Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
J Phys Chem C Nanomater Interfaces. Author manuscript; available in PMC 2011 March 5.
Published in final edited form as:
J Phys Chem C Nanomater Interfaces. 2010 March 5; 114(14): 6396–6400.
doi:  10.1021/jp100354z
PMCID: PMC2873216

Dissolving Ag from Au-Ag Alloy Nanoboxes with H2O2: A Method for Both Tailoring the Optical Properties and Measuring the H2O2 Concentration


This article describes a method for generating Au-based nanocages with controlled wall thickness, porosity, and optical properties by dissolving Ag from Au-Ag alloy nanoboxes with H2O2. It involves two steps: i) formation of Au-Ag alloy nanoboxes with some pure Ag left behind by titrating Ag nanocubes with aqueous HAuCl4; and ii) removal of Ag atoms from both the pure Ag remaining in the nanoboxes and the alloy walls via H2O2 etching. The optical properties of the resultant Au-Ag nanocages can be easily tailored by controlling the amount of H2O2 added into the reaction system. Due to the changes to the optical spectra, the Au-Ag alloy nanoboxes can also be employed to detect H2O2 with a more or less linear readout in the range of concentration from 5×10−2 M down to 5×10−7 M.

Keywords: Au-based nanoboxes, nanocages, dealloying, H2O2 detection


Metallic hollow nanostructures often exhibit exceptional optical, electronic, and catalytic properties as compared to their solid counterparts due to the increased surface area, reduced density, and broadly tunable localized surface plasmon resonance (LSPR).1 They have been actively explored for a range of applications, such as sensing, optical imaging, photothermal treatment, drug delivery, as well as catalysis.26 Galvanic replacement, driven by the electrical potential difference between two metals, provides a facile method for generating metallic hollow nanostructures. Several bimetallic and tri-metallic hollow nanostructures comprised of Ag and Au, Pd, or Pt have been produced through the galvanic replacement between Ag nanostructures and salt precursors of less reactive metals such as Au, Pd, and Pt.7,8 For example, Ag nanocubes can be titrated with an aqueous HAuCl4 solution to generate Au-Ag alloy nanoboxes (structures with hollow interiors) and nanocages (structures with both hollow interiors and porous walls) with a broad range of controllable compositions. During the structural evolution from nanocubes to nanoboxes and nanocages, the LSPR peaks shift from the visible to the near-infrared (NIR) region due to the change in wall thickness relative to the overall dimension of the nanostructures.9 In this process, the dissolution of Ag and deposition of Au occur simultaneously on the same nanocube during the reaction. This coupling makes it very difficult to tune and control the wall thickness and porosity of the resultant hollow nanostructures, especially during the dealloying stage. Recently, a two-step procedure was demonstrated to solve this problem by dealloying Ag from the pre-formed Au-Ag nanoboxes (pre-formed via the galvanic replacement reaction) with an etchant such as Fe(NO3)3, NH4OH, or thiols.10 Unlike HAuCl4, these wet etchants offer a tighter control over the wall thickness and porosity due to the elimination of Au deposition. In principle, any chemical with an electrochemical potential higher than that of Ag+/Ag can serve as an oxidative etchant in the second step.

Hydrogen peroxide (H2O2) is a powerful oxidizing agent. The standard redox potential of H2O2 is dependent on the pH value of the solution:11



Under both conditions, the electrochemical potentials are higher than that of Ag+/Ag (E0 = 0.7996 V).12 Therefore, H2O2 can serve as an effective etchant for elemental Ag in tailoring the optical properties of the Au-Ag alloy nanoboxes and nanocages. In addition, during the removal of Ag, the changes to optical spectra provide a simple assay for H2O2 detection based on a colorimetric approach. Hydrogen peroxide is involved in many biological processes and enzymatic reactions related to intracellular oxidative stress and ageing.1314 For example, the intracellular concentration of H2O2 has a profound impact on cellular physiological processes including proliferation, growth arrest, and apoptosis.1517 It also plays an important role in food processing as an antimicrobial agent18 and in atmospheric science as an oxidant related to the formation of acid rain.19 Therefore, determination of H2O2 concentration is essential to monitoring and understanding these biological and environmental processes.

In this study, we systematically investigated the etching of Au-Ag alloy nanoboxes, with H2O2 serving as an oxidant. The wall thickness of the resultant Au-based nanocages could be readily controlled by adding different amounts of H2O2 to the suspension of Au-Ag alloy nanoboxes that were, in turn, obtained by carrying out the galvanic replacement to different extents. The LSPR peaks of the Au-based nanocages could be conveniently tuned across the entire visible region into the near-infrared from 500 nm to 1000 nm. Based on this etching mechanism, we also demonstrated the use of the Au-Ag alloy nanoboxes as a new platform for the quantitative analysis of H2O2 using a spectroscopic method.

Experimental Section

Titration of Au-Ag Alloy Nanoboxes with H2O2

The Au-Ag alloy nanoboxes were synthesized using the galvanic replacement reaction between Ag nanocubes and HAuCl4 as described previously.10 Briefly, 15 mL of 1 mg/mL poly(vinyl pyrrolidone) (PVP, M.W. ≈ 55,000) aqueous solution was added to a 100-mL round-bottom flask and heated in an oil bath at 100 °C under stirring for 10 min. Subsequently, 100 µl of the as-synthesized Ag nanocubes of 45 nm in mean edge length (0.17 nM or 1.0×1014 nanocube/mL) was added to the flask. Two minutes later, 0.5 mM HAuCl4 aqueous solution was introduced into the flask using a syringe pump at a rate of 45 mL/h. The titration was stopped after the extinction peak of the solution had reached 460 nm. A similar procedure was also used to prepare Au-Ag nanoboxes with LSPR peaks being tuned to 520, 545, 675, and 745 nm, respectively.

For the reaction with H2O2, the as-prepared suspension of Au-Ag nanoboxes was added to a 20-mL glass vial under magnetic stirring at room temperature, followed by the addition of an excess amount of H2O2 solution. The reaction was allowed to proceed at room temperature for 10 min. To study the structural evolution during the reaction, the Au-Ag nanoboxes with an LSPR peak at 520 nm were titrated with different amounts of H2O2 solution and aliquots were taken from the solution at different time points. The products were washed using a procedure reported in a previous publication.20

Detection of H2O2 with the Au-Ag Nanoboxes

The Au-Ag nanoboxes with an LSPR peak at 545 nm were used to demonstrate H2O2 detection. In a typical assay, a series of H2O2 solutions were prepared with concentrations in the range of 1×10−6 to 1×10−1 mol/L (M). Then, 0.5 mL of the H2O2 solutions having various concentrations were separately added to 2-mL plastic cuvettes, followed by the introduction of 0.5 mL of the Au-Ag nanoboxes, resulting in final H2O2 concentrations of 5×10−7 M to 5×10−2 M. The solution in each cuvette was mixed on a shaker at room temperature for 30 min before a spectrum was taken using a UV-vis-NIR spectrometer.


The sample for scanning electron microscopy (SEM) was prepared by dropping a few microliters of the particle suspension on a small piece of silicon substrate, followed by drying under ambient conditions. The SEM images and the energy-dispersive X-ray (EDX) spectra were captured using a field-emission microscope (FEI, Nova NanoSEM 230) operated at accelerating voltages of 10–20 kV. The sample for transmission electron microscopy (TEM) was prepared by replacing the silicon substrate with a carbon-coated copper grid. The TEM images were captured using a FEI G2 Spirit Twin electron microscope operated at 120 kV. The high-resolution TEM images were recorded using a JEOL 2100F electron microscope operated at 200 kV. The UV-vis-NIR spectra were taken using a Cary 50 spectrophotometer (Varian, Palo Alto, CA) or a Lambda 950 spectrophotometer (Perkin Elmer, Waltham, MA).

Results and Discussion

The essence of this study is the formation of Au-based nanocages by dissolving Ag from the Au-Ag alloy nanoboxes with an excess amount of H2O2. Prior to the addition of H2O2, the Au-Ag nanoboxes were prepared using the galvanic replacement reaction between Ag nanocubes and HAuCl4. Figure 1, A–D, shows TEM images of the Au-Ag alloy nanoboxes with increased void sizes that were synthesized by titrating suspensions of Ag nanocubes (of the same volume) with different amounts of HAuCl4. During the replacement reaction, three Ag atoms were removed while one Au atom was formed and deposited on the surface of the Ag nanocube. As a result, a pit was initially formed on the surface of the nanocube. It is known that Ag atoms can diffuse into the lattice of Au to form a thin layer of Au-Ag alloy at the surface.21 At an intermediate stage, the pit could evolve into a void inside the nanocube, resulting in the formation of a Au-Ag nanobox with some pure Ag remaining in the Au-Ag alloy nanobox.

Figure 1
TEM images of Au-Ag nanoboxes before and after etching with an excess amount of H2O2: (A–D) the Au-Ag nanoboxes with their LSPR peaks being tuned to 460, 520, 675 and 745 nm, respectively, prior to the addition of H2O2, and (E–H) the corresponding ...

In the first set of experiments, an excess amount of H2O2 was introduced as an oxidant to etch away elemental Ag from both the remaining pure Ag and the Au-Ag alloy walls (Figure 1, E–H). The dissolution of Ag is based on the following reaction at pH = 6:


The remaining pure Ag inside each Au-Ag alloy nanobox and the Ag in the Au-Ag alloy walls could both be dissolved without additional deposition of Au atoms. As a result, the wall thickness and porosity of the Au nanocages can be more easily controlled. Depending on the amount of HAuCl4 added in the first step, the wall thickness of the resultant nanocages was tuned to 2.5, 3.5, 4.5 and 6.5 nm, respectively (Figure 1, E–H). Note that the nanocages in Figure 1, E and F, were slightly distorted because their walls were too thin to survive the strong capillary force induced by solvent evaporation during TEM sample preparation. These nanostructures were also more porous than those shown in Figure 1, G and H, because of their further reduced wall thickness. EDX analysis indicates that the percentages of Au atoms in the nanocages were 23, 44, 46 and 66%, respectively, for the samples shown in Figure 1, E–H. This result suggests that H2O2 could not completely remove Ag atoms from the Au-Ag alloy walls, which is different from our previous study that involved the use of Fe(NO3)3 as an etchant.10 This incompleteness might be due to the oxidation of surface Ag atoms or the formation of a by-product such as AgOH or Ag2O that might hinder the further removal of Ag from the alloy walls. Additionally, it might also be related to the change in redox potential as the composition of the Au-Ag alloy nanoboxes was altered.

Figure 2 shows a TEM image and a typical high-resolution TEM image taken from the corner of a nanocage. The well-resolved, continuous lattice fringe (2.0 Å in spacing) parallel to the edges of the nanocage indicates that the nanocage was a single-crystal made of Ag-Au alloy. Comparing to the previous work based on Fe(NO3)3,10 there are three major advantages for using H2O2 as an etchant: i) the wall thickness of the nanocage can be conveniently tuned down to 2.5 nm, with an accuracy of ±0.5 nm; ii) the Au nanocage retained its well-defined, box-type shape in stead of being transformed into a frame-like structure; and iii) the elimination of a possible source of contamination from the Fe ions that might compromise the performance of Au nanocages in various applications.

Figure 2
(A) TEM image of a representative nanocage (from the sample shown in Figure 1F) synthesized by partially dissolving Ag from the Au-Ag nanobox with excess H2O2. (B) High-resolution TEM image of the area boxed in (A).

We then studied the optical properties of the as-prepared Au-Ag alloy nanostructures by UV-vis-NIR spectroscopy. Figure 3 shows the UV-vis-NIR spectra of Au-Ag nanoboxes before and after treatment with an excess amount of H2O2, corresponding to the structures shown in Figure 1. The LSPR peaks of the Au-Ag alloy nanoboxes in Figure 1, A–D, were located at 460, 520, 675, and 745 nm, respectively. After the treatment with excess H2O2, the peaks were red-shifted to 1210, 925, 790, and 758 nm, respectively, for the samples in Figure 1, E–H. The corresponding intensities of the peaks were reduced from 1.0 to 0.13, 0.18, 0.58 and 0.74, respectively. This trend is in agreement with our previously published calculation based on discrete-dipole-approximation (DDA): that is, the LSPR peak tends to red-shift as the wall thickness is reduced, and the intensity of the peak drops as the walls became thinner and more porous.22 Additionally, the full width at half maximum (FWHM) of the LSPR peak became broader as the wall thickness decreased. The increased broadening can be attributed to a combination of surface scattering of electrons, as well as increased radiation damping for the nanoboxes.23 This effect is more pronounced when the wall thickness is below 5 nm.

Figure 3
UV-vis-NIR spectra of the samples that correspond to the Au-Ag alloy nanoboxes and nanocages shown in Figure 1. The spectra in black are for the starting Au-Ag nanoboxes and the spectra in red are for the resulting Au-based nanocages after a reaction ...

To fully understand the etching process, we also studied the structural evolution from the nanobox to nanocage during the titration with H2O2. Figure 4 shows TEM images of Au-Ag alloy nanoboxes with an LSPR peak at 520 nm after treatment with different amounts of H2O2: 0, 10, 20, 30, 40, and 50 µL of 10 mM H2O2, respectively. As more H2O2 was titrated into the reaction, the void size in the interior of each nanobox gradually increased due to the dissolution of the remaining Ag (Figure 4, B and C). After the pure Ag had been completely removed from the inside, the H2O2 started to dissolve Ag atoms from the alloyed walls. As a result, the walls became porous, resulting in the formation of Au-based nanocages. For the sample shown in Figure 4F, the nanocages contained 44% Au and 56% Ag. Figure 5 shows the corresponding UV-vis-NIR spectra for the samples shown in Figure 4, clearly indicating that the LSPR peaks were shifted to 617, 691, 850, 901, and 963 nm.

Figure 4
TEM images of Au-based nanocages after a suspension of Au-Ag alloy nanoboxes (with an LSPR peak at 520 nm) had been titrated with different volumes of 10 mM H2O2: (A) 0, (B) 10, (C) 20, (D) 30, (E) 40, and (F) 50 µL.
Figure 5
UV-vis spectra that correspond to different stages of the titration process shown in Figure 4. The spectra were normalized to unity.

The spectral changes observed for the Au-Ag alloy nanoboxes can be applied to develop a colorimetric method for the detection of H2O2. After the addition of H2O2, the absorbance spectrum of the nanoboxes displayed two major changes: i) drop of the LSPR peak intensity, and ii) red-shift of the peak position. We tested the sensitivity of detection for Au-Ag alloy nanoboxes with their LSPR peaks being tuned to 545, 660 and 750 nm, respectively. Figure 6A shows the time-dependent study with H2O2 treatment to determine the best reaction time for the assay. At a concentration of 0.1 M, the intensity of the peak decreased rapidly in the first 20 min and levered off in 30 min after the addition of H2O2. This observation suggested that the reaction between H2O2 and Ag were completed in 30 min. We therefore chose 30 min to be the incubation time for the assay. In a typical procedure, 0.5 mL H2O2 aqueous solution at a concentration ranging from 0.1 M to 10−6 M was added into each cuvette, followed by the addition of 0.5 mL suspension of the Au-Ag alloy nanoboxes, resulting in final H2O2 concentrations in the range of 5×10−2 M to 5×10−7 M. After incubation for 30 min on a shaker, a UV-vis-NIR spectrum was measured for each reaction. The intensity of the absorbance peak at 545 nm decreased gradually as a function of H2O2 concentration (Figure 6B). In the case of the other two samples of Au-Ag alloy nanoboxes with the LSPR peaks at 660 and 750 nm, the intensity change for the absorbance peak was very small when the concentration of H2O2 was low, possiblely due to the presence of a large portion of Ag in the alloy walls (Figure S1). As a result, we chose Au-Ag alloy nanoboxes with the LSPR peak at 545 nm for the detection of H2O2. We plotted the absorbance change (ΔAbs) as a function of p[M], where p[M] is defined as the negative decimal logarithm of the H2O2 concentration in an aqueous solution (Figure 6C). The ΔAbs linearly decreased with p[M] in the concentration range from 5×10−2 to 5×10−7 M, ΔAbs = 0.28 − 0.03 × p[M], R2 = 0.99. This linear dependence offers a simple method for reading out the concentration of H2O2. The detection of H2O2 is based on an electrochemical process, in which the remaining pure Ag in the Au-Ag alloy box was oxidized by the H2O2, resulting in changes to the optical spectrum. In other words, the intensity of the absorbance peak decreases at 545 nm due to the dissolution of pure Ag, accompanied by the rise of absorbance peak at longer wavelengths because the structure became more empty. This method may provide some advantages over the currently existing methods,2426 for example, the possibility for a direct, quantitative measurement of H2O2 without adding special reagents, easiness of operation, and relatively low cost.

Figure 6
H2O2 detection based on changes to the optical properties of Au-Ag nanoboxes. (A) Plots of intensity of absorbance at 545 nm as a function of incubation time. (B) UV-vis-NIR spectra of the Au-Ag nanoboxes (with an LSPR at 545 nm) after incubation with ...


We have demonstrated a facile method for preparing Au-based alloy nanocages with well-controlled wall thickness and porosity by gradually dissolving the Ag from Au-Ag alloy nanoboxes with H2O2. For the nanocages with an inner edge length of 45 nm, the wall thickness could be conveniently tuned from 6.5 nm to 2.5 nm with the corresponding LSPR peaks being red-shifted from 758 to 1210 nm. Addition of an excess amount of H2O2 resulted in the formation of porous nanocages with relatively sharp corners, rather than the formation of nanoframe as in the case of etching with Fe(NO3)3. By using this method, we can easily produce the well-defined nanocages with ultrathin walls which could be a better candidate for catalysis.27 Additionally, when a specific amount of Au-Ag alloy nanoboxes was used, the intensity of the LSPR peak decreased linearly as the concentration of H2O2 was increased in the range of 5×10−7 to 5×10−2 M, suggesting that the Au/Ag alloy nanoboxes can potentially serve as new platform for H2O2 detection.

Supplementary Material



This work was supported in part by a 2006 Director’s Pioneer Award from the NIH (DP1 OD000798) and startup funds from Washington University in St. Louis. As a visiting student from the University of Science and Technology of China, Q.Z. was also partially supported by the China Scholarship Council. Part of the work was performed at the Nano Research Facility (NRF), a member of the National Nanotechnology Infrastructure Network (NNIN), which is supported by the NSF under award no. ECS-0335765. We thank Prof. Richard Loomis for allowing us to use the UV-vis-NIR spectrometer in his laboratory.


1. Skrabalak SE, Chen J, Sun Y, Lu X, Au L, Cobley CM, Xia Y. Acc. Chem. Res. 2008;41:1587–1595. [PMC free article] [PubMed]
2. Sun Y, Xia Y. Anal. Chem. 2002;74:5297–5305. [PubMed]
3. Chen J, Saeki F, Wiley BJ, Cang H, Cobb MJ, Li Z-Y, Au L, Zhang H, Kimmey MB, Li X, Xia Y. Nano Lett. 2005;5:473–477. [PubMed]
4. Chen J, Wang D, Xi J, Au L, Siekkinen A, Warsen A, Li Z-Y, Zhang H, Xia Y, Li X. Nano Lett. 2007;7:1318–1322. [PMC free article] [PubMed]
5. Hirsch LR, Gobin AM, Lowery AR, Tam F, Drezek RA, Halas NJ, West JL. Ann. Biomed. Eng. 2006;34:15–22. [PubMed]
6. Cobley CM, Campbell DJ, Xia Y. Adv. Mater. 2008;4:748–752. [PMC free article] [PubMed]
7. Chen J, McLellan JM, Siekkinen A, Xiong Y, Li Z-Y, Xia Y. J. Am. Chem. Soc. 2006;128:14776–14777. [PMC free article] [PubMed]
8. Chen J, Wiley B, McLellan J, Xiong Y, Li Z-Y, Xia Y. Nano Lett. 2005;5:2058–2062. [PubMed]
9. Sun Y, Xia Y. Nano Lett. 2003;3:1569–1572.
10. Lu X, Au L, McLellan J, Li Z-Y, Marquez M, Xia Y. Nano Lett. 2007;7:1764–1769. [PMC free article] [PubMed]
11. Tratnyek PG, Macalady DL. Handbook of Property Estimation Methods for Chemicals: Environmental and Health Sciences. 2000
12. Som T, Karmakar B. Nano Res. 2009;2:607–616.
13. Fang J, Sawa T, Akaike T, Maeda H. Cancer Res. 2002;62:3138–3143. [PubMed]
14. Delaunay A, Pflieger D, Barrault M, Vinh J, Toledano MB. Cell. 2002;111:471–481. [PubMed]
15. Finkel T, Holbrook NJ. Nature. 2000;408:239–247. [PubMed]
16. Deng X, Gao F, May WS., Jr Blood. 2003;102:3179–3185. [PubMed]
17. Veal EA, Day AM, Morgan BA. Mol. Cell Rev. 2007;26:1–14. [PubMed]
18. Welk A, Meller C, Schubert R, Schwahn C, Kramer A, Below H. BMC Microbiol. 2009;9:1–8. [PubMed]
19. Lee M, Heikes BG, O'Sullivan DW. Atmos. Environ. 2000;34:3475–3494.
20. Skrabalak SE, Au L, Li X, Xia Y. Nat. Protoc. 2007;2:2183–2190.
21. Sun Y, Xia Y. J. Am. Chem. Soc. 2004;126:3892–3901. [PubMed]
22. Chen J, Wiley B, Li Z-Y, Campbell D, Saeki F, Cang H, Au L, Lee J, Li X, Xia Y. Adv. Mater. 2005;17:2255–2261.
23. Hu M, Petrova H, Siekkinen AR, Chen J, McLellan JM, Li Z-Y, Marquez M, Li X, Xia Y, Hartland GVJ. J. Phys. Chem. B. 2006;110:19923–19928. [PubMed]
24. Hurdis E, Romeyn H., Jr Anal. Chem. 1954;26:320–325.
25. Gou S, Wang E. Anal. Chim. Acta. 2007;598:181–192. [PubMed]
26. Hanaoka S, Lin J, Yamada M. Anal. Chim. Acta. 2001;426:57–64.
27. Zeng J, Zhang Q, Chen J, Xia Y. Nano Lett. 2010;10:30–35. [PubMed]