1 Introduction

The importance of knowing how to prepare gold nanoparticles (AuNPs) with different sizes and shapes resides in their potential application in materials science, medicine, biological processes and catalysis [1,2,3]. In the case of heterogeneous catalysis, the catalytic activity and selectivity of these nanoparticles depend on their size and shape, the chemical medium to which they are exposed during reaction and, if they are immobilized on a support, the interaction between the nanoparticles and the support [4]. A convenient method to prepare metal nanoparticles is the use of plant extracts, as they contain a complex mixture of biomolecules, resulting from the plant metabolism, which are able to reduce the precursor metal salt to metal nanoparticles while acting as the same time as capping agents that would protect the nascent metal aggregates from excessive growing. This overall process is called bioreduction [5,6,7,8]. Typically, bioreduction by plant extract involves mixing the extract with an aqueous solution of the relevant metallic salt. The reaction occurs at room temperature and is usually completed in a few minutes. In view of the number of chemicals involved, the bioreduction process is relatively complex. Therefore, the challenge is to prepare stable gold nanoparticles with controlled sizes and shapes. However, the complexity of the bioreduction process together with the complex chemical composition of the plant extracts makes difficult to predict in advance the nature of the gold entities resulting from the use of specific plant extracts. For these reasons, a variety of plants have been and are currently being explored as bioreduction agents for the preparation of metal nanoparticles, particularly for the synthesis of gold nanoparticles [9, 10]. Moreover, the methodology commonly used to carry out the bioreduction process starts from an aqueous solution of tetrachloroauric acid, to which the plant extract acting as both reducing and capping agent is added, what usually results in AuNPs of relatively large size [11]. In our case, as we are interested in catalytic applications of gold entities, we are aiming at the synthesis of small-size Au aggregates. On this regard, it has been shown elsewhere [12] that if a solution of gold in a mixture of ammonium chloride and concentrated nitric acid (this mixture is usually called aqua regia) is contacted with rosemary oil in a two-liquid phase system, gold nanoclusters and/or AuNPs are formed as a function of the specific synthesis procedure. The aqua regia solution is strongly oxidant and we believe this would contribute to reduce the average size of the nascent Au entities during bioreduction as compared with those formed by the aforementioned conventional methods that use a solution of tetrachloroauric in water. Therefore, in order to accomplish our goal of favoring as much as possible the formation of gold nanoclusters but not AuNPs, we have also used in this work a solution of gold in aqua regia. In addition, this new approach has been explored here by treating the Au solution in aqua regia with aqueous extracts of two different Mexican plants, Toad Leaf (Eryngium heterophyllum, SF) and Cuachalalate (Amphipterygium adstringens, CF), acting as bioreductors but forming when added to the Au solution a single phase. To the best of our knowledge, these two plants have never been used yet as bioreductors for the synthesis of metal clusters. These plants that grow in Mexico have medicinal uses, and they should have chemical compounds interesting for the reduction of metallic salts to the corresponding metal, in this case, gold. Cuachalalate (A. adstringens) is a tree that grows in the south–south west tropical forest of Mexico. It has been used since pre-Columbian times as a medicinal plant to treat some stomach and intestinal illness such as gastritis, ulcers and colitis an also to help cicatrization of skin [13]. At the present time, it is also investigated for anti-cancer activity, which makes it one of the most important plants in the Mexican herbarium, mainly used by drinking the aqueous extract prepared from its bark. Anti-inflammatory properties have been correlated to its high content in masticadienonic acid (C30H46O3), which is a triterpene, and some of its derivatives [14]. Toad Leaf (E. heterophyllum) is an herbaceous plant distributed from southeastern Arizona to western Texas and south of Mexico. It has been used in traditional medicine to treat arthritis, inflammatory processes and diabetes. Currently, it is investigated as an effective agent to reduce triglyceride and cholesterol levels in blood [15]. The aqueous and ethanolic extract from the aerial part of the dry plant are commonly prepared as medicinal beverages. To the best of our knowledge, full chemical characterization of the extracts of this plant has not yet been reported. However, chemical reactions for the detection of secondary metabolites suggested the presence of alkaloids, coumarins and phenolic compounds in the aqueous extract of Toad Leaf [15]. Therefore, both plants extracts contain complex organic compounds that would have the capability of acting as bioreductors of metallic salts, and in particular of gold. In addition to the preparation of gold nanoclusters or nanoparticles, it is important to note that once they have been obtained, it would be convenient to immobilize these gold entities in a material with large surface area and appropriated porosity so that they can be used as heterogeneous catalysts. For this reason, in this work the support that was used to immobilize the gold entities was the SBA-15 ordered mesoporous material functionalized with mercaptopropyl groups. SBA-15 contains a hexagonal array of mesopores of about 6–8 nm, large enough to accommodate Au nanoclusters and small AuNPs, and to facilitate the diffusion of even relatively large substrates. The resulting Au/SBA-15 materials have been used as catalysts for the selective oxidation of cyclohexene in liquid phase with molecular oxygen at atmospheric pressure. Au nanoparticles supported on graphite, silica and TiO2 [16, 17] and Au nanoclusters supported on thiol-functionalized mesoporous materials prepared from different synthesis routes have been used as catalysts in this reaction [18,19,20].

2 Experimental

2.1 Extraction Method of the Plants

2.1.1 Cuachalalate Tree

50 g of Cuachalalate bark were divided in small pieces (~ 2 cm) and then placed into a cellulose Soxhlet thimble. Then, extraction was performed by re-circulating 250 mL of a boiling 70/30 water/methanol solution from a round bottom flask in a typical Soxhlet system during 6 h. The extract was roto-evaporated under vacuum at 60 °C until a volume of approximately 40 mL was reached and then collected for the synthesis of nanoparticles as described below (Sect. 2.3).

2.1.2 Toad Leaf

Steam distillation was performed by placing 100 g of small pieces (~ 2 cm) of the aerial part of the plant in a 2 L glass alembic, connected in the bottom to a 3 L round bottom flask containing 2 L of distilled water. The top of the alembic was connected to a Liebig condenser through a still head. Steam from the boiling water produced in the round bottom flask was passed through the plant, vaporizing and carrying over the desired oil to the condenser. Once there, steam and vaporized oil were condensed and collected. The oil phase was separated by decantation. This procedure was repeated two more times (using fresh plant each time) in order to obtain ~ 40 g of the extract to be used for the synthesis of nanoparticles as described below (Sect. 2.3).

2.2 Synthesis of Thiol-Containing SBA-15

Propyl-thiol mesoporous SBA-15 was prepared from a gel with molar composition: 1.0 TEOS:0.06 MPTMS:0.0186 P123:6.42 HCl:180 H2O, where TEOS stands for tetraethyl orthosilicate (Sigma-Aldrich, > 99%); MPTMS for 3-mercaptopropyltrimethoxysilane (Sigma-Aldrich, 95%); P123 for Pluronic 123, the triblock co-polymer PEO20PPO70PEO20, m.w. ∼5800 (Sigma-Aldrich); and HCl for hydrochloric acid (Panreac, 37 wt%), according to the procedure described in [21], as follows: 125 mL of 1.9 M HCl were placed in a 500 mL plastic bottle provided with a cover having a hole for allowing the insertion of a PTFE (polytetrafluoroethylene) stirrer blade, and 4 g of P123 were added. Then, the bottle was heated at 40 °C in a silicone oil bath, and 8.2 mL of TEOS were added under continuous stirring. After 45 min, 412 μL of MPTMS were added, and the mixture was stirred for 22 h. Then, it was poured into a stainless-steel autoclave provided with a Teflon liner, and heated statically at 100 °C for 24 h. After that, the autoclave was cooled and its content filtered, washed with ethanol and dried at room temperature overnight. The dried sample was treated with ethanol (200 mL of ethanol/g of sample) under stirring in a 1 L round-bottom flask at 90 °C for 24 h in order to remove the surfactant, and then dried at room temperature. The resulting sample is denoted SH-SBA-15.

2.3 Synthesis of the Gold Nanoparticles

A gold lump (0.967 g Johnson-Matthey, 99.99%) was dissolved under gentle agitation in 150 g of aqua regia. The latter was prepared by dissolving ammonium chloride (Sigma-Aldrich, > 98 wt%) in concentrated nitric acid (Panreac, 65 wt%), in a ratio of 1:4 w/w and heated in silicone bath at 50 °C. This gold solution was then divided into two parts (71 g each) and the toad extract (40 g) was added to one part, and the Cuachalalate extract (40 g) was added to the other 71 g of the gold solution. Reaction solutions were left for 4 days. After this time, 10 g of each reaction solution was taken, 40 mL of distilled water was added, and then to each of them 5 g of the respective extracts were added. To immobilize the gold nanoparticles, 300 mg of extracted SH-SBA-15 were added to 40 g of each of the gold solutions, and stirred for 3 h at room temperature. Then, the solid was recovered by centrifugation, washed with ethanol repeatedly and dried at room temperature. The sample obtained from the Toad Leaf extract was denoted as SF-SH-S (SF), and that from Cuachalalate as CF-SH-S (CF).

2.4 Catalytic Tests

The two gold-containing SF-SH-S and CF-SH-S materials described above were used as catalysts for the liquid phase oxidation of cyclohexene using molecular oxygen as oxidant at atmospheric pressure, in the presence of a small amount of toluene as solvent and TBHP as initiator. A sample of SH-SBA-15 material that contains no gold was also tested for comparison purposes.

The oxidation reaction of cyclohexene with oxygen was performed at atmospheric pressure under stirring in liquid phase [18], in a four-neck 50 mL flask placed in a silicone oil bath to maintain the reaction temperature at 65 °C. The reaction mixture was made of 0.049 mol of cyclohexene (4.055 g, Sigma-Aldrich, 99%), 0.4055 g of octane (10 wt% referred to cyclohexene; Sigma-Aldrich, > 99%), 0.2027 g of a tert-butylhydroperoxide solution (TBHP, 5 wt% referred to cyclohexene; ∼ 5.5 M in decane, Sigma-Aldrich), 3.041 g of toluene (75 wt% of the cyclohexene; Panreac, > 99.5%). To this mixture, 0.05 g of catalysts were added. O2 (air liquid, purity 99.999%) was bubbled through the reaction mixture (1.8 mL/min). The catalyst was previously activated at 100 °C under N2 inside the reaction flask. The reaction products were analyzed by GC by using a Varian CP-300 instrument provided with a Varian VF-1 ms dimethylpolysiloxane column, 15 m of length and 0.25 mm of i.d.

2.5 Characterization Techniques

Powder X-ray diffraction was done using a PANalytical X’pert Pro instrument (Cu Kα radiation). The gold content of the solid was determined by inductively coupled plasma (ICP-OES) spectrometry with an ICP Winlab Optima 3300 DV Perkin-Elmer spectrometer. Thermogravimetric analyses were performed in a Perkin-Elmer TGA7 instrument, in an air flow of 40 mL/min, with a heating ramp from 25 to 900 °C at 20 °C/min. CHNS elemental analyses were obtained in a LECO CHNS-932 analyser provided with an AD-4 Perkin-Elmer scale. Nitrogen adsorption–desorption isotherms where measured in a Micromeritics ASAP 2420 apparatus at the temperature of liquid nitrogen (− 196 °C). Prior to analysis, the samples were degassed in situ at 70 °C in vacuum for 16 h. Surface areas were determined using the BET method. The pore volume and the average pore diameter were calculated by applying the BJH method to the adsorption branch of the isotherm. Diffuse reflectance UV–visible (UV–vis) spectra were recorded on a Cary 5000 Varian spectrophotometer equipped with an integrating sphere with the synthetic polymer spectralon as reference. Transmission electron microscopy studies (TEM) were carried out using a JEOL 2100F electron microscope operating at 200 kV in conventional and STEM mode. The catalysts were dispersed in ethanol and dropped onto a holey carbon copper grid for the observations.

3 Results and Discussion

3.1 Characterization of the Gold-Containing Samples

After contacting SH-SBA-15 with the solution containing the extract of each of the plants, as explained in the experimental part, the presence of gold in the resulting samples was corroborated. The gold contents were 1.1 and 0.7 wt% for the SF and CF samples (Table 1), respectively, showing that the extract from Toad plant is more efficient in incorporating gold into the SH-SBA-15 support. From the X-ray diffraction patterns of the samples it is verified that the SBA-15 phase remains after immobilization of gold, because the characteristic (100), (110) and (200) diffraction peaks of SBA-15 materials were observed, indicating the presence of well-ordered solids (Fig. 1). The d (100) spacing and hence the unit cell is very similar in the two samples (Table 1). The adsorption/desorption N2 isotherms (Fig. 2) are characteristic of well-ordered mesoporous SBA-15 materials, with a surface area between 480 and 590 m2/g, pore volume between 0.64 and 0.78 cm3/g, while the pore diameter is practically the same, 5.7–5.8 nm (Table 1). The reported average pore size of a pure silica SBA-15 material which does not contain mercaptopropyl groups is 7.0 nm [22]. Therefore, this reduction in the pore size from 5.8 to 7.0 nm evidences that the mercaptopropyl groups are anchored in the surface of the pores, which decreases the effective pore opening. Indeed, chemical analysis evidences the presence of sulfur (1.6 wt%) and carbon (Table 2) which indicates the actual presence of mercaptopropyl groups. The thermogravimetric analysis (TGA), Fig. 3, shows a weight loss at T < 140 °C corresponding to the desorption of water, and two weight losses centered at 230 °C and in the range 320–400 °C. Weight losses at T > 300 °C are generally attributed to a partial oxidation of the initial SH groups to sulfonic and/or disulfide species, while materials that contain only SH groups usually show a sharp weight loss centered at nearly 300 °C corresponding to the decomposition/desorption of these groups [23]. However, no such sharp weight loss is clearly observed in the TG of the two Au-containing materials. More recently, TG patterns of S-bearing SBA-15 samples similar to those of the Au-materials synthesized in the present work have been associated to major oxidation of the initial SH groups to SO3− groups [19]. Therefore, in the gold-containing materials prepared from plants extracts that were in contact for 4 days with the [AuCl4]− solution, the original mercaptopropyl groups would have been oxidized in considerable extension.

Table 1 Structural and textural properties of the samples
Full size table
Fig. 1
figure 1

XRD pattern of the CF-SH-S (CF) and SF-SH-S (SF) samples

Full size image
Fig. 2
figure 2

N2 adsorption/desorption isotherms of the samples CF-SH-S (CF) and SF-SH-S (SF)

Full size image
Table 2 Chemical composition (wt%) and TG results of the two catalysts
Full size table
Fig. 3
figure 3

TG curves of samples: CF-SH-S and SF-SH-S

Full size image

From the chemical and TG analysis, Table 2, it can be seen that C/S ratio is 12 in both samples, this being a value higher than that reported for mesoporous materials functionalized with mercaptopropyl groups [21, 23]. This high amount of carbon may be due to the presence of some methoxy groups of the silane precursors that were not hydrolyzed during the process of synthesis, and also to the partial covering of the surface of the material with ethoxy groups due to the prolonged time of reflux with ethanol required to remove the surfactant [19]. However, some residual organic compounds that could have been incorporated into the solid during the immobilization process could eventually contribute also to the high carbon content. Figure 4 shows the UV–vis absorption spectra of the catalysts, where it can be observed in both samples the presence of a band around 350 nm, which have been associated to Au clusters stabilized by thiol groups and with a diameter of approximately 0.8 nm [24]. No band at 520 nm characteristic of surface plasmon resonance of AuNPs is observed in the CF-SH-S sample, indicating the absence of gold nanoparticles. However, in the spectrum of SF-SH-S sample (inset) a band at 547 nm of very weak intensity is detected, a value that is somewhat close to the characteristic plasmon of AuNPs (520 nm) [25]. This suggests that a minor amount of the total gold of the sample could be present as nanoparticles, but Au clusters would nevertheless predominate by far. Figure 4 also collects a STEM image of the CF-SH-S sample, which shows the hexagonal arrangement of pores in a honeycomb-like structure characteristic of SBA-15. The image allows the observation of a strong contrast particularly on the walls of the channels, where the small dots of clearer contrast are attributed to the presence of highly dispersed gold entities in the SBA-15 mesoporous materials [12, 26], but no AuNPs were observed.

Fig. 4
figure 4

UV–vis absorption spectra of the samples before and after reaction (top). SF-R and CF-R correspond to catalysts after reaction. STEM image of sample CF-SH-S (down)

Full size image

3.2 Catalytic Reactions

The two gold-containing SF-SH-S and CF-SH-S materials described above were used as catalysts for the liquid phase oxidation of cyclohexene using molecular oxygen as oxidant at atmospheric pressure, in the presence of a small amount of toluene as solvent and TBHP as initiator. The two catalysts show activity in this reaction, Fig. 5, which is clearly higher than that of the Au-free SH-SBA-15 sample used as reference, which evidences that the enhanced activity is due to the presence of gold in the two catalysts. SF sample reaches a conversion of nearly 40% at 2 days of reaction time, while that of CF sample is 22% at the same reaction time. The higher activity of the SF sample is explained by its higher gold content. The variation of conversion as a function of the reaction time can be related to differences in the surface properties of the gold clusters, which could be governed by their size but also by their effective charge and the chemical interactions stablished with the sulphur-containing groups of the support. The conversion of cyclohexene increases rapidly up to 28 h, but the reaction rate is much lower beyond that point. This slowdown of the activity could be explained by a quickly aggregation of the nanoclusters in the reaction medium to form nanoparticles, which are not very active. Indeed, it has been observed that the color of the catalysts changes during the reaction, from white to pale yellowish before reaction, to pink and then to purple as the reaction proceeds. This observation evidences that the particle size of the gold increases in the reaction medium, as it is confirmed by UV–vis spectroscopy of the catalysts after reaction (Fig. 4), as both present an absorption band about 530 nm, while that at 350 nm of the fresh catalysts attributed to Au clusters has vanished. This aggregation of Au clusters immobilized on SBA-15 containing sulfonic groups have also been reported previously, and it has been attributed to the relatively low interaction strength between these groups and the Au clusters [18,19,20]. An increase of the size of Au nanoparticles during reaction in the benzyl alcohol oxidation has been also reported [27].

Fig. 5
figure 5

Conversion % vs reaction time of the catalysts

Full size image

Five reaction products were identified: cyclohexene epoxide, and cyclohexanediol from the double-bond oxidation, and 2-cyclohexen-1-ol, 2-cyclohexen-1-one and 2-cyclohexenyl hydroperoxide, resulting from the allylic oxidation of the cyclohexene ring. The cyclohexene conversion was calculated from the yields of the five products.

Figure 6 shows the selectivity of the five reaction products with the catalysts CF-SH-S and SF-SH-S. As it can be seen in this figure, the allylic oxidation pathway is dominant, and the selectivity to the corresponding products is nearly 85%. The predominance of the allylic oxidation is in agreement with what has been previously reported for 1% Au/carbon catalysts containing 5–50 nm gold nanoparticles [16]. For the CF-SH-S catalyst, the major reaction product is the enone, which remains stable as the conversion increases. The hydroperoxide selectivity increases from very low values at the beginning of reaction, to nearly 30% at 20% conversion, to decrease rapidly beyond that point. Opposite to the hydroperoxide, the enol selectivity is very high at low conversion, but then decreases and stabilizes around 17% for conversion > 20%. The enone/enol molar ratio is always > 1 and it increases with the conversion, reaching a value greater than 2 as the conversion approaches to 20%. The selectivity pattern of the SF catalyst is quite similar to that of CF, save for the smoother change in hydroperoxide selectivity as a function of conversion. It is interesting to compare the selectivity pattern of these catalysts with that of those synthesized also from a similar gold solution but using Rosemary essential oil as reductant and capping agent in a two-liquid phases system, immobilized on the same support [18]. In this case, the selectivity to the hydroperoxide is very high at low conversion but decreases as the reaction proceeds, being nevertheless the major product for conversion of 25%. The selectivity to enone follow the opposite trend. These results suggest that the reaction mechanism would be dependent upon the procedure used to prepare the gold entities, and eventually also of the nature of the functional groups to which they are attached in the support.

Fig. 6
figure 6

Selectivity vs conversion of catalysts CF-SH-S (top) and SF-SH-S (down)

Full size image

4 Conclusions

Gold entities have been prepared by using aqueous extracts of the Mexican medicinal plant Toad Leaf and the bark of a Mexican tree, Cuachalalate, as reductant and capping agents of gold dissolved in aqua regia. These gold entities have been immobilized on SH-SBA-15 mesoporous hybrid materials functionalized with mercaptopropyl groups. The structural characteristics of the SBA-15 material are not affected by the immobilization procedure of the gold entities; however, the thiol groups are partially oxidized during the impregnation process depending on the specific synthesis parameters. On the other hand, the amount of gold immobilized also depends on the nature of the vegetable extracts, being in the range 0.7–1.1 wt%. The gold is present mainly as nanoclusters, although the formation of minor amount of gold nanoparticles is favored in the catalyst with the highest Au content, 1.1 wt%, prepared from Toad Leaf. The two gold-containing catalysts are active in the aerobic oxidation of cyclohexene under mild conditions, and the activity increases with the gold content. The products of allylic oxidation of the cyclohexene ring are dominant. It has been found for both catalysts that the gold nanoclusters initially present evolve toward nanoparticles during the catalytic reaction. Moreover, the selectivity pattern is quite different than that of catalysts prepared from a similar gold solution by a two-liquid phase method using rosemary essential oil. This suggests that the reaction mechanism can be affected by the synthesis method and/or the nature of the functional groups linking the gold entities to the support, which would be worth of exploring.