1 Introduction

During the last few decades, Au nanoparticles have attracted great attention due to their size- and shape-dependent characteristics that make them interesting for a variety of applications [1,2,3,4,5]. One of the important applications of gold nanoparticles is their use as a catalyst for various reactions. In the 1970s Bond et al. showed the unexpected catalytic activity of Au on boehmite for the hydrogenation of alkyne in the presence of alkenes [6]. This was followed by the impressive work of Haruta et al. on the low-temperature oxidation of CO [7], and Hutchings on the hydrochlorination of ethyne using Au as a catalyst of choice in the 1980s [8]. More recently, Au nanoparticles have begun to be employed in various industries, especially the chemical and pharmaceutical industries [9].

The catalytic performance of Au nanoparticles depends sensitively on their size and shape [10]. The surface area and surface energy play a crucial role in the comparative catalytic performance of the various Au nanoparticle morphologies. Therefore, there is great interest in optimizing the synthesis of Au nanoparticles for their best catalytic performance. The anisotropic nanoparticles e.g. NRs are especially interesting owing to their higher surface area and the presence of a high concentration of low coordination atoms which act as catalytic hotspots. Various comparative studies for morphology-dependent catalysis have been conducted, but often with conflicting results. Kundu et al. showed that gold nanospheres were catalytically most active than Au NRs and nanoplates for the model reaction of catalytic reduction of 4-NP to 4-AP [11]. On the other hand, Jiji et al. showed that Au NRs were catalytically more active than gold nanospheres [12]. However, in addition to the morphology, the types of facets covering the surface for a certain particle shape can also have a marked influence on the catalytic activity. For example, Zhang et al. showed that the presence of high-indexed facets on gold nanoparticles enhanced their catalytic performance due to the presence of a higher fraction of surface atoms with low coordination [13].

Gold NRs are known to consist of higher energy side facets and lower energy end facets [14]. Therefore, the relative surface area of the side and end facets, which changes with the change in their aspect ratio, can affect their catalytic efficiency. Herein, we show that an increase in the aspect ratio of Au NRs can significantly increase their catalytic activity. We demonstrate the catalytic conversion of 4-NP to 4-AP and the catalytic degradation of MB and MO. Gold NPs of different aspect ratios were synthesized with the seed-mediated method. The aspect ratio was simply controlled by tuning the pH of the reaction medium. The NRs having a higher aspect ratio were found to be more effective for the catalytic reduction of 4-NP, MB and MO.

2 Materials and Methods

2.1 Materials

Chemicals used for the synthesis are cetyltrimethylammonium bromide (CTAB), chloroauric acid, sodium borohydride (NaBH4), ascorbic acid, nitric acid (HNO3) and silver nitrate (AgNO3). For washing purposes deionized (DI) water, and aqua regia were used.

2.2 Synthesis of Au NRs by Seed Mediated Method

For the synthesis of Au NRs, we used a seed-mediated method. It is a two-step process.

2.2.1 Preparation of Gold Seeds

100 μl of 0.025 M HAuCl4 solution was added into (10 ml, 0.1 M) CTAB solution. Then, 450 μl of 0.02 M freshly prepared NaBH4 was added under vigorous stirring. After the addition of NaBH4, the color of the solution turned brown, indicating particle formation. The seed solution was aged at 35 °C for 2–3 h before use.

2.2.2 Preparation of Growth Solution

Three Au NRs samples were prepared. For each one, 100 μl of 0.025 M HAuCl4 solution was added into (5 ml, 0.1 M) CTAB solution contained in a clean scintillation vial. Subsequently, 5 μl of 0.1 M AgNO3 was added. Next, 1 ml and 1.5 ml of 0.1 M HNO3 was added to two different vials, and the samples were labeled as R2 and R3, respectively. No acid was added to the third sample (labeled as R1). Subsequently, 35 μl of freshly prepared ascorbic acid (0.1 M) was added to each sample. For R1, the sample became colorless instantly while for R2 and R3 they became colorless within a few mins due to the slower reaction kinetics owing to the relatively lower pH of these growth solutions. Finally, 10 μl of seed solution was added to each vial. The resultant mixture was kept undisturbed at 30 °C for overnight to grow. The appearance of reddish color in the growth solution indicated the growth of the seeds. In this case, as well, the color appeared later for the samples with lower pH of the growth solution due to the presence of acid.

2.3 Characterization

The AuNR formation was initially confirmed with UV–Vis spectroscopy. To investigate the size and shape of AuNRs, transmission electron microscopy (TEM) was carried out. The catalytic activity of synthesized AuNRs was monitored by a UV–Vis spectrophotometer.

2.4 Catalysis

To assess the effect of different AR on the catalytic properties of Au NRs, we investigate the catalytic reduction of 4-nitrophenol (4-NP), methylene blue (MB) and methyl orange (MO). For this purpose, we carried out an experiment in which 1 ml of 0.1 M NaBH4 solution was added to 3 ml of either 5 × 10–5 M 4-NP, MO or MB. As discussed in the subsequent section, the yield i.e. conversion of Au ions to Au(0) is different for three Au NRs samples, therefore, we added different volumes of each AuNR sample in the reaction medium to keep the concentration of added Au atoms the same in all cases. For 4-NP and MO, 19 μl, 21 μl and 23 μl of R1, R2 and R3 solutions, respectively, were added. For MB, 37 μl, 42 μl and 46 μl of R1, R2 and R3, respectively, have been added. The Au NRs solution in all cases was added under vigorous stirring to make sure the efficient mixing of the catalysts. However, after a few seconds, the stirring was stopped. The reaction was observed with a UV–Vis spectrometer.

Fig. 1
figure 1

UV–Vis spectra of Au NRs

Full size image

3 Results and Discussions

Figure 1 shows the UV–Vis absorption spectra of the synthesized Au NRs. For all samples, the characteristic transverse and longitudinal plasmon peaks are present. Moreover, there is a clear red shift in the longitudinal plasmon peak position with an increase in the concertation of HNO3 in the growth solution. The red shift in the longitudinal plasmon peak indicates the increase in the aspect ratio of NRs [15].

Another consequence of the lower reducing power of ascorbic acid is the reduction in the Au ions to Au(0) conversion yield. As discussed before, the seed-mediated method of Au NRs does not provide 100% conversion of Au ions to Au(0). The concentration of Au(0) atoms in the product can be found by measuring the absorbance value at 400 nm [16]. It is well established that the absorbance value at 400 nm is entirely due to the interbond transitions. An absorbance value of 1 corresponds to the molarity of 0.4 mM of Au(0) [16]. The estimated concentration of Au(0) in our case is 0.0628 mM, 0.0552 mM and 0.0508 mM for R1, R2 and R3. A slightly lower molar concentration for R2 and R3 is due to the added acid, which increases the overall volume of the Au NRs suspension. The Au ion to Au NRs conversion yield of about 13% for all samples.

Fig. 2
figure 2

TEM images of samples R1 (a), R2 (b) and R3 (c); the scale bar is 50 nm in all cases

Full size image

Figure 2a–c depicts the TEM images of samples R1, R2 and R3, respectively. It can be easily seen that most of the synthesized Au NRs have a uniform rod-shaped morphology. The yield of nanorods is greater than 90% for all three samples. The average lengths of Au NRs are about 38 ± 6, 68 ± 12 and 71 ± 13 nm and the width are determined to be 16 ± 2, 15 ± 1.5 and 13 ± 1.9 nm. There is a clear increase in length and decrease in the width of NRs with increasing HNO3 concentration in the growth solution. Consequently, the aspect ratio of Au NRs also increases with the increase in HNO3 concentration. The average aspect ratios for Au NRs are approximately 2.3, 4.6 and 5.4 for samples R1, R2 and R3, respectively. Therefore, the TEM results are consistent with the UV–Vis spectra.

The mechanism of the formation of Au NRs is debated in the literature. However, it is widely believed that with the addition of HAuCl4 solution in the CTAB, chloride ions in the AuCl4− are replaced with bromide ions. The resultant AuBr4− makes a complex with cations of the surfactant i.e. CTA+ [16]. This results in the cathodic shifting of the redox potential of Au ions. Under these reaction conditions, ascorbic acid can only partially reduce Au(III) ions to Au(I) [17,18,19]. With the addition of seeds, Au(I) ions reduce to Au(0) on top of the seeds due to the catalytic effect provided by the seeds. The synthesized seeds by our method are single-crystal CTAB capped with a size of 2–4 nm, as described in the previous reports [16]. These seeds grow by atom-by-atom deposition of Au monomers preferably on the end facets. There are different opinions about the type of side and end facets with which Au nanorods are bounded. But it seemed to be accepted now that the side facets are {520}, which have relatively higher energy compared to the end {111} and {100} facets [14]. Therefore, the side facets are more prone to the preferential adsorption of the surfactant, or the underpotential deposition of Ag, which directs the deposition of Au monomers to the end facets resulting in anisotropic growth. With the decrease in the pH of the reaction medium, the reducing power of the ascorbic acid decreases [20, 21]. Consequently, slower growth kinetics favors the preferred deposition of Au monomers onto the end facets, rather than the side facet, leading to the formation of Au NRs with higher aspect ratios.

3.1 Catalysis Reduction Reactions Using Au NRs

The catalytic reduction of 4-NP, MB and MO was carried out using Au NRs as catalysts. 4-NP is a phenolic compound also known as 4-hydroxynitrobenzene or p-Nitrophenol [22]. In general, it is used in the production of drugs (e.g. Acetaminophen), ethyl and methyl parathion, pesticide, antifungal agent, and dyes. 4-NP is an intermediate in the manufacturing of paracetamol [23]. In an aqueous environment, 4-NP shows a peak at 400 nm which decreases rapidly in the presence of NaBH4 and Au NRs catalyst with time. Reduction of 4-NP results in the formation of 4-aminophenol (4-AP). The product peak for 4-AP appears gradually at 300 nm [24]. Figure 3a–c depicts the UV–Vis spectra of the reduction reaction. The peak at 400 nm decreases with time; there is an initial slower decrease which accelerates within a few mins. This could be due to the removal of the CTAB from the Au NRs. Initially, Au NRs are covered by the bilayer of the CTAB. It has been reported that in the presence of a high concentration of NaBH4 in the solution, which is relevant in this case, BH4− ions can replace the CTAB coting from the Au NRs [25]. Therefore, at the start, the reduction of 4-NP is slower but with the passage of time and the CTAB bilayer is removed from the Au NRs, more active surfaces are available for the reduction reaction to occur. Therefore, there is an increase in the catalytic degradation rate.

Fig. 3
figure 3

Catalytic reduction of 4-NP using sample R1 (a), R2 (b) and R3 (c) as a catalyst; natural logarithm of relative intensity vs time plot (d)

Full size image

To find out the relative performance of catalysts, we carried out the linear fitting of the ln (C/C0) versus the time plot (Fig. 3d). Here C is the instantaneous intensity and C0 is the initial intensity of 4-NP. The apparent rate constants for R1–R3 were estimated as 0.17447, 0.307 and 0.909 min−1, respectively. This indicates that with an increase in the aspect ratio of Au NRs, their catalytic capability increases. There is a five-fold increase in the apparent rate constant with the increase in aspect ratio from 2.3 to 5.4.

Methylene Blue (MB) is a cationic, thiazine dye that is one of the most frequently used substances in the dye industry [26, 27]. In an aqueous environment, MB shows maximum absorbance at 664 nm. In oxidized form it has an intense deep blue color; the reduced form leukomethylene blue (LMB) is a colorless product [28, 29]. The catalytic degradation of MB into LMB was investigated in the presence of NaBH4, and by employing Au NRs as a catalyst.

The absorption peak at 664 nm decreased gradually with time as shown in Fig. 4a–c. Similar to the previous case, there is an initially slower degradation which accelerates with time, probably due to the removal of the CTAB from the Au NRs. Rate constants for sample.

Fig. 4
figure 4

Catalytic degradation of MB using samples R1 (a), R2 (b) and R3 (c) as a catalyst; natural logarithm of relative intensity vs time plot (d)

Full size image

R1, R2 and R3 were measured as 0.17692, 0.25597 and 0.44892 min−1, respectively, by making a linear fit to the plot of Fig. 4d. Therefore, the degradation of dye is faster for Au NRs of higher aspect ratios. In this case, there is a 3.5 order increase in the rate of reaction when the aspect ratio of Au NRs was increased from 2.3 to 5.4.

Finally, we carried out the catalytic degradation of MO which is a common anionic azo dye [21, 30]. It is water-soluble and used to stain silk, wool, etc. Methyl orange is widely used in various fields from textiles to the pharmaceutical industry. But it is also a carcinogenic dye, therefore, its degradation into non-toxic dyes is crucial [31]. In an aqueous environment, MO shows maximum absorbance at 464 nm. This absorption peak at 464 nm decreases rapidly in the presence of NaBH4 and Au NRs as shown in Fig. 5a–c which indicates its reductive degradation. A new peak at 248 nm appears which increases in intensity with time. This verifies the successful degradation of MO into two or more molecules with amine (–NH2) [32]. Figure 5d depicts the variation of ln(C/C0) with time. In this case, as well, there is a slight change in the slope with time, which could be due to the removal of the CTAB bilayer with the time that exposes the active sites and accelerates catalysis, as discussed before. Rate constants for samples R1, R2 and R3 were 0.367, 0.429 and 0.462 min−1, respectively. Although in this case there is not a marked difference in the degradation rates, they still increase with the increase in the aspect ratio of NRs.

Fig. 5
figure 5

Catalytic degradation of MO using samples R1 (a), R2 (b) and R3 (c) as a catalyst; natural logarithm of relative intensity vs time plot (d)

Full size image

3.2 Discussion

The surface energy of the gold nanoparticles depends sensitively on the type of facets. As mentioned before, the side facets of Au NRs have higher energy {520} compared to the end {111} and {100} facets which have relatively lower energy [14]. We have estimated the surface area of the side and end facets using a model depicted in Fig. 6. The AuNR can be approximated to consist of a cylinder with hemispherical endcaps. The surface area of the cylinder represents the surface area of the side facets, and the surface area of the hemispherical endcaps represents the surface area of the end facets. Below is the equation used to estimate the surface areas.

$$A = A_{cylinder} + A_{endcaps}$$
(1)
$$A = 2\pi \left( {d/2} \right)l + 4\pi \left( {d/2} \right)^{2}$$
(2)

where Acylinder and Aendcaps are the surface areas of the cylinder and endcaps, respectively. Moreover, l and d are the length of the cylindrical part and the diameter of the endcaps, respectively, which were calculated using TEM images. We have also estimated the total surface area of the NRs used for catalysis. Table 1 shows the respective values of the facets for the AuNRs samples. It is interesting to see that the overall surface area of the Au NRs decreased for R2, compared to R1, yet the catalytic activity of R2 is higher than R1 for all reactions. This can be rationalized by considering the relative surface area of the high-energy side facets for the two samples, which is much higher for R2 compared to R1. Therefore, it is concluded from this result that the availability of the high-energy surface can be a significant factor in the catalytic efficiency of the nanorods.

Fig. 6
figure 6

Nanorod’s model used for the surface areas of calculation of side and end facets

Full size image
Table 1 The surface area of Au NRs used for the catalytic reduction of 4-NP and MO. For MB, the values are twice as large as the ones given below
Full size table

As discussed previously, many researchers have reported the use of NPS for catalytic reduction of various reactions with contradictory results. However, there are several assumptions in those comparative studies that may lead to these contradictions. For example, in some reports, it has been assumed that there is a complete conversion of Au ions into Au NPs and the molar concentration of the NP is assumed to be the same as that of the starting precursor molarity [11, 12]. This is usually not the case as the 100% conversion efficiency of Au ions to Au(0) is not achievable under most of the experimental conditions, and it even changes with a slight change in the reaction conditions e.g. with the variation in the pH of the growth solution, as seen in this study. Therefore, it is important to verify the Au content of each morphology before the comparative study. The current study also suggests that, in addition to the relative surface area, effect of the surface energy should be taken into account.

4 Conclusion

We have shown that the aspect ratio of Au NRs can be controlled with the variation of the pH of the growth solution. The slower growth kinetics favored the deposition of the Au atoms on the end facets of growing NRs and consequently resulted in the fabrication of higher aspect ratio Au NRs. The relative fraction of the high-energy side facets increases with the increase in the aspect ratio, which has a marked influence on the catalytic performance of Au NRs. The catalytic reduction of 4-NP, MB and MO by Au NRs depended sensitively on the aspect ratios of Au NRs. Nanorods with higher aspect ratios were more effective for catalysis in all cases due to the higher surface area of the high-energy side facets.