Formation of PdPt Alloy Nanodots on Gold Nanorods - American

Feb 18, 2011 - ma-tion of PdPt alloy nanodots on gold nanorods (Au@PdPt NRs). Well- defined alloy ..... ship with Pd/Pt ratio. In comparison with pure...
0 downloads 0 Views 1MB Size
ARTICLE pubs.acs.org/Langmuir

Formation of PdPt Alloy Nanodots on Gold Nanorods: Tuning Oxidase-like Activities via Composition Ke Zhang,†,§ Xiaona Hu,†,§ Jianbo Liu,†,§ Jun-Jie Yin,*,‡ Shuai Hou,†,§ Tao Wen,†,§ Weiwei He,† Yinglu Ji,† Yuting Guo,† Qi Wang,† and Xiaochun Wu*,† †

CAS Key Laboratory of Standardization and Measurement for Nanotechnology, National Center for Nanoscience and Technology, Zhongguancun, Beiyitiao No. 11, Beijing, 100190, P. R. China ‡ Center for Food Safety and Applied Nutrition, Food and Drug Administration, College Park, Maryland 20740, United States § Graduate School of the Chinese Academy of Sciences, Beijing 100190, P. R. China

bS Supporting Information ABSTRACT: The island growth mode of Pt was employed to guide the forma-tion of PdPt alloy nanodots on gold nanorods (Au@PdPt NRs). Welldefined alloy nanodots, with tunable Pd/Pt ratios from 0.2 to 5, distribute homogeneously on the surface of the Au NR. Formation of nanodots shell leads to the red-shift and broadening of the longitudinal surface plasmon resonance (LSPR) band of the Au NRs. The Au@PdPt alloy NRs exhibit catalytic activity toward oxidation of often-used chromogenic substrates by dissolved oxygen under mild conditions, suggesting a new type of oxidase mimics. Composition dependence catalytic activity is observed for the oxidation of ascorbic acid (AA) and 3,30 ,5,50 -tetramethylbenzidine (TMB) and for the reduction of p-nitrophenol. For AA and TMB, catalytic activity enhances quickly at lower Pd/Pt ratios and tends to saturate at higher Pd/Pt ratios. For p-nitrophenol reduction, catalytic activity shows a nice linear relationship with Pd/Pt ratio owing to much higher catalytic activity of Pd. In conclusion, proper alloying of Pd and Pt presents an effective route to tailor the catalytic activity. Interesting, alloy nanodots can also catalyze the oxidation of Fe (II) to Fe (III) by dissolved oxygen. Thus, based on the competitive oxidation of TMB and Fe (II), selective detection of the latter can be achieved.

’ INTRODUCTION Size, shape, structure and composition are critical factors in determining optical, electronic, magnetic and catalytic properties of nanoparticles (NPs), especially for noble metal nanocrystals (NCs).1 For instance, by forming rod-like structure, the surface plasmon resonance (SPR) feature of Au NPs can be easily tuned from visible to near-infrared spectral region by simply changing aspect ratios (length versus diameter), thus extending their potentials in biomedical applications.2 By forming proper bimetallic composite nanostructures, such as Pd-Pt nanodendrites, enhanced catalytic activity and tolerability for oxygen reduction have been exhibited.3 Apart from achieving improved performance by tailoring above parameters, another interesting aspect is to find new functionalities of existed nanomaterials. One typical sample is Fe3O4 magnetic NPs. Generally considered to be biologically and chemically inert, they have been, however, discovered to possess intrinsic peroxidase-like activity and can be used in traditional enzyme linked immunosorbent assay (ELISA) by replacing horseradish peroxidase (HRP).4 Afterward, the enzyme mimics have been reported from other nanomaterials, such as peroxidase-like activity from FeS nanosheets, graphene oxides and single wall carbon nanotube,5 oxidase-like activity from CeO2 NPs,6 and SOD/catalase-like activity from small Pt NPs.7 These r 2011 American Chemical Society

NPs mimics demonstrate great promises for various biodetections and biomedical applications. In comparison with natural enzyme, they own the advantages of controlled synthesis in lowcost, tunability in catalytic activity, and high stability against stringent conditions. Recently, we found that AgPt alloy nanospheres and AgPd alloy NCs also exhibited peroxidase-like activity toward typical HRP substrates such as o-phenylenediamine (OPD), 3,30 ,5,50 tetramethylbenzidine (TMB) and 2,20 -azinobis(3-ethylbenzthiazoline-6-sulfonic acid) diammonium salt (ABTS).8 One advantage of alloy NPs is that the catalytic activity is readily tailored by varying alloying composition. However, one drawback for Ag related alloy NPs is that alloy NPs with a high Ag percentage are unstable in the presence of H2O2, thus limiting the tunability of catalytic activity. This inspires us to explore the possible substitutes of Ag, such as Pd, which are more noble than silver. In comparison with solid structure, formation of a porous nanostructure is an effective route to improve the catalytic properties due to enlarged surface area. In our previous work, it was found that among Au, Ag, Pd, and Pt, Pt exhibited an island Received: November 17, 2010 Revised: January 20, 2011 Published: February 18, 2011 2796

dx.doi.org/10.1021/la104566e | Langmuir 2011, 27, 2796–2803

Langmuir growth mode on Au NRs, which is also verified by other groups.9 Well-defined Pt nanodots formed a shell with nanoscale pores on the Au rod.9d Such a structure is desired for the catalysis. This growth mode was successfully employed to the formation of PtAg alloy nanodots on the Au rod.10 Herein, this strategy is further extended to the combination of Pd and Pt. Well-defined PdPt alloy nanodots with Pd/Pt ratio from 0.2 to 5 are obtained. Different from Ag, alloying with Pd enhances the catalytic oxidation of ascorbic acid (AA) by dissolved oxygen. Similar to CeO2 NPs, PdPt nanoalloys also catalytically oxidize ABTS, OPD, and TMB in the presence of dissolved oxygen under mild conditions, thus acting as an oxidase mimic. Catalytic kinetics can be described by Michaelis-Menten equation. Taking TMB as chromogenic substrates, increasing Pd percentage in alloy, Michaelis constant (Km) shows a little decrease while catalytic constant (Kcat) increases slightly. Intriguing, the nanoalloys also accelerate the conversion of Fe2þ ions to Fe3þ ions at similar reaction conditions. On the basis of the competitive oxidation of TMB and Fe(II), selective detection of the latter is achieved. In order to further explore the relationship between the alloy composition and catalytic property and possible selectivity produced by the combination of two metals, the catalytic reduction of p-nitrophenol is employed. For p-nitrophenol reduction, catalytic activity shows a nice linear relationship with Pd/Pt ratio, indicating much higher catalytic activity of Pd.

’ EXPERIMENTAL SECTION Sodium borohydride (NaBH4), hydrogen tetrachloroaurate (III) trihydrate (HAuCl4 3 3H2O), cetyltrimethylammonium bromide (CTAB), silver nitrate (AgNO3), L-ascorbic acid (AA), palladium chloride (PdCl2) and potassium tetrachloroplatinate(II) (K2PtCl4), poly(styrenesulfate) (PSS), o-phenylenediamine (OPD), 3,30 ,5,50 -tetramethylbenzidine (TMB) and 2,20 -azino-bis(3-ethylbenzthiazoline-6sulfonic acid) diammonium salt (ABTS) were all purchased from Alfa Aesar and used as received. Ferrous(II) chloride tetrahydrate (FeCl2 3 4H2O), manganese(II) chloride tetrahydrate (MnCl2 3 4H2O), nickel(II) chloride hexahydrate (NiCl2 3 6H2O), cobalt(II) chloride hexahydrate (CoCl2 3 6H2O), sulphuric acid (H2SO4) and p-nitrophenol were bought from Sinopharm Chemical Reagent Co., Ltd. Milli-Q water (18 MΩ 3 cm) was used for all solution preparations. Preparation of Au@PdPt Alloy NRs. CTAB-capped Au NRs, prepared using the well-developed seed-mediated growth method,11 were employed as seeds to initiate the codeposition of PdPt shell. The 2 mM H2PdCl4 or PtCl42- aqueous solution was prepared as following: 0.035 g of PdCl2 or 0.0688 g of K2PtCl4 was dissolved in HCl (2 mL, 0.2 M) solution and diluted to 100 mL with water. Before the growth of Pt (Pt and Pd, or pure Pd), Au NRs were purified by centrifugation at 8422 g for 7 min twice. After that, a suspension of 1 mL purified Au NRs (0.5 mM according to Au atoms) was mixed with 40 μL of PtCl42- (2 mM) and a fixed volume of 2 mM H2PdCl4. To this mixture, a 10 times excess of AA (0.1 M) was added and followed by vigorously shaking. The final volume was adjusted to 3 mL by water. By changing the volume (8-200 μL) of added H2PdCl4, different Pd/Pt ratios can be obtained. The reaction took place in several minutes accompanying with a color change from brown to gray and finished after several hours. After 12 h, 200 μL of 0.1 M CTAB was added. The suspensions were centrifuged at 9562 g for 8 min and the precipitate was collected and diluted with H2O to a final concentration of 0.36 nM (according to Au NRs). For TMB oxidation and Fe(II) ion detection, the surface of NRs was further modified with PSS via electrostatic assembly in order to stabilize them in buffer solutions. The preparation of PSS-coated NRs is as follows: 1.0 mL of NRs suspension was purified by centrifuging (9562 g

ARTICLE

8 min). The precipitate was then redispersed in 1.0 mL of water and mixed with 50 μL of 20 mg mL-1 PSS solution (containing 60 mM NaCl). The suspension was placed in a 30 °C water bath for at least 3 h. After that, one time centrifugation (9562g, 8 min) was executed to remove the excessive PSS and the precipitate was redispersed in 100 μL water for further use. Catalytic Oxidation of AA. A 1 mL aliquot of 0.36 nM (according to the concentration of NRs) NRs suspension was diluted with H2O to 3 mL. Then, 5 μL of 0.1 M AA solution was added. UV-vis absorption spectra were recorded at 0.5 min interval to monitor the catalytic kinetics. Catalytic Oxidation of TMB. A 10 μL aliquot of TMB (17 mM) was diluted by 3 mL of PBS (pH = 4.5). Then, 20 μL of PSS-coated NRs (3.6 nM) was added into the solution. UV-vis-NIR spectra were recorded immediately after the addition of NRs at an interval of 0.5 min. The reaction temperature was kept at 30 °C. Apparent kinetic parameters were calculated based on the Michaelis-Menten equation υ = Vmax  [S]/(Km þ [S]), where υ is the initial velocity, Vmax is the maximal reaction velocity, [S] is the concentration of substrate and Km is the Michaelis constant. Detection of Fe2þ. A 10 μL aliquot of TMB (17 mM) and Fe2þ ions with different concentrations (0.04-20.9 μM) was mixed with 3 mL of PBS solution (pH = 4.5). Then, 20 μL of Au@PdPt alloy NRs (Pd/Pt = 1) suspension ([PSS-S4] = 3.6 nM) was added to initiate competition reaction. UV-vis-NIR spectra were recorded at 10 min after the addition of the NRs. The reaction temperature was kept at 30 °C. For possible interfering ions like Mn2þ, Ni2þ, and Co2þ, the similar process was applied except the concentration of metal ions was kept constant at 20.9 μM. Catalytic Reduction of p-Nitrophenol. A 9.0 mL aliquot of 3.7  10 -4 M p-nitrophenol solution and 1 mL of 1.2 M NaBH4 solution were first mixed and stirred for 10 min at room temperature. Then, 0.5 mL of this mixture was diluted with 2.5 mL of H2O. Finally, 50 μL of 0.36 nM Au@PdPt NRs suspension was injected to initiate the reaction. Reaction kinetics was recorded with a 0.5 min interval. Instruments. UV-visible absorption spectra were recorded from Perkin-Elmer Lambda 950 or Cary 50 in the case of scan kinetics. For transmission electron microscopy (TEM, Tecnai G2 F20 U-TWIN) and scanning electron microscopy (SEM, Hitachi S-4800) measurements, one more centrifugation (9562g, 5 min) was applied. High-resolution transmission electron microscopy (HRTEM) images and selected-area electron diffraction pattern were captured at an accelerating voltage of 200 kV. Scanning transmission electron microscopy (STEM) measurement and elemental maps were carried out under a high-angle annular dark field (HAADF) mode from the same microscope. ESR Measurements. Catalase (from bovine liver), iron(II) sulfate heptahydrate (FeSO4 3 7H2O), hydrogen peroxide, and ZnO were purchased from Sigma Chemical Co. (St. Louis, MO). 5-Diethoxyphosphoryl-5-methyl-1-pyroline-N-oxide (DEPMPO) was from Radical Vision (Marseille, France). Fenton Sequence. Different concentrations of Au@Pt NRs were added to 0.1 mM FeSO4 3 7H2O and 0.5 mM H2O2 in the classical Fenton reaction. Then 50 mM DMPO was used as the spin trap. To assess whether the NRs scavenged hydroxyl radicals or interacted with the Fenton reagents, the NRs were first mixed with either iron or hydrogen peroxide for different time periods. Samples were put in 50 μL glass capillary tubes and placed in the ESR cavity. Spectra were recorded under the following conditions: 20 mW microwave power; 1 G field modulation; 100 G scan width. ZnO/UV. In order to determine whether the Au@Pt NRs scavenged hydroxyl radicals, the ZnO/UV system was used. Under UV light (365 nm), ZnO produces hydroxyl radicals. Different concentrations of the NRs were added to 0.5 mg/mL ZnO and 10 mM DEPMPO, put in quartz capillary tubes, and placed in the ESR cavity. To confirm that 2797

dx.doi.org/10.1021/la104566e |Langmuir 2011, 27, 2796–2803

Langmuir

ARTICLE

H2O2 was not an intermediate or byproduct in the production of hydroxyl radicals by ZnO, experiments were repeated with different concentrations of catalase. ESR spectra were recorded after 5 min exposure to UV light at 365 nm under the following conditions: 20 mW microwave power; 1 G field modulation; 100 G scan width. ESR experiments were run on a Bruker EMX spectrometer. A light system consisting of a xenon lamp coupled with a Schoeffel monochromator was used to generate UV light at 365 nm for the ZnO/UV experiments.

’ RESULTS AND DISCUSSION Formation of Well-Defined PdPt Alloy Nanodots on the Au NR. Figure 1A shows typical SEM images of Au@PdPt NRs

formed by coreduction of Pt 2þ and Pd 2þ ions. The Au NRs template has a smooth surface (Figure 1A (a)). When depositing pure Pt, as expected, Pt nanodots evolve from the surface of the Au NR due to the island growth mode (Figure 1A(b)).9d In the case of Pt and Pd coreduction, similarly, well-dispersed bimetallic nanodots form. And the density of dots increases with increasing Pd/Pt ratios. In contrast, at the same deposition condition, without the involvement of Pt, we obtain Pd nanoislands with irregular shapes (Figure 1A(h) and Figure S1A, Supporting Information). The Pd/Au molar ratio obtained from EDX analysis shows a nice linear relationship with the calculated one, indicating a quantitative conversion of Pd ions to Pd atoms (Figure 1A(i)). Obviously, with the guidance of Pt, Pt and Pd bimetallic nanodots with well-controlled “dot” morphology can be obtained. Because of the high dielectric sensitivity of the Au rod, tiny changes of the shell structure can be readily seen from UV-vis-NIR absorption spectra (Figure 1B). The original Au NRs show a sharp longitudinal surface plasmon resonance (LSPR) around 800 nm. After depositing Pt (S1), the LSPR band shows an obvious red-shift (76 nm) and a medium broadening. For Au@PdPt NRs, at lower Pd/Pt ratios of 0.2 (S2) and 0.5 (S3), a further small red-shift continues. Obviously, these nanorods have a well-defined NIR plasmonic feature. For these NRs, the broadening may come from damping of LSPR band by the PdPt nanoalloys, the lower monodispersity of the core/shell structure in comparison with the Au nanorod core due to the polydispersity in terms of shell thickness and porosity. The redshift mainly comes from the porous structure. As demonstrated previously, formation of a porous nanostructure can lead to redshifting of SPR band for Au nanocages.12 In contrast, by forming solid Pd shell, blue-shift of LSPR band with increasing Pd/Au ratio was observed.9b Further increasing Pd/Pt ratio to 1 (S4), 2 (S5), and 5 (S6), severe broadening to the longer wavelength appears, accompanying strong suppression of the LSPR intensity. For these NRs, the broadening and red-shift mainly arise from the formation of small aggregates of the NRs. HRTEM image exhibits that the nanodots shell has a lot of nanoscale pores. Zoom-in image shows a well-defined lattice image (Figure 2b). The corresponding fast Fourier transform (FFT) pattern (Figure 2c) demonstrates that only one set of diffraction spots ({002}) appear, indicating that the electron beam is aligned in the [001] direction. The HRTEM images exhibit that PdPt shell is a single crystal. The single crystalline nature of the shell also indicates that Pd and Pt form an alloy. The STEM image shows a clear core/shell structure. Element mapping shows a homogeneous distribution of Pd and Pt in shell, supporting the formation of PdPt alloy (Figure 3). The formation of alloy is also substantiated by XPS measurements

Figure 1. (A) SEM images of (a) Au NRs, (b) [email protected] NRs (S1), (c-g) Au@PdPt NRs with Pd/Pt ratios of 0.2 (S2), 0.5 (S3), 1 (S4), 2 (S5), and 5 (S6), and (h) [email protected] NRs (S7). The Pt/Au ratio is fixed at 0.16. The scale bar is 60 nm and applies to all images. (i) Pd/Au ratio from EDX analysis versus added Pd/Au ratio. (B) UV-vis-NIR absorption spectra of the corresponding nanorods. The absorbance is normalized at 400 nm.

(Figure S2, Supporting Information). Increasing Pd amount, the intensities of Pt4f /Au4f and Pd3d /Au4f both gradually increase, indicating codeposition of Pd and Pt. The growth mode of one metal on another is mainly determined by the lattice mismatch and interactions between the two metals.13 Generally, three growth modes are observed, the layered growth mode (Frank-van der Merwe (F-M) mode), the island growth (Volmer-Weber (V-W) mode), and the intermediate mode (Stranski-Krastanow (S-K) mode).13 In the case of Au@Pt, the lattice mismatch (3.8%) between Pt (0.392 nm) and Au (0.408 nm) is relatively large but still within the limit of 5% lattice mismatch. However, the interaction among Pt atoms is much stronger than that between Pt and Au. Overall, Pt prefers an island growth mode on Au. This growth behavior has been demonstrated previously both on gold nanospheres and 2798

dx.doi.org/10.1021/la104566e |Langmuir 2011, 27, 2796–2803

Langmuir

Figure 2. HRTEM image (a) of Au@PdPt alloy NR with Pd/Pt ratio of 5 (S6). A zoom-in image of red square in (a) and the FFT pattern of green square in (a) is shown in (b), and (c), respectively. (d) Schematic diagram of Au@PdPt alloy NR.

Figure 3. STEM image and corresponding element maps of S6.

nanorods.9a,9d,13,14 For Au@Pd, 4.5% lattice mismatch is a little larger than that of Au@Pt. However, the interaction among Pd atoms is similar to that between Pd and Au. This leaves the space to tune growth mode by the manipulation of deposition conditions. In fact, both layered and island growth modes have been observed for Au and Pd combination.9a,9b,9e In our previous work, with a high concentration of CTAB (0.1M) in growth solution, deposition of Pd on the Au NR follows the F-M mode: Au@Pd nanobars with a smooth Pd surface are obtained.9b,9e In

ARTICLE

contrast, by increasing Pd/Au ratio (>3.2) or lowering CTAB concentration (