The instability of silver nanoblocks under atmospheric conditions is investigated. The localized surface plasmon resonance band of the silver nanoblocks shows a red shift, broadening, and damping with increasing storage time under atmospheric conditions. The change in spectral properties of silver nanoblocks is considered to be due to sulfidation of silver and structural breakage of silver nanoblock based on scanning electron microscope observation and numerical simulation. The effect of aspect ratio of silver nanoblocks on the change in spectral properties of the nanoengineered silver blocks is also discussed.
© 2011 OSA
Properties of metals, including optical, electronic, magnetic, and catalytic characters can be tailored when their size is brought down to nano-scale [1,2]. Among various metal nanostructures, silver nanoparticles (NPs) have obtained the most attention due to their localized electric field enhancement which is stronger than that of the other noble metal NPs . In addition, silver NPs can respond in the violet region of the visible spectrum due to their peak amplitude in vacuum at a higher frequency . Chu et al.  probed surface plasmons (SP) in individual Ag NP in the ultra-violet (UV) spectral regime. Silver NPs have also been applied for enhancement of luminescence intensity [6–8], surface enhanced Raman scattering (SERS) [9,10], and enhancement of photovoltaic conversion efficiency of solar cells , which are utilized in the fields of biosensors, imaging, optoelectronic devices, new energy, etc.
However, the utility of silver for commercial nanophotonics applications is not as popular as gold so far, although silver is much cheaper than gold, because nanoengineered silver shows poor chemical stability in ambient atmosphere at room temperature, leading to red shift, broadening and damping of the localized surface plasmon resonance (LSPR) band. The mechanism of change in spectral properties of silver NPs has been discussed before [12,13]. The authors ascribed the change in spectral properties mainly to sulfidation of silver, but they neglected the effect of structural change of silver nanoblock (NB) on the change in spectral properties.
In this investigation, we study the stability of silver NBs fabricated by electron beam lithography (EBL) [14–16], and explore the mechanism of change in spectral properties of silver NBs. Different from the previous researches, the present investigation focuses on the effect of structural change on the change in spectral properties. Based on direct scanning electron microscope (SEM) observation and theoretical simulation, we propose that as well as sulfidation of silver, the structural breakage of silver NB also plays an important role in the change of spectral properties. In addition, as the EBL fabrication method has advantages of high resolution and easily shape controlling ability, we fabricate silver NBs with different aspect ratio (R) defined as the ratio of the length to the width of the oblong NB in the horizontal plane, for the purpose of discussing the effect of NB shape on the change in spectral properties of silver NBs.
The silver NBs with different R (R = 1, 2.2, 3.2, 4.3, 5) were fabricated on conventional glass substrates with high-resolution EBL method. The size of NBs with different R in the horizontal plane were 160 nm × 160 nm (R1), 110 nm × 240 nm (R2.2), 90 nm × 290 nm (R3.2), 80 nm × 340 nm (R4.3), and 77 nm × 385 nm (R5). The height of NB with different R was 40 nm and the interval distance of neighboring NBs was 200 nm.
For extinction spectra in visible region, a collimated beam from a halogen lamp spectrally filtered to a 350-950 nm wavelength range, was focused onto a sample mounted on an Olympus optical microscope (BX-51). Collimation was performed using the condenser lens of the microscope. The transmitted beam was collected by a microscope objective lens (40 × , N.A. = 0.75). The resultant beam was subsequently coupled into a Hamamatsu Photonics multichannel photodetector (PMA-11). Infrared extinction spectra were measured by a Fourier-transform infrared spectrometer with a microscope attachment (IRT-3000). The surface morphology of silver NBs was characterized using SEM (JSM-6700FT, JEOL, Japan). Platinum coating was deposited on the sample before SEM observation to improve the conductivity. The chemical composition of the sample was investigated by energy dispersive spectroscope (EDS, EMAX 350) attached to the SEM. The height of silver NBs was measured by atomic force microscope (AFM, VN-8000, KEYENCE, Japan).
Finite-difference time-domain (FDTD) simulation method (Lumerical FDTD Solution 7.0) was adopted to simulate the extinction spectra of silver NBs without and with Ag2S coverage, and of as-made and broken silver NBs. To calculate the extinction spectra, the following parameters were used: refractive index of glass substrate was 1.5, and the mesh accuracy was 1 nm. The optical constants for silver, titanium, and Ag2S were taken from Johnson and Christy , Palik  and Bennett et al. , respectively.
3. Results and discussion
Figure 1 shows the SEM images of as-made silver NBs (R4.3) (a) and the one after storage for 2 months in air (b). After 2 months, the silver NBs structure is broken. Some silver NPs in the NBs are not in the original position, especially the silver NPs on the edges of individual NB. Since the structure of silver NBs stored for 2 months is not the same as that of as-made, we study the spectral properties of silver NBs of as-made and being stored for different time.
Figure 2 shows the experimental [(a), (b)] and calculated [(c), (d)] extinction spectra of LSPR bands of longitudinal (L) [(a), (c)] and transverse (T) modes [(b), (d)] of as-made silver NBs with different R (R = 1, 2.2, 3.2, 4.3, 5) under polarized irradiation. With increasing R, the peaks of L mode LSPR bands move gradually from 650 to 1500 nm in visible and near infrared region, showing a red shift, while the peaks of T mode LSPR bands move from 650 to 450 nm in visible region, showing a blue shift. The calculated extinction spectra of either L or T mode match well with the experimental result.
Figure 3 shows the L mode LSPR bands of silver NBs with different R (R = 2.2, 3.2, 4.3, 5) stored for different time in air. All of them show a red shift, broadening, and damping with increasing storage time. The L mode LSPR band almost disappears after storage for 3 months until R = 4.3 and 5. We find that the broadening and damping of LSPR bands are faster when R increases. The T mode LSPR bands of silver NBs with different R show the same trend.
In accordance with the earlier investigations [12,13], the silver is easy to be corroded through sulfidation by the hydrogen sulfide gas (H2S) and carbonyl sulfide (OCS) in air. The EDS pattern (shown in Fig. 4 ) collected from the region marked by the rectangle in the inserted SEM image of silver NB arrays stored in air for 2 months reveals the presence of sulfur, verifying the sulfidation of the sample [12,13]. The signal of platinum is due to the platinum coating deposited before SEM observation.
Figure 5 shows the experimental extinction spectra of silver NBs stored for different time and calculated extinction spectra of silver NBs covered by Ag2S of different thickness (e.g. L mode, R4.3). The total height of silver and silver sulfide is set as the same as the one of as-made silver NB (40 nm). We can see that the peak positions are in good agreement with those of silver NBs covered by Ag2S of different thickness. With increasing height of Ag2S, the calculated LSPR band also shows a red shift and damping. This result indicates that sulfidation of silver is probably the reason for spectral red shift and damping. However, we find out that the calculated spectra do not show the broadening of LSPR bands, which is different from the experimental result. Therefore, we suggest that sulfidation of silver is not the only significant reason for spectral change.
Figure 1 clearly shows the structural breakage of silver NB. Thus, we propose that some silver atoms may diffuse into the neighboring area to form some silver NPs or silver NPs may be blown off to the interval area of the neighboring NBs, resulting in broadening of LSPR band.
In order to prove our assumption, two models (model 1 and 2) are set to simulate the structural breakage condition of silver NB. The schematic picture is shown in Fig. 6(a) . The shape and size of silver NBs in models 1 and 2 are similar to the first and third silver NBs (R4.3) stored for 2 months, as shown in SEM image [Fig. 1(b): upper row, from left to right]. There are some hemispheres with irregular radius ranged from 6 to 14 nm, distributing randomly around the broken NBs. We can see that the real shape of NBs stored for 2 months are not oblong anymore, but we approximately simplify them as oblong with four round corners. For convenience of simulation, we assume that the same breakage happens to every NB. From SEM images, we find that there are some voids formed on silver NBs, causing the nonuniformity of the silver NB height. In addition, the total height of silver NBs reduces with increasing storage time, which is proved by AFM characterization. It may be caused by the diffusion of silver atoms and silver NPs being blown off. Hence, for simplification of simulation, we set the height of silver NBs as uniform as 30 nm (thinner than 40 nm). The volume sum of small hemispheres and the pieces of fragments is set as the same as that of as-made silver NB (R4.3, 80 × 340 × 40 nm3).
Figure 6(b) shows the calculated extinction spectra of silver NBs (R4.3) of as-made and model 1 and 2. We can see that the LSPR bands of models 1 and 2 show a blue shift, broadening, and damping. Besides the spectral broadening, the LSPR band of either models 1 or model 2 shows blue shift, compared with that of as-made NB structure. We suggest that the blue shift is due to the shortening of NBs. For the NB of same area and height, the one with lower R (i.e., the shorter NB), shows blue shift of L mode LSPR band compared with the one with higher R (i.e., the longer NB) . If the silver NB structure maintains as that of as-made NB, and the height changes from 40 nm to 30 nm, the L mode LSPR band shows red shift. However, the LSPR bands of models 1 and 2 with 30 nm height show blue shift. It means that the structural breakage of silver NB plays a more important role in spectral peak shift than decrease of NB height. Consequently, we conclude that sulfidation of silver causes spectral red shift and damping, while structural breakage of silver NB causes spectral blue shift, broadening and damping.
In reality, sulfidation of silver and structural breakage of silver NB affect the spectral properties of silver NBs simultaneously. Thus, we simulate the extinction spectra of silver NBs under the both influence. We take model 1 covered by Ag2S for example. The total height of silver and silver sulfide is still set as 30 nm.
Figure 7 shows the experimental L mode extinction spectra of silver NBs (R 4.3) with increasing storage time and calculated extinction spectra of model 1 covered by Ag2S of different thickness. We can see that the calculated extinction spectra show a red shift, broadening, and damping as well as the experimental result. The peak of calculated LSPR band almost matches the experimental result. However, the broadening of calculated LSPR band is not as large as experimental result. That is due to the assumption of the same model happening to every block. However, the real structural breakage condition of silver NB is more complicated and irregular. Therefore, the broadening of experimental LSPR band is larger than that of calculated result.
As we mentioned above, the broadening and damping of LSPR bands were faster with higher R. Since the structural breakage of silver NB leads to LSPR band broadening, we suggest that the NB with high R is more fragile than that with low R. In our opinion, for the NB of same area and height, the one with higher R, has smaller cross section (the plane which is perpendicular to the longitudinal direction of NB, parallel to the transverse and normal direction of NB) than that with lower R. Therefore, it is easier for the NB with higher R to be fragmented into pieces due to smaller cross section.
In conclusion, we investigate the spectral properties of nanoengineered silver blocks and the mechanism of their instability. The LSPR band of the silver NBs shows a red shift, broadening, and damping with increasing storage time under atmospheric conditions. For spectral peak shift, structural breakage of silver NB causes blue shift, whereas sulfidation of silver causes red shift, and the latter one plays the leading role in spectral peak shift. Spectral broadening is due to structural breakage of silver NB. Spectral damping is ascribed to both of them. As a result, we suggest that besides the sulfidation of silver, the structural breakage of silver NB is also an important reason for spectral change. Moreover, the broadening and damping of LSPR band are faster with higher R. To solve the problem of instability of silver NPs, coating titanium on their top is an effective way to prevent them from corrosion .
This work was financially supported by the National Natural Science Foundation of China (Grant Nos. 50672087, 50872123, and 50802083). H. M. also acknowledges funding from the Ministry of Education, Culture, Sports, Science, and Technology of Japan; KAKENHI Grants-in-Aid (19049001) for Scientific Research on the Priority Area “Strong Photon-Molecule Coupling Fields for Chemical Reactions” (470), and a Grant-in-Aid from Hokkaido Innovation through Nanotechnology Support (HINTS).
References and links
1. S. R. Emory, W. E. Haskins, and S. Nie, “Direct observation of size-dependent optical enhancement in single metal nanoparticles,” J. Am. Chem. Soc. 120(31), 8009–8010 (1998), http://www.etseq.urv.es/DEQ/Doctorat/index/web_nanobiotech/handouts/Lecture_16_Handout_12_Nanoparticles.pdf. [CrossRef]
2. M. A. El-Sayed, “Small is different: shape-, size-, and composition-dependent properties of some colloidal semiconductor nanocrystals,” Acc. Chem. Res. 37(5), 326–333 (2004), http://pubs.acs.org/doi/abs/10.1021/ar020204f. [CrossRef] [PubMed]
3. U. Kreibig and M. Vollmer, Optical Properties of Metal Clusters (Springer-Verlag, 1995).
4. G. Mie, “Beiträge zur Optik trüber Medien, speziell kolloidaler Metallösungen,” Ann. Phys. 330(3), 377–445 (1908). [CrossRef]
5. M. W. Chu, P. Sharma, C. P. Chang, S. C. Liou, K. T. Tsai, J. K. Wang, Y. L. Wang, and C. H. Chen, “Probing surface plasmons in individual Ag nanoparticles in the ultra-violet spectral regime,” Nanotechnology 20(23), 235705 (2009), http://iopscience.iop.org/0957-4484/20/23/235705. [CrossRef] [PubMed]
6. T. D. Corrigan, S. Guo, R. J. Phaneuf, and H. Szmacinski, “Enhanced fluorescence from periodic arrays of silver nanoparticles,” J. Fluoresc. 15(5), 777–784 (2005), http://www.springerlink.com/content/fk5844x114t27q1n/fulltext.pdf. [CrossRef] [PubMed]
7. H. Mertens and A. Polman, “Plasmon-enhanced erbium luminescence,” Appl. Phys. Lett. 89(21), 211107 (2006), http://apl.aip.org/resource/1/applab/v89/i21/p211107_s1. [CrossRef]
8. K. Ueno, S. Juodkazis, V. Mizeikis, K. Sasaki, and H. Misawa, “Clusters of closely spaced gold nanoparticles as a source of two-photon photoluminescence at visible wavelengths,” Adv. Mater. (Deerfield Beach Fla.) 20(1), 26–30 (2008), http://onlinelibrary.wiley.com/doi/10.1002/adma.200602680/pdf. [CrossRef]
9. M. Futamata, Y. Y. Yu, T. Yanatori, and T. Kokubun, “Closely adjacent Ag nanoparticles formed by cationic dyes in solution generating enormous SERS enhancement,” J. Phys. Chem. C 114(16), 7502–7508 (2010), http://pubs.acs.org/doi/pdf/10.1021/jp9113877. [CrossRef]
10. Y. Yokota, K. Ueno, and H. Misawa, “Highly controlled surface-enhanced Raman scattering chips using nanoengineered gold blocks,” Small 7(2), 252–258 (2011), http://onlinelibrary.wiley.com/doi/10.1002/smll.201001560/abstract. [CrossRef] [PubMed]
11. S. S. Kim, S. I. Na, J. Jo, D. Y. Kim, and Y. C. Nah, “Plasmon enhanced performance of organic solar cells using electrodeposited Ag nanoparticles,” Appl. Phys. Lett. 93(7), 073307 (2008), http://apl.aip.org/resource/1/applab/v93/i7/p073307_s1. [CrossRef]
12. M. D. McMahon, R. Lopez, H. M. Meyer III, L. C. Feldman, and R. F. Haglund Jr., “Rapid tarnishing of silver nanoparticles in ambient laboratory air,” Appl. Phys. B 80(7), 915–921 (2005), http://www.springerlink.com/content/w3kn5l6627160652/fulltext.pdf. [CrossRef]
13. W. Cao and H. E. Elsayed-Ali, “Stability of Ag nanoparticles fabricated by electron beam lithography,” Mater. Lett. 63(26), 2263–2266 (2009). [CrossRef]
14. K. Ueno, V. Mizeikis, S. Juodkazis, K. Sasaki, and H. Misawa, “Optical properties of nanoengineered gold blocks,” Opt. Lett. 30(16), 2158–2160 (2005), http://www.opticsinfobase.org/ol/abstract.cfm?uri=ol-30-16-2158. [CrossRef] [PubMed]
15. K. Ueno, S. Juodkazis, M. Mino, V. Mizeikis, and H. Misawa, “Spectral sensitivity of uniform arrays of gold nanorods to dielectric environment,” J. Phys. Chem. C 111(11), 4180–4184 (2007), http://pubs.acs.org/doi/abs/10.1021/jp068243m. [CrossRef]
16. K. Ueno, S. Juodkazis, T. Shibuya, Y. Yokota, V. Mizeikis, K. Sasaki, and H. Misawa, “Nanoparticle plasmon-assisted two-photon polymerization induced by incoherent excitation source,” J. Am. Chem. Soc. 130(22), 6928–6929 (2008), http://pubs.acs.org/doi/abs/10.1021/ja801262r. [CrossRef] [PubMed]
18. D. W. Lynch and W. R. Hunter, “Titanium (Ti),” in Handbook of Optical Constants of Solids, E. D. Palik, ed. (Academic, 1998).
19. J. M. Bennett, J. L. Stanford, and E. J. Ashley, “Optical constants of silver sulfide tarnish films,” J. Opt. Soc. Am. 60(2), 224–232 (1970), http://www.opticsinfobase.org/abstract.cfm?id=53944. [CrossRef]
20. L. Wang, W. Xiong, Y. Nishijima, Y. Yokota, K. Ueno, H. Misawa, G. Bi, and J. R. Qiu, “Controlling plasmonic spectral properties of engineered silver nanorods by using titanium coating,” IEEE Photon. Technol. Lett. (submitted).