Seed-Mediated Synthesis of Gold Nanorods at Low Concentrations of CTAB.

Although gold nanorods capped with hexadecyltrimethylammonium bromide (CTAB) have been prepared through the seed-mediated method for their use in diagnostics and therapeutics, the toxicity of AuNRs@CTAB limits their practical applications in the biomedical field. In this work, the synthesis and tuning of gold nanorods at very low concentrations of CTAB (as low as 0.008 M) was successfully achieved by using the seed-mediated method. Furthermore, we managed to optimize the growth conditions by changing the amount of seeds, AgNO3, and/or HCl. At low CTAB concentrations, gold nanorods with tunable size and aspect ratio, high monodispersity, and high purity were obtained and studied by UV-vis spectroscopy, transmission electron microscopy, and Mie-Gans theoretical calculations. This work revealed a method of nanoparticle growth that may be extended to synthesize other nanomaterials such as Ag, Cu, Pd, and Pt at such low CTAB concentrations.

Introduction

Gold nanorods (AuNRs) have attracted many researchers in the recent two decades because of their special characteristics and wide applications, including chemical and biological sensing,[1−3] surface-enhanced Raman scattering (SERS),[4−6] anticancer therapy,[7−9] and imaging.[10−12] Optically, the interaction of free electrons with light at different wavelengths leads to the localized surface plasmon resonance (LSPR) effect.[13,14] It can be characterized by the scattering and absorption peaks corresponding to various wavelength ranges in the spectrum.[15] Furthermore, experimental results and theoretical calculations show that the LSPRs are directly related to the aspect ratio (length/diameter) of AuNRs (LSPR along the longitudinal direction contributes the most), and one can significantly control the aspect ratio of AuNRs by changing the amount of the reactant through the process of synthesis, which creates a wide spectral range from visible (∼600 nm) to the near-infrared region.[16−19]

Several synthesis methods of AuNRs have been developed, including chemical reduction,[20] electrochemical reduction,[21] and photochemical reduction.[22] A general and most used method is the seed-mediated method: Au3+ is fast reduced by sodium borohydride and turned into gold nanoparticles in the form of seeds in an aqueous hexadecyltrimethylammonium bromide (CTAB) solution. After the mild reduction of Au3+ in the growth solution by ascorbic acid, Au0 directionally grows onto the seeds. This method was first proposed by El-Sayed and Nikoobakht in 2003[19] and further developed by Murray,[16] Murphy,[23] Wang,[18] and Liz-Marzán[24] groups. Through the process of AuNR synthesis, the concentration of CTAB has a great influence on the quality of the obtained products, including monodispersity, aspect ratio, and purity.[13] For this reason, suitable concentrations of CTAB should be considered to obtain a high quality and tunability of the AuNRs. Traditionally, a high concentration of CTAB (0.1 M) has been used in the seed-mediated growth processes.[19] In the work of Murray et al.,[16] the concentration of CTAB was lowered to 0.037 M by using binary surfactant mixtures, that is, the sodium oleate (NaOL) was introduced to function as another surfactant together with CTAB. This method greatly improved the dimensional tunability, monodispersity, and purity. Besides AuNRs, CTAB has been widely used in the synthesis of Cu,[25] Pd,[26] Pt,[27] and Ag[28] nanoparticles. By changing the types and concentration of reducing agents,[29] capping agents, and stabilizers,[30] different sizes and shapes of nanocrystals could be obtained. CTAB played the role of a capping and structure-directing agent and enhanced the anisotropic growth.[31,32] However, what cannot be ignored is the toxicity of CTAB combined with gold nanorods, which can cause serious damage to the biological cells and tissues. Ray et al.[33] carried out a series of experiments indicating that only a few cells (∼10%) could survive in the AuNRs@CTAB condition. Moreover, they mentioned that citrate-capped Au nanoparticles of different sizes are not inherently toxic to human skin cells, but AuNRs are highly toxic due to the presence of CTAB as a coating material. Guadagnini et al.[34] also demonstrated that some other nanoparticles like Fe3O4 coated with sodium oleate led to an unexpected increase in cytotoxicity. When the concentration of Fe3O4@NaOL was higher than 7 μg/cm3, the metabolic activities of cells dropped precipitously. For reducing/removing the surfactants, e.g., CTAB and NaOL on the surface of AuNRs, some surface modification methods were carried out including AuNRs@mesoporous SiO2[35] and AuNRs@poly(ethylene glycol) (PEG).[36] The advantage of using the mesoporous SiO2 layer is that it enhances the stability of the AuNRs in solution.

Here, we report the seed-mediated growth of AuNRs at very low concentrations of CTAB (as low as 0.008 M, ∼20% of that in the previous work[16]) by using a binary surfactant (CTAB + NaOL). Simultaneously, the concentration of NaOL was also dropped to 5 mM (∼60% of that in the previous work[16]). Furthermore, we managed to optimize the growth conditions at low CTAB concentrations by changing the volumes of the seed solution, AgNO3, and HCl. We characterized the obtained AuNRs by optical extinction spectra, transmission electron microscopy (TEM), and Mie–Gans calculations.

Results and Discussion

AuNRs were synthesized by a modified seed-mediated method with binary surfactant mixtures.[16]Figure shows a schematic process of the AuNR synthesis at 0.008 M CTAB. Firstly, the Au seeds were prepared by the reduction of HAuCl4 using a strong reducing agent (NaBH4). After aging for 30 min, the Au seeds were added to the growth solution containing the mixture of CTAB, NaOL, HAuCl4, AgNO3, and HCl. Finally, a mild reducing agent (ascorbic acid) was added and the solution was left undisturbed for nanorod growth (for more experimental details, see the Experimental Section and Supporting Information (SI)). The obtained AuNRs were characterized by optical extinction spectroscopy and TEM.

Figure 1

Schematic representation of the AuNR synthesis process at low CTAB concentrations.

Figure a–e shows the TEM images of the AuNRs synthesized at low concentrations of CTAB (0.008 and 0.010 M). The amounts of NaOL, AgNO3, HCl, and seeds are fully detailed in Table . By changing the growth condition, the aspect ratio of the obtained AuNRs could be tuned from 1.9 to 4.1. We determined the length, diameter, and aspect ratio of each sample by measuring around 200 (The maximum from Figure is 300, determined by the AuNRs appearing in the TEM images.) nanorods from the corresponding TEM images. The average length, diameter, and aspect ratio are shown in Table . It can be observed from Figure that although the CTAB concentration is very low, high-purity AuNRs with monodispersed shape were still obtained in this growth condition. Figure S1 shows a large-area TEM image of the same samples as in Figure e. Due to the high monodispersity, some of the nanorods self-assembled side-by-side on the copper grid. The inset in Figure S1 shows an optical image of the sample in a 40 mL vial (the total volume of the solution is ∼25 mL).

Figure 2

Morphology characterization of the AuNRs synthesized at low CTAB concentrations. (a–e) TEM images of the AuNRs synthesized at low concentrations of CTAB ((a) 0.008 M. (b–e) 0.01 M). The insets on the top right of (a)–(e) are the average aspect ratios of the AuNRs shown in the TEM images. Here, we mostly varied the AgNO3 concentration and HCl addition to obtain different sizes of AuNRs, which are fully detailed in Table . (f) A high-resolution TEM (HRTEM) image of the AuNR. The sample is the same as in (e). The nanorod has the crystallographic ⟨001⟩ direction along the long axis of the rod and is observed to lie on the grid in a [110] projection, which is indicated by the white arrows in (f).

Table 1

Average and Standard Deviations of the AuNR Size Parameters (Length, Diameter, and Aspect Ratio) Measured from the TEM Images and LSPR of the AuNR Samples Shown in Figure

CTAB (M)	NaOL (M)	HCl (mL)	AgNO3 (mM)	seed (mL)	length (nm)	diameter (nm)	aspect ratio	LSPR (nm)	figure number
0.008	0.005	0.3	0.144	0.04	64.1 ± 7.7	34.9 ± 4.6	1.9 ± 0.2	615	Figure 2a
0.01	0.005	0.3	0.032	0.04	116.7 ± 11.6	56.1 ± 5.6	2.1 ± 0.3	684	Figure 2b
0.01	0.005	0.2	0.144	0.04	112.6 ± 9.4	36.8 ± 3.1	3.1 ± 0.4	789	Figure 2c
0.01	0.005	0.3	0.240	0.04	87.7 ± 10.1	23.7 ± 2.2	3.7 ± 0.5	846	Figure 2d
0.01	0.005	0.3	0.048	0.04	105.9 ± 9.0	26.2 ± 2.1	4.1 ± 0.5	869	Figure 2e

To characterize the crystalline structure of the AuNRs, we performed high-resolution transmission electron microscopy (HRTEM) characterization for the sample in Figure e. Figure f shows a HRTEM image of the AuNR. The AuNR is a single crystal of face-centered cubic (fcc) structure, with one of the main axes oriented along the length of the rod. It is observed to lie on the copper grid in a [110] orientation, which is in agreement with previous reports.[37−39]

Figure shows the extinction spectra of the AuNRs in Figure . The LSPR band wavelengths are at 615, 684, 789, 846, and 869 nm. There is still a wide range of tunability (from 615 to 869 nm) at the very low CTAB concentrations (0.008–0.010 M). The width of the LSPR peaks, which can be characterized by full width at half-maximum (fwhm), is quite sharp for each AuNR sample (e.g., fwhm = 52 nm, or 0.171 eV for the LSPR at 615 nm (black curve); fwhm = 99 nm, or 0.263 eV for the LSPR at 684 nm (red curve); fwhm = 126 nm, or 0.251 eV for the LSPR at 789 nm (blue curve); fwhm = 189 nm, or 0.328 eV for the LSPR at 846 nm (magenta curve); fwhm = 146 nm, or 0.240 eV for the LSPR at 869 nm (green curve)). The narrow fwhm further confirms that the obtained AuNRs have a high monodispersity.

Figure 3

Normalized extinction spectra of the AuNRs shown in Figure a–e. The corresponding LSPR peaks are characterized in Table . The spectra are normalized to the extinction maximum.

To explore the size tunability of the AuNRs, we performed a series of experiments using different volumes of the seed solution at a fixed CTAB concentration (0.01 M). The other reactants were also kept unchanged and are fully detailed in Table . Figure a–e shows TEM images of the AuNRs synthesized with different amounts of seed particles. When the amount of seed particles was reduced, both the length and diameter of the AuNRs increased. When fewer seeds are available in the growth solution, the ratio of gold precursors/seeds increases. As every single seed binding site receives more number of gold precursors, it leads to the appearance of thicker and bigger AuNRs. A former research carried out by Murray et al.[16] also showed that less seed amounts in the final growth solution tended to produce thicker AuNRs using binary surfactants. Figure f shows the corresponding extinction spectra of the AuNRs shown in Figure a–e. Adding more seeds (from 0.005 to 0.16 mL) leads to an increase of the aspect ratio from 2.1 to 4.9 (Figure a–e). The increase of the aspect ratio is also reflected in a red shift of the LSPR peak from 709 to 896 nm (Figure f).

Table 2

Average and Standard Deviations of the AuNR Size Parameters (Length, Diameter, and Aspect Ratio) Measured from the TEM Images and LSPR of the AuNR Samples Shown in Figure

CTAB (M)	NaOL (M)	HCl (mL)	AgNO3 (mM)	seed (mL)	length (nm)	diameter (nm)	aspect ratio	LSPR (nm)	figure number
0.01	0.005	0.3	0.144	0.005	130.1 ± 13.0	60.8 ± 4.0	2.2 ± 0.3	709	Figure 4a
0.01	0.005	0.3	0.144	0.01	108.3 ± 11.4	42.3 ± 3.0	2.6 ± 0.4	736	Figure 4b
0.01	0.005	0.3	0.144	0.02	102.9 ± 9.7	35.2 ± 4.1	3.0 ± 0.5	765	Figure 4c
0.01	0.005	0.3	0.144	0.04	71.7 ± 9.2	20.9 ± 2.3	3.5 ± 0.7	793	Figure 4d
0.01	0.005	0.3	0.144	0.16	60.4 ± 11.3	12.9 ± 2.5	4.8 ± 1.0	896	Figure 4e

Figure 4

AuNRs synthesized by using different amounts of seed solution. (a–e) TEM images of AuNRs synthesized with different amounts of seeds. The insets on the top right of (a)–(e) show the length and diameter of the AuNRs. With the increase in amount of the seed particles, the size (both length and diameter) of the AuNRs reduced while the aspect ratio increased. (f) Normalized extinction spectra of the AuNRs shown in (a)–(e). The corresponding LSPR peaks are characterized in Table . The spectra are normalized to the extinction maximum.

To investigate the plasmonic properties theoretically, we used the Mie–Gans theory[40] to calculate the extinction cross section of a single nanorod for the samples shown in Figures and 4. We assumed that the AuNR has a spheroidal shape. Then, the extinction cross section can be obtained by an analytical expression (see eqs –6 in the Experimental Section). In the calculation, the size parameters were used as shown in Tables and 2. More calculation details can be found in the Experimental Section. Figure S2 shows the extinction spectra of AuNRs by Mie–Gans calculation. The LSPR red-shifted as the aspect ratio increased. The characteristics of the calculated extinction spectra are in good agreement with the experimental spectra. However, the widths of the LSPR in the calculated spectra are sharper and the LSPR position is slightly different compared to the experimental data, which can be explained by the polydispersity of the experimental obtained nanorods, which was not taken into account in the single-particle calculations.

To fully control the synthesis of the AuNRs, we further investigated the influence of AgNO3 and HCl on AuNR synthesis. First, we fixed the other parameters and only changed the concentration of AgNO3 (from 0.016 to 0.48 mM) in the synthesis. All of the samples were synthesized with 0.01 M CTAB concentration. The experimental parameters are listed in Table S1, and the extinction spectra are shown in Figure S3. There was an obvious increasing trend of the LSPR peaks from 653 to 889 nm at low concentrations of AgNO3 (0.016–0.048 mM). However, at higher concentrations of AgNO3 (>0.048 mM), the LSPR peak varied between 762 and 870 nm, showing an irregular change.

To evaluate the quality of AuNRs, Q and A1/A2 factors were defined. Here, where ω0 = 2πc/λ0, ω1 = 2πc/λ1, ω2 = 2πc/λ2, c is the velocity of light in vacuum, λ0 is the LSPR peak, and λ1 and λ2 are the wavelengths at half maximum of the LSPR peak. The parameter Q is widely used in electronics to describe the filter’s filtering/passing capability for specific signal frequencies. The larger the parameter Q, the stronger the strobe ability of the circuit for a specific frequency of a signal and the greater the obstruction ability for other frequencies of the signal, which is reflected in the narrow bandwidth of the waveform. By transformation, the x-axis of the spectrum can be transformed into the frequency (ω = 2πc/λ) and Q can be used to describe the bandwidth of the LSPR peaks instead of only the full width at half maximum. It is more precise and objective. Therefore, Q is a parameter that can reflect the monodispersity of AuNRs. In A1/A2, A1 is the absorbance of the longitudinal LSPR peak and A2 is the absorbance of the transverse LSPR/Au spheres peak. Khlebstov et al.[17] used A1/A2 to measure the purity of AuNRs. If a large quantity of Au spheres is obtained, the A1/A2 will be much lower than that for the normal AuNR samples. Although it is difficult to observe the trend of Q vs the AgNO3 concentration, all the samples have high Q values (>4), as shown in Figure a. For the A1/A2 factor, it can be found that from 0.05 to 0.35 mM the A1/A2 values are high, indicating that the number of gold nanospheres in the synthesis process is relatively lower, which suggests a higher purity of the AuNRs. For the synthesis of AuNRs, the presence of Ag+ is vital. It is believed to bond selectively to the {110} facets of Au nanocrystals and slow down the growth rate. Moreover, the tuning of AuNRs by adding different amounts of seed particles could not be achieved in the absence of Ag+, leading to the appearance of Au particles with many different shapes.[41,42] The above series of experiments aimed at exploring a AgNO3 concentration that is more suitable for the synthesis of AuNRs at a low concentration of CTAB (0.01 M). In our study, 0.05–0.35 mM AgNO3 concentration was found to achieve a higher monodispersity and purity in the process of AuNR synthesis.

Figure 5

Quality analysis of the AuNRs synthesized at different concentrations of AgNO3. (a) The parameter “Q” of gold NRs that were synthesized at AgNO3 concentrations ranging from 0.016 to 0.48 mM. (b) A1/A2 in the same conditions as in (a). The amounts of other reactants are shown in Table S3. The LSPR peaks from 0.016 to 0.48 mM AgNO3 concentration are listed in Table S1.

In addition to the AgNO3, we also studied the effect of HCl in the synthesis. The experimental parameters are listed in Table S2, and the extinction spectra of the samples are shown in Figure S4. The LSPR peaks of 0.15, 0.2, 0.25, 0.3, 0.4, and 0.5 mL HCl addition amounts were 826, 782, 798, 864, 809, and 816 nm, respectively. Similar to the results of AgNO3 mentioned above, there is no obvious trend of AuNR LSPR peaks with a single variation of HCl at a low concentration of CTAB (Figure S4a). When we further reduced the amount of HCl (0.1 and 0.15 mL of HCl in a 25 mL volume of the final solution), pure AuNRs could not be obtained (Figure S4b). Figure shows the monodispersity and purity of AuNRs that were synthesized in different amounts of HCl through the parameters “Q” and A1/A2. It can be observed in Figure that there is a good fit of the amounts of HCl (from 0.2 to 0.3 mL) for obtaining high-quality (both monodispersity and purity) AuNRs in this condition.

Figure 6

Quality analysis of the AuNRs synthesized with different amounts of HCl. (a) The parameter “Q” of gold NRs that were synthesized in 0.15–0.5 mL of HCl. (b) A1/A2 in the same conditions as in (a). The amounts of other reactants are listed in Table S3. The LSPR peaks from 0.15 to 0.5 mL HCl volume are listed in Table S2.

In the synthesis processes of AuNRs, HCl is introduced to adjust the pH. Series of studies have been done on the effect of pH in the process of AuNR synthesis, indicating that the good-fitting amount of HCl is essential for a good aspect ratio, monodispersity, and even shape of AuNRs. Researchers have proposed a hypothesis that the {110} facet of AuNRs would be packed densely by CTAB molecular bilayers and coated with AgBr, and the electrostatic attractive force between the bilayers and the AuNR surface is highly sensitive to the pH. With the increase of pH, a lesser number of CTAB–AgBr bilayers are found on the AuNR surface, and most AuNRs grow into different shapes.[43] However, whether this fits to the actual situation remains to be proved through further experimental work.

To test if this method can be scaled up, we carried out a 500 mL growth solution synthesis. The two samples shown in Figure S5 have exactly the same parameters as in Figure e. Both samples had a brownish color, which is the typical color for AuNRs. Furthermore, we carried out experiments at a much lower CTAB concentration and found that the 0.008 M concentration is very close to its limit. When we dropped the CTAB concentration to 0.006 and 0.005 M, the extinction spectra and optical images (Figure S6) confirmed that it failed to obtain high-quality AuNRs. A former work from Xu et al.[44] has proved that the key ingredient in directing the anisotropic growth of AuNRs is the complex CTA–Br–Ag+. Thus, when the CTAB concentration is lowered to below 0.008 M, the concentration of the produced CTA–Br–Ag+ is too low to ensure stable growth in this condition.

It should be noted that both similarities and differences exist when the results for the samples synthesized at a low CTAB concentration are compared with former works. One of the most notable differences is that the samples have become more pH sensitive (shown in Figure ) and changes to the added amount of HCl affect the monodispersity and purity of the obtained AuNRs. After a series of experiments, we figured out that it is necessary to reduce the concentration of NaOL to 5 mM as well as vary the concentration of AgNO3 to maintain a good monodispersity and tunability of the AuNRs (the NaOL concentrations in the previous works are 8 and 10 mM[16]). The differences mentioned above may due to the presence of CTAB, which preferentially binds to the middle of the AuNRs and renders a stabilizing power to the {100} facet that is proportional to the CTAB concentration.[13] When the CTAB concentration goes down, the stabilizing power is weakened and the growth condition becomes more sensitive to the synthesis parameters.

Conclusions

We have explored the seed-mediated synthesis and tuning of AuNRs at very low concentrations of CTAB by using CTAB + NaOL as the binary surfactant. The AuNRs can be obtained at as low as 0.008 M CTAB concentration, which is much lower than those in previous reports, e.g., 0.1 M from the El-Sayed group[19] and 0.037 M from the Murray group.[16] At 0.01 M CTAB concentration, we optimized the growth condition and successfully obtained AuNRs with size tunability, high monodispersity, and purity. Lesser seed amounts produced thicker nanorods. Moreover, AuNRs with higher quality were obtained at ∼0.05–0.35 mM AgNO3 concentrations and 0.2–0.3 mL volume of HCl (in 25 mL growth solution). Compared to other methods, we directly reduced the CTAB concentration on the AuNR surface in the synthesis process while maintaining a high monodispersity and purity of the final products, which made it a direct template for practical applications. This method may improve the efficiency (e.g., simplify the wash cycles), lower the toxicity of the products, and can be extended to the synthesis of other nanomaterials such as Ag, Cu, Pd, and Pt at such low CTAB concentration.

Experimental Section

Materials

All chemicals were obtained from commercial suppliers and used without further purification. Hexadecyltrimethylammonium bromide (CTAB, >99.0%), chloroauric acid (HAuCl4), and l-ascorbic acid (>99.99% metals basis) were purchased from Shanghai Macklin Biochemical Technology Co., Ltd. Silver nitrate (AgNO3, >99.8%), sodium oleate (NaOL, >99.88%), and sodium borohydride (NaBH4, >98%) were purchased from Shanghai Aladdin Biochemical Technology Co., Ltd.

Synthesis of Gold Nanorods

Seed solution: 0.25 mL of 10 mM HAuCl4 was added to 10 mL of 0.1 M CTAB solution. Then, 0.6 mL of 10 mM NaBH4 (freshly prepared with cold water) was added under vigorous stirring. The color of the solution changed from yellow to brownish yellow. The seed solution was aged at 30 °C for 30 min before use.

The growth solution was prepared by adding 2.5 mL of 0.1 M CTAB in a 40 mL scintillation vial. Then, 0.037 g of NaOL was dissolved in 21.25 mL of warm water (45–50 °C) in the same scintillation vial. When the solution cooled down to 30 °C, 0.9 mL of 4 mM AgNO3 solution was added. The mixture was kept undisturbed at 30 °C for 15 min, after which 0.25 mL of 10 mM HAuCl4 was added. The solution became colorless after 90 min of stirring. To adjust the pH, 0.3 mL of HCl (37 wt % in water) was added. 0.075 mL of 64 mM ascorbic acid and 0.04 mL of seed solution were added after 15 min. Finally, the resultant solution was left undisturbed at 30 °C for 12 h to grow. The final CTAB concentration was 0.01 M. The experimental parameters are listed in Tables and 2, S1, and S2.

Characterization

Optical extinction spectra were recorded with a UV-1900i Spectrophotometer (Shimadzu, Japan) with a 10-mm optical path. Transmission electron microscopy (TEM) images were obtained with a HT-7700 microscope (Hitachi, Japan) operating at 100 kV. High-resolution TEM (HRTEM) images were obtained using a Tecnai 30F TEM (FEI) with a 300 kV acceleration voltage. The particle sizes of the AuNRs were measured from the TEM images, whereby >100 particles were measured for each sample.

To prepare the TEM samples, 1.5 mL of AuNR solution was centrifuged at 6000 rpm for 5 min. Then, the supernatant was removed and 1.5 mL of water was added (the concentration of CTAB was always kept higher than its CMC, ∼1 mM) and ultrasonicated for 10 min. The processes were repeated for three times. Finally, 5 μL of the solution in a centrifuge tube was drop-cast onto a carbon-coated 300 square-mesh copper grid and dried slowly in air.

Calculation of the Extinction Cross Section of the AuNRs

The extinction cross sections were calculated using the Mie–Gans theory.[40] Assuming AuNR has a spheroidal shape, the absorption (Cabs), scattering (Csca), and extinction cross sections (Cext) were obtained using the following analytical expressionswhere α is the polarizability corresponding to the long-axis excitation and is the wave number. According to the Classius–Mossoti relation,[45] the polarizability α can be expressed aswhere V is the AuNR volume, ε0 is the vacuum permittivity, εAu is the complex dielectric function of gold, and εm is the medium dielectric constant. L is the depolarization factor corresponding to the longitudinal direction and can be expressed as a function of the AuNR aspect ratio ARIn the calculations, the medium dielectric constant of water εm = 1.78 was used and εAu was taken from Ciesielski et al.[46] The length and diameter of the AuNRs were obtained from the measurements of the TEM images (shown in Tables and 2). The calculated extinction spectra are shown in Figure S2.