2184 Chem. Mater. 2010, 22, 2184–2189 DOI:10.1021/cm901791u
Thermal Stability of Silver Nanorod Arrays Kelsey R. Beavers, Nicole E. Marotta, and Lawrence A. Bottomley* School of Chemistry & Biochemistry, Georgia Institute of Technology, Atlanta, Georgia 30332-0400 Received June 25, 2009. Revised Manuscript Received December 16, 2009
Silver nanorod arrays fabricated by glancing angle deposition have found application as substrates for surface enhanced Raman scattering. A critical issue in the continued development of these substrates is their stability over time. The thermal stability of the arrays was systematically investigated. Thermally induced changes in surface morphology were evaluated using scanning electron microscopy and X-ray diffraction and correlated with changes in SERS enhancement. The findings presented herein suggest that the shelf life of silver nanorod arrays is limited by migration of silver on the surface. This rate of migration is enhanced by the presence of a surface layer of chemisorbed oxygen. Introduction Surface-enhanced Raman scattering (SERS) is a powerful analytical tool for determining analyte concentration and characterizing the vibrational, rotational, and other low-frequency modes of molecules and molecular ensembles.1-7 One impediment to its application in biosensing is the difficulty involved in manufacturing SERS substrates with reproducible performance characteristics.8,9 Recently, substrates comprised of silver nanorod arrays have been shown to be highly uniform in structure, easy to *Author to whom correspondence should be addressed. E-mail: Bottomley@ gatech.edu.
(1) Moskovits, M. Rev. Mod. Phys. 1985, 57(3), 783–826. (2) McCreery, R. L. Raman Spectroscopy for Chemical Analysis; New York, 2000. (3) Aroca, R. Surface Enhanced Vibrational Spectroscopy; New York, 2006. (4) Kneipp, K. Phys. Today 2007, 60(11), 40–46. (5) Willets, K. A.; Van Duyne, R. P. Annu. Rev. Phys. Chem. 2007, 58, 267–97. (6) Scaffidi, J. P.; Gregas, M. K.; Seewaldt, V.; Vo-Dinh, T. Anal. Bioanal. Chem. 2009, 393(4), 1135–1141. (7) Vo-Dinh, T.; Yan, F.; Wabuyele, M. B., Top. Appl. Phys. 2006, 103 (Surface-Enhanced Raman Scattering), 409-426. (8) Stiles, P. L.; Dieringer, J. A.; Shah, N. C.; Van Duyne, R. P. Annu. Rev. Anal. Chem. 2008, 1, 601–626. (9) Lal, S.; Grady, N. K.; Kundu, J.; Levin, C. S.; Lassiter, J. B.; Halas, N. J. Chem. Soc. Rev. 2008, 37(5), 898–911. (10) Chaney, S. B.; Shanmukh, S.; Dluhy, R. A.; Zhao, Y. P. Appl. Phys. Lett. 2005, 87(3), 031908/1–031908/3. (11) Driskell, J. D.; Shanmukh, S.; Liu, Y.; Chaney, S. B.; Tang, X. J.; Zhao, Y. P.; Dluhy, R. A. J. Phys. Chem. C 2008, 112(4), 895–901. (12) Leverette, C. L.; Dluhy, R. A. Colloids Surf., A 2004, 243(1-3), 157–167. (13) Leverette, C. L.; Jacobs, S. A.; Shanmukh, S.; Chaney, S. B.; Dluhy, R. A.; Zhao, Y. P. Appl. Spectrosc. 2006, 60(8), 906–913. (14) Leverette, C. L.; Shubert, V. A.; Wade, T. L.; Varazo, K.; Dluhy, R. A. J. Phys. Chem. B 2002, 106(34), 8747–8755. (15) Liu, Y.; Fan, J.; Zhao, Y. P.; Shanmukh, S.; Dluhy, R. A. Appl. Phys. Lett. 2006, 89(17), 173134/1–173134/3. (16) Liu, Y. J.; Zhang, Z. Y.; Zhao, Q.; Dluhy, R. A.; Zhao, Y. P. J. Phys. Chem. C 2009, 113(22), 9664–9669. (17) Liu, Y. J.; Zhang, Z. Y.; Zhao, Q.; Dluhy, R. A.; Zhao, Y. P. Appl. Phys. Lett. 2009, 94(3), 033103–3. (18) Zhao, Y. P.; Chaney, S. B.; Shanmukh, S.; Dluhy, R. A. J. Phys. Chem. B 2006, 110(7), 3153–3157. (19) Zhao, Y. P.; Li, S. H.; Chaney, S. B.; Shanmukh, S.; Fan, J. G.; Dluhy, R. A.; Kisaalita, W. J. Electron. Mater. 2006, 35(5), 846– 851.
pubs.acs.org/cm
fabricate, and provide unprecedented signal enhancements.10-19 The nanorod arrays have been fabricated by glancing angle deposition (GLAD).18-28 GLAD is a physical vapor deposition technique that creates thin films consisting of columnar microstructures as a result of atomic shadowing.29-31 Applications of these arrays into SERS-based biosensing are beginning to emerge.32-37 One of our research objectives is the development of a SERS-based nucleic acid hybridization assay using Ag (20) Brett, M. J.; Seto, M. W.; Sit, J. C.; Harris, K. D.; Vick, D.; Robbie, K., Proc. SPIE-Int. Soc. Opt. Eng. 1999, 3790 (Engineered Nanostructural Films and Materials), 114-118. (21) Robbie, K.; Brett, M. J. J. Vac. Sci. Technol., A 1997, 15(3, Pt. 2), 1460–1465. (22) Robbie, K.; Shafai, C.; Brett, M. J. J. Mater. Res. 1999, 14(7), 3158–3163. (23) Robbie, K.; Sit, J. C.; Brett, M. J. J. Vac. Sci. Technol., B: Microelectron. Nanometer Struct.--Process., Meas., Phenom. 1998, 16(3), 1115–1122. (24) Singh, J. P.; Karabacak, T.; Ye, D. X.; Liu, D. L.; Picu, C.; Lu, T. M.; Wang, G. C. J. Vac. Sci. Technol., B: Microelectron. Nanometer Struct.--Process., Meas., Phenom. 2005, 23(5), 2114–2121. (25) Sit, J. C.; Vick, D.; Robbie, K.; Brett, M. J. J. Mater. Res. 1999, 14(4), 1197–1199. (26) Tait, R. N.; Smy, T.; Brett, M. J. Thin Solid Films 1993, 226(2), 196– 201. (27) Zhao, Y. P.; Chaney, S. B.; Zhang, Z. Y. J. Appl. Phys. 2006, 100(6), 063527/1–063527/8. (28) Zhao, Y. P.; Ye, D. X.; Wang, G. C.; Lu, T. M. Nano Lett. 2002, 2(4), 351–354. (29) Abelmann, L.; Lodder, C. Thin Solid Films 1997, 305(1-2), 1–21. (30) Amassian, A.; Kaminska, K.; Suzuki, M.; Martinu, L.; Robbie, K. Appl. Phys. Lett. 2007, 91(17), 173114/1–173114/3. (31) Vankranenburg, H.; Lodder, C. Mater. Sci. Eng. R 1994, 11(7), 295–354. (32) Driskell, J. D.; Seto, A. G.; Jones, L. P.; Jokela, S.; Dluhy, R. A.; Zhao, Y. P.; Tripp, R. A. Biosens. Bioelectron. 2008, 24(4), 917– 922. (33) Driskell, J. D.; Shanmukh, S.; Liu, Y. J.; Hennigan, S.; Jones, L.; Zhao, Y. P.; Dluhy, R. A.; Krause, D. C.; Tripp, R. A. IEEE Sens. J. 2008, 8(5-6), 863–870. (34) Shanmukh, S.; Jones, L.; Driskell, J.; Zhao, Y. P.; Dluhy, R.; Tripp, R. A. Nano Lett. 2006, 6(11), 2630–2636. (35) Shanmukh, S.; Jones, L.; Zhao, Y.-P.; Driskell, J. D.; Tripp, R. A.; Dluhy, R. A. Anal. Bioanal. Chem. 2008, 390, 1551–1555. (36) Tripp, R. A.; Dluhy, R. A.; Zhao, Y. P. Nano Today 2008, 3(3-4), 31–37. (37) Zhao, Y. P.; Shanmukh, S.; Liu, Y. J.; Chaney, S. B.; Jones, L.; Dluhy, R. A.; Tripp, R. A. Proc. SPIE-Int. Soc. Opt. Eng. 2006, 6324 (Plasmonics: Nanoimaging, Nanofabrication, and Their Applications II), 63240M/1-63240M/9.
Published on Web 03/02/2010
r 2010 American Chemical Society
Article
Chem. Mater., Vol. 22, No. 7, 2010
nanorod array substrates. A critical aspect of this application involves heating and cooling of nucleic acids on the array to promote formation of the hybrid and minimize mismatches. We found a decrease in SERS performance upon thermal cycling, prompting further investigation into the effect of temperature on Ag nanorod structure. Our analysis of the thermal stability of Ag nanorod arrays is reported herein; the application of Ag nanorod arrays to DNA hybridization will be presented separately. Experimental Section Substrate Fabrication. Two methods were used to fabricate the SERS active nanorod arrays. All SERS substrates were fabricated from standard 25 75 mm glass microscope slides. The process began by immersing each slide in “piranha” solution38 for 30 min, copious rinsing with deionized water, and drying under a flowing nitrogen gas stream. Method One. Slides were placed in a CVC-601 DC Sputtering tool (Consolidated Vacuum Corp., Rochester, NY), and thin films of titanium (50 nm) followed by silver (500 nm) were deposited onto one side of the slide at an angle of incidence at 90°. The titanium under layer promotes adhesion of the silver to the glass surface. The substrates were then removed from the DC sputtering tool and positioned in a CVC-SC5000 electron beam evaporator (Consolidated Vacuum Corp., Rochester, NY). Silver (99.99% from Kurt J. Lesker, Clairton, PA) was deposited onto the slides at a deposition rate of 3 A˚/s, a starting pressure of 1 10-6, and at a flux angle of 86° until the apparent deposition thickness on the quartz crystal thickness monitor reached a reading of 1.5 μm. Since the monitor is located closer to the source metal, this value is not representative of nanorod length or average film thickness. The substrates were allowed to cool under vacuum for at least 10 min prior to backfilling the chamber with nitrogen and removal from the tool. Method Two. SERS substrates were prepared by Dr. Hsiao Yun Chu of Prof. Yiping Zhao’s group in physics at the University of Georgia by electron beam evaporation. The apparatus used enabled the substrates to remain in vacuum for the entire fabrication process. Glass microscope slides were mounted on a sample holder and placed directly above the source. Thin films of Ti (50 nm) and Ag (500 nm) were deposited on the sample surface at normal incidence. Then stepper motors rotated the substrate to an incident angle of 86° from normal, and Ag nanorods were deposited. Following deposition using either method, substrates were packaged in UniMailer slide holders (VWR Scientific Inc., West Chester, PA) and stored in a desiccator until use. Post fabrication, each Ag nanorod substrate was cut into eight 12.5 mm 18.75 mm chips. Chips were placed on a preheated hot plate, held at a fixed temperature for prescribed amounts of time, and then cooled under ambient conditions. Substrate Characterization. Substrates were physically characterized by scanning electron microscopy (SEM). Micrographs were obtained using a Zeiss SEM Ultra60 scanning electron microscope (Carl Zeiss SMT Inc., Peabody, MA) using an accelerating voltage of 5 eV. To minimize charging artifacts, slides were attached to the sample holder using double-sided copper tape. The SERS activity was evaluated using a Kaiser Optical Systems’ Holoprobe 785 spectrometer (Kaiser Optical (38) ”Piranha” solution must be handled with care; it is extremely oxidizing, reacts violently with organics, and should only be stored in loosely tightened containers to avoid pressure buildup.
2185
Systems, Inc., Ann Arbor, MI) using an excitation wavelength of 785 nm, a 10 objective, an integration time of 10 s, and a surface power of 4.5 mW. Rhodamine 6G (R6G) was used as the SERS probe for all SERS studies and comparisons. This analyte was used as received (Tokyo Kasei Kogyo Co. Ltd., Toshima, Japan). Stock solutions of 1.0 mM R6G were prepared in deionized water; 5 μL aliquots were applied to each chip following thermal treatment and allowed to evaporate. Test solutions were prepared by serial dilution of each stock. X-ray diffraction (XRD) was performed using a Panalytical X’Pert Pro MultiPurpose Diffractometer (MPD) with an Anton Paar HTK1200 furnace. Samples were heated in air or under nitrogen, sampling XRD patterns every 5 min.
Results Figure 1a presents an SEM micrograph typical of Ag nanorod arrays prepared using method one and stored under ambient conditions. Image analysis revealed a consistent growth angle of 57 ( 0.5° and nanorod lengths of 450 ( 15 nm across the middle of the slide. Figure 1b depicts the SERS spectrum obtained for R6G. This intensity equates to a SERS enhancement factor39,40 of ∼107.41 When a chip from the same substrate was held at 75 °C for just 5 min, both coarsening of the Ag nanostructure (Figure 1c) and a 10-fold decrease in SERS enhancement were observed (Figure 1d). This decrease in enhancement with surface reorganization parallels the relationship previously observed between SERS intensity and Ag nanorod length.11 The substrate used in acquiring the data presented in Figure 1 was fabricated using method one in which a break in vacuum and exposure to oxygen occurred between deposition of the under layers and Ag nanorods. Figure 2a shows an SEM micrograph typical of Ag nanorod arrays prepared using method two and stored at room temperature (25 °C). With this method, both the under layers and the nanorod arrays are deposited without breaking vacuum. Image analysis revealed a consistent growth angle of 57 ( 0.5° and nanorod lengths of 870 ( 25 nm across the middle of the slide. A comparison of Figures 1a and 2a reveals that the Ag nanorods fabricated using method two have a higher aspect ratio than those fabricated using method one. Nanostructures fabricated with method two give SERS intensities for R6G of over 50,000 counts, as shown in Figure 2b. This equates to a SERS enhancement factor of ∼108.11 When a chip from this substrate was held at 75 °C for 30 min, no change in SERS signal intensities was found. Similarly, no change in SERS intensity was observed when a chip was held at 125 °C for 5 min. However, when this chip was held at 125 °C for 30 min, some coarsening of the surface was observed (Figure 2c). The resultant deformation of the nanostructure led to a ∼30% decrease in overall SERS enhancement, as shown in Figure 2d. When a third (39) SERS enhancement factors were determined by the method of Le Ru and co-workers.{Le Ru, 2007 #350}. (40) Le Ru, E. C.; Blackie, E.; Etchegoin, P. G. J. Phys. Chem. C 2007, 111, 13794–13803. (41) Marotta, N. E.; Barber, J. R.; Dluhy, P. R.; Bottomley, L. A. Appl. Spectrosc. 2009, 63(10), 1101–1106.
2186
Chem. Mater., Vol. 22, No. 7, 2010
Beavers et al.
Figure 1. SEM micrographs of an Ag nanorod array substrate #1 before (a) and after (c) heating at 75 °C for 5 min. Scale bars in the images denote 1.0 μm. SERS spectra of 1.0 10-5 M R6G obtained on this substrate is shown before (b) and after (d) heating.
chip from this substrate was held at 150 °C for 30 min, coarsening of the nanorod surface similar in extent to that shown above in Figure 1d resulted (see Figure 2e). The original nanorod structures are no longer existent, having deformed into globular mounds. The corresponding SERS spectrum (Figure 2f) presents SERS enhancement so diminished that peak heights are approximately 90% smaller than those obtained from SERS spectra of the unheated Ag nanorod array. Figure 3 presents a comparison of SERS intensity as a function of temperature for substrates prepared by the aforementioned methods. Due to the variation in enhancement between the two, intensities were normalized to those observed for the 611 cm-1 band of R6G on chips held at ambient temperature. Substrates exposed to oxygen prior to nanorod fabrication undergo a sharp decrease in SERS enhancement when maintained at temperatures above 60 °C for only 5 min. In contrast, substrates exposed to oxygen after nanorod fabrication exhibit a decrease in SERS intensity when maintained at temperatures above 100 °C for 30 min. Silver surfaces are known to undergo rearrangement at temperatures well below their bulk melting point.42,43 Degradation of SERS enhancements following thermal (42) Morgenstern, K.; Rosenfeld, G.; Lægsgaard, E.; Besenbacher, F.; Comsa, G. Phys. Rev. Lett. 1998, 80(3), 556–559. (43) Pedemonte, L.; Tatarek, R.; Bracco, G. Phys. Rev. B 2002, 66(4), 5.
annealing of substrates has also been previously reported. Liu and co-workers have found a 20% decrease in SERS response when substrates prepared by electrochemical roughening were held at temperatures greater than 125 °C.44,45 Whitney et al. reported that shifts in the local surface plasmon resonance of silver nanoparticles were dependent upon annealing temperature.46,47 Previous studies of the thermal stability of structured Ag surfaces were undertaken at higher temperatures. To our knowledge, this is the first report that the thermal stability of Ag nanorod arrays depends upon the method of preparation. X-ray diffraction (XRD) was performed to determine the change in the crystal structure upon surface reorganization. Peaks for the (111), (220), and (200) crystal directions were present in the initial spectrum taken before heating. As the temperature of the substrate was increased to 150 °C, the peak corresponding to the (111) direction increased by ∼30%. No difference in the rate of change or appearance of the final XRD spectrum was observed when a fresh chip was heated under a nitrogen (44) Liu, Y.-C.; Yang, K.-H.; Hsu, T.-C. J. Raman Spectrosc. 2009, 40(8), 903–907. (45) Liu, Y.-C.; Hsu, T.-C.; Tsai, J.-F. J. Phys. Chem. C 2007, 111(28), 10570–10574. (46) Litorja, M.; Haynes, C. L.; Haes, A. J.; Jensen, T. R.; Van Duyne, R. P. J. Phys. Chem. B 2001, 105(29), 6907–6915. (47) Whitney, A. V.; Elam, J. W.; Stair, P. C.; Van Duyne, R. P. J. Phys. Chem. C 2007, 111(45), 16827–16832.
Article
Chem. Mater., Vol. 22, No. 7, 2010
2187
Figure 2. SEM micrographs of an Ag nanorod array substrate #2 (a) before and (c) after heating at 125 °C for 5 min and (e) after heating at 150 °C for 5 min. Scale bars in the images denote 1.0 μm. SERS spectra of 1.0 10-5 M R6G obtained on this substrate is shown before (b) and after (d and f) heating.
atmosphere. The XRD results demonstrate the intrinsic instability of the nanorod structure and its tendency to reorganize to a more stable structure.48
migrate at room temperature.48-68 This migration is rapid69 and anisotropic,70-72 and results from diffusion and coalescence of Ag islands (i.e., Smoluchowski ripening)
Discussion Our findings demonstrate that Ag nanorod arrays are thermally unstable. Silver surface atoms are known to (48) Morgenstern, K. Phys. Status Solidi B 2005, 242(4), 773–796. (49) Esser, M.; Morgenstern, K.; Rosenfeld, G.; Comsa, G. Surf. Sci. 1998, 402-404, 341–345. (50) Kaganer, V. M.; Ploog, K. H.; Sabelfeld, K. K. Phys. Rev. B 2006, 73(11), 115425–8. (51) Layson, A. R.; Evans, J. W.; Fournee, V.; Thiel, P. A. J. Chem. Phys. 2003, 118(14), 6467–6472. (52) Layson, A. R.; Evans, J. W.; Thiel, P. A. Phys. Rev. B: Condens. Matter Mater. Phys. 2002, 65(19), 193409/1–193409/4. (53) Layson, A. R.; Thiel, P. A. Surf. Sci. Lett. 2001, 472, L151–L156. (54) Morgenstern, K.; Braun, K.-F.; Rieder, K.-H. Phys. Rev. Lett. 2002, 89(22), 226801/1–226801/4. (55) Morgenstern, K.; Laegsgaard, E.; Besenbacher, F. Mater. Res. Soc. Symp. Proc. 2001, 648 (Growth, Evolution and Properties of Surfaces, Thin Films and Self-Organized Structures), P1 5/1-P1 5/11.
(56) Morgenstern, K.; Laegsgaard, E.; Besenbacher, F. NATO Sci. Ser. II: Math. Phys. Chem. 2001, 29 (Collective Diffusion on Surfaces: Correlation Effects and Adatom Interactions), 201-212. (57) Morgenstern, K.; Laegsgaard, E.; Besenbacher, F. Phys. Rev. Lett. 2001, 86(25), 5739–5742. (58) Morgenstern, K.; Laegsgaard, E.; Besenbacher, F. Phys. Rev. Lett. 2005, 94(16), 166104/1–166104/4. (59) Morgenstern, K.; Rieder, K.-H.; Fiete, G. A. Phys. Rev. B: Condens. Matter Mater. Phys. 2005, 71(15), 155413/1–155413/12. (60) Morgenstern, K.; Rosenfeld, G.; Comsa, G. Phys. Rev. Lett. 1996, 76(12), 2113–16. (61) Morgenstern, K.; Rosenfeld, G.; Comsa, G. Surf. Sci. 1999, 441 (2-3), 289–300. (62) Morgenstern, K.; Rosenfeld, G.; Comsa, G.; Sorensen, M. R.; Hammer, B.; Laegsgaard, E.; Besenbacher, F. Phys. Rev. B: Condens. Matter Mater. Phys. 2001, 63(4), 045412/1–045412/5. (63) Morgenstern, K.; Rosenfeld, G.; Poelsema, B.; Comsa, G. Phys. Rev. Lett. 1995, 74(11), 2058–61. (64) Semin, D. J.; Lo, A.; Roark, S. E.; Skodje, R. T.; Rowlen, K. L. J. Chem. Phys. 1996, 105(13), 5542–5551.
2188
Chem. Mater., Vol. 22, No. 7, 2010
Figure 3. Plot of the normalized SERS intensity for the 611 cm-1 band versus temperature for the substrate prepared using method one (blue triangle) and two (red square). The data points listed in this figure are the average intensities (with standard deviations indicated by the error bars) observed at five or more locations on the same chip.
as well as diffusive mass transfer from smaller to larger islands (i.e., Oswald ripening) to minimize the free energy at step edges.52,66,67,73 Heating increases the rate of nanostructure coarsening.50,64,74-77 Several groups have studied the coarsening of Ag using scanning tunneling microscopy.64,68 Semin et al.64 found that the average Ag island radii increased by ∼35% when held at 42 °C for 4 h. They showed that postdeposition coarsening is a thermally activated process with an activation energy of just 13 ( 2 kcal/mol. Our Ag nanorods increased in radius within minutes upon heating. Upon exposure of Ag to air, a layer of chemisorbed oxygen is formed.52,78-87 Layson et al. have shown that (65) Stoldt, C. R.; Caspersen, K. J.; Bartelt, M. C.; Jenks, C. J.; Evans, J. W.; Thiel, P. A., Mater. Res. Soc. Symp. Proc. 2000, 619 (Recent Developments in Oxide and Metal Epitaxy: Theory and Experiment), 15-25. (66) Stoldt, C. R.; Jenks, C. J.; Thiel, P. A.; Cadilhe, A. M.; Evans, J. W. J. Chem. Phys. 1999, 111(11), 5157–5166. (67) Thiel, P. A.; Evans, J. W. Cluster diffusion, coalescence, and coarsening in metal(100) homoepitaxial systems. In Series on Directions in Condensed Matter Physics; Zhang, Z., Lagally, M. G., Eds.; World Scientific Publishing Company: River Edge, NJ, 1998; Vol. 14, pp 384-402. (68) Thiel, P. A.; Evans, J. W. J. Phys. Chem. B 2000, 104(8), 1663–1676. (69) Thiel, P. A.; Shen, M.; Liu, D. J.; Evans, J. W. J. Phys. Chem. C 2009, 113(13), 5047–5067. (70) Winegard, W. C. Acta Metall. 1953, 1(2), 230. (71) Ferrando, R.; Treglia, G. Phys. Rev. B 1994, 50(16), 12104–12117. (72) Yu, B. D.; Scheffler, M. Phys. Rev. Lett. 1996, 77(6), 1095–1098. (73) Wen, J. M.; Evans, J. W.; Bartelt, M. C.; Burnett, J. W.; Thiel, P. A. Phys. Rev. Lett. 1996, 76(4), 652–5. (74) Mougin, K.; Zheng, Z.; Piazzon, N.; Gnecco, E.; Haidara, H. J. Colloid Interface Sci. 2009, 333(2), 719–724. (75) Kizuka, T.; Ichinose, H.; Ishida, Y. J. Jpn. Inst. Met. 1991, 55(3), 233–240. (76) Nandipati, G.; Shim, Y.; Amar, J. G.; Karim, A.; Kara, A.; Rahman, T. S.; Trushin, O. J. Phys.: Condens. Matter 2009, 21(8), 084214. (77) Rorak, S. E.; Lo, A.; Skodje, R. T.; Rowlen, K. L. Nanostructured Materials 1997, 679, 152–168. (78) Menzel, E.; Menzel-Kopp, C. Surf. Sci. 1964, 2, 376–80. (79) Weaver, J. F.; Hoflund, G. B. Chem. Mater. 1994, 6(10), 1693– 1699. (80) Vattuone, L.; Gambardella, P.; Valbusa, U.; Rocca, M. In HREELS study of O-2 molecular chemisorption on Ag(001); 1997; Elsevier Science Bv: 1997; pp 671-675. (81) Franchy, R.; Bartolucci, F.; de Mongeot, F. B.; Cemic, F.; Rocca, M.; Valbusa, U.; Vattuone, L.; Lacombe, S.; Jacobi, K.; Tang, K. B. K.; Palmer, R. E.; Villette, J.; Teillet-Billy, D.; Gauyacq, J. P. J. Phys.: Condens. Matter 2000, 12(6), R53–R82.
Beavers et al.
oxygen exposure activates low-temperature coarsening.52,53 Molecular oxygen adsorbs preferentially at kink sites on the silver surface where the activation barrier to dissociation is lowered.80,81,84,87-90 Migration of silver atoms occurs upon dissociation of oxygen from the surface.68,69,81,88 Thus, the greater the abundance of kink sites on the surface, the greater the extent of silver atom migration. As our XRD results have shown, nanorods are polycrystalline. In addition to the formation of silveroxide, exposure to air can result in adsorption of contaminants. Depending upon their chemical composition, contaminants can hasten the rate of migration91 and/or provide spectral interference for SERS sensing applications. Substrates exposed to atmospheric oxygen during fabrication (method one) began surface restructuring at lower temperatures and shorter exposure times than those held in vacuum for the entire process (method two). One rationalization for this observed difference is that the oxygen underlayer impacts nucleation. Layson et al.51 have demonstrated that when silver is deposited onto a layer of chemisorbed oxygen, silver atom mobility is increased resulting in the formation of larger islands with a concomitant decrease in island density. They propose several mechanisms that might account for this phenomenon. Fabrication of our Ag nanorod arrays by either method begins with nucleation. In the GLAD process,23 nanorods grow because the rate of Ag deposition exceeds the rate of adatom migration, and because the nanorods are self-shadowing. If the Ag islands formed during nucleation on the chemisorbed oxygen layer (during method one) redistribute and coalesce on the surface by the mechanisms proposed by Layson and co-workers, the base of the nanorods would broaden, resulting in a more conical-shaped nanorod. In contrast, in the absence of an oxide underlayer, a more cylindrical-shaped nanorod is anticipated. Comparison of the shape of the nanorods formed by the two methods (see Figures 1a and 2a) is suggestive of oxygen-induced coarsening of the nucleates. The difference in nanorod shape may also be due to (82) Messerli, S.; Schintke, S.; Morgenstern, K.; Nieminen, J.; Schneider, W. D. Chem. Phys. Lett. 2000, 328(4,5,6), 330–336. (83) Savio, L.; Vattuone, L.; Rocca, M.; De Renzi, V.; Gardonio, S.; Mariani, C.; del Pennino, U.; Cipriani, G.; Dal Corso, A.; Baroni, S. Surf. Sci. 2001, 486(1-2), 65–72. (84) Loffreda, D.; Dal Corso, A.; Baroni, S.; Savio, L.; Vattuone, L.; Rocca, M. Surf. Sci. 2003, 530(1-2), 26–36. (85) Savio, L.; Vattuone, L.; Rocca, M. Phys. Rev. B 2003, 67(4), 045406(5). (86) Caspersen, K. J.; Liu, D.-J.; Bartelt, M. C.; Stoldt, C. R.; Layson, A. R.; Thiel, P. A.; Evans, J. W. Comput. Mater. Chem. 2004, 91– 124. (87) Savio, L.; Vattuone, L.; Rocca, M. Appl. Phys. A: Mater. Sci. Process. 2007, 87(3), 399–404. (88) Valbusa, U.; De Mongeot, F. B.; Rocca, M.; Vattuone, L. Vacuum 1998, 50(3-4), 445–450. (89) Vattuone, L.; Gambardella, P.; Burghaus, U.; Cemic, F.; Cupolillo, A.; Valbusa, U.; Rocca, M. J. Chem. Phys. 1998, 109(6), 2490– 2502. (90) Savio, L.; Giallombardo, C.; Vattuone, L.; Kokalj, A.; Rocca, M. In Oxygen interaction at Ag(511): from chemisorption to the initial stages of oxide formation; 2008; Iop Publishing Ltd.: 2008. (91) Shen, M. M.; Liu, D. J.; Jenks, C. J.; Thiel, P. A.; Evans, J. W. J. Chem. Phys. 2009, 130(9), 094701(13).
Article
differences in substrate temperature during deposition.92,93 In essence, Ag nanorods are kinetic, not thermodynamic products. Oswald and Smoluchowski ripening coarsens the surface and promotes the return of nanorods to their more thermodynamically stable form. Exposure to air hastens silver atom migration and surface contamination. Thus, while the shelf life of the arrays can be improved by storing them at low temperature under vacuum or inert atmosphere, restructuring of the Ag surface is (92) Steele, J. J.; Brett, M. J. J. Mater. Sci.: Mater. Electron. 2007, 18(4), 367–379. (93) Neither evaporator is equipped with a substrate temperature monitor or means to control the substrate temperature during deposition. Thus, the actual temperature during deposition is unknown. A higher deposition temperature is expected for substrates prepared by method one since the platen that holds the substrates in the evaporator is closer to the silver source compared to the evaporator used for preparing substrates by method two.
Chem. Mater., Vol. 22, No. 7, 2010
2189
unavoidable. The ultimate impact of this process is the concomitant decrease in surface enhancement when these arrays are used as SERS substrates in sensing applications. Acknowledgment. We thank Prof. Mohan Srinivasarao and Dr. Matija Crne from the School of Polymer and Textile Fiber Engineering at Georgia Tech for extended use of their Raman spectrometer as well as Dr. Hsiao Yun Chu and Prof. Yiping Zhao of the University of Georgia for providing us with the Ag nanorod substrates prepared by method two. We also thank Melanie Kirkham and Prof. Robert L. Snyder of the School of Materials Science and Engineering at Georgia Tech for their assistance in acquiring the XRD data. K.R.B. gratefully acknowledges the support of the Beckman Foundation Undergraduate Scholarship program at Georgia Tech. Financial support of this research was provided by the Georgia Tech/UGA Biomedical Research program and the Georgia Research Alliance VentureLab program.
