A halogen-free synthesis of gold nanoparticles using gold(III) oxide
- 1.5k Downloads
Gold nanoparticles are one of the most used nanomaterials. They are usually synthesized by the reduction of gold(III) chloride. However, the presence of halide ions in the reaction mixture is not always welcome. In some cases, these ions have detrimental influence on the morphology and structure of resulting nanoparticles. Here, we present a simple and halogen-free procedure to prepare gold nanoparticles by reduction of gold(III) oxide in neat oleylamine. The method provides the particles with an average size below 10 nm and dispersity of tens of percent. The process of nanoparticle formation was monitored using UV–Vis spectroscopy. The structure and chemical composition of the nanoparticles was determined by SEM, XPS and EDX. We also proposed the mechanism of reduction of gold(III) oxide based on MS, IR and NMR data. Importantly, the synthetic protocol is general and applicable for the preparation of other coinage metal nanoparticles from the corresponding metal oxides. For instance, we demonstrated that the absence of halogen enables efficient alloying of metals when preparing gold–silver bimetallic nanoparticles.
KeywordsGold nanoparticles Gold(III) oxide Oleylamine Bimetallic nanoparticles
Gold nanoparticles (AuNPs)—due to chemical robustness, easy preparation and functionalization—are often used as a model system for studying nanoscale phenomena (Daniel and Astruc 2004; Hakkinen 2012; Kalsin et al. 2006; Nealon et al. 2012; Sashuk 2012; Sashuk et al. 2012, 2013; Teulle et al. 2015; Wells et al. 2015; Zhao et al. 2016). Moreover, the pristine chemical and physical properties of nanoscopic gold provide additional appeal for applications such as catalysis or sensing (Giljohann et al. 2010; Jans and Huo 2012; Kale et al. 2014; Mikami et al. 2013; Saha et al. 2012; Stratakis and Garcia 2012; Yeh et al. 2012). AuNPs are routinely synthesized by reducing gold halides, typically gold(III) chloride (Bhargava et al. 2005; Brust et al. 1994; Dozol et al. 2013; Jana et al. 2001; Jana and Peng 2003; Kimling et al. 2006; Lee et al. 2010; Leff et al. 1996; Martin et al. 2010; Perrault and Chan 2009; Zhao et al. 2013). The role of halogen in this process is generally overlooked since the size and shape—two basic parameters of the NP—can be controlled by reductants and ligands employed (Scarabelli et al. 2014; Zhang et al. 2014). Though, the halide ions present in a solution can affect the morphology and surface chemistry of NPs (Li et al. 2013; Rai et al. 2006; Singh et al. 2007). The presence of halides in a reaction mixture can also influence the final composition of NPs, e.g. Au–Ag alloys (Rajendra et al. 2015). Hence, we wondered whether gold oxide could be an alternative to gold chloride. The both gold precursors are commercially available on comparable prices that were an additional reason to pursue the research. Meisel and co-workers have recently shown that gold(III) oxide undergoes decomposition to gold colloids upon reduction by molecular hydrogen (Merga et al. 2010). The method, however, requires a special set-up by virtue of the risk of gas explosion that restricts significantly its practical utility. Also, the as-prepared NPs are quite large (20–100 nm) displaying a moderate degree of monodispersity. From application point of view, the smaller nanoparticles with narrow size distribution would be more desirable because of, e.g. more pronounced intrinsic properties, better resistance against aggregation and suitability for studying self-assembly processes.
Herein, we present a simple and convenient method to obtain gold nanoparticles by reduction of gold(III) oxide with oleylamine. Remarkably, the amine (Mourdikoudis and Liz-Marzán 2013; Yu et al. 2014) acts as all-in-one reagent with functions of reductant, ligand and reaction medium. The reaction yields NPs with a small mean size (5–9 nm) and reasonable polydispersity (<1.5). Thus, the method offers a specific size range not accessible from classical oleylamine reduction route based on gold(III) chloride. For instance, the NPs obtained under solution processing conditions were usually above 9 nm (Aslam et al. 2004; Fanizza et al. 2013; Hiramatsu and Osterloh 2004; Lakshminarayana and Qing-Hua 2009; Shen et al. 2008). On other hand, mechanochemical synthesis performed in bulk afforded ultra-small particles with diameters between 1 and 5 nm (Rak et al. 2014). The use of gold(III) oxide affects not only the size of NPs but also enables to preserve the initial metal ratio when preparing Au–Ag alloy NPs. In contrast, the alloys obtained from gold(III) chloride are characterized by reduced silver content owing to solubility issues.
All chemicals were purchased from commercial suppliers and used without further purification: Au2O3 (99 %, ABCR), Ag2O (99 %, Sigma-Aldrich), HAuCl4·3H2O (Sigma-Aldrich), AgNO3 (99 %, Alfa Aesar), oleylamine (technical grade, 70 %, Sigma), octylamine (for synthesis, Merck), dodecylamine (98 %, Aldrich), hexadecylamine (98 %, Aldrich), octadecylamine (98 %, Alfa Aesar), triphenylphosphine (95 %, Fluka) and 1-undecanthiol (98 %, Aldrich). The solvents were of analytical grade quality and degassed by freeze-pomp-thaw technique prior to use: toluene, chlorobenzene, chloroform (ChemPur), decane, tetradecane (Aldrich) and 1,2-dichlorobenzene (ROTH). Silicon wafers were received from ITME (Warsaw). TEM grids were purchased from Ted Pella Inc.
NMR spectra were recorded on Bruker (400 MHz) instrument. GC–MS analyses were performed on PerkinElmer Clarus 680/600S. MS spectra were recorded on Maldi SYNAPT G2-S HDMS (Waters) spectrometer. UV–Vis spectra were recorded using Evolution220 spectrophotometer from Thermo Scientific. XPS spectra were recorded on PHI 5000 VersaProbe X-ray photoelectron spectrometer using an Al KR X-ray source. FT-IR spectra were recorded on Jasco 6200 instrument. SEM, STEM and EDX were recorded on FEI Nova NanoSEM 450.
General procedure for the synthesis and characterization of NP dispersions
Metal precursor(s) (0.01 mmol) and oleylamine (3 mmol, 1 mL) were loaded into 10 mL Schlenk tube. The tube was evacuated and filled with argon three times. If necessary, the degassed solvent was added. The tube was immersed into a pre-heated oil bath, and the suspension was stirred with 1 cm cylindrical bar at a speed of 1400 rpm. The samples for analyses were prepared as following. UV–Vis: aliquots of 1 % (v/v) of original volume were taken over the course of the reaction using automatic or glass pipette and diluted with chloroform; SEM, EDX and XPS: aliquots of 5 % (v/v) of original volume taken from the reaction mixture were diluted with 1:1 (v/v) EtOH-MeOH and agitated on a laboratory mixer at 240 rpm for 1 h. If necessary, the dispersion was centrifuged at 2000–3500 rpm up to 10 min followed by the decantation of supernatant. The procedure was repeated up to 5 times. The NP sediment was dissolved in chloroform and deposited on a silicon wafer or TEM grid. GC–MS, IR, NMR and MS samples were obtained from the supernatant by evaporation of solvent on rotavap.
Results and discussion
We explored potential reductants of gold(III) oxide with a focus on substances capable of serving as ligands for AuNPs. The best results were obtained in organic media by using fatty amines. Other ligands, for example, thiols and phosphines produced polydisperse and aggregated NPs. The temperature regime was crucial to obtain high-quality NPs and secure them against aggregation. The reduction by amines required elevated temperatures (>110 °C) otherwise the reaction was sluggish furnishing NP aggregates. The NPs were also prone to aggregation at low metal–amine ratios. The NP surface was not protected enough by amine ligands leading to uncontrolled growth and sintering of NPs. The effective passivation of NP surface was only attained at the ratios above 1:80. The correlation between the chain length of amine, and the size of NPs was not as apparent as for NPs derived from gold(III) chloride (Marchetti et al. 2011). The NP sizes were in range of 6–9 nm irrespective of aliphatic amine (C12–C18) used. In turn, the alkyl chain of octylamine was too short to effectively stabilize the metal core of NPs. The reaction proceeded equally well in aromatic and aliphatic hydrocarbon solvents that is consistent with previous reports on the reduction of gold(III) chloride (Wu et al. 2013). The reaction also took place without solvent, in particular when employing oleylamine. The latter was also superior in view of uniformity of NPs formed and therefore served us as a model to study the reduction of gold(III) oxide in more detail. The obtained nanoparticles are readily redispersible in nonpolar solvents, especially in chloroform and can be further functionalized with organic thiols (1-undecanthiol, 11-mercaptoundecyl-N,N,N-trimethylammonium bromide).
In summary, we developed a new method for obtaining noble metal NPs from corresponding oxides by reducing with aliphatic amines. The NPs obtained by this method are usually smaller than those prepared using halogen-containing metal precursors. For example, the reduction of gold(III) oxide yields sub-10 nm particles with good monodispersity. The lack of halogen has also influence on the final composition of NPs. The NPs made of gold and silver are alloyed better than those obtained in the presence of halogen. Finally, and importantly, the presented approach adheres well to green chemistry principles. The synthesis is performed in the absence of organic solvents and the reagents used are non-toxic (gold(III) oxide) and manufactured from natural oils (oleylamine).
This project was funded by National Science Center (Grant Sonata UMO-2011/01/D/ST5/03518). We thank Grzegorz Sobczak and Tomasz Wojciechowski for measurements of some SEM and EDX data.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.