Journal of Colloid and Interface Science 440 (2015) 236–244
Contents lists available at ScienceDirect
Journal of Colloid and Interface Science www.elsevier.com/locate/jcis
Facile preparation of gold nanocages and hollow gold nanospheres via solvent thermal treatment and their surface plasmon resonance and photothermal properties Haifei Wang, Jing Han, Wensheng Lu ⇑, Jianping Zhang, Jinru Li, Long Jiang ⇑ Beijing National Laboratory for Molecular Science, Key Laboratory of Colloid, Interface and Thermodynamics, Institute of Chemistry, Chinese Academy of Sciences, Beijing 100190, PR China
a r t i c l e
i n f o
Article history: Received 17 July 2014 Accepted 3 November 2014 Available online 11 November 2014 Keywords: Core–shell nanostructures Hollow gold nanostructures Solvent thermal treatment Surface plasmon resonance (SPR) absorptions Photothermal conversion property
a b s t r a c t Although template etching method is one of the most common ways of preparation of hollow gold nanostructures, this approach still requires further improvements to avoid the collapse of gold shells after the cores were removed. In this work, an improved template etching method, with which hollow gold nanostructure is fabricated by etching Polystyrene (PS) cores from PS@Au core–shell nanospheres with solvent thermal treatment in N,N-Dimethylformamide (DMF), is demonstrated. When PS cores were removed by a thermal treatment process, gold nanoshells reconstruct and the collapse of the nanoshells is avoided. Gold nanocages and hollow gold nanospheres are easily obtained from the various structures of PS@Au core–shell nanospheres. These hollow nanostructures represent special near infrared (NIR) optical property and photothermal property. Compared with hollow gold nanospheres, the gold nanocages show higher temperature increase at the same particle concentration. Ó 2014 Elsevier Inc. All rights reserved.
1. Introduction Hollow gold nanostructures (especially gold nanocages and hollow gold nanospheres) exhibit unique optical property due to their tunable surface plasmon resonance (SPR) absorptions, and have attracted extensive research attentions [1–3]. The SPR absorptions can be tuned from the visible region to the near-infrared (NIR) region by changing the size and thickness of gold shells [4–6]. The hollow gold nanostructures have potential applications in cell imaging [7–9], biosensor [10], photothermal therapy of cancers [11,12], controllable drug release [13,14] and catalysis [15–17]. Currently, a series of methods on preparation of hollow gold nanostructures with tunable optical property have been researched in some groups. The most common approaches for the preparation of hollow gold nanostructure involve growing gold nanoparticles on soft surfactant aggregates templates [18], Abbreviations: SPR, surface plasmon resonance; NIR, near infrared; PS, polystyrene; DMF, N,N-Dimethylformamide; PEI, poly (ethyleneimine); PVP, polyvinypyrrolidone; SEM, scanning electron microscopy; TEM, transmission electron microscopy; HRTEM, high-resolution transmission electron microscopy; XRD, power X-ray diffraction; DLS, dynamic light scattering; UV–vis–NIR, ultraviolet– visible–near infrared. ⇑ Corresponding authors. E-mail addresses: [email protected] (W. Lu), [email protected] (L. Jiang). http://dx.doi.org/10.1016/j.jcis.2014.11.004 0021-9797/Ó 2014 Elsevier Inc. All rights reserved.
galvanic replace reaction by using sacrificial metal templates [4,19,20], and selective etching or calcination gold core–shell nanostructures [21–24]. As to the selective etching and calcination methods, hollow gold nanostructures were usually fabricated by depositing gold shells on PS sphere templates, and then selectively removed templates via calcination or chemical etching. PS sphere templates have easily functionalized surface and broad size range and have therefore been frequently used [25,26], and the way of fabricating gold core–shell nanostructures on colloidal surfaces has been widely researched since Halas’ pioneering work on preparation of gold core–shell nanospheres [3]. However, the gold shells of hollow nanostructures would be collapsed or destroyed after the PS cores were selectively removed. The main challenge in the preparation of hollow gold nanoshells is the difficulty of removing PS templates. There are reports on successfully removing PS templates by calcination at high temperature (above 300 °C), but the processes are rather complex because the gold shells would be destroyed if they are not protected before calcination [27–30]. Other method is dissolution of PS core into an organic solvent. Even though PS can be easily dissolved in Tetrahydrofuran (THF) and Dimethylformamide (DMF), gold shell would collapse after the PS cores were removed due to its low mechanical strength [31,32]. To date, most PS@metal core–shell nanostructures have been applied without core removed, and caused performance
H. Wang et al. / Journal of Colloid and Interface Science 440 (2015) 236–244
degradation of the core–shell nanostructures [33]. Therefore, it is essential to develop an effective and facile method of fabricating hollow gold nanostructures. Compared with the calcination method, the chemical etching method will be a promising route to remove PS templates at low temperature, if troubles of shell collapse are solved. As well known, metal nanoparticles grow larger by Ostwald ripening at the expense of smaller ones due to chemical potential energy difference. Moreover, metal nanoparticles would reconstruct at lower temperature than the melting point of blocks. There are reports on reconstructing gold nanoparticles upon the thermal treatment [34–38]. In our work, to achieve perfect hollow gold shells, the thermal treatment process was used in solvent etching method to solve the trouble of gold shell collapse. When PS@Au core–shell nanospheres were treated by heating in DMF solvent, PS cores were etched out and gold nanoshells reconstructed to improve their mechanical strength, so the gold nanoshells would not collapse after PS cores were removed. Herein, we report on the preparation of gold nanocages and hollow gold nanospheres from PS@Au core–shell nanospheres by solvent thermal treatment in DMF. In this process, gold nanoshells reconstruct in Ostwald ripening process by thermal treatment (below 120 °C). Hollow gold nanostructures are obtained without shells collapse. Moreover, various shell structures of hollow nanostructures (nanocages or hollow nanospheres) are controllably obtained from variable surface structures of PS@Au core–shell nanospheres. The SPR features of these hollow nanostructures are investigated, and they have strong SPR absorption in NIR region. In addition, these hollow gold nanostructures have stable photothermal effect, and hollow gold nanocages have higher temperature increase than that of hollow gold nanospheres at the same particle concentration. 2. Experimental section 2.1. Materials PS spheres (about 260 nm in diameter) were used as received. HAuCl44H2O was obtained from Shanghai Reagent Company. Poly (ethyleneimine) branched (PEI, Mn = 10,000) was purchased from Sigma–Aldrich. N,N-Dimethylformamide (DMF) and potassium carbonate (K2CO3), ammonium hydroxide (NH3H2O, 28–30%), formaldehyde (HCHO 37%), sodium borohydride (NaBH4), and polyvinypyrrolidone (PVP, Mw = 30,000) were received from Beijing Chemical Company. All chemicals were analytical grade and used without further purification. Deionized water used for all experiments was treated with a Millipore water purification system (Millipore Corp.). 2.2. Preparation of the PS@Au core–shell nanospheres The PS spheres were modified with PEI layer via electrostatic interaction before used. 2.5 mL suspension of negatively charged sulfate-stabilized PS spheres (87 mg mL 1) was added to excess positive charged PEI solution (10 mL, 0.6 mg mL 1), and was mixed for 30 min. The PEI modified PS spheres were purified by repeated centrifuging (10,000 rpm, 20 min). Then the PEI modified positive charged PS spheres (PS-PEI) were redispersed into 25 mL water (8.7 mg mL 1). PVP protected gold nanoparticles were prepared according to the literature [39] and were used as seeds. Gold seeds were absorbed on modified PS spheres as following: 200 lL, PS-PEI was added to 8 mL newly prepared gold sol quickly. After 30 min, the excess gold sol was removed by centrifuging (8000 rpm, 5 min) and washed one time. Then, gold seeds on PS spheres (PS@Au seeds) were obtained. PS@Au core–shell nanospheres were prepared via a seeding-mediated growth approach in 0.3 mM HAuCl4 growth solution. Growth solutions were prepared as
237
following [40]: 50 mg of K2CO3 was dissolved in 200 mL water. After 10 min of stirring, 1.41 mL of HAuCl4 (48 mM) was added. The solution became colorless after 30 min and was aged for 24 h in the dark for use in subsequent steps. PS@Au core–shell nanospheres were fabricated by adding former PS@Au seeds to 10 mL growth solution with PVP (1 mg mL 1) as stabilizer, and 50 lL NH3H2O (28–30%) was added to adjust pH. Then, 60 lL formaldehyde was added and reacted for 12 h. Different PS@Au core–shell nanostructures were obtained by adjusting the volume of HAuCl4 (5 mL, 10 mL, 15 mL, 20 mL, 25 mL, and 40 mL) and formaldehyde (Table S1 in Supporting Information). Finally, PS@Au was washed three times by centrifuging (5000 rpm, 5 min), and redispersed into water. 2.3. Preparation of Au nanocages and hollow Au nanospheres Hollow gold nanostructures were obtained by removing the PS templates in heated DMF. PS@Au core–shell nanospheres (prepared in 20 mL growth solution) were added into 20 mL DMF with stirring. With the heating for 12 h at 120 °C, the PS templates were removed and hollow gold nanostructures were obtained. Gold nanocages and hollow gold nanospheres were prepared from variable PS@Au core–shell nanostructures. Gold nanostructures were washed with DMF to remove the PS in solution by centrifuging (5000 rpm, 5 min) three times, and redispersed into water. Effects of temperature and time of thermal treatment were also investigated. 2.4. Photothermal conversion analysis The photothermal experiments of gold nanocages and hollow gold nanospheres were carried at the same particle concentration. The same amounts of PS spheres were used at all the core–shell nanostructures, so the corresponding hollow gold nanostructures were the same particle concentration. The suspensions of gold nanocages and hollow gold nanospheres were irradiated at a power density of 0.5 W cm 2 by 808 nm laser. The laser spot was altered to cover the surface of the suspensions. The temperature increases of the suspensions were recorded as a function of the irradiation time. Stability test was carried by recording the temperature increase after irradiating the suspensions for 5 min, and re-irradiating the suspensions after their temperature deceases to room temperature. This process was repeated for 8 times. 2.5. Characterization Scanning electron microscopy (SEM) images were obtained by using a Hitachi S-4800 field emission microscope operated at 10 kV. Transmission electron microscopy (TEM) images were obtained with a JEOL model JEM 1011 operated at 100 kV and high-resolution transmission electron microscopy (HRTEM) images were obtained by using a JEOL model JEM 2100F operated at 200 kV. UV–vis–NIR spectra measurements were characterized on U-2800 (400–1000 nm) and lambda 950 (400–1400 nm) (PerkinElmer UV WinLab). Size and zeta potential analysis were recorded at 25 °C on Malvern ZetaSizer Instruments. Power X-ray diffraction (XRD) measurements were recorded on an EMPYREAN Diffractometer system with Cu Ka radiation (k1 = 1.540598 Å, k2 = 1.544426 Å, intensity ratio k2/k1 = 0.5) with samples on glass holder. 3. Results and discussion 3.1. Preparation of PS@Au core–shell nanospheres Scheme 1 illustrates the preparation procedure of hollow gold nanostructures. PS@Au core–shell nanostructures are fabricated
238
H. Wang et al. / Journal of Colloid and Interface Science 440 (2015) 236–244
via seeding-mediated approach firstly. After removing PS cores in DMF at 120 °C, gold nanocages and hollow gold nanospheres are obtained. PS spheres are modified with PEI for better absorption of gold seeds on PS surfaces. The surface potential reverses to positive after PS spheres successfully modified with PEI, and decreases when PVP stabilized Au nanoparticles (3 nm) absorbed on the modified PS surfaces as seeds, as evidenced by the results of zeta potential measurements (Table S2). Structures of PS@Au core–shell nanospheres are controlled by varying volume of growth solution as listed at Table S1. Fig. 1 shows SEM images of PS@Au core–shell nanospheres obtained by controlling growth solution volume. As shown in Fig. 1, the coverage of gold shell on PS spheres increases with increasing the volume of growth solution from 5 mL to 40 mL. The sizes of PS@Au core–shell nanospheres measured from the SEM images are 274.0 nm, 280.8 nm, 282.3 nm, 283.9 nm, 286.3 nm and 300.0 nm respectively, which are proportional to the volume of growth solution. Fig. S1 represents the relationship between diameters of PS@Au core–shell nanospheres and growth solution volume, and indicates that three stages happen in the process. The UV–vis extinction spectra (Fig. S2) also show that the extinction peaks red shift as the volumes of growth solution increase. And the shift of extinction peaks is coincident to former report [40]. Besides, gold nanoparticles grow well-distributed on the PS surface, and PS@Au core–shells are well uniform as shown in SEM images. Next, gold nanocages and hollow gold nanospheres are fabricated by removing PS templates of PS@Au core–shell nanospheres with thermal treatment in DMF. Three different types of PS@Au nanospheres are used in the next experiments. As shown from the contrast of TEM images in Fig. 2, these PS@Au nanospheres have clearly structure characteristics and are denoted to PS@Au (I) (Fig. 2a), PS@Au (II) (Fig. 2b) and PS@Au (III) (Fig. 2c). PS@Au (I) is obtained from 10 mL growth solution, which shows gold nanoparticles dispersed on PS without connection and has low coverage of gold nanoparticles on PS surfaces. PS@Au (II) is obtained from larger growth solutions (20 mL), and gold seeds grew to form shells on PS surfaces, however, there are still many gaps on the shell surfaces. PS@Au (III) is obtained from 40 mL growth solution and it has a continuous shell without gaps. The magnified TEM images show clearly that many irregular gaps on the shell of PS@Au (II), and continuous but rough shell on PS@Au (III), as indicated by arrows in Fig. 2 (d and e). The sizes of PS@Au (I), PS@Au (II) and PS@Au (III) are 280.8 nm, 283.9 nm and 300 nm. And the shells thicknesses of PS@Au (II) and PS@Au (III) are 10.2 nm and 15.7 nm. Dynamic light scattering (DLS) measurements also prove that they are well uniform (Fig. S3).
3.2. Preparation of hollow gold nanostructures As shown from the contrast in Fig. 2, the gaps spread all over the gold shells of PS@Au core–shell nanospheres, which decrease their shells mechanical strength and gold shells would collapse after PS cores are removed in DMF directly (Fig. S4). However, hollow nanostructures are obtained after thermal treatment of former PS@Au core–shell nanospheres in DMF at 120 °C for 12 h. As shown in Fig. 3, TEM and SEM images indicate that PS templates are removed after thermal treatment in DMF. SEM image of Fig. 3a shows that PS@Au (I) collapse like a shrunken gold capsule and cannot form hollow shells after thermal treatment at 120 °C. Even though gold nanoparticles have lower melt point than gold bulk, they still cannot melt at such low treatment temperature. As a result, the dispersed and discontinuous gold nanoparticles on PS@Au (I) cannot be connected and gold nanoshells collapse after thermal treatment. Gold nanocages were obtained after thermal treatment of PS@Au (II) in DMF, as shown in Fig. 3c. As observed from SEM images, many irregular holes are distributed over the smooth gold shell surfaces. The sizes decrease by about 35 nm from 283.9 nm (for PS@Au (II)) to 249.6 nm (for nanocages), and the nanoshells thicknesses increase approximately from 10.2 nm to 18.9 nm, as measured from SEM (Fig. 3c) and TEM (Fig. 3g) images. It proves that gold shells of PS@Au (II) reconstruct notably with size decreasing and thickness increasing upon thermal treatment. As a result, the reconstruction contributes to lower their surface-to-volume ratio and minimize the total surface energy after thermal treatment. Upon thermal treatment, gold nanoparticles deposited on PS@Au (II) may be energetically metastable and have a large driving force to reduce the surface area in order to minimize the total energy thus decreasing their coverage. By comparison of the shell characteristics before and after thermal treatment, these holes likely initiate along the gaps of gold nanoshells due to their shells reconstructions. Fig. 3(e and f) shows that the smooth hollow gold nanospheres were obtained after thermal treatment of PS@Au (III). The sizes of hollow gold nanospheres slightly decrease about 3.4 nm from 300 nm to 296.6 nm, as measured from SEM images. From the contrast of magnified TEM images (Fig. 3h), the thickness of hollow gold nanospheres slightly increases to 19.6 nm. These prove that continuous shells of PS@Au (III) just reconstruct surface to smooth without destroying gold shells. On the basis of the above results, hollow gold nanocages and nanospheres are well prepared by thermal treatment of PS@Au (II) and PS@Au (III). Besides, these PS@Au nanostructures were also treated in DMF at 150 °C, as shown in Fig. S5. The structures of hollow gold nanostructures are affected by thermal treatment
Scheme 1. Schematic illustration of the preparation of hollow gold nanostructures from PS@Au core–shell nanospheres.
H. Wang et al. / Journal of Colloid and Interface Science 440 (2015) 236–244
239
Fig. 1. SEM images of different surface structures of PS@Au core–shell nanospheres fabricated in various volume of growth solution (a): 5 mL, (b): 10 mL, (c): 15 mL, (d): 20 mL, (e): 25 mL and (f): 40 mL; Scale bars: 200 nm.
Fig. 2. TEM images of PS@Au core–shell nanospheres with different surface structure: (a) PS@Au (I); (b) PS@Au (II); (c) PS@Au (III), scale bars: 200 nm. Insert: magnified TEM images of PS@Au (II) (d) and PS@Au (III) (e), scale bars: 20 nm.
240
H. Wang et al. / Journal of Colloid and Interface Science 440 (2015) 236–244
Fig. 3. SEM and TEM images of gold nanostructures obtained by thermal treatment of PS@Au core–shell nanospheres in DMF at 120 °C for 12 h. (a) and (b) gold nanoparticle capsules (after thermal treatment of PS@Au (I)); (c) and (d) gold nanocages (after thermal treatment of PS@Au (II)); (e) and (f) hollow gold nanospheres (after thermal treatment of PS@Au (III)), scale bars: 200 nm. Inserts: magnified TEM images of gold nanocages (g) and hollow gold nanospheres (h), scale bars: 20 nm.
temperature, as proved by the changes in size and morphology. More specifically, for 150 °C thermal treatment, larger irregular gold nanoparticles and larger holes on nanocages are obtained as compared to those formed from thermal treatment of PS@Au (I) and PS@Au (II) at 120 °C. In contrast, the same sphere-shaped nanostructures are obtained after thermal treatment of PS@Au (III) at 120 °C and 150 °C, respectively. This result may due to their compact shell structures, thus the hollow gold nanoshells can be unaffected even at 150 °C. The effect of treating temperature and time on morphology of products will be discussed in this part. Taking PS@Au (II) as an example, Fig. 4 shows the SEM and TEM images of nanocages after thermal treatment of PS@Au (II) nanostructures at different temperature and time. As shown on the SEM images at Fig. 4(a–c), the gold nanoshells change clearly as thermal treatment temperature at 100 °C, 120 °C and 150 °C for 12 h. Firstly, the sizes of gold nanocages are 258.1 nm (Fig. 4a), 249.6 nm (Fig. 4b) and 228.2 nm (Fig. 4c) respectively, which decrease as thermal treatment temperature increasing from 100 °C to 150 °C. Secondly, the sizes of holes increase and the amounts of holes decrease as treating temperature increasing. As shown in the SEM images (Fig. 4(a–c)), many small holes at Fig. 4a and just several large holes at Fig. 4c on the cages which are obtained at 100 °C and 150 °C, respectively. Therefore, the shell reconstructions depend highly on the treatment temperature. It is also proved by the contrasts of TEM images (Fig. 4(d–f)). Fig. 4(g–l) shows the effect of treating time on the morphologies of gold nanocages at 120 °C. As clearly shown from
the SEM and TEM images, the sizes of nanocages decrease with the increase of treatment time. Besides, the sizes of holes increase and the amounts of holes decrease at longer treatment time. The results of time effect are similar to those of temperature. On the basis of the above results, the reconstructions are dramatically dependent both on thermal treatment temperature and time. By precisely controlling the thermal treatment conditions, it is possible to control the morphology of gold nanoshells. Consistent with the shape evolution observed by SEM and TEM, the evolution of UV–vis spectra of the nanocages solutions with the increase of treatment temperature and time is also recorded in Fig. S6. Obviously, after PS cores are removed by thermal treatment, SPR extinction spectra of gold nanocages start from 520 nm and their start absorption wavelength is shorter than that of PS@Au (II) (start from 600 nm). In addition, with increasing thermal treatment time (Fig. S6, a), we observe the emergence of a new absorption peak at 760 nm. On the other hand, the SPR extinction spectra broaden as increase the treatment temperature. As reported, the SPR absorption frequency is dependent highly on the nanoshells structures (sizes and shapes) and the SPR peaks change as their shells reconstructing. Therefore, the SPR results also confirm that the gold nanoshells reconstruction is affected by the treatment conditions. As shown from the above discussions, thermal treatment temperature is crucial to their shell structures of hollow nanostructures. After treated in room temperature, gold shell would not reconstruct and collapse after PS cores were removed. Therefore, heating is the key point to avoid the gold shell to collapse. Meanwhile, the
H. Wang et al. / Journal of Colloid and Interface Science 440 (2015) 236–244
241
Fig. 4. Effects of temperature and time on fabricating nanocages and hollow nanospheres. (a)–(f): SEM and TEM images of gold nanocages obtained by thermal treatment PS@Au (II) in DMF for 12 h at different temperatures: (a) and (d) 100 °C; (b) and (e) 120 °C; (c) and (f) 150 °C. (g)–(l): SEM and TEM images of gold nanocages obtained by thermal treatment PS@Au (II) in DMF at 120 °C for different times: (g) and (j) 4 h; (h) and (k) 8 h; (i) and (l) 24 h. Scale bars: 200 nm (a–i); 100 nm (j–l).
heating temperature must be well controlled in a suitable condition. As shown from the above results, hollow gold nanospheres obtained from high heating temperature would change their shell structures intensively. The reconstructions are attributed to the Ostwald ripening process. The gold shell reconstructions on the Ostwald ripening process depend highly on the thermal treatment temperature and time, and the process intensity increases with the increase of the treatment temperature and time. Therefore, we think that gold shells of PS@Au core–shell nanospheres would not be melted at the low treating temperature, however, gold shells would reconstruct via Ostwald ripening process and increase their mechanical strength. The most significant changes are crystallinity improvement of the gold shell after the Ostwald ripening process. Crystallinity characteristics of shells are analyzed by the examinations of high-resolution transmission electron microscopy (HRTEM) and X-ray diffraction (XRD) before and after thermal treatment. Fig. 5 shows the HRTEM images of PS@Au (II), PS@Au (III) before thermal treatment and corresponding gold nanocages, hollow gold nanospheres. The HRTEM images indicate that gold shells of PS@Au (II) and PS@Au (III) contain many small single Au crystals areas (Fig. 5a and b), which reveal that the gold shells of PS@Au (II) and PS@Au (III) appear to be polycrystalline in structure. While
the shells of gold nanocages and hollow gold nanospheres consist of large single Au crystals areas (Fig. 5c and d) after thermal treatment at 120 °C in DMF. It is clear that gold shells show good crystallization characteristics after thermal treatment as proved in these results. The fast Fourier transform (FFT) patterns of selected panes (Fig. 5, inserts) also confirm that. As shown from the spot FFT patterns, which represent that the gold shell crystallization characteristics improve after thermal treatment. Crystallization characteristics of PS@Au nanostructures are also analyzed by further examinations of XRD. Fig. 6 represents the XRD patterns of the PS@Au nanostructures and corresponding hollow gold nanostructures after thermal treatment. The XRD patterns exhibit the peaks at 2h angles of 38.21, 44.37, 64.67, and 77.64, corresponding to the reflections of (1 1 1), (2 0 0), (2 2 0), and (3 1 1) respectively, and these are in good agreement with the value in the literature (JCPDS NO. 04-0784). The PS@Au (I) nanostructures have poor diffraction peaks before and after thermal treatment, because of their small size and low loading capacity on PS surface [41]. The XRD patterns (Fig. 6a) of untreated PS@Au nanostructures show broad bands at all peaks, which indicate that as-fabricated PS@Au nanostructures are poor crystallization before thermal treatment. After thermal treatment of the PS@Au (II) and PS@Au (III), the intensities of diffraction peaks increase and peak widths
242
H. Wang et al. / Journal of Colloid and Interface Science 440 (2015) 236–244
Fig. 5. HRTEM analysis of the edges of PS@Au (II) (a), PS@Au (III) (b) before thermal treatment and nanocages (c) and hollow nanospheres (d). Scale bar: 5 nm. Inserts: FFT patterns of selected area shown in pane.
decrease, as shown in Fig. 6b. These indicate that the crystallization of hollow gold nanocages and gold nanospheres improve, as their shell reconstruct in thermal treatment. XRD analyses indicate that the reconstructions improve their crystallization after thermal treatment, which are well coincident to HRTEM results. 3.3. The SPR and photothermal property of gold nanocages and hollow gold nanospheres Gold core–shell nanospheres are used in a variety of applications based on their strong SPR absorption in NIR region. In order to investigate the SPR evolution in details, UV–vis–NIR spectra
are recorded in Fig. 7 and Fig. S7. As shown in Fig. S7, the SPR of PS@Au (I) represents a broad absorption band from 600 nm to 1000 nm, and center at 800 nm (Fig. S7, black). After thermal treatment, the SPR absorption band blue shift and we observe the emergence of a new broad absorption from 500 nm to 800 nm (Fig. S7, red). After thermal treatment, the original PS@Au (I) nanoshells plasmon feature disappeared, leaving only the broad absorption band from 500 nm to 800 nm, characteristic of large irregular spherical gold nanoparticles. The SEM and TEM images taken of final treatment product (Fig. 3a and b) also confirm that the original PS@Au (I) nanoshells have been destroyed. As shown in Fig. 7, the UV–vis–NIR spectra of PS@Au (II), PS@Au (III) and
H. Wang et al. / Journal of Colloid and Interface Science 440 (2015) 236–244
243
Fig. 6. XRD analyses of PS@Au (I) (black), PS@Au (II) (red) and PS@Au (III) (blue) before (a) and after (b) thermal treatment in DMF at 120 °C for 12 h. Length of the bars represent 1000 (a.u.) of intensity. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
Fig. 7. Normalized extinction spectra of synthesized PS@Au core–shell nanospheres (a) before thermal treatment and hollow gold nanostructures and (b) obtained after thermal treatment. (a), PS@Au (II) (red), and PS@Au (III) (blue); (b), gold nanocages (red) and hollow gold nanospheres (blue). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
corresponding hollow gold nanostructures are recorded. The characteristic SPR peaks of PS@Au (II) and PS@Au (III) (Fig. 7a) show platform at the wavelength longer than 800 nm. Besides, PS@Au (III) shows a weak shoulder absorption peak at 810 nm. After thermal treatment of PS@Au (II), we observe the emergence of two new absorption peaks of gold nanocages at 750 nm and 1070 nm (Fig. 7b, red). This is likely due to the removing of PS cores and the reconstruction of gold nanoshells. After thermal treatment of PS@Au (III), the absorption peak of hollow gold nanospheres slightly blue-shifts to 800 nm as compared to the peak of PS@Au (III) (810 nm) and we also observe the emergence of absorption peak at 1120 nm in NIR region. Previous studies have determined that the absorption peak of gold nanoshells blue shifts with increasing shell thickness [40]. The blue-shifted SPR results confirm that gold nanoshells thicknesses increase after thermal treatment of PS@Au (III), and the results are consistent with the TEM images (Fig. 3h).
The hollow gold nanocages and nanospheres show special NIR optical property, and can be used as photosensitizer. In our experiments, their photothermal conversion property were investigated via laser irradiation at k = 808 nm at a power density of 0.5 W cm 2 for various time. Fig. 8a shows temperature increase of gold nanocages and hollow gold nanospheres by irradiation at the laser power density of 0.5 W cm 2. It indicates that the suspension temperatures of both gold nanocages and hollow gold nanospheres increase after laser irradiation. It also represents clearly that the temperature of gold nanocages suspension increases fast by 4– 6 °C within one minute, and the temperature increase (DT) of gold nanocages is higher than that of hollow gold nanospheres after irradiation for 10 min at same concentration of particles. Gold nanocages show better photothermal conversion property. The suspensions of gold nanocages show higher temperature increase compared with that of gold nanospheres. This is likely due to the presence of holes on the gold nanoshells surface. In the presence
Fig. 8. (a) The plots of temperature increase (DT) of suspensions of gold nanocages (red) and hollow gold nanospheres (blue) as a function of irradiation time at laser power density of 0.5 W cm 2. (b) Stability test of photothermal property of gold nanocages (red) and hollow gold nanospheres (blue). Temperature increase (DT) after irradiation for 5 min at laser power density of 0.5 W cm 2. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
244
H. Wang et al. / Journal of Colloid and Interface Science 440 (2015) 236–244
of gold nanocages, the temperature increases dramatically by 11– 12 °C as irradiation for 10 min. As a result of denaturation of biomolecules at high temperature above 42 °C, cells and tissues can be irreversible damaged. Therefore, gold nanocages suspensions have ability to kill cancer tissues with NIR laser irradiation. In addition, for a photosensitizer to be useful, it should be stable under repeated NIR laser irradiation. To evaluate the stability of photothermal property, we carried out the additional experiment to irradiate their suspensions under 808 nm laser cycled for eight times. Fig. 8b shows that both gold nanocages and hollow gold nanospheres have stable photothermal property after eight repeats. 4. Conclusions Hollow gold nanostructures (nanocages and nanospheres) were fabricated by removing PS core of PS@Au in DMF solvent. The shell structures were controlled by the PS@Au core–shell structures, treatment temperature and time. These hollow gold nanostructures could be obtained easily by thermal treatment below 120 °C over 4 h. Hollow gold nanocages were obtained by thermal treatment of PS@Au (II) with gaps on gold shells, and hollow gold nanospheres were fabricated by thermal treatment of PS@Au (III) with continuous gold shell. These hollow nanostructures represented special NIR optical property. Both the gold nanocages and the gold nanospheres efficiently convert 808 nm laser into heat, and have stable photothermal property. Moreover, due to the holes on gold nanoshells surfaces, gold nanocages show higher temperature rise than that of gold nanospheres at the same particle concentration. Acknowledgment The authors would like to acknowledge funding from the National Natural Science Foundation of China (Grant Nos. 20903106, 20933007, 91127012, 21161130521, 21321063), and the Fund of Chinese Academy of Sciences (CMS-PY-201302). Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.jcis.2014.11.004. References [1] S.E. Skrabalak, J. Chen, Y. Sun, X. Lu, L. Au, C.M. Cobley, Y. Xia, Acc. Chem. Res. 41 (2008) 1587.
[2] E.C. Dreaden, A.M. Alkilany, X. Huang, C.J. Murphy, M.A. El-Sayed, Chem. Soc. Rev. 41 (2012) 2740. [3] S. Oldenburg, R. Averitt, S. Westcott, N. Halas, Chem. Phys. Lett. 288 (1998) 243. [4] H.-P. Liang, L.-J. Wan, C.-L. Bai, L. Jiang, J. Phys. Chem. B 109 (2005) 7795. [5] P.K. Jain, M.A. El-Sayed, J. Phys. Chem. C 111 (2007) 17451. [6] E. Prodan, C. Radloff, N.J. Halas, P. Nordlander, Science 302 (2003) 419. [7] M.P. Melancon, M. Zhou, C. Li, Acc. Chem. Res. 44 (2011) 947. [8] Y. Wang, Y. Liu, H. Luehmann, X. Xia, D. Wan, C. Cutler, Y. Xia, Nano Lett. 13 (2013) 581. [9] J. Chen, F. Saeki, B.J. Wiley, H. Cang, M.J. Cobb, Z.-Y. Li, L. Au, H. Zhang, M.B. Kimmey, X. Li, Nano Lett. 5 (2005) 473. [10] Y. Zhang, F. Xu, Y. Sun, C. Guo, K. Cui, Y. Shi, Z. Wen, Z. Li, Chem. Eur. J. 16 (2010) 9248. [11] S.E. Skrabalak, J. Chen, L. Au, X. Lu, X. Li, Y. Xia, Adv. Mater. 19 (2007) 3177. [12] J. Chen, C. Glaus, R. Laforest, Q. Zhang, M. Yang, M. Gidding, M.J. Welch, Y. Xia, Small 6 (2010) 811. [13] J. You, G. Zhang, C. Li, ACS Nano 4 (2010) 1033. [14] M.S. Yavuz, Y. Cheng, J. Chen, C.M. Cobley, Q. Zhang, M. Rycenga, J. Xie, C. Kim, K.H. Song, A.G. Schwartz, L.V. Wang, Y. Xia, Nat. Mater. 8 (2009) 935. [15] M.A. Mahmoud, R. Narayanan, M.A. El-Sayed, Acc. Chem. Res. 46 (2013) 1795. [16] M.A. Mahmoud, F. Saira, M.A. El-Sayed, Nano Lett. 10 (2010) 3764. [17] M.A. Mahmoud, D. O’Neil, M.A. El-Sayed, Chem. Mater. 26 (2013) 44. [18] W. Wang, Y. Han, M. Tian, Y. Fan, Y. Tang, M. Gao, Y. Wang, ACS Appl. Mater. Interfaces 5 (2013) 5709. [19] Y. Sun, Y. Xia, Science 298 (2002) 2176. [20] H.N. Xie, I.A. Larmour, Y.C. Chen, A.W. Wark, V. Tileli, D.W. McComb, K. Faulds, D. Graham, Nanoscale 5 (2013) 765. [21] F. Caruso, R.A. Caruso, H. Mohwald, Science 282 (1998) 1111. [22] A. Imhof, Langmuir 17 (2001) 3579. [23] S.-W. Kim, M. Kim, W.Y. Lee, T. Hyeon, J. Am. Chem. Soc. 124 (2002) 7642. [24] Z. Chen, P. Zhan, Z. Wang, J. Zhang, W. Zhang, N. Ming, C.T. Chan, P. Sheng, Adv. Mater. 16 (2004) 417. [25] P. Jiang, J.F. Bertone, V.L. Colvin, Science 291 (2001) 453. [26] Y. Xia, W. Lu, L. Jiang, Nanotechnology 21 (2010) 085501. [27] P.M. Arnal, M. Comotti, F. Schueth, Angew. Chem. Int. Ed. 45 (2006) 8224. [28] B.G. Prevo, J.C. Fuller, O.D. Velev, Chem. Mater. 17 (2005) 28. [29] H. Yang, Y. Liu, Q. Shen, L. Chen, W. You, X. Wang, J. Sheng, J. Mater. Chem. 22 (2012) 24132. [30] Z. Liang, A. Susha, F. Caruso, Chem. Mater. 15 (2003) 3176. [31] D.G. Shchukin, T. Shutava, E. Shchukina, G.B. Sukhorukov, Y.M. Lvov, Chem. Mater. 16 (2004) 3446. [32] C. Mangeney, S. Bousalem, C. Connan, M.-J. Vaulay, S. Bernard, M.M. Chehimi, Langmuir 22 (2006) 10163. [33] Y. Xia, J. Li, L. Jiang, J. Colloid Interface Sci. 377 (2012) 34. [34] Z. Wu, C. Dong, Y. Li, H. Hao, H. Zhang, Z. Lu, B. Yang, Angew. Chem. Int. Ed. 52 (2013) 9952. [35] Y.X. Zhang, H.C. Zeng, Langmuir 24 (2008) 3740. [36] H. Sun, M. Yu, G. Wang, X. Sun, J. Lian, J. Phys. Chem. C 116 (2012) 9000. [37] G. Andreasen, P. Schilardi, O. Azzaroni, R. Salvarezza, Langmuir 18 (2002) 10430. [38] C. Damle, K. Biswas, M. Sastry, Langmuir 17 (2001) 7156. [39] W. Cui, J. Li, Y. Zhang, H. Rong, W. Lu, L. Jiang, Nanomed.: Nanotechnol. Biol. Med. 8 (2012) 46. [40] W. Shi, Y. Sahoo, M.T. Swihart, P. Prasad, Langmuir 21 (2005) 1610. [41] D.V. Leff, P.C. Ohara, J.R. Heath, W.M. Gelbart, J. Phys. Chem. 99 (1995) 7036.
Contents lists available at ScienceDirect
Journal of Colloid and Interface Science www.elsevier.com/locate/jcis
Facile preparation of gold nanocages and hollow gold nanospheres via solvent thermal treatment and their surface plasmon resonance and photothermal properties Haifei Wang, Jing Han, Wensheng Lu ⇑, Jianping Zhang, Jinru Li, Long Jiang ⇑ Beijing National Laboratory for Molecular Science, Key Laboratory of Colloid, Interface and Thermodynamics, Institute of Chemistry, Chinese Academy of Sciences, Beijing 100190, PR China
a r t i c l e
i n f o
Article history: Received 17 July 2014 Accepted 3 November 2014 Available online 11 November 2014 Keywords: Core–shell nanostructures Hollow gold nanostructures Solvent thermal treatment Surface plasmon resonance (SPR) absorptions Photothermal conversion property
a b s t r a c t Although template etching method is one of the most common ways of preparation of hollow gold nanostructures, this approach still requires further improvements to avoid the collapse of gold shells after the cores were removed. In this work, an improved template etching method, with which hollow gold nanostructure is fabricated by etching Polystyrene (PS) cores from PS@Au core–shell nanospheres with solvent thermal treatment in N,N-Dimethylformamide (DMF), is demonstrated. When PS cores were removed by a thermal treatment process, gold nanoshells reconstruct and the collapse of the nanoshells is avoided. Gold nanocages and hollow gold nanospheres are easily obtained from the various structures of PS@Au core–shell nanospheres. These hollow nanostructures represent special near infrared (NIR) optical property and photothermal property. Compared with hollow gold nanospheres, the gold nanocages show higher temperature increase at the same particle concentration. Ó 2014 Elsevier Inc. All rights reserved.
1. Introduction Hollow gold nanostructures (especially gold nanocages and hollow gold nanospheres) exhibit unique optical property due to their tunable surface plasmon resonance (SPR) absorptions, and have attracted extensive research attentions [1–3]. The SPR absorptions can be tuned from the visible region to the near-infrared (NIR) region by changing the size and thickness of gold shells [4–6]. The hollow gold nanostructures have potential applications in cell imaging [7–9], biosensor [10], photothermal therapy of cancers [11,12], controllable drug release [13,14] and catalysis [15–17]. Currently, a series of methods on preparation of hollow gold nanostructures with tunable optical property have been researched in some groups. The most common approaches for the preparation of hollow gold nanostructure involve growing gold nanoparticles on soft surfactant aggregates templates [18], Abbreviations: SPR, surface plasmon resonance; NIR, near infrared; PS, polystyrene; DMF, N,N-Dimethylformamide; PEI, poly (ethyleneimine); PVP, polyvinypyrrolidone; SEM, scanning electron microscopy; TEM, transmission electron microscopy; HRTEM, high-resolution transmission electron microscopy; XRD, power X-ray diffraction; DLS, dynamic light scattering; UV–vis–NIR, ultraviolet– visible–near infrared. ⇑ Corresponding authors. E-mail addresses: [email protected] (W. Lu), [email protected] (L. Jiang). http://dx.doi.org/10.1016/j.jcis.2014.11.004 0021-9797/Ó 2014 Elsevier Inc. All rights reserved.
galvanic replace reaction by using sacrificial metal templates [4,19,20], and selective etching or calcination gold core–shell nanostructures [21–24]. As to the selective etching and calcination methods, hollow gold nanostructures were usually fabricated by depositing gold shells on PS sphere templates, and then selectively removed templates via calcination or chemical etching. PS sphere templates have easily functionalized surface and broad size range and have therefore been frequently used [25,26], and the way of fabricating gold core–shell nanostructures on colloidal surfaces has been widely researched since Halas’ pioneering work on preparation of gold core–shell nanospheres [3]. However, the gold shells of hollow nanostructures would be collapsed or destroyed after the PS cores were selectively removed. The main challenge in the preparation of hollow gold nanoshells is the difficulty of removing PS templates. There are reports on successfully removing PS templates by calcination at high temperature (above 300 °C), but the processes are rather complex because the gold shells would be destroyed if they are not protected before calcination [27–30]. Other method is dissolution of PS core into an organic solvent. Even though PS can be easily dissolved in Tetrahydrofuran (THF) and Dimethylformamide (DMF), gold shell would collapse after the PS cores were removed due to its low mechanical strength [31,32]. To date, most PS@metal core–shell nanostructures have been applied without core removed, and caused performance
H. Wang et al. / Journal of Colloid and Interface Science 440 (2015) 236–244
degradation of the core–shell nanostructures [33]. Therefore, it is essential to develop an effective and facile method of fabricating hollow gold nanostructures. Compared with the calcination method, the chemical etching method will be a promising route to remove PS templates at low temperature, if troubles of shell collapse are solved. As well known, metal nanoparticles grow larger by Ostwald ripening at the expense of smaller ones due to chemical potential energy difference. Moreover, metal nanoparticles would reconstruct at lower temperature than the melting point of blocks. There are reports on reconstructing gold nanoparticles upon the thermal treatment [34–38]. In our work, to achieve perfect hollow gold shells, the thermal treatment process was used in solvent etching method to solve the trouble of gold shell collapse. When PS@Au core–shell nanospheres were treated by heating in DMF solvent, PS cores were etched out and gold nanoshells reconstructed to improve their mechanical strength, so the gold nanoshells would not collapse after PS cores were removed. Herein, we report on the preparation of gold nanocages and hollow gold nanospheres from PS@Au core–shell nanospheres by solvent thermal treatment in DMF. In this process, gold nanoshells reconstruct in Ostwald ripening process by thermal treatment (below 120 °C). Hollow gold nanostructures are obtained without shells collapse. Moreover, various shell structures of hollow nanostructures (nanocages or hollow nanospheres) are controllably obtained from variable surface structures of PS@Au core–shell nanospheres. The SPR features of these hollow nanostructures are investigated, and they have strong SPR absorption in NIR region. In addition, these hollow gold nanostructures have stable photothermal effect, and hollow gold nanocages have higher temperature increase than that of hollow gold nanospheres at the same particle concentration. 2. Experimental section 2.1. Materials PS spheres (about 260 nm in diameter) were used as received. HAuCl44H2O was obtained from Shanghai Reagent Company. Poly (ethyleneimine) branched (PEI, Mn = 10,000) was purchased from Sigma–Aldrich. N,N-Dimethylformamide (DMF) and potassium carbonate (K2CO3), ammonium hydroxide (NH3H2O, 28–30%), formaldehyde (HCHO 37%), sodium borohydride (NaBH4), and polyvinypyrrolidone (PVP, Mw = 30,000) were received from Beijing Chemical Company. All chemicals were analytical grade and used without further purification. Deionized water used for all experiments was treated with a Millipore water purification system (Millipore Corp.). 2.2. Preparation of the PS@Au core–shell nanospheres The PS spheres were modified with PEI layer via electrostatic interaction before used. 2.5 mL suspension of negatively charged sulfate-stabilized PS spheres (87 mg mL 1) was added to excess positive charged PEI solution (10 mL, 0.6 mg mL 1), and was mixed for 30 min. The PEI modified PS spheres were purified by repeated centrifuging (10,000 rpm, 20 min). Then the PEI modified positive charged PS spheres (PS-PEI) were redispersed into 25 mL water (8.7 mg mL 1). PVP protected gold nanoparticles were prepared according to the literature [39] and were used as seeds. Gold seeds were absorbed on modified PS spheres as following: 200 lL, PS-PEI was added to 8 mL newly prepared gold sol quickly. After 30 min, the excess gold sol was removed by centrifuging (8000 rpm, 5 min) and washed one time. Then, gold seeds on PS spheres (PS@Au seeds) were obtained. PS@Au core–shell nanospheres were prepared via a seeding-mediated growth approach in 0.3 mM HAuCl4 growth solution. Growth solutions were prepared as
237
following [40]: 50 mg of K2CO3 was dissolved in 200 mL water. After 10 min of stirring, 1.41 mL of HAuCl4 (48 mM) was added. The solution became colorless after 30 min and was aged for 24 h in the dark for use in subsequent steps. PS@Au core–shell nanospheres were fabricated by adding former PS@Au seeds to 10 mL growth solution with PVP (1 mg mL 1) as stabilizer, and 50 lL NH3H2O (28–30%) was added to adjust pH. Then, 60 lL formaldehyde was added and reacted for 12 h. Different PS@Au core–shell nanostructures were obtained by adjusting the volume of HAuCl4 (5 mL, 10 mL, 15 mL, 20 mL, 25 mL, and 40 mL) and formaldehyde (Table S1 in Supporting Information). Finally, PS@Au was washed three times by centrifuging (5000 rpm, 5 min), and redispersed into water. 2.3. Preparation of Au nanocages and hollow Au nanospheres Hollow gold nanostructures were obtained by removing the PS templates in heated DMF. PS@Au core–shell nanospheres (prepared in 20 mL growth solution) were added into 20 mL DMF with stirring. With the heating for 12 h at 120 °C, the PS templates were removed and hollow gold nanostructures were obtained. Gold nanocages and hollow gold nanospheres were prepared from variable PS@Au core–shell nanostructures. Gold nanostructures were washed with DMF to remove the PS in solution by centrifuging (5000 rpm, 5 min) three times, and redispersed into water. Effects of temperature and time of thermal treatment were also investigated. 2.4. Photothermal conversion analysis The photothermal experiments of gold nanocages and hollow gold nanospheres were carried at the same particle concentration. The same amounts of PS spheres were used at all the core–shell nanostructures, so the corresponding hollow gold nanostructures were the same particle concentration. The suspensions of gold nanocages and hollow gold nanospheres were irradiated at a power density of 0.5 W cm 2 by 808 nm laser. The laser spot was altered to cover the surface of the suspensions. The temperature increases of the suspensions were recorded as a function of the irradiation time. Stability test was carried by recording the temperature increase after irradiating the suspensions for 5 min, and re-irradiating the suspensions after their temperature deceases to room temperature. This process was repeated for 8 times. 2.5. Characterization Scanning electron microscopy (SEM) images were obtained by using a Hitachi S-4800 field emission microscope operated at 10 kV. Transmission electron microscopy (TEM) images were obtained with a JEOL model JEM 1011 operated at 100 kV and high-resolution transmission electron microscopy (HRTEM) images were obtained by using a JEOL model JEM 2100F operated at 200 kV. UV–vis–NIR spectra measurements were characterized on U-2800 (400–1000 nm) and lambda 950 (400–1400 nm) (PerkinElmer UV WinLab). Size and zeta potential analysis were recorded at 25 °C on Malvern ZetaSizer Instruments. Power X-ray diffraction (XRD) measurements were recorded on an EMPYREAN Diffractometer system with Cu Ka radiation (k1 = 1.540598 Å, k2 = 1.544426 Å, intensity ratio k2/k1 = 0.5) with samples on glass holder. 3. Results and discussion 3.1. Preparation of PS@Au core–shell nanospheres Scheme 1 illustrates the preparation procedure of hollow gold nanostructures. PS@Au core–shell nanostructures are fabricated
238
H. Wang et al. / Journal of Colloid and Interface Science 440 (2015) 236–244
via seeding-mediated approach firstly. After removing PS cores in DMF at 120 °C, gold nanocages and hollow gold nanospheres are obtained. PS spheres are modified with PEI for better absorption of gold seeds on PS surfaces. The surface potential reverses to positive after PS spheres successfully modified with PEI, and decreases when PVP stabilized Au nanoparticles (3 nm) absorbed on the modified PS surfaces as seeds, as evidenced by the results of zeta potential measurements (Table S2). Structures of PS@Au core–shell nanospheres are controlled by varying volume of growth solution as listed at Table S1. Fig. 1 shows SEM images of PS@Au core–shell nanospheres obtained by controlling growth solution volume. As shown in Fig. 1, the coverage of gold shell on PS spheres increases with increasing the volume of growth solution from 5 mL to 40 mL. The sizes of PS@Au core–shell nanospheres measured from the SEM images are 274.0 nm, 280.8 nm, 282.3 nm, 283.9 nm, 286.3 nm and 300.0 nm respectively, which are proportional to the volume of growth solution. Fig. S1 represents the relationship between diameters of PS@Au core–shell nanospheres and growth solution volume, and indicates that three stages happen in the process. The UV–vis extinction spectra (Fig. S2) also show that the extinction peaks red shift as the volumes of growth solution increase. And the shift of extinction peaks is coincident to former report [40]. Besides, gold nanoparticles grow well-distributed on the PS surface, and PS@Au core–shells are well uniform as shown in SEM images. Next, gold nanocages and hollow gold nanospheres are fabricated by removing PS templates of PS@Au core–shell nanospheres with thermal treatment in DMF. Three different types of PS@Au nanospheres are used in the next experiments. As shown from the contrast of TEM images in Fig. 2, these PS@Au nanospheres have clearly structure characteristics and are denoted to PS@Au (I) (Fig. 2a), PS@Au (II) (Fig. 2b) and PS@Au (III) (Fig. 2c). PS@Au (I) is obtained from 10 mL growth solution, which shows gold nanoparticles dispersed on PS without connection and has low coverage of gold nanoparticles on PS surfaces. PS@Au (II) is obtained from larger growth solutions (20 mL), and gold seeds grew to form shells on PS surfaces, however, there are still many gaps on the shell surfaces. PS@Au (III) is obtained from 40 mL growth solution and it has a continuous shell without gaps. The magnified TEM images show clearly that many irregular gaps on the shell of PS@Au (II), and continuous but rough shell on PS@Au (III), as indicated by arrows in Fig. 2 (d and e). The sizes of PS@Au (I), PS@Au (II) and PS@Au (III) are 280.8 nm, 283.9 nm and 300 nm. And the shells thicknesses of PS@Au (II) and PS@Au (III) are 10.2 nm and 15.7 nm. Dynamic light scattering (DLS) measurements also prove that they are well uniform (Fig. S3).
3.2. Preparation of hollow gold nanostructures As shown from the contrast in Fig. 2, the gaps spread all over the gold shells of PS@Au core–shell nanospheres, which decrease their shells mechanical strength and gold shells would collapse after PS cores are removed in DMF directly (Fig. S4). However, hollow nanostructures are obtained after thermal treatment of former PS@Au core–shell nanospheres in DMF at 120 °C for 12 h. As shown in Fig. 3, TEM and SEM images indicate that PS templates are removed after thermal treatment in DMF. SEM image of Fig. 3a shows that PS@Au (I) collapse like a shrunken gold capsule and cannot form hollow shells after thermal treatment at 120 °C. Even though gold nanoparticles have lower melt point than gold bulk, they still cannot melt at such low treatment temperature. As a result, the dispersed and discontinuous gold nanoparticles on PS@Au (I) cannot be connected and gold nanoshells collapse after thermal treatment. Gold nanocages were obtained after thermal treatment of PS@Au (II) in DMF, as shown in Fig. 3c. As observed from SEM images, many irregular holes are distributed over the smooth gold shell surfaces. The sizes decrease by about 35 nm from 283.9 nm (for PS@Au (II)) to 249.6 nm (for nanocages), and the nanoshells thicknesses increase approximately from 10.2 nm to 18.9 nm, as measured from SEM (Fig. 3c) and TEM (Fig. 3g) images. It proves that gold shells of PS@Au (II) reconstruct notably with size decreasing and thickness increasing upon thermal treatment. As a result, the reconstruction contributes to lower their surface-to-volume ratio and minimize the total surface energy after thermal treatment. Upon thermal treatment, gold nanoparticles deposited on PS@Au (II) may be energetically metastable and have a large driving force to reduce the surface area in order to minimize the total energy thus decreasing their coverage. By comparison of the shell characteristics before and after thermal treatment, these holes likely initiate along the gaps of gold nanoshells due to their shells reconstructions. Fig. 3(e and f) shows that the smooth hollow gold nanospheres were obtained after thermal treatment of PS@Au (III). The sizes of hollow gold nanospheres slightly decrease about 3.4 nm from 300 nm to 296.6 nm, as measured from SEM images. From the contrast of magnified TEM images (Fig. 3h), the thickness of hollow gold nanospheres slightly increases to 19.6 nm. These prove that continuous shells of PS@Au (III) just reconstruct surface to smooth without destroying gold shells. On the basis of the above results, hollow gold nanocages and nanospheres are well prepared by thermal treatment of PS@Au (II) and PS@Au (III). Besides, these PS@Au nanostructures were also treated in DMF at 150 °C, as shown in Fig. S5. The structures of hollow gold nanostructures are affected by thermal treatment
Scheme 1. Schematic illustration of the preparation of hollow gold nanostructures from PS@Au core–shell nanospheres.
H. Wang et al. / Journal of Colloid and Interface Science 440 (2015) 236–244
239
Fig. 1. SEM images of different surface structures of PS@Au core–shell nanospheres fabricated in various volume of growth solution (a): 5 mL, (b): 10 mL, (c): 15 mL, (d): 20 mL, (e): 25 mL and (f): 40 mL; Scale bars: 200 nm.
Fig. 2. TEM images of PS@Au core–shell nanospheres with different surface structure: (a) PS@Au (I); (b) PS@Au (II); (c) PS@Au (III), scale bars: 200 nm. Insert: magnified TEM images of PS@Au (II) (d) and PS@Au (III) (e), scale bars: 20 nm.
240
H. Wang et al. / Journal of Colloid and Interface Science 440 (2015) 236–244
Fig. 3. SEM and TEM images of gold nanostructures obtained by thermal treatment of PS@Au core–shell nanospheres in DMF at 120 °C for 12 h. (a) and (b) gold nanoparticle capsules (after thermal treatment of PS@Au (I)); (c) and (d) gold nanocages (after thermal treatment of PS@Au (II)); (e) and (f) hollow gold nanospheres (after thermal treatment of PS@Au (III)), scale bars: 200 nm. Inserts: magnified TEM images of gold nanocages (g) and hollow gold nanospheres (h), scale bars: 20 nm.
temperature, as proved by the changes in size and morphology. More specifically, for 150 °C thermal treatment, larger irregular gold nanoparticles and larger holes on nanocages are obtained as compared to those formed from thermal treatment of PS@Au (I) and PS@Au (II) at 120 °C. In contrast, the same sphere-shaped nanostructures are obtained after thermal treatment of PS@Au (III) at 120 °C and 150 °C, respectively. This result may due to their compact shell structures, thus the hollow gold nanoshells can be unaffected even at 150 °C. The effect of treating temperature and time on morphology of products will be discussed in this part. Taking PS@Au (II) as an example, Fig. 4 shows the SEM and TEM images of nanocages after thermal treatment of PS@Au (II) nanostructures at different temperature and time. As shown on the SEM images at Fig. 4(a–c), the gold nanoshells change clearly as thermal treatment temperature at 100 °C, 120 °C and 150 °C for 12 h. Firstly, the sizes of gold nanocages are 258.1 nm (Fig. 4a), 249.6 nm (Fig. 4b) and 228.2 nm (Fig. 4c) respectively, which decrease as thermal treatment temperature increasing from 100 °C to 150 °C. Secondly, the sizes of holes increase and the amounts of holes decrease as treating temperature increasing. As shown in the SEM images (Fig. 4(a–c)), many small holes at Fig. 4a and just several large holes at Fig. 4c on the cages which are obtained at 100 °C and 150 °C, respectively. Therefore, the shell reconstructions depend highly on the treatment temperature. It is also proved by the contrasts of TEM images (Fig. 4(d–f)). Fig. 4(g–l) shows the effect of treating time on the morphologies of gold nanocages at 120 °C. As clearly shown from
the SEM and TEM images, the sizes of nanocages decrease with the increase of treatment time. Besides, the sizes of holes increase and the amounts of holes decrease at longer treatment time. The results of time effect are similar to those of temperature. On the basis of the above results, the reconstructions are dramatically dependent both on thermal treatment temperature and time. By precisely controlling the thermal treatment conditions, it is possible to control the morphology of gold nanoshells. Consistent with the shape evolution observed by SEM and TEM, the evolution of UV–vis spectra of the nanocages solutions with the increase of treatment temperature and time is also recorded in Fig. S6. Obviously, after PS cores are removed by thermal treatment, SPR extinction spectra of gold nanocages start from 520 nm and their start absorption wavelength is shorter than that of PS@Au (II) (start from 600 nm). In addition, with increasing thermal treatment time (Fig. S6, a), we observe the emergence of a new absorption peak at 760 nm. On the other hand, the SPR extinction spectra broaden as increase the treatment temperature. As reported, the SPR absorption frequency is dependent highly on the nanoshells structures (sizes and shapes) and the SPR peaks change as their shells reconstructing. Therefore, the SPR results also confirm that the gold nanoshells reconstruction is affected by the treatment conditions. As shown from the above discussions, thermal treatment temperature is crucial to their shell structures of hollow nanostructures. After treated in room temperature, gold shell would not reconstruct and collapse after PS cores were removed. Therefore, heating is the key point to avoid the gold shell to collapse. Meanwhile, the
H. Wang et al. / Journal of Colloid and Interface Science 440 (2015) 236–244
241
Fig. 4. Effects of temperature and time on fabricating nanocages and hollow nanospheres. (a)–(f): SEM and TEM images of gold nanocages obtained by thermal treatment PS@Au (II) in DMF for 12 h at different temperatures: (a) and (d) 100 °C; (b) and (e) 120 °C; (c) and (f) 150 °C. (g)–(l): SEM and TEM images of gold nanocages obtained by thermal treatment PS@Au (II) in DMF at 120 °C for different times: (g) and (j) 4 h; (h) and (k) 8 h; (i) and (l) 24 h. Scale bars: 200 nm (a–i); 100 nm (j–l).
heating temperature must be well controlled in a suitable condition. As shown from the above results, hollow gold nanospheres obtained from high heating temperature would change their shell structures intensively. The reconstructions are attributed to the Ostwald ripening process. The gold shell reconstructions on the Ostwald ripening process depend highly on the thermal treatment temperature and time, and the process intensity increases with the increase of the treatment temperature and time. Therefore, we think that gold shells of PS@Au core–shell nanospheres would not be melted at the low treating temperature, however, gold shells would reconstruct via Ostwald ripening process and increase their mechanical strength. The most significant changes are crystallinity improvement of the gold shell after the Ostwald ripening process. Crystallinity characteristics of shells are analyzed by the examinations of high-resolution transmission electron microscopy (HRTEM) and X-ray diffraction (XRD) before and after thermal treatment. Fig. 5 shows the HRTEM images of PS@Au (II), PS@Au (III) before thermal treatment and corresponding gold nanocages, hollow gold nanospheres. The HRTEM images indicate that gold shells of PS@Au (II) and PS@Au (III) contain many small single Au crystals areas (Fig. 5a and b), which reveal that the gold shells of PS@Au (II) and PS@Au (III) appear to be polycrystalline in structure. While
the shells of gold nanocages and hollow gold nanospheres consist of large single Au crystals areas (Fig. 5c and d) after thermal treatment at 120 °C in DMF. It is clear that gold shells show good crystallization characteristics after thermal treatment as proved in these results. The fast Fourier transform (FFT) patterns of selected panes (Fig. 5, inserts) also confirm that. As shown from the spot FFT patterns, which represent that the gold shell crystallization characteristics improve after thermal treatment. Crystallization characteristics of PS@Au nanostructures are also analyzed by further examinations of XRD. Fig. 6 represents the XRD patterns of the PS@Au nanostructures and corresponding hollow gold nanostructures after thermal treatment. The XRD patterns exhibit the peaks at 2h angles of 38.21, 44.37, 64.67, and 77.64, corresponding to the reflections of (1 1 1), (2 0 0), (2 2 0), and (3 1 1) respectively, and these are in good agreement with the value in the literature (JCPDS NO. 04-0784). The PS@Au (I) nanostructures have poor diffraction peaks before and after thermal treatment, because of their small size and low loading capacity on PS surface [41]. The XRD patterns (Fig. 6a) of untreated PS@Au nanostructures show broad bands at all peaks, which indicate that as-fabricated PS@Au nanostructures are poor crystallization before thermal treatment. After thermal treatment of the PS@Au (II) and PS@Au (III), the intensities of diffraction peaks increase and peak widths
242
H. Wang et al. / Journal of Colloid and Interface Science 440 (2015) 236–244
Fig. 5. HRTEM analysis of the edges of PS@Au (II) (a), PS@Au (III) (b) before thermal treatment and nanocages (c) and hollow nanospheres (d). Scale bar: 5 nm. Inserts: FFT patterns of selected area shown in pane.
decrease, as shown in Fig. 6b. These indicate that the crystallization of hollow gold nanocages and gold nanospheres improve, as their shell reconstruct in thermal treatment. XRD analyses indicate that the reconstructions improve their crystallization after thermal treatment, which are well coincident to HRTEM results. 3.3. The SPR and photothermal property of gold nanocages and hollow gold nanospheres Gold core–shell nanospheres are used in a variety of applications based on their strong SPR absorption in NIR region. In order to investigate the SPR evolution in details, UV–vis–NIR spectra
are recorded in Fig. 7 and Fig. S7. As shown in Fig. S7, the SPR of PS@Au (I) represents a broad absorption band from 600 nm to 1000 nm, and center at 800 nm (Fig. S7, black). After thermal treatment, the SPR absorption band blue shift and we observe the emergence of a new broad absorption from 500 nm to 800 nm (Fig. S7, red). After thermal treatment, the original PS@Au (I) nanoshells plasmon feature disappeared, leaving only the broad absorption band from 500 nm to 800 nm, characteristic of large irregular spherical gold nanoparticles. The SEM and TEM images taken of final treatment product (Fig. 3a and b) also confirm that the original PS@Au (I) nanoshells have been destroyed. As shown in Fig. 7, the UV–vis–NIR spectra of PS@Au (II), PS@Au (III) and
H. Wang et al. / Journal of Colloid and Interface Science 440 (2015) 236–244
243
Fig. 6. XRD analyses of PS@Au (I) (black), PS@Au (II) (red) and PS@Au (III) (blue) before (a) and after (b) thermal treatment in DMF at 120 °C for 12 h. Length of the bars represent 1000 (a.u.) of intensity. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
Fig. 7. Normalized extinction spectra of synthesized PS@Au core–shell nanospheres (a) before thermal treatment and hollow gold nanostructures and (b) obtained after thermal treatment. (a), PS@Au (II) (red), and PS@Au (III) (blue); (b), gold nanocages (red) and hollow gold nanospheres (blue). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
corresponding hollow gold nanostructures are recorded. The characteristic SPR peaks of PS@Au (II) and PS@Au (III) (Fig. 7a) show platform at the wavelength longer than 800 nm. Besides, PS@Au (III) shows a weak shoulder absorption peak at 810 nm. After thermal treatment of PS@Au (II), we observe the emergence of two new absorption peaks of gold nanocages at 750 nm and 1070 nm (Fig. 7b, red). This is likely due to the removing of PS cores and the reconstruction of gold nanoshells. After thermal treatment of PS@Au (III), the absorption peak of hollow gold nanospheres slightly blue-shifts to 800 nm as compared to the peak of PS@Au (III) (810 nm) and we also observe the emergence of absorption peak at 1120 nm in NIR region. Previous studies have determined that the absorption peak of gold nanoshells blue shifts with increasing shell thickness [40]. The blue-shifted SPR results confirm that gold nanoshells thicknesses increase after thermal treatment of PS@Au (III), and the results are consistent with the TEM images (Fig. 3h).
The hollow gold nanocages and nanospheres show special NIR optical property, and can be used as photosensitizer. In our experiments, their photothermal conversion property were investigated via laser irradiation at k = 808 nm at a power density of 0.5 W cm 2 for various time. Fig. 8a shows temperature increase of gold nanocages and hollow gold nanospheres by irradiation at the laser power density of 0.5 W cm 2. It indicates that the suspension temperatures of both gold nanocages and hollow gold nanospheres increase after laser irradiation. It also represents clearly that the temperature of gold nanocages suspension increases fast by 4– 6 °C within one minute, and the temperature increase (DT) of gold nanocages is higher than that of hollow gold nanospheres after irradiation for 10 min at same concentration of particles. Gold nanocages show better photothermal conversion property. The suspensions of gold nanocages show higher temperature increase compared with that of gold nanospheres. This is likely due to the presence of holes on the gold nanoshells surface. In the presence
Fig. 8. (a) The plots of temperature increase (DT) of suspensions of gold nanocages (red) and hollow gold nanospheres (blue) as a function of irradiation time at laser power density of 0.5 W cm 2. (b) Stability test of photothermal property of gold nanocages (red) and hollow gold nanospheres (blue). Temperature increase (DT) after irradiation for 5 min at laser power density of 0.5 W cm 2. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
244
H. Wang et al. / Journal of Colloid and Interface Science 440 (2015) 236–244
of gold nanocages, the temperature increases dramatically by 11– 12 °C as irradiation for 10 min. As a result of denaturation of biomolecules at high temperature above 42 °C, cells and tissues can be irreversible damaged. Therefore, gold nanocages suspensions have ability to kill cancer tissues with NIR laser irradiation. In addition, for a photosensitizer to be useful, it should be stable under repeated NIR laser irradiation. To evaluate the stability of photothermal property, we carried out the additional experiment to irradiate their suspensions under 808 nm laser cycled for eight times. Fig. 8b shows that both gold nanocages and hollow gold nanospheres have stable photothermal property after eight repeats. 4. Conclusions Hollow gold nanostructures (nanocages and nanospheres) were fabricated by removing PS core of PS@Au in DMF solvent. The shell structures were controlled by the PS@Au core–shell structures, treatment temperature and time. These hollow gold nanostructures could be obtained easily by thermal treatment below 120 °C over 4 h. Hollow gold nanocages were obtained by thermal treatment of PS@Au (II) with gaps on gold shells, and hollow gold nanospheres were fabricated by thermal treatment of PS@Au (III) with continuous gold shell. These hollow nanostructures represented special NIR optical property. Both the gold nanocages and the gold nanospheres efficiently convert 808 nm laser into heat, and have stable photothermal property. Moreover, due to the holes on gold nanoshells surfaces, gold nanocages show higher temperature rise than that of gold nanospheres at the same particle concentration. Acknowledgment The authors would like to acknowledge funding from the National Natural Science Foundation of China (Grant Nos. 20903106, 20933007, 91127012, 21161130521, 21321063), and the Fund of Chinese Academy of Sciences (CMS-PY-201302). Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.jcis.2014.11.004. References [1] S.E. Skrabalak, J. Chen, Y. Sun, X. Lu, L. Au, C.M. Cobley, Y. Xia, Acc. Chem. Res. 41 (2008) 1587.
[2] E.C. Dreaden, A.M. Alkilany, X. Huang, C.J. Murphy, M.A. El-Sayed, Chem. Soc. Rev. 41 (2012) 2740. [3] S. Oldenburg, R. Averitt, S. Westcott, N. Halas, Chem. Phys. Lett. 288 (1998) 243. [4] H.-P. Liang, L.-J. Wan, C.-L. Bai, L. Jiang, J. Phys. Chem. B 109 (2005) 7795. [5] P.K. Jain, M.A. El-Sayed, J. Phys. Chem. C 111 (2007) 17451. [6] E. Prodan, C. Radloff, N.J. Halas, P. Nordlander, Science 302 (2003) 419. [7] M.P. Melancon, M. Zhou, C. Li, Acc. Chem. Res. 44 (2011) 947. [8] Y. Wang, Y. Liu, H. Luehmann, X. Xia, D. Wan, C. Cutler, Y. Xia, Nano Lett. 13 (2013) 581. [9] J. Chen, F. Saeki, B.J. Wiley, H. Cang, M.J. Cobb, Z.-Y. Li, L. Au, H. Zhang, M.B. Kimmey, X. Li, Nano Lett. 5 (2005) 473. [10] Y. Zhang, F. Xu, Y. Sun, C. Guo, K. Cui, Y. Shi, Z. Wen, Z. Li, Chem. Eur. J. 16 (2010) 9248. [11] S.E. Skrabalak, J. Chen, L. Au, X. Lu, X. Li, Y. Xia, Adv. Mater. 19 (2007) 3177. [12] J. Chen, C. Glaus, R. Laforest, Q. Zhang, M. Yang, M. Gidding, M.J. Welch, Y. Xia, Small 6 (2010) 811. [13] J. You, G. Zhang, C. Li, ACS Nano 4 (2010) 1033. [14] M.S. Yavuz, Y. Cheng, J. Chen, C.M. Cobley, Q. Zhang, M. Rycenga, J. Xie, C. Kim, K.H. Song, A.G. Schwartz, L.V. Wang, Y. Xia, Nat. Mater. 8 (2009) 935. [15] M.A. Mahmoud, R. Narayanan, M.A. El-Sayed, Acc. Chem. Res. 46 (2013) 1795. [16] M.A. Mahmoud, F. Saira, M.A. El-Sayed, Nano Lett. 10 (2010) 3764. [17] M.A. Mahmoud, D. O’Neil, M.A. El-Sayed, Chem. Mater. 26 (2013) 44. [18] W. Wang, Y. Han, M. Tian, Y. Fan, Y. Tang, M. Gao, Y. Wang, ACS Appl. Mater. Interfaces 5 (2013) 5709. [19] Y. Sun, Y. Xia, Science 298 (2002) 2176. [20] H.N. Xie, I.A. Larmour, Y.C. Chen, A.W. Wark, V. Tileli, D.W. McComb, K. Faulds, D. Graham, Nanoscale 5 (2013) 765. [21] F. Caruso, R.A. Caruso, H. Mohwald, Science 282 (1998) 1111. [22] A. Imhof, Langmuir 17 (2001) 3579. [23] S.-W. Kim, M. Kim, W.Y. Lee, T. Hyeon, J. Am. Chem. Soc. 124 (2002) 7642. [24] Z. Chen, P. Zhan, Z. Wang, J. Zhang, W. Zhang, N. Ming, C.T. Chan, P. Sheng, Adv. Mater. 16 (2004) 417. [25] P. Jiang, J.F. Bertone, V.L. Colvin, Science 291 (2001) 453. [26] Y. Xia, W. Lu, L. Jiang, Nanotechnology 21 (2010) 085501. [27] P.M. Arnal, M. Comotti, F. Schueth, Angew. Chem. Int. Ed. 45 (2006) 8224. [28] B.G. Prevo, J.C. Fuller, O.D. Velev, Chem. Mater. 17 (2005) 28. [29] H. Yang, Y. Liu, Q. Shen, L. Chen, W. You, X. Wang, J. Sheng, J. Mater. Chem. 22 (2012) 24132. [30] Z. Liang, A. Susha, F. Caruso, Chem. Mater. 15 (2003) 3176. [31] D.G. Shchukin, T. Shutava, E. Shchukina, G.B. Sukhorukov, Y.M. Lvov, Chem. Mater. 16 (2004) 3446. [32] C. Mangeney, S. Bousalem, C. Connan, M.-J. Vaulay, S. Bernard, M.M. Chehimi, Langmuir 22 (2006) 10163. [33] Y. Xia, J. Li, L. Jiang, J. Colloid Interface Sci. 377 (2012) 34. [34] Z. Wu, C. Dong, Y. Li, H. Hao, H. Zhang, Z. Lu, B. Yang, Angew. Chem. Int. Ed. 52 (2013) 9952. [35] Y.X. Zhang, H.C. Zeng, Langmuir 24 (2008) 3740. [36] H. Sun, M. Yu, G. Wang, X. Sun, J. Lian, J. Phys. Chem. C 116 (2012) 9000. [37] G. Andreasen, P. Schilardi, O. Azzaroni, R. Salvarezza, Langmuir 18 (2002) 10430. [38] C. Damle, K. Biswas, M. Sastry, Langmuir 17 (2001) 7156. [39] W. Cui, J. Li, Y. Zhang, H. Rong, W. Lu, L. Jiang, Nanomed.: Nanotechnol. Biol. Med. 8 (2012) 46. [40] W. Shi, Y. Sahoo, M.T. Swihart, P. Prasad, Langmuir 21 (2005) 1610. [41] D.V. Leff, P.C. Ohara, J.R. Heath, W.M. Gelbart, J. Phys. Chem. 99 (1995) 7036.
