Bottom-up synthesis of indium ( 0 ) nanoparticles and its application for the allylation of benzaldehyde . An experimental and theoretical study

In recent years, the synthesis of metal nanoparticles has attracted significant attention because of their unique properties. Although many strategies for the preparation of noble- and transition-metal nanoparticles have been published, the synthesis of indium nanoparticles (InNPs) has been scarcely reported. Most of the bottom-up methods require the use of indium salts and strong reducing agents such as sodium metal, zinc power, alkalides/electrides, or decomposition of organometallic complexes. Regrettably, some of them provide little control over particle size and size distribution of the InNPs, and the presence of stabilizing agents is generally mandatory. Recently we have been working on the simple, mild, and efficient synthesis of very reactive, monodisperse (4.0 ± 1.5 nm) spherical InNPs by fast reduction of indium(III) chloride with lithium powder and a catalytic amount of 4,4´-di-tert-butylbiphenyl (DTBB) in THF at room temperature, and in the absence of any anti-agglomeration additive or ligand. We explored the above mentioned InNPs-based reactive system for the allylation of a variety of aldehydes and ketones in a one-pot procedure, by adding allyl bromide over a suspension of InNPs followed by the addition of the corresponding carbonyl compound. For most of the compounds tested, the homoallylic alcohol product was obtained in good yields. In order to extend the scope of this indium-mediated protocol, herein, we report a comparative study on the allylation of benzaldehyde with a series of substituted allyl bromides promoted by both InNPs and In powder, in THF at room temperature. As expected, indium nanoparticles demonstrated to be much more efficient than granular indium yielding the corresponding allylic alcohols almost quantitatively. Additionally, computational studies have been applied in order to explain the differences in reactivity observed. Based on the experimental data and the DFT studies, we have proposed a possible reaction mechanism that implies the formation of a cyclic six-membered Zimmermann-Traxler-type transition state and the participation of allylindiums sesquihalides as intermediates in the indium-mediated allylation.


Introduction
In recent years, the synthesis of metal nanoparticles has attracted significant attention because of their unique properties.Although many strategies for the preparation of noble-and transition-metal nanoparticles have been published, the synthesis of indium nanoparticles (InNPs) has been scarcely reported.Most of the bottom-up methods require the use of indium salts and strong reducing agents such as sodium metal, zinc power, alkalides/electrides, or decomposition of organometallic complexes. 1grettably, some of them provide little control over particle size and size distribution of the InNPs, and the presence of stabilizing agents is generally mandatory.Recently, we have been working on the simple, mild, and efficient synthesis of very reactive, monodisperse (4.0 ± 1.5 nm) spherical InNPs by fast reduction of indium(III) chloride with lithium powder and a catalytic amount of 4,4´-di-tert-butylbiphenyl (DTBB) in THF at room temperature, and in the absence of any anti-agglomeration additive or ligand.We explored the above mentioned InNPs-based reactive system for the allylation of a variety of aldehydes and ketones in a one-pot procedure, by adding allyl bromide over a suspension of InNPs followed by the addition of the corresponding carbonyl compound.For most of the compounds tested, the corresponding homoallylic alcohol was obtained as the major reaction product in good yields. 2 order to extend the scope of this indium-mediated protocol, and prompted by our interest in the search of new and useful synthetic applications of metal NPs, herein, we report a comparative study on the allylation of benzaldehyde with a series of substituted allyl bromides promoted by both InNPs and In powder, under mild reaction conditions.Additionally, computational studies have been applied in order to explain the differences in reactivity observed.

Results and discussion
As shown in Scheme 1, we initiated our studies with the indium powder and InNPs mediated allylation of benzaldehyde with four different allyl bromides, following the standard conditions previously reported for this type of reactions, in THF, working at room temperature and under N 2 atmosphere.The indium nanoparticles (InNPs) were generated in situ by reduction of commercially available InCl 3 with an excess of lithium sand (3.5:1 M ratio relative to the corresponding indium trichloride) and a catalytic amount of DTBB (10 mol% referred to the indium salt), using tetrahydrofuran (THF) as solvent, at room temperature and under a nitrogen atmosphere.This methodology allowed the generation of well defined, monodisperse, spherical nanoparticles with a particle size distribution of ca.4.0±0.5 nm.

Scheme 1. "Indium(0)" mediated allylation of benzaldehyde
As can be seen in Table 1, the reaction of a series of allyl bromides and benzaldehyde in the presence of a suspension of In powder, gave the corresponding homoallylic alcohols 1 in poor to good yields, while the same allylation reactions promoted by InNPs, prepared in situ as described above, yielded the corresponding allylic alcohols almost quantitatively.
e Together with starting substrate.
Assuming that an allyl-indium intermediate is formed in these carbonyl allylation reactions 3 , we tested this reaction under the optimized conditions but stirring the corresponding indium reagent (InNPs suspension or In powder) with prenyl bromide for 60 min, before the addition of the carbonyl compound in order to favor the formation of this plausible intermediate.Thus, after 4 h reaction time, the -adduct 4 1-phenyl-2,2dimethylbut-3-en-1-ol (1b) was obtained in 65% yield in the presence of indium powder, and in 95% yield when the reaction was mediated by InNPs (Table 1, entry 2).
Although the yield of the allylated product was quite similar to that obtained using allyl bromide, the reaction time was markedly longer when prenyl bromide was used as the starting allyl halide.The less reactivity observed for this substrate could be attributed to the steric hindrance caused by methyl substituents attached to the allyl moiety, which could affect, somehow, the formation of the homoallylic alcohol.Using methallyl bromide, the yield of the corresponding alcohol 1c (1-phenyl-3-methylbut-3-en-1-ol) was markedly better by using InNPs compared to that obtained using indium powder (99% to 10%).Furthermore, in the latter reaction, a considerable amount of the starting benzaldehyde was recovered at the final reaction time (2 h) (Table 1, entry 3).The crotylation of benzaldehyde mediated by indium power gave 1-phenyl-2-methylbut-3en-1-ol (1d) with moderated yield (58%) together with starting carbonyl compound, whereas in the presence of InNPs afforded the corresponding -coupling homoallylic alcohol 1d almost quantitatively after 2 h.The diasterereoselectivity observed for this reaction was slightly better than that reported previously by other authors 4 (67:33 syn/anti) and the syn-isomer was dominant (Table 1, entry 4).
These experimental observations prompted us to study in more detail the process implied in the InNPs-mediated allylation of benzaldehyde using density functional theory (DFT).The ab initio calculations were performed with the B3LYP 5 DFT 6 functional and the LANL2DZ basis set.
As already mentioned above, the reactivity of the allyl-and prenyl bromides was very different, both giving the corresponding homoallylic alcohol in excellent yields (98 and 95% respectively) but the allylation with prenyl bromide being notably slower.Taking this into account, and considering results reported by other authors, 7 we evaluated the potential energy surfaces (PES) for the mentioned process by considering the formation of an initial complex between benzaldehyde and the allyl-indium intermediate, a six member cyclic chair-like transition state (TS) and a final complex, as can be seen in the Scheme 2 for the preparation of 1a as representative compound.An auxiliary bromide atom was used as ligand for the indium atom to obtain a simplified neutral model system. 7Scheme 2. Representative structures for the obtention of 1a included in the computational analysis As shown in Figures 1 and 2, the formation of the final complex of 1b takes place with higher activation energy than that of 1a (9.0 and 5.2 Kcal/mol, respectively).In addition, for product 1a the process occurs exothermically (-5.2 Kcal/mol), while for 1b the corresponding final complex requires +1.2 Kcal/mol.This thermochemical observation could explain the differences found in the reactivity between allyl bromide and prenyl bromide in the allylation of benzaldehyde mediated by InNPs.

Conclusions
We have developed a simple and convenient methodology for the synthesis of very reactive, well defined, monodisperse, spherical InNPs with a particle size distribution of ca.4.0±0.5 nm, through the SET reduction of indium(III) chloride in the presence of lithium sand and a catalytic amount of DTBB, under mild conditions and in the absence of any anti-agglomeration additive or ligand.These in situ prepared InNPs have demonstrated to be efficient for the synthesis of homoallylic alcohols from benzaldehyde using a series of substituted allyl bromides as allylating agents.As expected, InNPs demonstrated to be much more efficient than granular indium yielding the corresponding allylic alcohols almost quantitatively and at shorter reaction times.
On the other hand, DFT calculations have shown to be a successful approach for studying the allylindium intermediates as well as to explain the experimental results.To conclude, based on the experimental data and DFT studies, we have proposed a possible reaction mechanism that implies the formation of -coupling products via a cyclic six-membered Zimmermann-Traxler-type transition state, and the participation of allyl-indium sesquihalides as intermediates.

General
All moisture sensitive reactions were carried out under a nitrogen atmosphere.
Other solvents used were treated prior to use by standard methods. 8All starting materials were of the best available grade (Aldrich, Merck) and were used without further purification.Column chromatography was performed with Merck silica gel 60 (0.040-0.063 mm, 240-400 mesh).Thin layer chromatography (TLC) was performed on precoated silica gel plates (Merck 60, F254, 0.25 mm).All the starting carbonyl and allyl compounds, included in Table 1 were purchased from commercial sources (Aldrich, Fluka, Merck), and used as received.

Instrumentation and analysis
Nuclear magnetic resonance (NMR) spectra were recorded on a Bruker ARX-300 spectrophotometer using CDCl 3 as solvent and tetramethylsilane (TMS) as internal reference.Mass spectra (EI) were obtained at 70 eV on a Hewlett Packard HP-5890 GC/MS instrument equipped with a HP-5972 selective mass detector.The purity of volatile compounds and the chromatographic analyses (GC) were determined with a GC Shimadzu (GC-14B) with a flame ionization detector equipped with a HP-5MS column (30 m × 0.25 mm × 0.25 μm) using nitrogen as carrier gas.
Representative procedure for the allylation of carbonyl compounds promoted by In powder.Synthesis of homoallylic alcohol 1a A suspension of indium powder (-100 mesh, 115 mg, 1 mmol) and allyl bromide (181.3 mg, 0.128 mL, 1.5 mmol) in THF (4 mL) was stirred at room temperature under nitrogen atmosphere for 30 minutes.Subsequently, benzaldehyde (53 mg, 0.051 mL, 0.5 mmol) in THF (1 mL) was added by syringe.After total conversion of the starting material (TLC, GC), the resulting suspension was diluted with diethyl ether (10 mL) and treated with 10% HCl and water.The combined extracts were dried over anhydrous Na 2 SO 4 , and evaporated (20 mbar).The resulting residue was purified by flash column chromatography (silica gel, hexane-ethyl acetate) to give the corresponding homoallylic alcohol 1a (68%).
Representative procedure for the allylation of carbonyl compounds promoted by InNPs.

Synthesis of homoallylic alcohol 1a
A mixture of lithium powder (24.5 mg, 3.5 mmol), DTBB (26.7 mg, 0.1 mmol) and InCl 3 (221 mg, 1.0 mmol) in THF (3 mL) was stirred at room temperature under nitrogen atmosphere.The reaction mixture, which was initially dark green, changed to black, indicating the formation of the indium nanoparticles (InNPs).Then, allyl bromide (181.3 mg, 0.128 mL, 1.5 mmol) in THF (1 mL) was added by syringe, and the suspension was stirred for 30 minutes.Subsequently, benzaldehyde (53 mg, 0.051 mL, 0.5 mmol) in THF (1 mL) was added by syringe.After total conversion of the starting material (TLC, GC), the resulting suspension was diluted with diethyl ether (10 mL) and treated with 10% HCl and water.The combined extracts were dried over anhydrous Na 2 SO 4 , and evaporated (20 mbar).The resulting residue was purified by flash column chromatography (silica gel, hexane-ethyl acetate) to give the corresponding homoallylic alcohol 1a (86%).

Computational Procedure
The calculations were performed with Gaussian09. 9The initial conformational analysis of selected compounds was performed with the semiempirical AM1 method.The geometry of the most stable conformers thus obtained was used as starting point for the B3LYP studies.The zero point energy corrections were made at the LANL2DZ level for the thermodynamic quantities.The exploration of the potential surface was carried out varying the selected coordinate with full optimization for the remainder degrees of freedom.The characterization of all stationary points was done by Hessian matrix calculations of geometries obtained with full optimization for a minimum and by using the TS methodology for a transition state.Figures were built with the GaussView program.

Figure 1 .
Figure 1.Gas Phase B3LYP/LANL2DZ potential energy profile for the formation of 1a

Figure 2 .
Figure 2. Gas Phase B3LYP/LANL2DZ potential energy profile for the formation of 1b