Controlled Nucleation and Growth of Surface-Confined Gold

Au-Ring Microelectrodes for Monitoring Transmitters Released from Single Cells .... The Journal of Physical Chemistry C 0 (proofing), ..... Sensor...
3 downloads 0 Views 400KB Size
Download PDF
Anal. Chem. 2001, 73, 2843-2849

Controlled Nucleation and Growth of Surface-Confined Gold Nanoparticles on a (3-aminopropyl)trimethoxysilane-Modified Glass Slide: A Strategy for SPR Substrates Yongdong Jin, Xiaofeng Kang, Yonghai Song, Bailin Zhang, Guangjin Cheng, and Shaojun Dong*

Laboratory of Electroanalytical Chemistry, Changchun Institute of Applied Chemistry, Chinese Academy of Sciences, Changchun 130022, China

The thickness of the gold film and its morphology, including the surface roughness, are very important for getting a good, reproducible response in the SPR technique. Here, we report a novel alternative approach for preparing SPR-active substrates that is completely solution-based. Our strategy is based on self-assembly of the gold colloid monolayer on a (3-aminopropyl)trimethoxysilane-modified glass slide, followed by electroless gold plating. Using this method, the thickness of films can be easily controlled at the nanometer scale by setting the plating time in the same conditions. Surface roughness and morphology of gold films can be modified by both tuning the size of gold nanoparticles and agitation during the plating. Surface evolution of the Au film was followed in real time by UVvis spectroscopy and in situ SPRS. To assess the surface roughness and electrochemical stability of the Au films, atomic force microscopy and cyclic voltammetry were used. In addition, the stability of the gold adhesion is demonstrated by three methods. The as-prepared Au films on substrates are reproducible and stable, which allows them to be used as electrodes for electrochemical experiments and as platforms for studying SAMs. Optical surface plasmon resonance (SPR) spectroscopy is a powerful tool for in situ real-time characterization of solid/liquid interfaces. In the past decade, SPR has found increasingly widespread use for the study of interactions of biological molecules in optical biosensors.1 The refractive index measurement of SPR is the basis of the recently introduced commercial BIACORE SPR instrument.2,3 The assembly or binding process can be monitored without the need to label the reactants with spectroscopic or radioactive probes, making this technique an ideal noninvasive, in situ method amenable to a wide range of biologically relevant molecules, including nucleic acids, proteins, lipids, and carbohydrates.4 In chemistry, SPR measurements have also been used to study Langmuir-Blodgett films,2,5,6 self-assembled organic * To whom correspondence should be addressed. Fax: +86-0431-5689711. E-mail: [email protected]. (1) Garland, P. B. Q. Rev. Biophys. 1996, 29, 91-117. (2) Lofa`s, S.; Malmqvist, M.; Ronnberg, I.; Stenberg, E.; Liedberg, B.; Lundstrom, I. Sens. Actuators, B. 1991, 5, 79. (3) Malmqvist, M. Nature 1993, 361, 186. 10.1021/ac001207d CCC: $20.00 Published on Web 05/10/2001

© 2001 American Chemical Society

monolayers,7,8 specifically and nonspecifically adsorbed biopolymers (including DNA),9-16 nonlinear optical thin films,17 and thin organic films at electrochemical interfaces.18,19 In practice, a thin film of Au (or other noble metal) with an ∼45- to 60-nm thickness20 deposited on a glass substrate is used for the excitation of surface plasmon modes in the common Kretschmann configuration. To our knowledge, so far, almost all SPR-active substrates are prepared by vacuum deposition or the sputtering of metal by using expensive instruments having timeconsuming operations. Moreover, noble metals such as Au and Ag do not adhere well to glass21 unless that glass is plasma-cleaned in the vacuum system just before laying down the metal. With significant exposure to water, the film will be damaged, limiting its applicability. To enhance the gold adhesion to the substrate, Cr (or other metals) was used as a undercoating layer, which brings some problems related to Cr (Ti, W) diffusion along grain boundaries with the noble-metal surface.22-25 This phenomenon (4) Georgiadis, R.; Peterlinz, K. P.; Peterson, A. W. J. Am. Chem. Soc. 2000, 122, 3166-3173. (5) Lawrence, C. R.; Martin, A. S.; Sambles, J. R. Thin Solid Films 1992, 208, 269. (6) Wijekoon, W. M. K. P. et al. Langmuir 1992, 8, 135. (7) Peterlinz, K. A.; Georgiadis, R. Langmuir 1996, 12, 4731. (8) Lang, H.; Duschl, C.; Gratzel, M.; Vogel, H. Thin Solid Films 1992, 210/ 211, 818. (9) Jordan, C. E.; Corn, R. M. Anal.Chem. 1997, 69, 1449. (10) Jordan, C. E.; Frutos, A. G.; Thiel, A. J.; Corn, R. M. Anal. Chem. 1997, 69, 4939. (11) Jordan, C. E.; Frey, B. L.; Kornguth, S.; Corn, R. M. Langmuir 1994, 10, 3642. (12) Mrksich, M.; Sigal, G. B.; Whitesides, G. M. Langmuir 1995, 11, 4383. (13) Peterlinz, K. A.; Georgiadis, R.; Herne, T. M.; Tarlov, M. J. J. Am. Chem. Soc. 1997, 119, 3401. (14) Spinke, J.; Liley, M.; Guder, H.-J.; Argermaier, L.; Knoll, W. Langmuir 1993, 9, 1821. (15) Thiel, A. J.; Frutos, A. G.; Jordan, C. E.; Corn, R. M.; Smith, L. M. Anal. Chem. 1997, 69, 4948. (16) Terrettaz, S.; Stora, T.; Duschl, C.; Vogel, H. Langmuir 1993, 9, 1361. (17) Hanken, D. G.; Corn, R. M. Anal. Chem. 1995, 67, 3767. (18) Hanken, D. G.; Naujok, R. R.; Gray, J. M.; Corn, R. M. Anal. Chem. 1997, 69, 240. (19) Hanken, D. G.; Corn, R. M. Anal. Chem. 1997, 69, 3665. (20) Raether, H. Surface Plasmons; Springer-Verlag: Berlin, 1988. (21) Janata, J. Principles of Chemical Sensors; Plenum Press: New York 1989. (22) Josowicz, M.; Janata, J.; Levy, M. J. Electrochem. Soc. 1988, 135, 112-115. (23) Tisone, T. C.; Drovek, J. J. Vac. Sci. Technol. 1972, 9, 271-275. (24) Ashwell, G. W. B.; Heckingbottom, R. J. Electrochem. Soc. 1981, 128, 649654.

Analytical Chemistry, Vol. 73, No. 13, July 1, 2001 2843

Figure 1. Assembly strategy for SPR substrates by electroless Au plating on surface-confined colloidal Au monolayer.

leads inevitably to the altering of surface properties of the gold and may affect electron transfer and adsorption process,22,26 making them unsuitable for electrochemical experiments. Although some other methods have been developed for enhancing adhesion between vapor-deposited Au films and glass or Si-based substrates,27,28 the relative difficulty of these methods in controlling the film thickness and morphology in combination with an expensive vacuum deposition system results in a major obstacle to generalizing SPR methodology in wet-chemical laboratories. In this paper, an alternative approach is presented that is based on self-assembly of a gold colloid monolayer on a APTMS-modified glass slide and followed by electroless gold plating (Figure 1). This strategy provides a completely solution-based method for the fabrication of SPR-active substrates. There are several reasons for us to propose this strategy: (1) Aqueous solutions of colloidal Au are easy to obtain and can be prepared in a wide range of diameters (2.5-120 nm) with relatively high monodispersity.29-31 As a result, surfaces made from colloidal Au have a tunable nanometer-scale roughness defined solely by the particle diameter. (2) A high-coverage Au colloid monolayer can be prepared by room-temperature particle growth of a small-diameter Au colloid monolayer via electroless plating.32 (3) Electroless gold plating has been extensively studied and is easy to perform in wetchemical laboratories.33 Most importantly, the film thickness of Au can be controlled by the initial particle diameter and the amount of Au3+ ions added or plated.32 As detailed below, this approach has yielded substrates that are SPR-active. The growth of the Au film in NH2OH/Au3+ mixtures can be easily followed by UV-vis spectrophotometry and in situ SPR spectroscopy. We also characterize the topography of the substrate by AFM and assess the substrate with regard to (1) the Au adhesion, and (2) the electrochemistry feasibility. The results demonstrate the superior characteristics of these substrates in comparison to those produced by vacuum deposition. The advantage of our method over that prepared by vacuum deposition is that ours is completely solution based and costeffective; thus, it suggests a widespread use for SPR substrates. (25) Holloway, P. H. Gold Bull. 1978, 12, 99-106. (26) Cohen, R. M.; Janata, J. J. Electroanal. Chem. Interfacial Electrochem. 1983, 151, 33-39. (27) Goss, C. A.; Charych, D. H.; Majda, M. Anal. Chem. 1991, 63, 85-88. (28) Baker, L. A.; Zamborini, F. P.; Sun, L.; Crooks, R. M. Anal. Chem. 1999, 71, 4403-4406. (29) (29) Collodal Gold: Principles, Methods, and Applications; Hayat, M. A., Ed.; Academic Press: San Diego, 1989, Vols. 1-2. (30) Frens, G. Nature, Phys. Sci. 1973, 241, 20-22. (31) Goodman, S. L.; Hodges, G. M.; Trejdosiewicz, L. K.; Livingston, D. C. J. Microsc. 1981, 123, 201-213. (32) Brown, K. R.; Natan, M. J. Langmuir 1998, 14, 726-728. (33) Hou, Z. Z.; Abbott, N. L.; Stroeve, P. Langmuir 1998, 14, 3287-3297 and references therein.

2844

Analytical Chemistry, Vol. 73, No. 13, July 1, 2001

EXPERIMENTAL SECTION Chemicals. All aqueous solutions were made using demineralized water, which was further purified with a Milli-Q system (Millipore). The following materials were obtained from Aldrich: H2O2, HAuCl4‚3H2O, trisodium citrate dihydrate, (3-aminopropyl)trimethoxysilane (APTMS), hydroxylamine hydrochloride, and NaBH4. CH3OH (spectrophotometric grade) was obtained from EM Sciences. K2Cr2O7, H2SO4, and glass microscope slides (∼1.8 × 1.8 cm) were obtained from China. All of the chemicals, unless mentioned otherwise, were of analytical grade and were used as received. Preparation of Colloidal Particles. All glassware used in the following procedures was cleaned in a bath of freshly prepared 3:1 HCl:HNO3 (aqua regia) and rinsed thoroughly in H2O prior to use. Preparation of Au “seed colloid” solution of ∼2.5-nmdiameter particles was performed by adding 1 mL of 1% aqueous HAuCl4‚3H2O to 100 mL of H2O with vigorous stirring, followed by the addition of 1 mL of 1% aqueous sodium citrate 1 min later. After an additional 1 min, 1 mL of 0.075% NaBH4 in 1% sodium citrate was added. The solution was stirred for 5 min and then stored at 4 °C until needed.34 The formation of Au colloidal particles was examined using the UV-vis spectrum, which produces a strong surface plasmon band at 508 nm (not shown) characteristic of monodispersed) colloidal Au. Other nanosized colloidal gold particles were prepared according to Frens.30 Substrate Preparation. Substrate preparation and cleanliness are of critical importance in the experiments. Glass microscope slides were cleaned prior to use by soaking in a K2Cr2O7:conc H2SO4 solution overnight to remove particulate material from the surface of substrate. After thoroughly rinsing with deionized water, the slides were placed in freshly prepared piranha solution (3:1 H2SO4:30% H2O2) at 70 °C for ∼20 min to remove organic impurities. (Caution: piranha solution is a powerful oxidizing agent and reacts violently with organic compounds. It should be handled with extreme care.) The slides were rinsed thoroughly with deionized water and CH3OH, then were placed in a dilute solution of APTMS (0.3 mL of APTMS in 3 mL of CH3OH) for 12 h and rinsed with copious amounts of CH3OH upon removal.35 The polymer-coated slides were subsequently immersed in colloidal gold solution for 12-18 h for Au nanoparticles assembling. The gold monolayer were rinsed with water and used immediately for plating. Electroless Gold Plating. Substrates having a monolayer of nanosized gold particles were immersed in 6 mL of aqueous 0.4 mM hydroxylamine hydrochloride and 0.1% HAuCl4‚3H2O.36 All glassware was thoroughly cleaned with aqua regia. The solution was agitated to ensure the formation of a homogeneous Au film. The substrates changed color from pink to purple to blue and, finally, to reflective gold. The typical plating time for a good SPR response is ∼10 min. After plating, the substrates were rinsed (34) Grabar, K. C.; Allison, K. J.; Baker, B. E.; Bright, R. M.; Brown, K. R.; Freeman, R. G.; Fox, A. P.; Keating, C. D.; Musick, M. D.; Natan, M J. Langmuir 1996, 12, 2353-2361. (35) Freeman, R. G.; Grabar, K. C.; Allison, K. J.; Bright, R. M.; Davis, J. A.; Guthrie, A. P.; Hommer, M. B.; Jackson, M. A.; Smith, P. C.; Walter, D. G.; Natan, M. J. Science 1995, 267, 1629-1632. (36) Menzel, H.; Mowery, M. D.; Cai, M.; Evans, C. E. Adv. Mater. 1999, 11, 131-134.

thoroughly with water, dried under a nitrogen stream, and were ready for SPR experiments. Postplating Treatments. The following postplating treatments were developed to clean the films of electroless gold without delaminating the gold from its substrates. Electroless gold substrates were annealed at 250 °C for 3 h under the protection of N2 and then were electrochemically cleaned in 0.1 M H2SO4 by potential cycling between -0.2 and 1.4 V until the voltammograms did not change with further cycling. The samples were then rinsed with water and used as SPR substrates. SPR Measurements. Both in situ and ex situ scanning SPR experiments were performed to assess the electroless gold film. For these experiments, the glass slide with or without the electroless gold film was pressed onto the base of a half-cylindrical lens (n ) 1.566) via an index-matching oil. Linearly p-polarized light having a wavelength of 650 nm from a diode laser was directed through the prism onto the gold film in the Kretschmann configuration. The intensity of the reflected light was measured as a function of the angle of incidence, θ, using a photodiode with a chopper/lock-in amplifier technique. For ex situ SPR experiments, the as-prepared Au/glass substrates were mounted against the Teflon cuvette using a Kalrez O-ring, which provided a liquidtight seal. The Teflon cuvette allowed for the simultaneous recording of electrochemical data and the application of a voltage to the sample. The electroless gold film on the glass slide used for the excitation of surface plasmon modes also served as the working electrode. For in situ SPR experiments, another Teflon cell having a 3-mL volume was used to ensure enough plating solution. A series of SPR curves over time were obtained upon addition of plating solution to cell. UV-Vis Spectroscopy. Spectrometric studies were carried out using a Tracor Northern TN-6500 rapid scan spectrophotometer (Middleton, WI) using a 1-cm-light-path quartz cuvette. The glass slide coated with gold particles was placed perpendicular to the light beam inside the cuvette and exposed to plating solution. AFM Characterization. Surface images were acquired in tapping mode under ambient conditions (Nanoscope IIIa; Digital Instruments, Inc.). Si3N4 cantilevers having integral tips (spring constant, 20-100 N/m) were used. Images were obtained by oscillating the cantilever slightly below its resonance frequency (typically, 200-300 kHz) and raster scanning across the surface. Cyclic Voltammetry. Electrochemical experiments were performed using a PAR 370 electrochemical system (EG+G, PAR; USA) in a small, homemade, three-electrode, flow-through cell. The reference electrode was a KCl-saturated Ag/AgCl electrode, and the counter electrode was platinum. The potential of the gold electrode was cycled between -0.2 and 1.4 V at 0.1 V/s. The geometric surface area of the working electrode was ∼0.40 cm2. Gold Film Adherence Test. Three methods were used to evaluate the adherence of the gold films to the glass substrate, which include the cleaning test, the sonication test, and the tape test. In the cleaning test, a series of cleaning baths were used. In the tape test, a piece of clear Scotch tape (3M) was pressed against the film and pulled away at a constant rate of ∼0.5 cm/s. The statistical data were obtained from >5 samples. After the tape test, the sonication test was performed by immersing the substrate in

Figure 2. In situ SPR spectra of glass slide/APTMS/12-nm-diameter colloidal Au after electroless Au plating in HAuCl4/NH2OH for b, 0 ; c, 2; d, 5; e, 10; f, 15; and g, 20 min. Curve a is the SPR spectrum before self-assembly of the Au colloids.

a beaker of deionized water. The beaker was then placed in a water bath for sonicating e10 min. RESULTS AND DISCUSSION Control of Film Roughness and Morphology. The dependence of the SPR response on surface roughness is important, but it has not yet been well-investigated. In principle, the SPR spectrum of a bare gold film demonstrates that the position of the resonance is mostly a function of the extinction coefficient, K; the width of the resonance is mainly determined by the value of the refractive index, n; and the depth of the spectrum is predominantly controlled by the thickness of the metal film, t.37 However, in practice, we found that a relatively homogeneous and continuous gold film with ∼50 nm thickness and a few spots of nanoscopic surface roughness are both needed to get good SPR response. This is the reason why not all gold films having the same thickness that are prepared by vacuum deposition have satisfactory SPR responses. It has long been known that noble metals, such as Ag and Au, are weakly bound to a glass surface.38 Because of the weak binding forces between the metal film and the substrate, the metal atoms initially accumulate as islands during vacuum deposition. Therefore, it is difficult to obtain a homogeneous gold film that has a suitable thickness and nanoscopic surface roughness. Our initial experiments using sodium citrate to prepare 12nm gold particles as plating seeds were not successful. Because the immobilized particles on the substrate are still covered with citrate ions,39 there always existed an electrostatic repulsive force between the colloidal particles. Therefore, a submonolayer of physically separated particles was formed. The micrograph demonstrates a disordered particle deposition with an interparticle spacing of 10-50 nm.40 In this condition, subsequent gold plating in NH2OH/Au3+ mixtures failed to create a homogeneous film with a suitable surface roughness. Figure 2 shows in situ SPR spectra versus time during the course of plating. The SPR spectra obviously shift gradually toward a deeper and narrower response, which is indicative of the growth of surface-confined colloidal gold (37) Salamon, Z.; Macleod, H. A.; Tollin, G. Biochim Biophys. Acta 1997, 1331, 117-129. (38) Ye, Q.; Fang, J.; Sun, L. J. Phys. Chem. B 1997, 101, 8221. (39) Weitz, D. A.; Lin, M. Y. Surf. Sci. 1985, 158, 147. (40) Sato, T.; Brown, D.; Johnson, B. F. G. Chem. Commun. 1997, 1007.

Analytical Chemistry, Vol. 73, No. 13, July 1, 2001

2845

particles in this plating solution.32 Further growth causes the particles to coalesce and form a coherent gold layer;36 however, the final curve is still a wide-width one and there is a wide adsorption during the whole angle scan scope, which may be caused by broader particle distribution and heterogeneous surface roughness. To further prepare a high-coverage Au colloid monolayer, we synthesized a gold seed (of ∼2.5 nm-diameter) colloid solution and followed by chemisorption of this seed colloid to APTMS monolayer on the substrate. As expected, a relatively highcoverage gold nanoparticle monolayer with well-defined order and roughness is confirmed by AFM measurements (Figure 4a-1). It is well-known that the AFM image is a tip-sample convolution and the individual colloid particles (especially those of small diameter and tightly packed) are not visible.41 Although the topology in Figure 4a-2 mainly reflects the contour of glass slide (not shown), the height of the feature (∼2-3 nm) is consistent with the known size of gold nanoparticles (Figure 4a-2). Indeed, one big particle is composed of many small particles, which can further be confirmed by an AFM image with an enlarged scale. For further characterization, we examined the change of SPR curves by in situ experimentation in the subsequent plating (Figure 3). The SPR response obtained was better than that of the 12-nm gold particles. Although, because of the growth of the Au film, the damping of SPR causes the curves to be shallower after 5 min of plating, some modification in width can still be seen when they are compared with Figure 2. We also monitored surface evolution of colloidal gold film by the NH2OH/Au3+ system followed by UV-vis spectrophotometry in real time. After the addition of 3 mL of 0.4 mM NH2OH and 0.1% HAuCl4‚3H2O to the quartz cuvette, the growth of the colloidal Au surface plasmon band42 at ∼565 nm is obvious. Upon further growth, the particles coalesce, the particle coverage increases, and a new feature corresponding to a collective particle surface plasmon oscillation grows at regions beyond ∼650 nm35 (not shown). However, the resulting Au film is still far from a satisfactory SPR response. In fact, an islandlike morphology with high surface coverage is formed, which was confirmed by AFM image (Figure 4b). After

trials and errors, we found that agitation of the plating solution is useful and very important for controlling the morphology of the plated gold film. As can be seen from the AFM image (Figure 4c), a relatively homogeneous surface with notably reduced surface roughness can be fabricated by agitation during the plating, although there were still some small, shallow pits on the surface of film. In addition, surface roughness of the film depended mainly on the surface roughness of the glass slide that is used and can be further reduced by annealing. Deposition Rate and Film Thickness Control. Electroless plating permits the deposition of metals from solution onto surfaces without the need to apply an external electrical potential.43 The method is based on the chemical reduction of metal salts to metals at surfaces and is easy to perform in wet-chemical laboratories. In the plating method, NH2OH is thermodynamically capable of reducing Au3+ to bulk metal; however, the reaction is dramatically accelerated at gold surfaces. Therefore, no new particle nucleation occurs, and all Au3+ ions go into production of larger particles if colloidal gold is present. This allows production of monolayers of larger colloidal gold particles that are not available by direct methods.44 Upon further growth, the particles coalesce and form a coherent gold layer that shows a significantly higher long-range roughness than a gold colloid monolayer.36 On the basis of colloidal gold surface-catalyzed reduction of Au3+ by NH2OH, the approach for the initial nanoparticles growing into larger particles or films was determined solely by the initial particle diameter and amount of Au3+ that was added.; however, it is noteworthy that the deposition rate is nonlinear. Such behavior can be clearly expressed by the Gibbs-Thomson equation, which is related to the excess free energy (∆E) of small metal particle with its dimensions (Γ):45,46 ∆E ) 2σVm /(ΓF), where Vm is the molar volume of the metal, σ is the specific free surface energy of the interface with the electrolyte, and F is the Faraday constant. According to this equation, the surface free energy will decrease as the dimensions of particle increase. This implies well that the deposition rate will slow, irrespective of the change of concentration during plating. Although the deposition rate is nonlinear, it is easy to control the film thickness by controlling the plating time for the same condition. Using 6 mL of a solution of 0.4 mM NH2OH and 0.1% HAuCl4‚3H2O as a plating bath, we found that a plating time of 10 min is suitable to obtain good SPR substrates with highly reproducible results (Figure 5). Because of the slow deposition rate near 50-nm thickness, the SPR response of the substrates prepared with plating time of 10 ( 1.0 min (corresponds to ∼45∼55 nm in thickness) is very similar to that of 10.0 min, which facilitates fabrication of reproducible SPR substrates. The longer the plating time, the thicker the gold film will be, and the more serious damping of SPR will occur, even disappear. Gold Adhesion. The APTMS/Au substrates freshly prepared with and without agitation were subjected to a variety of commonly employed cleaning methods. Complete adhesion was retained for the former after (a) rinse in water, (b) immersion in ethanol, (c) immersion in a 1:4 H2O2/concentrated H2SO4 (piranha) solution, and (d) immersion in 0.1 M NaOH. We note that adhesion was

(41) Grabar, K. C.; Brown, K. R.; Keating, C. D.; Stranick, S. J.; Tang, S.-l.; Natan, M. J. Anal. Chem. 1997, 69, 471. (42) Kreibig, U.; Vollmer, M. Optical Properties of Metal Clusters; SpringerVerlag: Berlin, 1995.

(43) Hajdu, J. B. Plating Surf. Finish. 1996, 83 (Sept.), 29-33. (44) Mulvaney, P.; Giersig, M. J. Chem. Soc., Faraday Trans. 1996, 92, 3137. (45) Cramp, J. H. W.; Hillson, P. J. J. Photogr. Sci. 1976, 24, 25-28. (46) Konstantinov, I.; Malinowaki, J. J. Photogr. Sci. 1975, 23, 1-5.

Figure 3. In situ SPR spectra of glass slide/APTMS/2.5-nmdiameter colloidal Au after electroless Au plating in HAuCl4/NH2OH for b, 0; c, 2; d, 5; e 10; f, 15; and g, 20 min. Curve a is the SPR spectrum before self-assembly of the Au colloids.

2846 Analytical Chemistry, Vol. 73, No. 13, July 1, 2001

Figure 4. AFM images acquired in ambient, tapping mode condition and line scans of a 2.5-nm-diameter colloidal Au monolayer on APTMScoated glass slide (a) as prepared, and (b) after immersion into a solution of 6 mL of 0.4 mM NH2OH/0.1% HAuCl4 for 10 min without agitation and (c) with agitation. The rms roughness values for b and c are 3.5 and 2.6 nm, respectively.

Analytical Chemistry, Vol. 73, No. 13, July 1, 2001

2847

Figure 5. SPR responses of electroless gold film with different plating times of a, 5; b, 10; and c, 15 min.

partly lost when the latter was immersed in piranha solution. The results are better than those of Au film on the substrate with a Cr underlayer prepared by vacuum deposition. Au films deposited on glass microscope slides having no adhesion layer (such as a Cr layer) often lift off when immersed in or rinsed with solvents. In the sonication test, the substrates were sonicated in water for e10 min. As a result, the films that were deposited with agitation endured sonication, but others deposited without agitation and by vacuum deposition lifted off or delaminated more or less during the sonication test. Peel testing is another qualitative method that is often used to assess the strength of adhesion between a thin film and its substrates.47 In this test, adhesive tape is firmly pressed into contact with a metal overlayer and then peeled off at a constant rate. The amount of metal transferred to the tape is a measure of the adhesive strength of the film. In the absence of an adhesion layer, vapor-deposited Au was always easily removed from the glass surface by peeling. Although an APTMS interlayer promoted gold adhesion, the results were the same for substrates freshly prepared by plating without agitation because of the porosity and roughness of these films. Assuming that agitation improves mass transport and helps to level the islands to a relatively homogeneous and less porous gold film, we found that better adhesion was obtained by this treatment, although ∼50% of the deposited film that was freshly prepared peeled off when the tape was pressed firmly to the surface and then pulled back, at a constant rate of ∼0.5 cm/s, with the leading edge of the tape parallel to the substrate surface. The substrates tested after several days of storage produced better results;