Aqueous-Phase Catalytic Chemical Reduction of p-Nitrophenol Employing Soluble Gold Nanoparticles with Different Shapes

Abstract

Gold nanoparticles with different shapes were prepared and used as catalysts in the reduction of p-nitrophenol (PNP) in the aqueous phase and in the presence of sodium borohydride (NaBH₄). Parameters such as the reaction temperature, substrate/NaBH₄ molar ratio, and substrate/gold molar ratio were tested and evaluated. In this paper, we compare the catalytic reactivities of gold nanorods (AuNRs) and gold nanospheres (AuNSs), both synthesized by the seed-mediated method in the presence of cetyltrimethyl ammonium bromide (CTAB). Physical-chemical parameters such as the apparent rate constant (k_app) and activation energy (E_a) of the reactions were obtained for both systems. We observed that the catalytic system based on AuNRs is the most active. These colloidal dispersions were investigated and fully characterized by ultraviolet-visible absorption spectroscopy (UV–Vis) and transmission electron microscopy (TEM).

1. Introduction

The study and development of new soluble nanocatalysts in water has attracted the attention of several research groups in recent years. Due to their large surface/volume ratio, nanoparticulated catalytic materials can be considered as better catalysts than their bulky form counterparts [1,2]. In recent years, economic and ecological priorities have pressed the scientific community to develop highly active catalytic systems in water for different reactions [3]. It is well known that gold nanoparticles (AuNPs) have received widespread attention due to their interesting physical and chemical properties, particularly in relation to their catalytic activity [4]. In its bulky forms, metallic gold presents low reactivity and has been overlooked for many years as a potential catalyst. However, it was more recently verified that gold nanoparticles are quite active catalysts for many reactions, such as chemical reductions and oxidations [5]. In fact, hydrogenation reactions are the most common reactions that have been conducted using colloidal or supported transition metal nanoparticles. Although gold-based colloids have been widely employed to prepare supported gold catalysts, only a few reports are available in which a liquid colloidal dispersion of AuNPs was employed as the catalytic system [6]. In this context, the synthesis of soluble AuNPs in a high yield with high selectivity in terms of size and shape and the study of their potential applications in catalysis remains the subject of many scientific studies.

It is important to remember that gold has a limited ability to chemisorb hydrogen [7,8] compared to the platinum group metals, and thus, it is a less active catalyst for reactions involving the dissociation of hydrogen [9,10]. Indeed, there are only few works in which molecular hydrogen is employed in the reduction of the chemical species using gold as the catalyst [11]. On the other hand, reduction reactions using NaBH₄ catalyzed by supported AuNPs are frequently found in the literature [12]. For example, AuNPs supported on MgO [13], TiO₂ [14], Fe₃O₄ [15], SiO₂ [16], and Fe₂O₃ [17] are used as catalysts in the reduction reactions of nitro compounds, such as p-nitrophenol, p-nitrotoluene, and their derivatives.

Beyond the homogeneous and heterogeneous classification of catalysts, the morphology [18,19] and size [20,21,22] of the catalytic particles are crucial factors for nanocatalysis. Indeed, changes in the size and shape of nanoparticles are commonly used to explain the catalytic performance of a specific catalytic material. Although some works revealed that the reactivity and selectivity of reduction in the presence of NaBH₄ and catalyzed by AuNPs are dependent on the size and shape of the nanoparticle employed [23,24], with this work we added new aspects that can shed more light on the influence that the size and shape of nanocatalysts has on the performance of a specific catalytic transformation.

Here, we have carried out a series of experiments in order to compare the catalytic reactivities of two AuNPs with different morphologies—gold nanorods (AuNRs) and gold nanospheres (AuNSs)—in the reduction of p-nitrophenol (PNP) in the aqueous phase in the presence of sodium borohydride (NaBH₄). From those experiments, physical-chemical parameters such as the apparent rate constant (k_app) and activation energy (E_a) of the reactions were obtained for both systems.

2. Results and Discussion

The syntheses of the colloidal AuNPs (spheres and rods) were carried out using the same method (seed-mediated) in the presence of cetyltrimethyl ammonium bromide (CTAB). This is an important issue, since the particle-coating molecules display an important role related to the access of the substrates to the catalytic surface. Both colloids were characterized by UV–Vis spectroscopy, and their particles were analyzed by TEM (Figure 1). The absorption spectra of both colloids are typical of systems containing rods and spheres; i.e., two bands with maximum absorptions at 515 and 702 nm, and one band with maximum absorption at 530 nm, respectively. The TEM images confirmed the following trend: AuNSs of approximately 20 nm and AuNRs with dimensions of approximately 40 nm × 12 nm (aspect ratio 3.3) were obtained.

2.1. The Catalytic Activity

In this study, p-nitrophenol (PNP) was converted into p-aminophenol (PAP) in the presence of NaBH₄ using AuNSs or AuNRs as nanocatalysts for this chemical transformation. The chemical reduction of PNP was easily followed by UV–Vis spectroscopy. The addition of a specific amount of the gold colloidal solution (catalyst) into the aqueous solution of PNP and NaBH₄ starts the reaction, and a sequence of UV–Vis spectra were taken, allowing the kinetics of the conversion of the PNP into PAP to be followed. With the addition of the catalyst, the absorption band at 400 nm related to the substrate (PNP) starts to decrease, while a new band is formed at 300 nm due to product (PAP) formation [13]. It is worth mentioning that due to the reaction conditions, the anionic forms of both PNP and PAP are, in fact, the ones measured. In Figure 2, we can observe a typical example of the trend of the absorption bands related to the consumption of PNP and the formation of PAP during the chemical reduction of the nitro functional group. Moreover, in the absence of the catalysts, no reaction was observed at reasonable times [23].

For comparison, we have found that both colloidal solutions evaluated are active as catalysts independent of the shape of the AuNPs present in the reduction of PNP in the presence of NaBH₄. Here, it is important to mention that only a few reports have discussed the interaction of molecular hydrogen with gold. Sault and coworkers have shown that no hydrogen chemisorption occurs on the surface [25]. On the other hand, Stobinski and coworkers [26] suggested that some dissociative chemisorption of hydrogen molecules can occur on low coordinated gold atoms (corner and edge sites). Nevertheless, there is a lack of information concerning quantitative data of the adsorption of hydrogen on gold surfaces and supported gold catalysts, but the previously mentioned reasons explain why chemical reductions catalyzed by gold using molecular hydrogen are extremely rare.

The catalytic reduction of PNP in the presence of NaBH₄—in which the reduction of the nitro functional group of the molecule is observed—is, in fact, a reaction model usually adopted for the evaluation of potential catalysts. Moreover, the reduction of nitro compounds is a fundamental transformation that occurs in many chemical inputs and intermediates at both laboratory and industrial levels [27]. This catalytic reaction normally occurs with an excess of the hydride source NaBH₄, since a considerable part of the activated hydride on the surface of the catalyst is lost as gaseous molecular hydrogen [20,28]. Figure 3 is an illustrative representation of the reaction, with the activation of the hydrides on the surface of the catalyst and their transfer to the nitro group. In fact, according to the literature, in the reduction of nitroarenes using NaBH₄ catalyzed by AuNPs, first, a B–H bond cleavage is the rate-determining step to give the [Au]-H species. This species is responsible for the reduction of PNP into the corresponding hydroxylamine [12].

In this context, we elaborated a series of experiments to investigate the relationship between the morphology of the nanocatalyst and the catalytic reactivity of the system.

2.2. Influence of the NaBH₄ Concentration

This study consisted of evaluating the effect of the modification of the concentration of the reducing agent NaBH₄ on the kinetics of the conversion of PNP using both catalytic systems. As we can see in

Table 1. Evaluation of the rate of conversion of PNP using different concentrations of NaBH₄ ¹.

Entry	[NaBH4] (mol·L−1)	NaBH4/PNP Molar Ratio	Conversion Time (s)
			AuNRs	AuNSs
1	0.10	500	50	266
2	0.05	250	58	340
3	0.025	25	155	555
4	0.0025	12.5	2314	7200

¹ Reaction conditions: 2.0 mL of the PNP solution (0.1 mol·L⁻¹), 1.0 mL of the NaBH₄ solution, 0.05 mL of the AuNPs solution (2.0 × 10⁻⁴ mol·L⁻¹), reaction temperature of 25 °C, and at atmospheric pressure. PNP, p-nitrophenol; AuNRs, gold nanorods; AuNSs, gold nanospheres.

, the higher the concentration of NaBH₄, the faster the reaction is for both systems studied here, confirming that hydride species are involved in the rate-determining step of the reduction reaction of PNP [12].

In this test, we only analyzed the time necessary to completely convert our substrate. The UV–Vis absorption spectrum as a function of the wavelength allows the conversion of PNP to PAP to be followed. The absorption band at 400 nm indicates the formation of p-nitrophenolate, which is formed from the alkaline solution after the addition of NaBH₄, and the yellow color of the solution is observed [29]. After the addition of the nanocatalyst, the absorbance band related to the PNP (anionic form) decreases over time, and gradually, a new peak at around 300 nm emerges, corresponding to the formation of PAP (anionic form). Visually, the conversion of PNP is characterized by the modification of the color of the solution from yellow to colorless [13,30].

2.3. The Reaction Temperature

In this study, we carried out a series of reactions of the reduction of PNP in the presence of AuNPs in order to obtain thermodynamic parameters that show how the reaction temperature can affect the kinetics of such a reaction. From Figure 4, it can be seen that the reaction temperature has an important influence. In fact, at higher reaction temperatures, the conversion of PNP is faster. One interesting observation is that when AuNSs are used as the catalyst, the conversion of PNP is significantly slower at 25 °C. This trend shows that the nanocatalyst morphology is indeed an important issue to consider.

To obtain more information about the kinetics of the conversion reaction of PNP, for both catalytic systems, we examined the correlation between ln([PNP]/[PNP]₀) and reaction time for all reaction temperatures studied here (see Figure 5 and Figure 6). Supposing, by approximation, that we are under a pseudo-first-order reaction regime for all reactions, linearity must be observed for each set of reaction conditions. From the slope of the lines, we obtain the apparent rate constant (k_app) of the reaction at each set of reaction conditions.

The apparent rate constants obtained for this set of reactions are summarized in

Table 2. Apparent rate constants (k_app) for the reactions carried out at different temperatures using AuNRs or AuNSs as nanocatalysts.

Temperature (°C)	kAuNRs (s−1)	ErrorAuNRs	kAuNSs (s−1)	ErrorAuNSs
25	1.0 × 10−3	±1.28 × 10−4	0.1 × 10−3	±8.24 × 10−6
35	3.0 × 10−3	±4.06 × 10−4	3.9 × 10−3	±2.74 × 10−4
45	14 × 10−3	±0.005	4.4 × 10−3	±2.9 × 10−4
55	18 × 10−3	-	6.9 × 10−3	±5.3 × 10−4

. One can see that at different reaction temperatures, different rate constants are obtained, and as the reaction temperature increases, the rate constant increases. Nevertheless, it is important to see that the apparent rate constants for the systems using AuNRs as the catalyst are higher. This trend was expected, since the times to reach full conversion of PNP were shorter when AuNRs were used as the catalyst compared to reactions carried out under the same reaction conditions using AuNSs (see Table 1).

Based on the Arrhenius’ equation [31], by plotting the graph of the rate constant for each catalytic system versus the corresponding 1/T (see Figure 7), one can obtain the activation energy (E_a, apparent) for the reduction reaction of PNP for each catalytic system. From the slopes of the lines, we obtained the following values for E_a: 70 and 105 kJ/mol for the catalytic systems based on AuNRs and AuNSs, respectively.

It is important to evaluate these results in terms of the particle morphology. As we adopted the seed-mediated method to prepare the AuNPs for both colloidal systems, it is expected that the number of AuNPs in both catalytic systems is the same, as well as the total number of gold atoms. Nevertheless, as they have different morphologies, they can also display different numbers of atoms at their surfaces. Spheres display the lowest relative number of atoms at the surface in comparison to any other geometry. Therefore, under our restrictions of the number of total atoms and number of particles (i.e., both are the same in both catalytic systems), AuNRs should present the higher reactivity. However, it is important to remember that the types of facets, edges, and vertexes of the nanocrystal formed also have an important role for the reactivity of the surface of the nanocatalyst employed [24].

3. Materials and Methods

3.1. General

HAuCl₄·3H₂O (99.9%), NaBH₄ (99%), (+)-l-ascorbic acid (99%), CTAB, and AgNO₃ (99%) were obtained from Sigma-Aldrich (St. Louis, MO, USA) and used as purchased. Distilled water was used to prepare all solutions. UV–Vis/near-IR spectra were recorded on a Shimadzu UV-3600 spectrophotometer (Shimadzu, Kyoto, Japan) using optical glass cells with a length of 1.0 cm. The set-up was configured to fix the baseline of the distilled water absorption band from 200 to 1000 nm. TEM analyses were performed on a FEI Tecnai 20 model electron microscope (FEI, Hillsboro, OR, USA) at an accelerating voltage of 200 kV, and the samples were prepared by the addition of a drop of the gold colloidal solution on a copper grid coated with a porous carbon film.

3.2. Seed-Mediated Synthesis

AuNPs prepared in this study were produced by a seeded-growth method adapted from the protocols developed by Murphy [32] and El-Sayed [33]. The method consists of preparing two solutions: (i) the seed solution; and (ii) the growth solution.

3.2.1. Seed Solution

In a 25 mL flask, 0.1 mL of an aqueous solution of HAuCl₄·3H₂O (0.025 mol·L⁻¹, 0.0025 mmol) was added to a 7.4 mL aqueous solution of CTAB (0.0676 mol·L⁻¹, 0.5 mmol). Then, under stirring, 0.6 mL of an ice-cold aqueous solution of NaBH₄ (0.01 mol·L⁻¹, 0.006 mmol) was added. The solution color immediately turned to brown. After 2 min, the system was left for at least 2 h without stirring before use.

3.2.2. Growth Solution

In a 25 mL flask, 0.2 mL of an aqueous solution of HAuCl₄·3H₂O (0.025 mol·L⁻¹, 0.005 mmol) was added to a 7.3 mL aqueous solution of surfactant (0.0685 mol·L⁻¹, 0.5 mmol). Then, 0.15 mL of an aqueous solution of silver nitrate (4 × 10⁻³ mol·L⁻¹) was added under stirring, followed by the addition of ascorbic acid (0.070 mL, 0.0788 M). The system turned to colorless, proving the reduction of Au³⁺ to Au⁺.

3.2.3. Growth Process

A 0.060 mL aliquot of seed particles was added to the freshly prepared growth solution. The solution was kept under stirring for just 10 s and then allowed to stand for 4 h without stirring prior to characterization to ensure system stability.

3.3. Reduction Process in the Presence of NaBH₄

All reactions were carried out in a 4 mL glass optical cuvette at atmospheric pressure. Under specific temperatures (25, 35, 45, and 55 °C), the reagents were added in this sequence: 2.0 mL of the PNP aqueous solution (0.1 mol·L⁻¹), 1.0 mL of the aqueous solution of NaBH₄ at specific concentrations (0.1, 0.05, 0.025, and 0.0025 mol·L⁻¹), and 0.05 mL of the AuNPs solution (2.0 × 10⁻⁴ mol·L⁻¹). The catalytic conversion of PNP was then analyzed by UV–Vis spectroscopy.

4. Conclusions

In this work, we have demonstrated that aqueous colloidal solutions based on soluble AuNRs or AuNSs—both coated with CTAB and produced by seed-mediated method—were catalytically active in the conversion of PNP, producing PAP in the presence of NaBH₄. Although the amount of gold used was the same for both catalytic systems, they displayed different catalytic proprieties. The reaction rates for both systems are very sensitive to the reaction temperature, but in all cases, under the same reaction conditions, the k_app for the system containing AuNRs as catalyst were higher. This trend was expected, since the time to reach the full conversion of PNP was shorter when AuNRs were used as catalysts. From our results, we could estimate the activation energy of the reduction process of PNP to produce PAP; i.e., 70 kJ/mol and 105 kJ/mol for the catalytic systems based on AuNRs and AuNSs, respectively. It is important to evaluate all results obtained here in light of the particle morphology. As we adopted the seed-mediated method with the same amount of gold to prepare the AuNPs for both colloidal systems, it is expected that the number of AuNPs in both catalytic systems are the same, as well as the total number of gold atoms. Nevertheless, as they have different morphologies, they can also display different numbers of gold atoms at their surface. Spheres display the lowest relative number of atoms at the surface in comparison to any other geometries. Therefore, under our restrictions of total number of atoms and number of particles, AuNRs should present the higher reactivity. However, it is important to remember that the type of facets, edges, and vertexes of the nanocrystal formed also display important role for the reactivity of the surface of the employed nanocatalyst.