Abstract. In this article we explore the reduction of tetra- chloroauric acid, HAuCl4, with hydrogen peroxide to form gold nanoparticles in the presence of ...
Nanoscience & Nanotechnology, 7 eds. E. Balabanova, I. Dragieva, Heron Press, Sofia, 2007
SYNTHESIS OF GOLD NANOPARTICLES VIA HYDROGEN PEROXIDE REDUCTION ENHANCED BY SONICATION N. Vaklev, P. Vasileva, C. Dushkin Laboratory of Nanoparticle Science and Technology, Department of General and Inorganic Chemistry, Faculty of Chemistry, University of Sofia, 1 James Bourchier Blvd., Sofia 1164, Bulgaria
peroxide amount at a given concentration are varied to study their effect on the NPs size and properties.
Abstract. In this article we explore the reduction of tetrachloroauric acid, HAuCl4 , with hydrogen peroxide to form gold nanoparticles in the presence of poly(vinyl alcohol) as a capping agent. Ultrasonic irradiation is employed to enhance the reduction and to ensure that the reagents are thoroughly mixed. This method leads to the formation of stable gold colloids. The particles have different sizes and shapes: near-spherical, triangular and hexagonal. The effect of the concentration of the hydrogen peroxide solution and the peroxide amount at a given concentration on the particle size is studied. The absorbance spectrum is simulated with the Mie-Drude model which agrees well with the experimental data in the region of the main surface plasmon resonance (SPR) band. Some assumptions about the reaction mechanism are made.
2.
Tetrachloroauric acid trihydrate (HAuCl4 .3H2 O) (purity 99%) was purchased from Panreac QuZmica S.A. and used without further treatment. Analytical grade 9.3 M (30%wt.) and 21.9 M (60%wt.) hydrogen peroxide were purchased commercially. Poly(vinyl alcohol) (PVA) with molecular weight 2 × 104 – 4 × 104 gmol−1 was used as a capping agent. Stock solutions of 1 mM tetrachloroauric acid, 2%wt. PVA and 2.1 M, 4.6 M hydrogen peroxide in double distilled water were prepared from the commercial chemicals. The used glassware was washed with “clean glass” solution (1%wt. solution of NH4 F.HF) and rinsed with double distilled water. 2.5 mL 1 mM gold precursor and 1 mL aqueous solution of PVA were thoroughly mixed upon sonication for 2 min. Then, the aliquot of H2 O2 stock solutions was added dropwise at gold to peroxide molar ratios between 1:3 and 1:350. The samples were sonicated in an ultrasound bath operating at a frequency of approximately 44 kHz at 60 W output power in order to enhance the reduction process and to ensure thorough mixing. Mechanical stirring was applied in addition to some samples, e.g. magnetic stirrer. With 2.1 M, 4.6 M and 9.3 M hydrogen peroxide solutions the reaction mixture changed from yellow to purplered and at the end of the reaction time it was purple-brown and turbid. The reaction time was 25–45 min. With the 21.9 M H2 O2 solution the reaction mixture became pale purple and the color intensified considerably within 5 min. The overall reaction time in this case was 5-10 min depending on the peroxide amount. The as prepared gold colloids were stable for half a year. UV-Visible absorbance spectra were recorded in a Jenway (model 6400) spectrophotometer. The Au NPs solution was transferred to a 4 mL cuvette and then the spectrum was recorded in the region of 360–1000 nm. The shape and size of the particles were determined from micrographs made by TEM Philips CM-10 at 100 kV.
Keywords: gold nanoparticles, hydrogen peroxide, ultrasonic irradiation, tetrachloroauric acid.
1.
Experimental
Introduction
Nano-sized noble metal particles have been studied extensively, because of their optical properties, high catalytic activities, and electronic properties. Many studies have been conducted in order to prepare size- and shape-controlled gold nanoparticles (Au NPs), e.g. chemical reduction in aqueous medium or in organic solvents with capping agents such as poly(vinyl-pyrrolidone) and poly(vinyl alcohol) [1], UV-reduction [2,3] and sonochemical reduction [4,5]. Although the synthesis of gold nanoparticles started approximately sixty years ago, most chemical methods have the disadvantage to produce gold sols with byproducts due to the used reducing agent. On the contrary, hydrogen peroxide allows the synthesis of water-dispersed gold nanoparticles without such impurities. Moreover, Sarma et al. described gold nanoparticle–polyaniline composite prepared using H2 O2 for both reduction of HAuCl4 and oxidation of aniline in the same aqueous medium [6]. They also reported starch mediated shape-selective synthesis of Au NPs using H2 O2 as a reducing agent in the presence of ultrasonic waves [7]. Zayats et al. reported on a catalytic growth of gold seeds using H2 O2 for the reduction of HAuCl4 and cetyltrimethylammonium chloride as surfactant [8]. In this paper we study the reduction of HAuCl4 with hydrogen peroxide in PVA aqueous solutions enhanced by ultrasound irradiation in order to prepare stable gold colloids. The concentration of the hydrogen peroxide and the
3.
Results
Most of the samples show two surface plasmon resonance (SPR) bands in the absorption spectra. The results are 70
1.0
Normalized Absorbance
Normalized absorbance
Synthesis of Gold Nanoparticles via Hydrogen Peroxide Reduction Enhanced by Sonication
0.8 0.6 0.4
2.1 mol/L 9.3 mol/L 21.9 mol/L
0.2 0.0
400
500
600
700
800
1.0 0.9
0.7 0.6 0.5
900 1000
400
λmax [nm]
540 3.5μL
experimental points trend line 8μL
10.5μL
535
44μL
530
4μL-30μL
525 520
40μL
0
5
10 15 c [mol/L]
500
550
20
25
Figure 2. Position of the main absorption maximum against the used peroxide concentration. The points at one and the same peroxide concentration correspond to different volumes (in μL) of the added reducing agent shown near the points.
90
600
b
85 FWHM [nm]
shown in Figure 1. The main absorption band due to the transverse plasmon resonance is well defined and has high intensity. It is reproducible under the same reaction conditions. The concomitant absorption band with lower intensity is due to deviations from the spherical shape for some particles or aggregates. The second absorption band coming from longitudinal plasmon resonance varies in shape, position and intensity. For instance, the spectra of three samples obtained with 21.9 M. hydrogen peroxide under identical conditions differ significantly in the second band; thus, it is not reproducible. The main absorption band and the concomitant one become blue-shifted with an increase in the peroxide concentration. Figure 2 shows the change in the position of main absorption maximum (λmax ) versus the hydrogen peroxide concentration and different gold to peroxide molar ratios at given peroxide concentration. It is clear that the main peak is blue–shifted for both an increase in the peroxide concentration and an increase in peroxide amount at a given concentration. The main absorption maximum (λmax ) varies
4μL
450
λ [nm]
Figure 1. UV-Visible absorption spectra of the Au NPs in aqueous PVA solution generated by H2 O2 reduction at different peroxide concentrations upon sonication. The solid lines show three spectra of three samples obtained with the same amount of 21.9 M peroxide.
545
a
0.35 mmol 0.66 mmol 0.88 mmol
0.8
λ [nm]
550
71
80 75 70 65 60 0.0
0.2
0.4 0.6 0.8 n(H2O2) [mmol]
1.0
Figure 3. (a) Main plasmon absorption maxima of some samples synthesized upon sonication at different amount of 21.9 M hydrogen peroxide solution. (b) Full width at a half maximum (FWHM) against the increase of the reducing agent amount in samples synthesized upon sonication with 21.9 M hydrogen peroxide solution.
in the range 542–522 nm which is attributed to a change in the particle size. Its position depends mainly on the peroxide concentration, since at 1:33 gold to peroxide molar ratio 9.3 M peroxide gives main peak at 538±1 nm whereas 21.9 M peroxide results in 525±1 nm. For the samples synthesized with 4–30 μL 21.9 M H2 O2 the main absorption maximum remain constant at 525±1 nm, but with 40 μL it blue-shifts to 522±1 nm. Furthermore, for these samples, the full width at a half maximum (FWHM) increases with increase in the amount of reducing agent, as could be observed from the broadening of the main maxima (see Figure 3). TEM image (Figure 4a) of the gold particles prepared with 21.9 M H2 O2 shows nanoparticles with a wide size distribution and various shapes. Most of them are spherical, but there are also pyramids, ellipsoids and few rods. Figure 4b shows a histogram describing the size distribution of the spherical or near-spherical particles. The radii for majority of particles are in the range 3–11 nm (Figure 4b); it is noteworthy that there are few bigger ones with radii 15–20 nm. Hence, the concomitant peak is most probably due to the absorbance of coexisting particles with triangular and hexagonal forms (e.g. longitudinal plasmon
72
N. Vaklev, P. Vasileva, C. Dushkin
Normalized Absorbance
experimental data Mie-Drude model
0.8 0.6 0.4 0.2
b
400
500
600 700 λ [nm]
800
900 1000
Figure 5. UV-Visible absorption spectrum of the sample in Figure 4. The spectrum is fitted with the Mie-Drude model for R = 6.5 nm and A = 0.63.
where ωp and ωs0 are the plasma frequency (2.175 × 1015 Hz) and the scattering rate (6.5 × 1012 Hz) for bulk gold, respectively [13]. Further, ωs is a size dependent correction for the bulk scattering rate:
1
3
5
7
9 11 13 15 17 19 21 R [nm]
Figure 4. (a) TEM micrograph of the Au NPs synthesized with 21.9 M solution of hydrogen peroxide upon sonication. (b) Particle size distribution of the sample in (a); only the particles with spherical or near-spherical shape are taken into account; the accuracy of the single measurement is 0.6 nm; the average radius is 7.0 nm; sample standard deviation σ = 4 nm.
resonance). So this peak can be used as a rough estimate for the nanoparticles morphology; the more red-shifted and stronger the peak is, the predominant the pyramids are [7]. 4.
1.0
0.0
16 14 12 10 8 6 4 2 0
Discussion
It is known that the SPR of spherical gold nanoparticles depends on their size, the dielectric constant of the medium, the stabilizing agent and the particle-particle interactions [9]. Generally when the particle size decreases, the SPR band is blue-shifted [10,11]; this holds to a great extent for non-interacting particles stabilized with weak-binding agents. These conditions ensure that the dominating effects are the particle size and the medium properties. The Mie theory predicts the absorption cross section in the case of small spherical particles [12]: σabs (ω) = 12πR3 ε3/2 m
ε2 ω c (ε1 + 2εm )2 + ε22
(1)
Here R is the average particles radius, εm is the dielectric constant of the medium (1.775 for water), ε1 and ε2 are the real and imaginary part of the dielectric function ε(ω), respectively. ω and c correspond to the wave frequency and velocity of light. A dielectric function based on the Drude model is used to simulate ε [12]: ωp2 ωp2 − (2) ε(ω) = εbulk (ω)+ ω(ω + iωs0 ) ω[ω + i(ωs0 + ωs )]
ωs = AvF /R,
(3)
where A is a dimensionless factor and vF is the Fermi velocity for the bulk metal (1.4 × 106 ms−1 ). In this model the FWHM (the scattering rate) is proportional to the function Γ given by [12]: Γ = ωs0 + AvF /R.
(4)
The model agrees well with the experimental data in the region of the main SPR band (Figure 5), whereas the agreement for the near UV and IR regions is poor. Since the particles are small the interband transitions in the gold crystal exhibit size dependent properties which cannot be taken into account in this model [12]. Thus there are bigger deviations in the UV region where the interband transitions are observed. In the near IR region the sample shows higher absorbance compared to the Mie theory due to aggregates and bigger non-spherical particles that absorb in that region. Nevertheless, the main SPR band is well described by the model and gives theoretical radius of 6.5 nm
3.0
only sonication sonication & stirring
2.5 Absorbance
Frequency [%]
a
2.0 1.5 1.0 0.5 0.0
400
500
600
700
λ [nm]
800
900 1000
Figure 6. UV-Visible absorption spectra of two samples synthesized with 21.9 M hydrogen peroxide solution applying different methods to homogenize the reaction mixture.
Synthesis of Gold Nanoparticles via Hydrogen Peroxide Reduction Enhanced by Sonication which agrees well with the value obtained from TEM of 7.0 nm for the average radius. The model showed good credibility and was used to analyze the increase in the FWHM at 21.9 M hydrogen peroxide (Figure 3b). Here again the general tendency was preserved since the analysis based on Eq. (4) [11] showed that the particle size decreases with an increase in the amount of hydrogen peroxide. To explain these trends we assume the following reaction scheme based on previous works [5,8,14]: • • • (i) H2 O2 /H2 O −sonication −−−−→ H + HO + HO2 • • • sonication PVA −−−−−→ H + HO + R (ii) • • • • PVA + H/ OH/HO2 → R + H2 /H2 O/H2 O2 (iii) •
0 − + 3R• + AuCl− 4 → 3R + Au + 4Cl + 3H 0
nAu → Aun (small seeds) Aun +
2mAuCl− 4
(iv) (v)
+ 3mH2 O2
− + (vi) −−−−→ Au(n+2m) +3mO2 +8mCl +6mH Ultrasonic irradiation creates short living bubbles in the water that collapse at extremely high temperature. This phenomenon called cavitation produces strong acoustic waves in the medium and this effect was employed to mix the reagents during the reaction. Further, the created high temperature decomposes the water, PVA and most probably the hydrogen peroxide into radicals, (i) and (ii). The short living inorganic radicals further react with PVA to produce more stable organic radicals, (iii), which are responsible for the reduction of the gold precursor later on, (iv) [5]. Thus, small seeds are formed, (v). Moreover, it was proven that the growth of these seeds in the presence of hydrogen peroxide is a self-catalytic process [8], i.e. the seeds’ surface catalyses the reduction of the precursor via hydrogen peroxide, (vi). So, the final size of the nanoparticles should depend on the amount of gold precursor surrounding them and on the initial size of the seeds. Then, more peroxide means that more seeds would form in the initial stage. Since the precursor amount is held constant for all experiments, its average amount available per seed decreases. This makes understandable why the increased amount of the reducing agent leads to a decrease in the particles size. The wide size distribution on Figure 4b can also be explained – the initial seeds’ size could not be uniform which eventually would lead to wide size distribution. The following experiments show where some of the seeds are formed. When 20 μL 21.9 M peroxide solution were poured on one place of the reaction mixture, clearly visible purple colour occurred immediately at that place due to gold nanoparticles formation; than, the reaction continued. In that case the result was a turbid brown solution. A second sample, prepared under the same conditions except that the peroxide was added under stirring, had already an intense purple colour. As a result the SPR-band of the first sample is red-shifted and with irregular shape (Figure 6). This means that bigger particles had formed in the first sample compared to the second one. Thus, one can speculate that gold seeds were formed where the peroxide was poured, because of the produced colour change there and then, due to applied cavitation, these big seeds miAun
73
grated into the volume and grew there along with the other seeds. This unexpected result comes to show that the way the reducing agent is added influences the result, which is important from practical point of view. Actually, it was found that volumes bigger than 16 μL for 21.9 M peroxide should not be added all at once; otherwise such effect occurs. 5.
Conclusions
In this research we have successfully applied the hydrogen peroxide reduction of HAuCl4 in PVA-aqueous solutions enhanced by ultrasound irradiation to prepare stable gold colloids. The gold nanoparticles of different shapes and sizes are obtained by varying of reducing agent’s concentration. Blue-shifted SPR band is observed when the concentration of the hydrogen peroxide solution and the peroxide amount at a given concentration are increased. The absorbance spectrum is simulated with the Mie-Drude model which agrees well with the experimental data in the region of the main SPR band. Some assumptions about the reaction mechanism are made which are in accord with the experimental facts. Acknowledgements The authors are greatly thankful to Mr. Philip Oshel from the Department of Biology, Central Michigan University, USA, for the TEM examination of the gold nanoparticles. The financial support of the Bulgarian Ministry of Education and Science by Project VUH-09/05 and COST D 43 Action of EC is acknowledged. References [1] T. Teranishi, M. Hosoe, T. Tanaka, M. Miyake, J. Phys. Chem. B 103 (1999) 3818. [2] Y. Zhou, S.H. Yu, C.Y. Wang, X.G. Li, Y.R. Zhu, Z.Y. Chen, Adv. Mater. 11 (1999) 850. [3] Y. Zhou, C.Y. Wang, Y.R. Zhu, Z.Y. Chen, Chem. Mater. 11 (1999) 2310. [4] K. Okitsu, A. Yue, S. Tanabe, H. Matsumoto, Y. Yobiko, Y. Yoo, Bull. Chem. Soc. Jpn. 75 (2002) 2289. [5] W. Chen, W. Cai, L. Zhang, G. Wang, L. Zhang, J. Colloid Interface Sci. 238 (2001) 291. [6] T.K. Sarma, D. Chowdhury, A. Paul, A. Chattopadhyay, Chem. Commun. (2002) 1048. [7] T.K. Sarma, A. Chattopadhyay, Langmuir 20 (2004) 3520. [8] M. Zayats, R. Baron, I. Popov, I. Willner, Nano Letters 5 (2005) 21. [9] W. Rechberger, A. Hohenau, A. Leitner, J.R. Krenn, B. Lamprecht, F.R. Aussenegg, Optics Commun. 220 (2003) 137. [10] C. Soennichsen, T. Franzl, T. Wilk, G. von Plessen, J. Feldmann, New J. Phys. 4 (2002) 93.1. [11] U. Kreibig, L. Genzel, Surface Sci. 156 (1985) 678. [12] H. Hoevel, S. Fritz, A. Hilger, U. Kreibig, M. Vollmer, Phys. Rev. B 48 (1993) 18178. [13] I. El-Kady, M.M. Sigalas, R. Biswas, K.M. Ho, C.M. Soukoulis, Phys. Rev. B 62 (2000) 15299. [14] T. Sau1, A. Pal, N.R. Jana, Z. Wang, T. Pal, J. Nanoparticle Res. 3 (2001) 257.
