observed depending of the dendron generation used, this parameter being related to their shape and number of surface functions. In some cases, the length of the alkyl chain located at the focal point of the dendrons was also found to have a small influence on the size and shapes of the crystallites.

HR-TEM of Ru@3a and Ru@3b showed small and well dispersed nanoparticles (Fig. 3) with an average diameter centered on approximately 2 nm (1.99 ± 0.27 nm and 2.09 ± 0.32 nm respectively, see ESI†).

Fig. 3 HRTEM imaging and diffraction patterns (lower inserts) of Ru@3a (A) and Ru@3b (B) from THF colloidal solutions.

Since the nanoparticles were non-agglomerated, it was possible to localize a series of isolated nanocrystallites to obtain the theoretical electron diffractogram (lower insets on Fig. 3). All the resulting diffraction patterns show two or three typical distances in 2.06, 2.14 and 2.25 Å, that correspond to (101), (100) and (002) crystal family planes attributed to zero-valent ruthenium in a hexagonal close-packed cell and unit symmetry described by the space group P6₃/mmc, and lattice parameters a = 2.705 Å and c = 4.281 Å, as stated in the PDF number 06-0663, provide by the (ICDD).

A similar trend was observed for the first generation composites Ru@6a and Ru@6b. The nanoparticles presented similar sizes (see Table 2), 2.10 ± 0.33 nm and 2.06 ± 0.33 nm, respectively (Fig. 4). Isolated nanoparticles were characterized as zero-valent ruthenium species with hexagonal close-packed cell (see ESI†). Again, no influence of the dendron length of the alkyl chain could be detected on the shape or size of the composites.

Fig. 4 HRTEM imaging of Ru@6a (A) and Ru@6b (B) from THF colloidal solutions. Red arrows are pointing at agglomerated NPs.

In contrast, slightly bigger average sizes were observed for second-generation composites Ru@9a and Ru@9b, as 2.44 ± 0.26 nm and 2.81 ± 0.41 nm, respectively. From the theoretical electron diffraction of the isolated particle images, it was also concluded that the particles are zero-valent ruthenium in a hexagonal close-packed cell (see ESI†). In this case, the mean diameter of the NPs was found to be larger for long alkyl chain dendrons. These composites stabilized with second-generation dendrons showed on HRTEM imaging regions presenting nanoparticles in close proximity (Fig. 5), but the composites were maintained in dispersion during more than 4 months without noticeable precipitation. It was also possible to dry them out and re-suspend them in non-polar or weakly polar organic solvents like THF, ether or toluene.

Fig. 5 HRTEM imaging of Ru@9a (A) and Ru@9b (B) from THF colloidal solutions.

Contrarily to nanoparticles obtained with generation 0 or 1 dendrons which would not fully precipitate upon the addition of large quantities of pentane on their THF colloidal solutions, the systems obtained with the second generation dendrons were the only ones allowing a complete precipitation upon the sole addition of pentane on THF colloidal suspensions, as expected from the solubility properties of the dendrons involved in the mechanochemical process.

For each composite, a further energy dispersive X-ray spectroscopy (EDX) analysis was performed (see ESI†) during the HR-TEM experiments. On each system, aside from Ru, P (from Ru@3a, b composites) or P and S (from Ru@6a, b and Ru@9a, b composites) elements were detected on the nanoparticles. The detection of these elements confirmed the presence of PPH dendron molecules interacting with ruthenium particles. The same result was found when the systems were analyzed by HR-TEM and EDX spectroscopy after they were used as catalysts, indicating that dendrons are strongly attached to the nanoparticles, even after the catalytic reactions.

Although all colloidal solutions were found to be stable over long periods of time without macroscopic signs of agglomeration, significant differences could be observed on HRTEM imaging according to the dendron generation. Indeed, the numbers of aggregated or agglomerated species were found to be fairly greater in the case of NPs stabilized by higher generation dendrons (Ru@9a–b > Ru@6a–b > Ru@3a–b). It can be assumed from the schematic 2D representation of the dendron coating (Fig. 1) that nanocomposite systems of comparable sizes may present less hindered surfaces when the dendron generation increases. Consequently, this lower steric hindrance could be more favorable to inter-composite interactions that may be responsible for increased agglomerations (Fig. 6).

Fig. 6 HRTEM imaging of Ru@9b from THF solution after one day (A) and one month (B) of storage under atmospheric condition.

In order to evaluate the relative importance of dendron shape, and also the role of surface functions and thiophosphoryl scaffold upon the formation and stability of the nanocomposites, two complementary experiments were designed, involving either triphenyl phosphine alone as stabilizing agent or 7a dendron with arylaldehyde groups on its surface instead of the stabilizing triarylphosphines. The nanocomposites Ru@PPh₃ obtained with triphenylphosphine according to the same procedure as the one used for other composites were highly agglomerated (Fig. 7), and it was not possible to suspend them completely in THF. This finding is relevant to the fact that steric hindrance, or kinetic stabilization, of the active surface of the composites is necessary to reinforce efficiently the chemical stabilization provided by the phosphine ligands. The same mechanochemical procedure applied to dendron 7a afforded well-dispersed, non-agglomerated Ru@7a NPs (Fig. 7), confirming that the phosphines are not the sole stabilizing entities of this PPH dendrons series. The thiophosphoryl scaffold actually plays an important role in the stabilization of the nanocomposites, however the presence of phosphine termini is also required to ensure a good stability, since the colloids Ru@7a precipitated from THF solutions in less than one week.

Fig. 7 HRTEM imaging of Ru@PPh₃ (A) and Ru@7a (B) from THF solution after one day of storage under atmospheric condition.

X-ray photoelectron spectroscopy (XPS) analysis

An XPS survey of all Ru–dendrons composites was performed and the only elements that could be unambiguously identified corresponded to the PPH dendron, namely phosphorus (2p and 2s, at 133 and 191 eV respectively), carbon (1s at 284 eV) and oxygen (1s and an auger line, at 532 and 981 eV respectively). In addition, silicon 2p and 2s cores pertaining to the quartz surfaces on which the sample were deposited could also be identified at 104 and 152 eV respectively (ESI†).

The presence of high quantities of organic ligands surrounding the nanocomposites precluded the detection the typical Ru peaks. Actually, even in the most favorable case of Ru@3a colloids prepared with the smallest organic ligands, the principal core signal for ruthenium, 3d, was overlapped with the peak of the carbon 1s (ESI†).⁸⁵

Powder X-ray diffraction analysis

Powder X-ray diffraction analysis of the Ru@3a composite was implemented in order to confirm the metallic nature of the composite cores. The initial diffractogram (ESI†), showed only a broad peak at around 43° 2θ and an additional shoulder at around 39° 2θ. This same diffractogram is obtained after an additional annealing at 400 °C, under a nitrogen atmosphere (ESI†).

Comparison of both curves with theoretical diffraction peaks expected for zero-valent ruthenium species (ESI†) shows in both diffraction pattern the presence of the same broad peak centered at around 43° 2θ enclosing the three principal crystal family diffraction peaks (1,0,0), (0,0,2) and (1,0,1). In this regard, one can assume that the PPH dendrons are protecting the Ru nanoparticles from agglomeration even at high temperatures. The small difference between those patterns before and after the heating process indicates a growth of the crystalline particles; therefore there is a better definition of the crystal planes. Conversely, a significant difference is observed before and after annealing in the case of nanoparticles stabilized with higher generation dendrons than Ru@3a (ESI†). This discrepancy may support the idea that the first and second generation PPH dendrons are not as good stabilizing agents as 3a and 3b generation 0 dendrons.

MAS NMR study on composites

The existence of an interaction between the synthesized dendrons and the nanoparticles was also studied by solid-state nuclear magnetic resonance at magic angle spinning (MAS NMR). According to a recent report by Chaudret et al. on the surface chemistry of triarylphosphine-stabilized Ru nanoparticle obtained by hydrogenation of an organometallic precursor, triphenylphosphine as well as triphenylphosphine oxide can both interact on the surface on Ru NPs with typical ³¹P NMR signature centered on ca. 45 ppm and 28 ppm respectively.⁸⁶ However, it must be noted that in this reported study, both signals are convoluted and appear on solid state NMR spectra as a broad composite signal centered on ca. 40 ppm. For composites Ru@3a, ¹H–³¹P cross polarized (CP) MAS NMR experiments show a set of signals centered on ca. 30 ppm. In the case of Ru@3b, two signals are clearly identified at 26 and 27 ppm (Fig. 8). Taking into account that in the case of nanocomposites Ru@3a and Ru@3b no signal could be detected in the 45 ppm region nor in the −5/−10 ppm region, the presence of trivalent phosphorus atoms, either free or coordinated to Ru atoms, was excluded. Additionally, high power decoupled (HP/DEC) ³¹P-⁸⁷ experiments allowing observing more mobile species confirmed the absence of trivalent phosphorus species. Consequently, it was concluded that phosphorus atoms interacting with Ru atoms were mostly phosphine oxide, and that all terminal phosphine of the dendrons are oxidized during the milling process. In the case of colloids Ru@6a–b stabilized by generation 1 dendrons, the same observations could be made while an additional signal at ca. 65 ppm was also detected for the P(S) divergent point of the PPH scaffold (Fig. 8).

Fig. 8 ¹H–³¹P CPMAS NMR spectrum of compounds Ru@3b, Ru@6b and Ru@9b.

The NMR study of nanocomposites Ru@7a obtained with first generation, aldehyde-terminated dendron 7a, gave important information regarding the role of the scaffold of the dendrons in the stabilization of the nanoparticles. ¹H– ³¹P CPMAS NMR performed rapidly after the preparation of the nanocomposites revealed the presence of several signals. A large signal at 64 ppm accounting for the P=S linkages of the divergent points which were not affected by the mechanochemical process, and a set of three overlapping large signals at 21.9 ppm, 9 ppm and −3 ppm (Fig. 9B). Analysis of the nanocomposites by liquid NMR in deuterated THF revealed the appearance of a small signal at −6 ppm, and after one day at room temperature, signals from the divergent point had completely vanished, and the signal at −6 ppm signal was the major signal (Fig. 9A). According to previously reported methods to synthesize dendrimers having P=O linkages at their divergent points,⁸⁸ this signal was attributed to the P=O linkage resulting from the transformation of the P=S, and the signal at 12 ppm was tentatively attributed to these P=O or P=S linkages interacting with ruthenium atoms.

Fig. 9 ¹H–³¹P CPMAS NMR spectrum of compound Ru@7b rapidly after preparation (B) and ³¹P liquid NMR⁸⁷ spectrum recorded after one day (A).

According to these observations on liquid NMR spectra, the signals at 9 and 21 ppm identified on CPMAS NMR spectrum of Ru@7a were assumed to be related to the interaction of the phosphorus moieties of the divergent points with the surface of the ruthenium composites. These results confirm the important stabilizing role of the scaffold, and consequently the presence of a large signal centered at 26 ppm on CP ³¹P–{¹H} MAS NMR spectra of Ru@6b (Fig. 8) nanocomposite could account not only for the interaction of terminal phosphine oxide with Ru surface atoms, but also phosphorus atom of the divergent points interacting with Ru atoms. This assumption is also supported by the fact that such dendrons are flexible enough to flatten on the surface of the nanocomposites to offer several sites of interactions.

Nanocomposite Ru@9a–b stabilized by generation 2 dendrons having 4 triarylphosphine on their surface presented comparable spectral features with a composite signal in the 25–30 ppm range that was deconvoluted (Fig. 10). This composite signal is composed of one major signal centered on 25.8 ppm, attributed to phosphine oxide or phosphine oxide interacting with Ru species and a smaller signal centered on 39.5 ppm attributed to trivalent phosphine interacting with Ru atoms located on the surface of the nanoparticles, in agreement with previous studies reported by the group of Chaudret.⁸⁶ This assumption is also supported by the fact that this signal was not detectable on the HP/DEC ³¹P NMR spectra which allows the observation of rather mobile species. Another signal at ca. 65 ppm for the P=S divergent points of the PPH scaffold is also observed, and the most intense signal of the spectra is centered on −6 ppm, accounting for unbound triarylphosphines and P=O divergent points similar to those observed in the case of Ru@7a.

Fig. 10 ¹H–³¹P CPMAS NMR spectrum of compound Ru@9b (blue line: original spectrum, red dotted line: fitting curve, green lines: deconvoluted curves).

This NMR study tends to indicate that in the case of generation 0 and 1 dendrons, all the surface functions are probably interacting with the surface of the nanoparticles as phosphine oxide, whereas in the case of generation 2 dendrons, some surface functions remain free and unbounded. The interaction of the internal P=S linkages located at the divergent points of the structure with surface Ru atoms of the nanoparticles is accompanied by a conversion to P=O linkages. The fact that in the case of larger dendrons some phosphines are not oxidized could also be related to the fact that despite the relative flexibility of the PPH skeleton, a generation 2 PPH dendron cannot fully flatten onto the surface of small Ru NPs, leaving some of its surface and inner functions free of interaction with ruthenium atoms.

Catalytic tests

All the freshly synthesized composites were tested as catalysts in the hydrogenation of styrene (Fig. 11), a model unsaturated substrate offering the possibility to evaluate the potential of our systems to reduce a terminal alkene and an arene. The hydrogenations were first performed in the presence of 0.5 mol% Ru and 3 bar H₂ in THF at 70 °C for 3.5 h. In these conditions, all the Ru nanocomposites converted styrene into ethylbenzene (A) quantitatively (yields of A > 99%). By decreasing the temperature to 40 °C, it was thus possible to compare the catalytic activity of Ru nanoparticles prepared from various phosphine dendrons of the zeroth (3a–b), first (6a–b) and second (9a–b) generations (Fig. 11). Whatever the length of the alkyl chain (12 or 16C), the systems involving second-generation dendrons C12-containing Ru@9a and C16-containing Ru@9b displayed higher catalytic activity compared to those involving smaller dendrons 3a–b and 6a–b. Ethylbenzene A could indeed be obtained in 96% yield with Ru@9a against 85–87% when using the nanoparticles prepared from dendrons 3a and 6a respectively. Along the same lines, C16-containing Ru@9b allowed to get ethylbenzene A in 94% yield compared to 83–84% with Ru composites involving 3b and 6b. This difference of reactivity is consistent with the fact that the surface of nanocomposite systems is less hindered when higher generation dendrons are involved, thus allowing easier access to the substrates. In this regard the relative lower stability of the colloidal systems obtained with the higher generation dendrons is highly related to their higher catalytic activity.

Fig. 11 Catalytic hydrogenation of styrene with Ru–dendron composites. [a] Standard conditions: 1 mmol styrene, 3 bar H₂, THF (6.3 mL), 3.5 h, 40 °C. [b] Yields of ethylbenzene A determined by GC analysis using 1,3,5-trimethoxybenzene as the standard (average of three runs). [c] >99% yield of A (<1% B) at 100 °C (instead of 40 °C) for 72 h (instead of 3.5 h).

It is worth noting that whatever the dendron, ethylbenzene was obtained as the almost exclusive product, ethylcyclohexane being observed in trace amounts even if the reaction time is set to 72 h and the temperature increased to 100 °C (Fig. 11, note c). It has been previously demonstrated that hydrogenation of aromatic rings requires the presence of faces on the surface of the nanoparticles while vinyl groups can be hydrogenated anywhere on the nanoparticle.⁸⁹ The absence of ethylcyclohexane even by forcing the reaction conditions might thus indicate that there is no face accessible to the aromatic rings on the Ru–dendrons composites. Such selective hydrogenation in mild conditions is of significant importance to the petrochemical industry, especially when petroleum feedstocks such as styrene are concerned. In addition, the composites reported here are complementary to other Ru nanoparticles enabling the complete hydrogenation of arenes into cyclohexyl rings.^90–93 Interestingly, the system Ru@9a could be recovered by precipitation, washings and filtration through cannula. The recycled Ru nanocomposites were reused twice and the catalytic activity was found to slightly decrease over the runs (Table 1). This observed loss of activity might be rather related to “mechanical” loss of Ru–dendrons composites during the filtration process. A deactivation process can however not be excluded with certainty although the overall appearance of the composites did not change considerably after one catalytic cycle according to the transmission electron microscopy analysis performed (Fig. 12).

Table 1 Catalytic hydrogenation of styrene with Ru@9a: recycling experiments^a,b

Run 1^a^,^b	Run 2^a^,^b	Run 3^a^,^b
a Standard conditions: 1 mmol styrene, 3 bar H2, THF, 3.5 h, 40 °C.b Yields of ethylbenzene A determined by GC analysis using 1,3,5-trimethoxybenzene as the standard.
96	85	78

---

Fig. 12 TEM imaging of Ru@9b before (A) and after (B) catalysis, and size report ((C), before catalysis in light grey and after catalysis in dark grey).

Indeed, in a typical example (Fig. 12) involving Ru@9b, the aspect of the system was slightly modified with increased agglomerations. In all cases, the mean diameters of the colloids were found to increase but not substantially (ca. 0.3 to 0.4 nm, Table 2).

Table 2 Nanoparticles sizes measured by TEM before and after catalysis

Composite	Mean diameter before catalysis (nm)	Mean diameter after catalysis (nm)
Ru@3a	2.00 ± 0.3	2.4 ± 0.5
Ru@3b	2.1 ± 0.3	2.4 ± 0.4
Ru@6a	2.1 ± 0.3	2.3 ± 0.6
Ru@6b	2.1 ± 0.3	2.4 ± 0.6
Ru@9a	2.4 ± 0.3	2.6 ± 0.7
Ru@9b	2.8 ± 0.4	3.1 ± 0.6

---

It can be assumed that the heating of the system could be responsible for this increase, which is merely due to a subtle mechanism that may resemble crystal maturation with a controlled percolation of ruthenium atoms or ions. This moderate increment is a further indirect evidence on the good stabilizing properties offered by the dendrons through their interactions with the Ru nanoparticles surface. It is important to note that this phenomenon was not dependent on the stability of the systems, as it was observed for all composites.

EDX analysis (ESI†) performed during HRTEM analysis confirmed that the size increase was not related to a complete reorganization of the nanoparticles, as the dendrons were found to coat the nanoparticles after the catalytic experiments. X-ray diffusion patterns also confirmed the subtle nature of the size increase as the Ru⁰ core of the nanoparticle was found to be conserved as a hexagonal close packed phase.

Conclusion

The easy and environment friendly mechanochemical Ru NPs synthesis along with the use of PPH dendrons as coatings leads to Ru nanoparticles in zero oxidation state which are air stable, and proved to be very good nanomaterials for hydrogenation catalysis. According to this strategy, nanocomposite systems of comparable sizes may present less hindered surfaces when the dendron generation increases. This lower steric hindrance is assumed to be responsible for lower stability of these systems, as depicted by a larger size of these systems due to favored nanoparticle contacts related to Ru nanoparticle growing. The length of the alkyl chain (12 or 16C) does not account for significant differences on the catalytic reactivity or in the Ru nanoparticles sizes. The systems involving second-generation dendrons C12-containing Ru@9a and C16-containing Ru@9b displayed higher catalytic activity compared to those involving smaller dendrons. This difference of reactivity is consistent with the fact that the surface of nanocomposite systems is less hindered when higher generation dendrons are involved, thus allowing easier access to the substrates. The relative lower stability of the colloidal systems obtained with the higher generation dendrons is consistent with their higher catalytic activity.

Additionally, we have shown that the stabilization of the nanoparticle involves both terminal phosphine oxide groups and the dendrimeric scaffold. As already demonstrated for other fields of applications in which dendrimers are interacting with living systems,⁹⁴ the surface functions should not be considered as a unique pivotal parameter ruling the nature of interactions: the nature of scaffold also plays a major role, either by orienting the surface function, or by providing its own chemical functions prone to provide specific effects. This point may appear rather obvious for systems that have been widely described for the obtaining of DENs, in which the dendrimer can be seen as a nano-container. Conversely, this observation is rather new in the field of dendrimer- or dendron-stabilized NPs in which colloids are surrounded by a soft shell of organic stabilizing ligands, these systems being reminiscent to small-ligand stabilized colloids.

Acknowledgements

Nidia G. García-Peña, acknowledges the mexican “Consejo Nacional de Ciencia y Tecnología” (CONACyT) for the scholarship granted during her PhD studies (scholarship holder number 216110, scholar fellowship number 315670). Prof. Rocío Redón also acknowledges the financial support by PAPIIT (IN117514) and CONACyT (167356) projects. The CNRS is also acknowledged for financial support. Yannick Coppel (CNRS, Toulouse) is kindly acknowledged for technical support (MAS NMR).

Notes and references

1. M. Q. Zhao, L. Sun and R. M. Crooks, J. Am. Chem. Soc., 1998, 120, 4877–4878.
2. L. Balogh and D. A. Tomalia, J. Am. Chem. Soc., 1998, 120, 7355–7356.
3. K. Esumi, A. Suzuki, N. Aihara, K. Usui and K. Torigoe, Langmuir, 1998, 14, 3157–3159.
4. L. M. Bronstein and Z. B. Shifrina, Chem. Rev., 2011, 111, 5301–5344.
5. Y. H. Niu and R. M. Crooks, C. R. Chim., 2003, 6, 1049–1059.
6. S. Berchmans, T. M. Vergheese, A. L. Kavitha, M. Veerakumar and V. Yegnaraman, Anal. Bioanal. Chem., 2008, 390, 939–946.
7. L. Jin, S. P. Yang, Q. W. Tian, H. X. Wu and Y. J. Cai, Mater. Chem. Phys., 2008, 112, 977–983.
8. R. M. Crooks, M. Q. Zhao, L. Sun, V. Chechik and L. K. Yeung, Acc. Chem. Res., 2001, 34, 181–190.
9. X. Y. Shi, S. H. Wang, S. D. Swanson, S. Ge, Z. Y. Cao, M. E. Van Antwerp, K. J. Landmark and J. R. Baker, Adv. Mater., 2008, 20, 1671–1678.
10. T. J. Daou, G. Pourroy, J. M. Greneche, A. Bertin, D. Felder-Flesch and S. Begin-Colin, Dalton Trans., 2009, 4442–4449.
11. A. K. Diallo, C. Ornelas, L. Salmon, J. R. Aranzaes and D. Astruc, Angew. Chem., Int. Ed., 2007, 46, 8644–8648.
12. J. A. Johnson, J. J. Makis, K. A. Marvin, S. E. Rodenbusch and K. J. Stevenson, J. Phys. Chem. C, 2013, 117, 22644–22651.
13. E. Badetti, A. M. Caminade, J. P. Majoral, M. Moreno-Manas and R. M. Sebastian, Langmuir, 2008, 24, 2090–2101.
14. H. Tanaka, S. Koizumi, T. Hashimoto, H. Itoh, M. Satoh, K. Naka and Y. Chujo, Macromolecules, 2007, 40, 4327–4337.
15. Y. Borodko, P. Ercius, V. Pushkarev, C. Thompson and G. Somorjai, J. Phys. Chem. Lett., 2012, 3, 236–241.
16. V. S. Myers, A. I. Frenkel and R. M. Crooks, Langmuir, 2012, 28, 1596–1603.
17. D. Yamamoto, S. Watanabe and M. T. Miyahara, Langmuir, 2010, 26, 2339–2345.
18. Y. L. Gu, H. Xie, J. X. Gao, D. X. Liu, C. T. Williams, C. J. Murphy and H. J. Ploehn, Langmuir, 2005, 21, 3122–3131.
19. N. Li, M. Echeverria, S. Moya, J. Ruiz and D. Astruc, Inorg. Chem., 2014, 53, 6954–6961.
20. N. Li, P. Zhao, N. Liu, M. Echeverria, S. Moya, L. Salmon, J. Ruiz and D. Astruc, Chem.–Eur. J., 2014, 20, 8363–8369.
21. R. J. Korkosz, J. D. Gilbertson, K. S. Prasifka and B. D. Chandler, Catal. Today, 2007, 122, 370–377.
22. X. Y. Liu, F. Cheng, Y. Liu, H. J. Liu and Y. Chen, J. Mater. Chem., 2010, 20, 360–368.
23. M. Franckevičius, A. Gustainytė, R. Kondrotas, R. Juškėnas, M. Marcos, J. L. Serrano, R. Vaišnoras and V. Gulbinas, J. Nanopart. Res., 2014, 16, 1–13.
24. A. Sutton, G. Franc and A. Kakkar, J. Polym. Sci., Part A: Polym. Chem., 2009, 47, 4482–4493.
25. A. Hameau, V. Colliere, J. Grimoud, P. Fau, C. Roques, A. M. Caminade and C. O. Turrin, RSC Adv., 2013, 3, 19015–19026.
26. X. Liu, W. Hao, C.-N. Lok, Y. C. Wang, R. Zhang and K. K. Y. Wong, J. Pediatr. Surg., 2014, 49, 1846–1851.
27. M. V. Gomez, J. Guerra, A. H. Velders and R. M. Crooks, J. Am. Chem. Soc., 2009, 131, 341–350.
28. C.-O. Turrin and A.-M. Caminade, in Dendrimers, ed. A. M. Caminade, C. O. Turrin, R. Laurent, A. Ouali and B. Delavaux-Nicot, John Wiley & Sons, Ltd, 2011, ch. 7, pp. 141–155,  DOI:10.1002/9781119976530.ch18.
29. N. Yan, C. X. Xiao and Y. Kou, Coord. Chem. Rev., 2010, 254, 1179–1218.
30. T. Naota, H. Takaya and S. I. Murahashi, Chem. Rev., 1998, 98, 2599–2660.
31. M. S. Sanford, J. A. Love and R. H. Grubbs, J. Am. Chem. Soc., 2001, 123, 6543–6554.
32. T. M. Trnka and R. H. Grubbs, Acc. Chem. Res., 2001, 34, 18–29.
33. S. E. Clapham, A. Hadzovic and R. H. Morris, Coord. Chem. Rev., 2004, 248, 2201–2237.
34. C. J. H. Jacobsen, S. Dahl, P. L. Hansen, E. Tornqvist, L. Jensen, H. Topsoe, D. V. Prip, P. B. Moenshaug and I. Chorkendorff, J. Mol. Catal. A: Chem., 2000, 163, 19–26.
35. T. Iwasita, Electrochim. Acta, 2002, 47, 3663–3674.
36. N. C. Antonels and R. Meijboom, Catal. Commun., 2014, 57, 148–152.
37. P. Bhattacharya, E. N. Nasybulin, M. H. Engelhard, L. Kovarik, M. E. Bowden, X. S. Li, D. J. Gaspar, W. Xu and J.-G. Zhang, Adv. Funct. Mater., 2014, 24, 7510–7519.
38. J. W. Kim and S. M. Park, Electrochem. Solid-State Lett., 2000, 3, 385–388.
39. I. L. Escalante García, J. Ledesma García, F. J. Rodríguez, T. W. Chapman and L. A. Godinez, ECS Trans., 2007, 3, 1–7.
40. A. Maximov, A. Zolotukhina, V. Murzin, E. Karakhanov and E. Rosenberg, ChemCatChem, 2015, 7, 1197–1210.
41. W. Yang and P. Xiaohong, Nano-Micro Lett., 2014, 6, 55–62.
42. K. A. Marvin, N. N. Thadani, C. A. Atkinson, E. L. Keller and K. J. Stevenson, Chem. Commun., 2012, 48, 6289–6291.
43. J. Peterson, V. Allikmaa, J. Subbi, T. Pehk and M. Lopp, Eur. Polym. J., 2003, 39, 33–42.
44. D. S. Deutsch, G. Lafaye, D. X. Liu, B. Chandler, C. T. Williams and M. D. Amiridis, Catal. Lett., 2004, 97, 139–143.
45. E. Murugan and I. Pakrudheen, Sci. Adv. Mater., 2015, 7, 891–901.
46. G. Lafaye, A. Siani, P. Marécot, M. D. Amiridis and C. T. Williams, J. Phys. Chem. B, 2006, 110, 7725–7731.
47. G. Lafaye, C. T. Williams and M. D. Amiridis, Catal. Lett., 2004, 96, 43–47.
48. N. Launay, A. M. Caminade, R. Lahana and J. P. Majoral, Angew. Chem., Int. Ed., 1994, 33, 1589–1592.
49. A. M. Caminade, R. Laurent, C. O. Turrin, C. Rebout, B. Delavaux-Nicot, A. Ouali, M. Zablocka and J. P. Majoral, C. R. Chim., 2010, 13, 1006–1027.
50. C. O. Turrin, V. Maraval, J. Leclaire, E. Dantras, C. Lacabanne, A. M. Caminade and J. P. Majoral, Tetrahedron, 2003, 59, 3965–3973.
51. O. Rolland, L. Griffe, M. Poupot, A. Maraval, A. Ouali, Y. Coppel, J. J. Fournié, G. Bacquet, C. O. Turrin, A. M. Caminade, J. P. Majoral and R. Poupot, Chem.–Eur. J., 2008, 14, 4836–4850.
52. A. Pérez-Anes, G. Spataro, Y. Coppel, C. Moog, M. Blanzat, C. O. Turrin, A. M. Caminade, I. Rico-Lattes and J. P. Majoral, Org. Biomol. Chem., 2009, 7, 3491–3498.
53. C. Larre, D. Bressolles, C. Turrin, B. Donnadieu, A. M. Caminade and J. P. Majoral, J. Am. Chem. Soc., 1998, 120, 13070–13082.
54. O. Rolland, C. O. Turrin, A. M. Caminade and J. P. Majoral, New J. Chem., 2009, 33, 1809–1824.
55. A.-M. Caminade, C.-O. Turrin and J.-P. Majoral, New J. Chem., 2010, 34, 1512–1524.
56. E. Martinez-Ferrero, G. Franc, S. Mazeres, C. O. Turrin, C. Boissiere, A. M. Caminade, J. P. Majoral and C. Sanchez, Chem.–Eur. J., 2008, 14, 7658–7669.
57. A. Hameau, V. Colliere, J. Grimoud, P. Fau, C. Roques, A. M. Caminade and C. O. Turrin, RSC Adv., 2013, 3, 19015–19026.
58. A.-M. Caminade, R. Laurent, A. Ouali and J. P. Majoral, Inorg. Chim. Acta, 2014, 409, 68–88.
59. G. Schmid, W. Meyer-Zaika, R. Pugin, T. Sawitowski, J. P. Majoral, A. M. Caminade and C. O. Turrin, Chem.–Eur. J., 2000, 6, 1693–1697.
60. G. Franc, E. Badetti, C. Duhayon, Y. Coppel, C.-O. Turrin, J.-P. Majoral, R.-M. Sebastian and A.-M. Caminade, New J. Chem., 2010, 34, 547–555.
61. A. Berger, R. M. Klein Gebbink and G. van Koten, in Dendrimer Catalysis, ed. L. Gade, Springer Berlin, Heidelberg, 2006, vol. 20, ch. 30, pp. 1–38.
62. J. C. Garcia-Martinez and R. M. Crooks, J. Am. Chem. Soc., 2004, 126, 16170–16178.
63. S. L. James, C. J. Adams, C. Bolm, D. Braga, P. Collier, T. Friscic, F. Grepioni, K. D. M. Harris, G. Hyett, W. Jones, A. Krebs, J. Mack, L. Maini, A. G. Orpen, I. P. Parkin, W. C. Shearouse, J. W. Steed and D. C. Waddell, Chem. Soc. Rev., 2012, 41, 413–447.
64. A. Nasser and U. Mingelgrin, Appl. Clay Sci., 2012, 67–68, 141–150.
65. V. Sepelak, A. Duvel, M. Wilkening, K.-D. Becker and P. Heitjans, Chem. Soc. Rev., 2013, 42, 7507–7520.
66. L. Takacs, Chem. Soc. Rev., 2013, 42, 7649–7659.
67. T. Friscic, Chem. Soc. Rev., 2012, 41, 3493–3510.
68. G. A. Bowmaker, Chem. Commun., 2013, 49, 334–348.
69. P. Chauhan and S. S. Chimni, Beilstein J. Org. Chem., 2012, 8, 2132–2141.
70. A. L. Garay, A. Pichon and S. L. James, Chem. Soc. Rev., 2007, 36, 846–855.
71. R. A. Buyanov and V. V. Molchanov, Russ. Chem. Rev., 2000, 69, 435–450.
72. C. Xu, S. De, A. M. Balu, M. Ojeda and R. Luque, Chem. Commun., 2015, 51, 6698–6713.
73. F. Delogu, Langmuir, 2012, 28, 10898–10904.
74. R. Redon, F. Ramirez-Crescencio and A. L. Fernandez-Osorio, J. Nanopart. Res., 2011, 13, 5959–5965.
75. N. G. García-Peña, R. Redón, A. Herrera-Gomez, A. L. Fernández-Osorio, M. Bravo-Sanchez and G. Gomez-Sosa, Appl. Surf. Sci., 2015, 340, 25–34.
76. F. Corvaisier, Y. Schuurman, A. Fecant, C. Thomazeau, P. Raybaud, H. Toulhoat and D. Farrusseng, J. Catal., 2013, 307, 352–361.
77. I. Favier, P. Lavedan, S. Massou, E. Teuma, K. Philippot, B. Chaudret and M. Gómez, Top. Catal., 2013, 56, 1253–1261.
78. F. Novio, D. Monahan, Y. Coppel, G. Antorrena, P. Lecante, K. Philippot and B. Chaudret, Chem.–Eur. J., 2014, 20, 1287–1297.
79. I. Angurell, C. O. Turrin, R. Laurent, V. Maraval, P. Servin, O. Rossell, M. Seco, A. M. Caminade and J. P. Majoral, J. Organomet. Chem., 2007, 692, 1928–1939.
80. C. Barriere, V. Latour, P. Fau, A.-M. Caminade and C.-O. Turrin, Tetrahedron Lett., 2012, 53, 1908–1911.
81. O. Herd, A. Heßler, M. Hingst, M. Tepper and O. Stelzer, J. Organomet. Chem., 1996, 522, 69–76.
82. L.-B. Xing, B. Yang, X.-J. Wang, J.-J. Wang, B. Chen, Q. Wu, H.-X. Peng, L.-P. Zhang, C.-H. Tung and L.-Z. Wu, Langmuir, 2013, 29, 2843–2848.
83. A. R. Borges, M. Hyacinth, M. Lum, C. M. Dingle, P. L. Hamilton, M. Chruszcz, L. Pu, M. Sabat and K. L. Caran, Langmuir, 2008, 24, 7421–7431.
84. A. Zoltán Juhász, Colloids Surf., A, 1998, 141, 449–462.
85. G. P. Rachiero, U. B. Demirci and P. Miele, Catal. Today, 2011, 170, 85–92.
86. T. Gutmann, E. Bonnefille, H. Breitzke, P.-J. Debouttiere, K. Philippot, R. Poteau, G. Buntkowsky and B. Chaudret, Phys. Chem. Chem. Phys., 2013, 15, 17383–17394.
87. W. Abdelmoez, M. Yasuda, H. Ogino, K. Ishimi and H. Ishikawa, Biotechnol. Prog., 2002, 18, 706–712.
88. M. L. Lartigue, N. Launay, B. Donnadieu, A. M. Caminade and J. P. Majoral, Bull. Soc. Chim. Fr., 1997, 134, 981–988.
89. E. Bonnefille, F. Novio, T. Gutmann, R. Poteau, P. Lecante, J.-C. Jumas, K. Philippot and B. Chaudret, Nanoscale, 2014, 6, 9806–9816.
90. V. Kelsen, A. Meffre, P.-F. Fazzini, P. Lecante and B. Chaudret, ChemCatChem, 2014, 6, 1714–1720.
91. D. González-Gálvez, P. Nolis, K. Philippot, B. Chaudret and P. W. N. M. van Leeuwen, ACS Catal., 2012, 2, 317–321.
92. P.-J. Debouttière, Y. Coppel, A. Denicourt-Nowicki, A. Roucoux, B. Chaudret and K. Philippot, Eur. J. Inorg. Chem., 2012, 2012, 1229–1236.
93. M. Guerrero, A. Roucoux, A. Denicourt-Nowicki, H. Bricout, E. Monflier, V. Collière, K. Fajerwerg and K. Philippot, Catal. Today, 2012, 183, 34–41.
94. A.-M. Caminade, S. Fruchon, C.-O. Turrin, M. Poupot, A. Ouali, A. Maraval, M. Garzoni, M. Maly, V. Furer, V. Kovalenko, J.-P. Majoral, G. M. Pavan and R. Poupot, Nat. Commun., 2015, 6, 7722.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra13709a

---