**4.2 Biomedical arrays**

602 Recent Advances in Nanofabrication Techniques and Applications

ZnO nanorods along the direction of the nanotubes. The quasi-perpendicular ZnO nanorods nucleate and grow uniformly and selectively throughout the superhydrophilic regions of the TiO2 nanotube surface by electric field assisted deposition at 90oC for 3 min, while no nanorods are observed in the superhydrophobic regions (Fig. 10a,b) (Lai et al., 2010a). The EDS spectra also reveal that the presence of Zn, Ti and O elements on the superhydrophilic regions, while the elemental components in the superhydrophobic areas are only Ti and O. The inset of Fig. 10b shows the hexagonal end facet of a vertically aligned ZnO nanorod with a diameter about 100–150 nm growing on top of the TiO2 nanotube array surface. Therefore, the density, size and orientation of ZnO nanorods are very sensitive to the presence of electric fields. A 3D AFM profile image (Fig. 10c) shows that the microscopic structure of the ZnO crystal deposition consisted in dense column arrays, which are induced and directed by the wettability template. The thickness of vertical ZnO nanorod film is in the range of 800–900 nm. Furthermore, the three dimensional confocal microscopy image (Fig. 10d) also shows that the growth of the ZnO nanorod pattern is identical with the

On the basis of the versatile superhydrophilic-superhydrophobic template, we can successfully control the growth of ZnO nanostructures in the superhydrophilic regions under mild reaction conditions and in the absence of seed and noble metal catalyst. In the superhydrophobic regions, the growth is suppressed. This special template can be utilized to generate different nanostructured ZnO patterns with clearly defined edges. Hence, it is expected that this novel micropatterned technique based on the superhydrophilicsuperhydrophobic template will become a powerful tool for fabricating various types of

Fig. 10. (a,b) Typical SEM images of the vertically aligned ZnO nanorods selectively grown on superhydrophilic patterning regions by the developed electric field assisted deposition

corresponding ZnO nanorod micropattern. The inset in (b) shows the higher magnified SEM image of a ZnO nanorod with hexagonal end facet. (c) 3D AFM image of the ZnO/TiO2 micropattern. (d) Confocal microscopy image of the perpendicular ZnO nanorod array.

technique at 90oC for 3 min. The inset in (a) shows the side view SEM image of the

superhydrophilic/superhydrophobic template.

micropatterned nanomaterials and devices.

The typical SEM image of TiO2 nanotube array surface before and after the deposition of OCP film by electrochemical technique for 5 min is shown in Fig. 11a,b. The superhydrophobic-superhydrophilic micropatterned TiO2 was used as a micro-template to selectively deposit nano-OCP crystals on the superhydrophilic regions by an electrochemical deposition to form a special micropatterned nano-OCP. The deposition electrolyte was consisted of 0.042 mol/L Ca(NO3)2 and 0.025 mol/L NH4H2PO4. The pH value was adjusted to approximately 4.2 with 0.05 mol/L NaOH solution. The precipitation was carried out galvanostatically at a cathodic current of 0.5 mA cm-2 under 67.5oC for a certain time (Wang et al., 2008). It can see that quasi-perpendicular ribbon-like crystals of several hundred nanometers in width are uniformly grown on the TiO2 nanotube array surface. Fig. 11c shows a typical fluorescence microscope of the superhydrophilic-superhydrophobic micropattern on TiO2 nanotube surface. As can be seen, the green dot patterns are clearly imaged through the fluorescence contrast between the UV-irradiated superhydrophilic and photomasked superhydrophobic regions. The photoirradiated dot exhibiting a uniformly stronger fluorescence against the surrounding dark background is due to the highly affinity to solution resulting in the absorption of fluorescent probes into the irradiated nanotube array films. Therefore, a clear well-defined fluorescence pattern based on the superhydrophobic-superhydrophilic pattern is obtained. Fig. 11d displays the identical patterning of OCP biomaterials deposited on the superhydrophilic-superhydrophobic patterns on TiO2 nanotube array surface. It is obvious that the size of the white OCP dots is equal to that of the superhydrophilic area on template, indicating the deposited regions were only located within the superhydrophilic dots where photocatalytic degradation of PTES-SAMs was performed.

Fig. 11. SEM images of the (a) TiO2 nanotube array film fabricated by electrochemical anodization; (b) OCP nanostructure layer on TiO2 nanotube array film by electrochemical deposition for 5 min. Optical fluorescence pattern of the superhydrophilicsuperhydrophobic template (c) and patterned OCP thin films selectively deposited in predefined superhydrophilic regions by electrochemical deposition for 5 min (d).

Extremely Wetting Pattern by Photocatalytic Lithography and Its Application 605

The *in vitro* experimental indicated that the superhydrophobic nanotube TiO2 layers exhibit a remarkable resistance to platelets attachment (Fig. 13) (Yang et al., 2010). It is indicated that the superhydrophobic nanotube TiO2 layers exhibit a remarkable resistance to platelets attachment. As shown in Fig. 13a,d,g, abundant platelets adhere to both the plain Ti surface and the superhydrophilic TiO2 nanotube layers after 30 min incubation. Comparatively, after 120 min incubation, a large number of platelets adhered and spread out on both the plain Ti surface (77 ± 7.4 per 5000 µm2, Fig. 13b,c) and the superhydrophilic surface which was obtained by exposing the TiO2 nanotubes under a UV irradiation (22 ± 1.5 per 5000 μm2, Fig. 13e,f), only very few of platelets (1 ± 0.8 per 5000 μm2) was found to adhere on the superhydrophobic TiO2 nanotube layers (Fig. 13h). Moreover, even though some platelets were occasionally seen attached on the superhydrophobic surface, they looked smooth without any growth of pseudopods (Fig. 13i), implying that the platelets adhered on the superhydrophobic TiO2 nanotube surface remain inactive and hardly grow and spread out for a long period. The quantities and morphologies of adhered platelets and their corresponding interactions on the different samples are illustrated in Figure 14. Therefore, the construction of superhydrophobic surface on biomedical implants could pave a way to

Fig. 14. Schematic illustration of the quantity and morphology of platelet and corresponding interactions on the three kinds of surfaces. (a) Plain Ti substrate; (b) Superhydrophilic TiO2

Based on photocatalytic lithography, we demonstrate a facile, rapid and practical approach to fabricate Ag nanoparticle (NP) patterns on TiO2 films by means of pulse-current electrodeposition technique (Huang et al., 2011). The size and density of as-deposited Ag NPs can be controlled by changing deposition charge density. Moreover, the resultant patterned Ag NP films exhibited particle size-as well as density-dependent UV-vis absorption and SERS enhancement effect. It was found that the patterned Ag NP films produced under the deposition charge density of 2.0 C cm−2 exhibited the intense UV-vis and Raman peaks. Two dimensional surface enhanced Raman scattering (SERS) mapping of

nanotubes; and (c) Superhydrophobic TiO2 nanotubes.

**4.3 Sensing devices** 

improve the blood compatibility of the biomedical devices and implants.

Fig. 12. SEM micrographs of MG-63 cells cultured on the patterned OCP coatings with different deposition time for 6 h, (a) 1 min; (b) 3 min; (c) 5 min; (d) higher magnification.

The *in-vitro* MG-63 cell tests were used to study the biological performance of the asobtained OCP micropatterns (Huang et al., 2010b). The results showed that MG-63 cells were found preferentially attached on the superhydrophilic regions with OCP thin films, while the superhydrophobic regions with the PTES monolayers can effectively prevented the adhesion of cells on the surface, indicating that the cells had the selective adhesion action on the tiny units of OCP films. Moreover, the cells adhered on the OCP film deposited for a longer period (5 min) are more active to spread on the OCP nanobelt covering surface. It is promising for developing a new cell chip for high throughput evaluation of the cell behaviors.

Fig. 13. SEM images of adhered platelets on various kinds of surfaces at 37◦C for different periods. (a-c) mechanically polished and cleaned Ti substrate; (d-f) superhydrophilic surface; (g-i) superhydrophobic surface; (a,d,g) 30 min; (b,e,h) 120 min; (c,f,i) magnified images of the corresponding images of (b,e,h).

The *in vitro* experimental indicated that the superhydrophobic nanotube TiO2 layers exhibit a remarkable resistance to platelets attachment (Fig. 13) (Yang et al., 2010). It is indicated that the superhydrophobic nanotube TiO2 layers exhibit a remarkable resistance to platelets attachment. As shown in Fig. 13a,d,g, abundant platelets adhere to both the plain Ti surface and the superhydrophilic TiO2 nanotube layers after 30 min incubation. Comparatively, after 120 min incubation, a large number of platelets adhered and spread out on both the plain Ti surface (77 ± 7.4 per 5000 µm2, Fig. 13b,c) and the superhydrophilic surface which was obtained by exposing the TiO2 nanotubes under a UV irradiation (22 ± 1.5 per 5000 μm2, Fig. 13e,f), only very few of platelets (1 ± 0.8 per 5000 μm2) was found to adhere on the superhydrophobic TiO2 nanotube layers (Fig. 13h). Moreover, even though some platelets were occasionally seen attached on the superhydrophobic surface, they looked smooth without any growth of pseudopods (Fig. 13i), implying that the platelets adhered on the superhydrophobic TiO2 nanotube surface remain inactive and hardly grow and spread out for a long period. The quantities and morphologies of adhered platelets and their corresponding interactions on the different samples are illustrated in Figure 14. Therefore, the construction of superhydrophobic surface on biomedical implants could pave a way to improve the blood compatibility of the biomedical devices and implants.

Fig. 14. Schematic illustration of the quantity and morphology of platelet and corresponding interactions on the three kinds of surfaces. (a) Plain Ti substrate; (b) Superhydrophilic TiO2 nanotubes; and (c) Superhydrophobic TiO2 nanotubes.

#### **4.3 Sensing devices**

604 Recent Advances in Nanofabrication Techniques and Applications

Fig. 12. SEM micrographs of MG-63 cells cultured on the patterned OCP coatings with different deposition time for 6 h, (a) 1 min; (b) 3 min; (c) 5 min; (d) higher magnification.

evaluation of the cell behaviors.

images of the corresponding images of (b,e,h).

The *in-vitro* MG-63 cell tests were used to study the biological performance of the asobtained OCP micropatterns (Huang et al., 2010b). The results showed that MG-63 cells were found preferentially attached on the superhydrophilic regions with OCP thin films, while the superhydrophobic regions with the PTES monolayers can effectively prevented the adhesion of cells on the surface, indicating that the cells had the selective adhesion action on the tiny units of OCP films. Moreover, the cells adhered on the OCP film deposited for a longer period (5 min) are more active to spread on the OCP nanobelt covering surface. It is promising for developing a new cell chip for high throughput

Fig. 13. SEM images of adhered platelets on various kinds of surfaces at 37◦C for different periods. (a-c) mechanically polished and cleaned Ti substrate; (d-f) superhydrophilic surface; (g-i) superhydrophobic surface; (a,d,g) 30 min; (b,e,h) 120 min; (c,f,i) magnified

Based on photocatalytic lithography, we demonstrate a facile, rapid and practical approach to fabricate Ag nanoparticle (NP) patterns on TiO2 films by means of pulse-current electrodeposition technique (Huang et al., 2011). The size and density of as-deposited Ag NPs can be controlled by changing deposition charge density. Moreover, the resultant patterned Ag NP films exhibited particle size-as well as density-dependent UV-vis absorption and SERS enhancement effect. It was found that the patterned Ag NP films produced under the deposition charge density of 2.0 C cm−2 exhibited the intense UV-vis and Raman peaks. Two dimensional surface enhanced Raman scattering (SERS) mapping of

Extremely Wetting Pattern by Photocatalytic Lithography and Its Application 607

interparticle spacing as well as the hot spots among the NPs (Felidj et al., 1999; Lu et al., 2005). The average surface enhancement factor for pyridine on the Ag NP film with a charge

Fig. 16a shows the typical SEM micrographs of the CdS nanosphere micropatterns after 3 min deposition on the superhydrophobic-superhydrophilic template of TiO2 nanotube films (Lai et al., 2010d). The bright rectangular areas corresponded to the deposition of CdS nanosphere crystals on the superhydrophilic regions. The boundary between the CdS pattern and the surrounding superhydrophobic regions is clearly visible at a higher magnification (inset). The dispersed CdS nanosphere crystals grew on the top of TiO2 nanotube arrays within the rectangular superhydrophilic region (Fig. 16c). Most of the crystals were less than 90 nm in diameter due to the confinement by the inner diameter of nanotube, though a few larger spheres (~120 nm) were seen across neighboring tube openings. While on the superhydrophobic areas (Fig. 16d), there was almost no CdS crystal. The high growth selectivity was also confirmed by the EDS analysis, revealing that CdS spheres easily nucleate and grow on the hydroxyl groups (–OH) terminated regions (Fig. 16e), but not on the –CF3 terminated areas (Fig. 16f). Since the difference of the water contact angle between the superhydrophilic and superhydrophobic regions is larger than 150°, electrolyte solution is preferentially presented on the uniform superhydrophilic dots. No water droplets go to the neighboring superhydrophobic regions. Although a few CdS particles resulted from homogeneous precipitation attached onto superhydrophobic surface due to van der Waals interactions and gravity, they can be easily removed by ultrasonication. Therefore, a clear and well-defined CdS pattern in line with the dimensions

of the superhydrophobic–superhydrophilic template has been obtained.

**(d) (e) (f)** 

Fig. 15. Typical element distribution maps of Ag (a),Ti (b), O (c), and optical microscopy(d) and the corresponding R6G SERS mapping (e) of the patterned Ag NP films with an area of 140 × 100 μm2 using the peak area at 614 cm−1 as the reference. (f) Raman spectra of the patterned Ag NP films from different charge density: curve A, bare subsrate; curve B, 0.5

C/cm2; C, 1.0 C/cm2; D, 2.0 C/cm2; and E, 2.5 C/cm2.

density of 2.0 C cm−2 was calculated to be 1*.*3×105.

Rhodamine 6G (R6G) on the patterned Ag NP films demonstrated a high throughput localized molecular adsorption and micropatterned SERS effect.

Furthermore, the elemental distributions of the as-prepared Ag NPs arrays were also observed by electron probe microanalyzer, which are shown in Fig. 15a-c. Figure 15a shows the Ag element distribution map. As shown in the map, the green dot patterns are clearly images obtained through element concentration contrast between the UV-irradiated superhydrophilic and photomasked superhydrophobic regions. The green dots exhibiting a uniform Ag concentration against the surrounding black regions indicate that Ag NPs are uniformly deposited and confined to the superhydrophilic regions. Figure 15b,c shows the element distribution maps of Ti and O, respectively, which are also in line with the dimensions of the photomask. The blue superhydrophilic regions (dot patterns) show lower Ti and O concentrations due to the preferential deposition of Ag NPs in the superhydrophilic areas, while the yellow and red superhydrophobic regions have higher Ti and O concentration. The corresponding line-scan signal intensity profiles of Ag, O, and Ti elements across the dot pattern (red line direction). The intense Ag signals in the superhydrophilic regions as well as the Ti and O signals in the superhydrophobic regions suggest that Ag NPs are deposited only in the dot areas and that the other regions are the exposed TiO2 nanotube films. The consistent intensity of the Ag signals indicates that Ag NPs are uniformly deposited on the superhydrophilic regions.

In addition, a two-dimensional point-by-point SERS mapping of the patterned Ag NP film whose deposited charge density is 2.0 C cm−2 was obtained using R6G as the probe molecule.

Figure 15d,e show the optical image and the corresponding SERS mapping image of the patterned Ag NP film. The mapping area was approximately 140 × 100 μm2 and the data acquisition time was 1 s. A signal to baseline from 594.0 to 623.4 cm−1 was chosen for the acquisition of the SERS mapping. The bright and dark areas respectively represent higher and lower intensity of the SERS signal. It is clear that the geometrically identical gray superhydrophilic areas (circle) with a strong SERS activity and the dark superhydrophobic areas without any SERS activity form a high-resolution SERS intensity distribution map. As can be seen from the SERS mapping results, most SERS peak area is in a very narrow intensity window as shown by the contrast in color codes. Furthermore, the SERS peak area is uniform over the superhydrophilic region with several high intensity spots represented by white color codes. On comparing the SERS mapping with the SEM detection, it is reasonable to conclude that the homogeneous SERS signal in the circle areas reflects the uniform dispersion of Ag NPs on the superhydrophilic areas. The high-resolution SERS intensity distribution and micropatterned SERS effect of the Ag NP film might make it potentially useful in high-throughput molecule detection and bio-recognition.

To gain insight into the dependence of SERS enhancement on the size and density of Ag NPs, the SERS spectra of R6G absorbed on the different patterned Ag NP films were detected, which are shown in Fig. 15f. Because of the small particle size and low density, which are not the optimum size and distribution for SERS, the signal enhancement is rather weak below the charge density of 0.5 C cm−2. The enhancement behavior of the substrate, however, is obviously improved under the charge density of 1.0 C cm−2. In particular, the patterned Ag NP film prepared under a charge density of 2.0 C cm−2 exhibits the highest intensity, which is attributed to large size and high density of Ag NPs, as shown by the SEM results. On increasing the charge density to 2.5 C cm−2, the signal becomes weaker. The sizecorrelated enhancement may be explained by the EM mechanism (Zeng et al., 2008). The intensity of the SERS signal might also be controlled by the NPs density, which changes the

Rhodamine 6G (R6G) on the patterned Ag NP films demonstrated a high throughput

Furthermore, the elemental distributions of the as-prepared Ag NPs arrays were also observed by electron probe microanalyzer, which are shown in Fig. 15a-c. Figure 15a shows the Ag element distribution map. As shown in the map, the green dot patterns are clearly images obtained through element concentration contrast between the UV-irradiated superhydrophilic and photomasked superhydrophobic regions. The green dots exhibiting a uniform Ag concentration against the surrounding black regions indicate that Ag NPs are uniformly deposited and confined to the superhydrophilic regions. Figure 15b,c shows the element distribution maps of Ti and O, respectively, which are also in line with the dimensions of the photomask. The blue superhydrophilic regions (dot patterns) show lower Ti and O concentrations due to the preferential deposition of Ag NPs in the superhydrophilic areas, while the yellow and red superhydrophobic regions have higher Ti and O concentration. The corresponding line-scan signal intensity profiles of Ag, O, and Ti elements across the dot pattern (red line direction). The intense Ag signals in the superhydrophilic regions as well as the Ti and O signals in the superhydrophobic regions suggest that Ag NPs are deposited only in the dot areas and that the other regions are the exposed TiO2 nanotube films. The consistent intensity of the Ag signals indicates that Ag

In addition, a two-dimensional point-by-point SERS mapping of the patterned Ag NP film whose deposited charge density is 2.0 C cm−2 was obtained using R6G as the probe molecule. Figure 15d,e show the optical image and the corresponding SERS mapping image of the patterned Ag NP film. The mapping area was approximately 140 × 100 μm2 and the data acquisition time was 1 s. A signal to baseline from 594.0 to 623.4 cm−1 was chosen for the acquisition of the SERS mapping. The bright and dark areas respectively represent higher and lower intensity of the SERS signal. It is clear that the geometrically identical gray superhydrophilic areas (circle) with a strong SERS activity and the dark superhydrophobic areas without any SERS activity form a high-resolution SERS intensity distribution map. As can be seen from the SERS mapping results, most SERS peak area is in a very narrow intensity window as shown by the contrast in color codes. Furthermore, the SERS peak area is uniform over the superhydrophilic region with several high intensity spots represented by white color codes. On comparing the SERS mapping with the SEM detection, it is reasonable to conclude that the homogeneous SERS signal in the circle areas reflects the uniform dispersion of Ag NPs on the superhydrophilic areas. The high-resolution SERS intensity distribution and micropatterned SERS effect of the Ag NP film might make it

potentially useful in high-throughput molecule detection and bio-recognition.

To gain insight into the dependence of SERS enhancement on the size and density of Ag NPs, the SERS spectra of R6G absorbed on the different patterned Ag NP films were detected, which are shown in Fig. 15f. Because of the small particle size and low density, which are not the optimum size and distribution for SERS, the signal enhancement is rather weak below the charge density of 0.5 C cm−2. The enhancement behavior of the substrate, however, is obviously improved under the charge density of 1.0 C cm−2. In particular, the patterned Ag NP film prepared under a charge density of 2.0 C cm−2 exhibits the highest intensity, which is attributed to large size and high density of Ag NPs, as shown by the SEM results. On increasing the charge density to 2.5 C cm−2, the signal becomes weaker. The sizecorrelated enhancement may be explained by the EM mechanism (Zeng et al., 2008). The intensity of the SERS signal might also be controlled by the NPs density, which changes the

localized molecular adsorption and micropatterned SERS effect.

NPs are uniformly deposited on the superhydrophilic regions.

interparticle spacing as well as the hot spots among the NPs (Felidj et al., 1999; Lu et al., 2005). The average surface enhancement factor for pyridine on the Ag NP film with a charge density of 2.0 C cm−2 was calculated to be 1*.*3×105.

Fig. 16a shows the typical SEM micrographs of the CdS nanosphere micropatterns after 3 min deposition on the superhydrophobic-superhydrophilic template of TiO2 nanotube films (Lai et al., 2010d). The bright rectangular areas corresponded to the deposition of CdS nanosphere crystals on the superhydrophilic regions. The boundary between the CdS pattern and the surrounding superhydrophobic regions is clearly visible at a higher magnification (inset). The dispersed CdS nanosphere crystals grew on the top of TiO2 nanotube arrays within the rectangular superhydrophilic region (Fig. 16c). Most of the crystals were less than 90 nm in diameter due to the confinement by the inner diameter of nanotube, though a few larger spheres (~120 nm) were seen across neighboring tube openings. While on the superhydrophobic areas (Fig. 16d), there was almost no CdS crystal. The high growth selectivity was also confirmed by the EDS analysis, revealing that CdS spheres easily nucleate and grow on the hydroxyl groups (–OH) terminated regions (Fig. 16e), but not on the –CF3 terminated areas (Fig. 16f). Since the difference of the water contact angle between the superhydrophilic and superhydrophobic regions is larger than 150°, electrolyte solution is preferentially presented on the uniform superhydrophilic dots. No water droplets go to the neighboring superhydrophobic regions. Although a few CdS particles resulted from homogeneous precipitation attached onto superhydrophobic surface due to van der Waals interactions and gravity, they can be easily removed by ultrasonication. Therefore, a clear and well-defined CdS pattern in line with the dimensions of the superhydrophobic–superhydrophilic template has been obtained.

Fig. 15. Typical element distribution maps of Ag (a),Ti (b), O (c), and optical microscopy(d) and the corresponding R6G SERS mapping (e) of the patterned Ag NP films with an area of 140 × 100 μm2 using the peak area at 614 cm−1 as the reference. (f) Raman spectra of the patterned Ag NP films from different charge density: curve A, bare subsrate; curve B, 0.5 C/cm2; C, 1.0 C/cm2; D, 2.0 C/cm2; and E, 2.5 C/cm2.

Extremely Wetting Pattern by Photocatalytic Lithography and Its Application 609

Extremely wetting micropattern (superhydrophilic/superhydrophobic) on TiO2 nanostructure surface by using SAM technique and photocatalytic lithography has been studied intensely as it provides a cost effective template to construct well defined functional composited pattern. Numerous potential applications have also been proposed and investigated in biomedical, sensors and micro-nano devices. We believe that the photocatatlytic lithography patterning technique presented in this chapter should be general to create micro-scale wetting pattern on other semiconductor substrates and these developments will open the door for more widespread application of the wetting pattern in

The authors thank the National Natural Science Foundation of China (grants 51072170, 21021002), the National Basic Research Program of China (grant 2007CB935603) and the National High Technology Research and Development Program of China (grant 2009AA03Z327), and the Environment and Water Industry Programme Office (EWI) under the National Research Foundation of Singapore (grant MEWR651/06/160) for their financial

Balaur, E., Macak, J.M., Taveira, L. & Schmuki, P. (2005). Tailoring the wettability of TiO2 nanotube layers. *Electrochem. Commun.,* Vol. 7, No. 10, 1066-1070, 1388-2481 Bearinger, J.P., Stone, G, Hiddessen, A.L., Dugan, L.C., Wu, L.G., Hailey, P., Conway, J.W.,

Bhawalkar, S.P., Qian, J., Heiber, M.C. & Jia, L. (2010). Development of a colloidal

Crawford, G.A. & Chawla, N. (2009). Porous hierarchical TiO2 nanostructures: Processing and microstructure relationships. *Acta Mater.,* Vol. 57, No. 3, 854-867, 1359-6454 Csucs, G., Kunzler, T., Feldman, K., Robin, F. & Spencer, N.D. (2003). Microcontact printing

Cui, B. & Veres, T. (2007). Fabrication of metal nanoring array by nanoimprint lithography

Falconnet, D., Koenig, A., Assi, T. & Textor M. (2004). A combined photolithographic and

in the biosciences. *Adv. Funct. Mater.,* Vol. 14, No. 8, 749-756, 1616-301X Felidj, N., Aubard, J. & Levi, G. (1999). Discrete dipole approximation for ultraviolet-visible

Kuenzler, T., Feller, L., Cerritelli, S. & Hubbell, J.A. (2009). Phototocatalytic lithography of poly(propylene sulfide) block copolymers: Toward high-throughput nanolithography for biomolecular arraying applications. *Langmuir,* Vol. 25, No. 2,

lithography method for patterning nonplanar surfaces. *Langmuir,* Vol. 26, No. 22,

of macromolecules with submicrometer resolution by means of polyolefin stamps.

(NIL) and reactive ion etching. *Microelectron. Eng.,* Vol. 84, No. 5-8, 1544-1547, 0167-

molecular-assembly approach to produce functional micropatterns for applications

extinction spectra simulation of silver and gold colloids. *J. Chem. Phys.,* Vol. 111,

**5. Summary and outlook** 

practical fields.

supports.

**7. References** 

9317

1238-1244, 0743-7463

16662-16666, 0743-7463

No. 3, 1195-1208, 0021-9606

*Langmuir,* Vol. 19, No. 15, 6104-6109, 0743-7463

**6. Acknowledgments** 

Fig. 16b shows the photocurrent spectra of the couple CdS/TiO2 nanotube array electrode prepared under different electrodeposition times. It is apparent that the pure TiO2 nanotube array samples have a photo-response wavelength lower than 400 nm due to its band-gap of 3.2 eV (curve a). The decoration of CdS nanospheres with a smaller energy band-gap (2.4 eV) can significantly extend the photo-response range from 380 nm to about 500 nm.Moreover, the CdS modified TiO2 nanotube array electrodes can also greatly increase the photocurrent response under UV light, especially for the samples obtained under 2 min electrodeposition (curve b), which thus would be the optimal deposition time. This is attributed to the uniform dispersed CdS nanospheres with suitable size decorated onto the TiO2 nanotubes. This allows for more efficient electron transfer and lower electron-hole recombination rate which leads to enhanced light harvesting at the directly grown CdS/TiO2 heterojunctions. With the increase of time (curve c and d), more CdS particles with bigger size started to randomly distribute on top of TiO2 nanotube arrays. Such composite nanostructures would weaken the light absorption of the uniform CdS/TiO2 heterojunction underlayer, which has resulted in a lower photocurrent in both UV and visible light region.

Fig. 16. (a) Typical SEM images of the CdS micropattern; (b) Photocurrent spectra of micropatterned CdS film on TiO2 nanotube array electrode. (curve a-d): pure TiO2; 2 min; 3 min; and 5 min. (c) Superhydrophilic region; (d) superhydrophobic region. EDX spectrum of the corresponding superhydrophilic (e) and superhydrophobic regions (f).
