A novel colloid probe preparation method based on chemical etching technique

A novel colloid probe preparation method based on chemical etching technique first-author corresp ‡Corresponding Author ‡Corresponding Author †E-mail:xugh@zju.edu.cn †E-mail:xugh@zju.edu.cn xugh@zju.edu.cn Department of Chemical Engineering, Zhejiang University, Hangzhou 310027, China Department of Chemical Engineering, Zhejiang University, Hangzhou 310027, China Several fundamental problems in hydrophobic force measurements using atomic force microscope (AFM) are discussed in this paper. A novel method for colloid probe preparation based on chemical etching technology is proposed, which is specially fit for the unique demands of hydrophobic force measurements by AFM. The features of three different approaches for determining spring constants of rectangular cantilevers, including geometric dimension, Cleveland and Sader methods are compared. The influences of the sizes of the colloids on the measurements of the hydrophobic force curves are investigated. Our experimental results showed that by selecting colloid probe with proper spring constant and tip size, the hydrophobic force and the complete hydrophobic interaction force curve can be measured by using AFM. Hydrophobic force Atomic force microscope (AFM) Colloid probe Chemical etching INTRODUCTION 1 Hydrophobic force is the unusually strong and long-range attractive force exhibited between hydrophobic surfaces in water (Israelachvili and Pashley, 1982 1982 ), which is among the most important nonspecific interactions in biological systems and plays a central role in many surface phenomena. Although many studies were carried out over the past 20 years, the nature of the hydrophobic force is still not clear, no single theory can account for all observed experimental behavior, even the experimental data on similar systems is often contradictory (Ishida et al., 2002 2002 ; Meyer et al., 2005 2005 ). Accurate measurement of interaction forces between different surfaces is important for a variety of natural phenomena and industrial processes. The emergence of the atomic force microscope (AFM) enables researchers to directly measure the interaction force between a sharp tip and a flat surface both in air and in water. But the measured force cannot be compared with theory, because the geometries of the tip are not regular or even too small to be known. After Ducker et al.( 1991 1991 ) used colloid probe technology to successfully measure the interaction forces between a colloid sphere and a flat surface in aqueous solutions, the measured forces could be theoretically analyzed and compared with theory for the first time. The measurement of hydrophobic force by AFM is somewhat difficult, since the cantilever might stick to the flat surface during the measurement due to the large adhesion forces if a small spring constant cantilever was used. Rabinovich and Yoon ( 1994 1994 ) solved this problem by using a stiffer rectangular silicon cantilever with a spring constant of 30~120 N/m. However, there are some sticky troubles in preparation of the colloid probes (Ducker et al., 1991 1991 ; Rabinovich and Yoon, 1994 1994 ). In the traditional colloid probe preparation processes, a silica glass bead is used as the colloid particle and mounted near the apex of the cantilever using an extremely small amount of epoxy resin by means of a micromanipulator. Due to the small size of the glass beads, normally in diameters of 7~30 μm, the adhesion process should be carefully controlled to make the glass bead just adhered to the right position on the cantilever without large tolerance to avoid possible errors in the following surface force measurement. It is a really hard job, and the success rate of the preparation is mainly dependent on the experiences of the operator. Another sticky business is the chemical modification of the colloid particles. To measure the hydrophobic force, both the originally hydrophilic glass bead and the flat sample surfaces, normally the silicon wafer chip used as the substrate, should be modified to be hydrophobic by molecular self-assembly technology, typically the OTS SAM (octadecyltrichlorosilane self-assembled monolayer). But, for a single glass bead with a diameter about 30 μm, the pre-cleaning treatments, chemical modification as well as the final cleaning procedures after the modification reaction are also arduous work. The hydrophobicity and quality of the hydrophobic SAM on the modified small bead surface cannot be guaranteed, and the quality flaws of the SAMs could finally spoil the accurate measurement of the surface forces. Here, we propose a novel method for colloid probe preparation based on the chemical etching technology, which potentially overcomes the shortcomings in the traditional processes and contributes greatly to the improvement of the surface force measurements by AFM. By using the chemical etched colloid probe, we measured the hydrophobic interactions by AFM. Our results showed that the hydrophobic force and the complete interaction force curve could be measured by using AFM. The calibration of the spring constants of the cantilevers is also a very important factor for surface force measurement. Three different calibration approaches, including the geometric dimension, Cleveland and Sader methods, were experimentally compared, the features of the three methods are discussed in this paper. EXPERIMENTAL DETAILS 1 Preparation of the flat hydrophobic sample surfaces 2 The hydrophobic sample surfaces were prepared by using OTS SAM technology with the experimental details being similar to those of Xu and Higashitani ( 2000 2000 ). Silicon (100) plates (1 cm×1 cm) were used as the solid substrates, which were cleaned by sonication in an ultrasonic cleaning tank in acetone, isopropanol (IPA) and distilled water for 5 min, respectively. After each sonication, the substrate plates were rinsed with distilled water, then finally sonicated in superpure de-ionized water (18.23 MΩ·cm) for another 5 min and dried in a stream of nitrogen gas. The well-cleaned silicon plates were immersed into a hot mixture solution composed of concentrated sulfuric acid and hydrogen peroxide (20 ml:10 ml) at 90 °C for 30 min, so that the silicon surface was oxidized to silicon dioxide. After the reaction, the silica plates were rinsed again with super-pure de-ionized water and dried in a stream of nitrogen gas. The well-cleaned and oxidized silica plates were immersed into a freshly prepared 1×10 −3 mol/L OTS-toluene solution at (18.0±0.1) °C for 30 min, and then withdrawn from the solution as quickly as possible. The modified silica plates were rinsed with toluene, acetone, chloroform and IPA in turn to remove any excess deposit, and dried in a stream of nitrogen gas. The average surface mean roughness of the OTS SAM was (0.23±0.02) nm/(5×5) μm 2 and the contact angle was 107°~110°. OTS (octadecyltrichlorosilane) was purchased from Acros Organics, USA, and used as received without further treatment. Toluene, chloroform, acetone, IPA, concentrated sulfuric acid and hydrogen peroxide were all of analysis reagent grade. Colloid probe preparation 2 Two different preparation methods were used in this experiment. Fig. 1a 1a shows the colloid probe prepared by the traditional process. After the modification, the glass bead (diameter of 27 μm, Duke Scientific, USA) was assumed to be coated with an OTS SAM, then adhered to a rectangular cantilever (spring constant of 40.7 N/m, MikroMash Co.) on the position just after the probe tip using a small amount of epoxy resin by means of a micromanipulator (PCS-5000 Series, Burleigh Instrument Inc., USA). Fig. 1b 1b shows the colloid probe made by the chemical etching technique, the cantilever was also from MikroMash Co. with a spring constant of 62.9 N/m. The chemical etching process under the operation of the micromanipulator is clearly shown in Fig. 2 2 . This skill is based on the technology of the wet isotropic etching of silicon, which is widely used in the MEMS (micro-electro mechanical systems) fabrication engineering for removal of work damaged surfaces, rounding of sharp anisotropically etched corners, removing of roughness, etc. (Madou, 1997 1997 ). The etchant used in this experiment was a mixture of HNO 3 -HF, which was designed according to the principle in (Schwartz and Robbins, 1976 1976 ). After the etching, the top of the probe tip became rounded and nearly a hemisphere with diameter of about 1.15 μm (Fig. 1b 1b ). The reproducibility of the etching process is good. Generally, after about 3~5 min etching, the diameters of the tops of the cantilever tips were about 1~3 μm. The etched cantilever probe was then washed in abundant de-ionized water to remove the residual etchant solution. The chemical modification and the following cleaning procedures for the chemically etched colloid probe were the same as those for flat hydrophobic sample surface. Hydrophobic force measurement 2 A multimode AFM (NanoIIIa, Digital Instrument, USA) was employed to measure the hydrophobic forces. All the measurements were conducted in de-ionized water (18.23 MΩ·cm) by using a liquid cell. Force measurements were conducted at room temperature ((20±2) °C), and always started about 30 min after the injection of the de-ionized water into the liquid cell. RESULTS AND DISCUSSION 1 Calibration of the spring constants of the cantilevers 2 Calibration of the spring constants of AFM cantilevers is very important in surface force measurements by AFM. For an end-loaded cantilevered beam of uniform rectangular cross section, the spring constant, k , could be calculated from the beam’s geometric dimensions, theoretically (Stokey, 1989 1989 ) $$k = Et^3 w/4l^3$$ where, E is Young’s Modulus, t thickness, w width and l length. This is exactly the case of the commercial rectangular silicon cantilevers usually used in the hydrophobic force measurements. Unfortunately, the thickness, t , of the cantilever is seldom uniform due to the limitation of the present micro-fabrication techniques (Fig. 3 3 ), so that measurement error cannot be avoided. Cleveland et al.( 1993 1993 ) considered the variations in the beam’s thickness and gave a relatively simple method to determine the spring constants of arbitrarily shaped thin cantilevers, specifically, by measuring the resonant frequency of the beam $$k = 2\pi ^3 l^3 w(\rho ^3 /E)^{1/2} v_o^3$$ here, ρ is the material density and ν o the resonant frequency, measurement of which, however, can be strongly influenced by the air damping effects, the gold coating on the cantilevers, and the load position of the added mass. To eliminating these influence factors, Sader et al.( 1995 1995 ; 1999 1999 ) proposed a new method, $$k = 0.1906\rho _f w^2 lQ_f \Gamma _i (\omega _f )\omega _f^2 ,$$ where, ρ f is the air density, Q f the quality factor in air, Γ i ( ω f ) the imaginary component of the hydrodynamic function, and ω f the fundamental radial resonant frequency in air. The technology and procedure of the Sader method is relatively much more complex. At present, Sader method is regarded as the standard calibration technique, but the other two methods are still used by some research groups due to its simplicity. Table 1 1 compares the three different calibration methods. Two rectangular cantilevers from MikroMash were carefully selected as the test samples (Fig. 3 3 ). Table 1 1 shows that compared to the Sader method, the relative errors of the dimension method are at least 8%, while the errors of the Cleveland method are about 3%~5%. The resonant frequency was measured in tapping mode in air by AFM The spring constants by Sader method were calibrated and offered by MikroMash Co. Calculated from Eq.( 2 2 ) with the bulk density of the crystal silicon being 2.9635×10 3 kg/m 3 and the bulk Young’s Modulus 1.9×10 11 N/m Calculated from Eq.( 1 1 ) The relative error is defined as | k Sader − k |/ k Sader ×100% 2 Cantilever No. 3 Dimensions 2 ν o a a (kHz) Sader b b 2 Cleveland c c 2 Geometric dimension d d l̄ (μm) w̄ (μm) t̄ (μm) k (N/m) k (N/m) Error e e (%) k (N/m) Error e e (%) 1 130 38 4.20 304.95 56.30 54.25 3.4 60.87 8.1 2 129 39 4.15 314.39 56.80 59.62 4.9 61.68 8.6 Hydrophobic forces measured by the glass bead probe and the chemical etching probe 2 Fig. 4 4 shows two typical retracting force curves measured under the same experimental conditions by the glass bead probe shown in Fig. 1a 1a and the chemical etching probe in Fig. 1b 1b , respectively. It is noted that, for the same OTS SAM surface systems in both cases, the adhesion forces measured by two probes are utterly different (Table 2 2 ). What is the real, or much more approaching to the real hydrophobic interaction? We do not know the strict criteria for the hydrophobic interaction up to now, but an obvious criterion is that, at least, the interfacial energy calculated from the measured adhesion force should be consistent with the literature values measured by the early pioneers by means of the traditional or classical methods. Probes Adhesion force F a / R (mN/m) Interfacial energy γ SL (mJ/m 2 ) Glass bead probe minus;94.76 10.6 Chem. etching probe −465.06 49.4 Ishida et al.( 2001 2001 ) ~ −270 ~ 28.7 According to JKR theory (Johnson et al., 1971 1971 ), to pull a solid (S) sphere of radius R off a flat solid (S) surface in a liquid (L), the required maximum force, i.e. the adhesion force, F a , is $$F_a /R = - 3\pi \gamma _{SL} ,$$ where, γ SL is the interfacial energy of the solid (S)-liquid (L) interface. The interfacial energies of typical hydrophobic surface-water interfaces are around 50 mJ/m 2 (Israelachvili, 1992 1992 ), e.g., for paraffin wax-water interface γ SL is 50 mJ/m 2 , for n -octane (saturated)-water interface, 51 mJ/m 2 , and octadecane (C 18 H 18 )-water interface, 52 mJ/m 2 . Obviously, the hydrophobic interaction measured by the chemical etching probe much more approaches to the real hydrophobic force. What is wrong with the glass bead probe? One of the direct factors arrived at by intuition is the big size of the colloid particle, which could result in a big hydrodynamic resistance during the force measurements in water. But, in the similar experiments by Ishida et al.( 2001 2001 ), the hydrodynamic effects was convincingly eliminated completely by means of a static method. However, the adhesion force, as well as the calculated interfacial energy of the OTS SAM in water was still much lower than the typical values (Table 2 2 ). The reasons may be the load position of the colloid particle and its big size. Sader et al.( 1995 1995 ) pointed out that the restoring force of the cantilever acting on an attached sphere is a strong function of the sphere position. When the colloid particle load is applied a distance Δ l away from the end tip, the spring constant k is related to the end tip spring constant k E by $$k = k_E (l/(l - \Delta l))^3 .$$ A glance at the numerous micrographs of colloid probes reveals that in many cases, these spheres are attached at least 10%~15% backward from the cantilever tip since the diameters of the used sphere is at least 7~30 μm. Therefore the real spring constant, as well as the measured forces, should be 30%~60% higher than that predicted by theory, which assumes that the load is at the end tip of the cantilever. Fortunately, our chemical etching probe technology minimizes the risks caused by the particle load position. CONCLUSION 1 A novel method for colloid probe preparation based on chemical etching technology was proposed, which is specially fit for the unique demands of the hydrophobic force measurement by using AFM. Our experiments showed that by using the chemical etching probe, accurate measurement of the hydrophobic interaction ensured. Our research also indicated that for the calibration of the spring constants of AFM cantilevers, compared to the Sader method, the relative errors of the dimension method are at least 8%, while the errors of the Cleveland method are about 3%~5%. References JP Cleveland , I Manne , D Bocek , PK Hansma . A nondestructive method for determining the spring constant of cantilevers for scanning force microscopy . Rev Sci Instrum , 1993 . 64 ( 2 ): 403 - 405 . DOI: 10.1063/1.1144209 http://doi.org/10.1063/1.1144209 . A nondestructive method for determining the spring constant of cantilevers for scanning force microscopy WA Ducker , TJ Senden , RM Pashley . Direct measurement of colloidal forces using an atomic force microscope . Nature , 1991 . 353 ( 6341 ): 239 - 241 . DOI: 10.1038/353239a0 http://doi.org/10.1038/353239a0 . Direct measurement of colloidal forces using an atomic force microscope N Ishida , M Sakamoto , M Miyahara , K Higashitani . Static method to evaluate interaction force by AFM . Journal of Colloid and Interface Science , 2001 . 235 ( 1 ): 190 - 192 . DOI: 10.1006/jcis.2000.7212 http://doi.org/10.1006/jcis.2000.7212 . Static method to evaluate interaction force by AFM N Ishida , M Sakamoto , M Miyahara , K Higashitani . Optical observation of gas bridging between hydrophobic surfaces in water . Journal of Colloid and Interface Science , 2002 . 253 ( 1 ): 112 - 116 . DOI: 10.1006/jcis.2002.8444 http://doi.org/10.1006/jcis.2002.8444 . Optical observation of gas bridging between hydrophobic surfaces in water JN Israelachvili . Intermolecular and Surface Forces , 1992 . 2nd Ed. : London Academic Press , 315 . Intermolecular and Surface Forces JN Israelachvili , RM Pashley . The hydrophobic interaction is long range, decaying exponentially with distance . Nature , 1982 . 300 ( 5890 ): 341 - 342 . DOI: 10.1038/300341a0 http://doi.org/10.1038/300341a0 . The hydrophobic interaction is long range, decaying exponentially with distance KL Johnson , K Kendall , AD Roberts . Surface energy and the contact of elastic solids . Proc Roy Soc, Ser A , 1971 . 324 301 - 313 . . Surface energy and the contact of elastic solids MT Madou . Fundamentals of Microfabrication , 1997 . : Boca Raton, FL CRC Press , . Fundamentals of Microfabrication EE Meyer , Q Lin , JN Israelachvili . Effects of dissolved gas on the hydrophobic attraction between surfactant-coated surfaces . Langmuir , 2005 . 21 ( 1 ): 256 - 259 . DOI: 10.1021/la048318i http://doi.org/10.1021/la048318i . Effects of dissolved gas on the hydrophobic attraction between surfactant-coated surfaces YI Rabinovich , RH Yoon . Use of atomic force microscope for the measurement of hydrophobic force between silanated silica plate and glass sphere . Langmuir , 1994 . 10 ( 6 ): 1903 - 1909 . DOI: 10.1021/la00018a048 http://doi.org/10.1021/la00018a048 . Use of atomic force microscope for the measurement of hydrophobic force between silanated silica plate and glass sphere JE Sader , I Larson , P Mulvaney . Method for the calibration of atomic force microscope cantilevers . Rev Sci Instrum , 1995 . 66 ( 7 ): 3789 - 3798 . DOI: 10.1063/1.1145439 http://doi.org/10.1063/1.1145439 . Method for the calibration of atomic force microscope cantilevers JE Sader , JWM Chon , P Mulvaney . Calibration of rectangular atomic force microscope cantilevers . Rev Sci Instrum , 1999 . 70 ( 10 ): 3967 - 3969 . DOI: 10.1063/1.1150021 http://doi.org/10.1063/1.1150021 . Calibration of rectangular atomic force microscope cantilevers B Schwartz , H Robbins . Chemical etching of silicon . J Electrochem Soc , 1976 . 123 ( 12 ): 1903 - 1909 . . Chemical etching of silicon WF Stokey . Shock and Vigration Handbook , 1989 . : New York McGraw-Hill , 7.1 - 7.44 . . Shock and Vigration Handbook GH Xu , K Higashitani . Formation of OTS self-assembled monolayer on glass surface investigated by AFM . Journal of Zhejiang University SCIENCE , 2000 . 1 ( 2 ): 162 - 170 . . Formation of OTS self-assembled monolayer on glass surface investigated by AFM We are most grateful to Prof. Xiaoxiang Zheng, Dr. Huijuan Yang, Ms. Guifeng Xiao in the Dept. of Biomedical Eng. and Instrument Science at Zhejiang University, and Prof. Juncai Sun, Ms. Sha Liu and Ms. Yaqing Shi in the Dept. of Materials at Dailian Maritime University. 10.1631/jzus.2006.B0304 Project (No. 20276057) supported by the National Natural Science Foundation of China 2005-07-26 2005-11-10 2006-03-13 2022-01-05 Copyright © 2006, Journal of Zhejiang University Science 2006 Journal of Zhejiang University-SCIENCE B (Biomedicine and Biotechnology) 04 07