Mechanical Engineering Journal

0123456789 Bulletin of the JSME Mechanical Engineering Journal Vol.3, No.2, 2016 Fabrication and evaluation of micro-structured reaction field with vertically aligned carbon nanotubes for micro bio-analysis device Yuma SUZUKI*, Ewelina PABJAŃCZYK-WLAZŁO**, Jungo ONODA*, Tetsuhide SHIMIZU* and Ming YANG* * Graduate School of System Design, Tokyo Metropolitan University 6-6 Asahigaoka, Hino-shi, Tokyo, 191-0065, Japan E-mail: yang@tmu.ac.jp ** Department of Material and Commodity Sciences and Textile Metrology, Lodz University of Technology Stefana Żeromskiego 116, 90-924 Łódź, Poland Received 14 October 2015 Abstract For high sensitivity in micro bio-analysis devices (MBD), the fabrication of the micro-structured reaction field using vertically aligned carbon nanotubes (VACNTs) which is pillar-structured by two methods was performed. The first method is the combination of photolithography and thermal chemical vapor deposition (CVD). The second method is the molding process of polydimethylsiloxane (PDMS) substrate with micro-pillars array and the transfer press of VACNTs synthesized by thermal CVD on PDMS substrate for lower cost in mass production compared with photolithography process. In the first method, circular-pattered metal film on silicon (Si) substrate as the catalyst for VACNTs synthesis was fabricated by photolithography and VACNTs-pillars array was successfully fabricated using the substrate with circular-pattered metal film by thermal CVD. Furthermore, the protein adsorption property of these structures was evaluated as the reaction field of MBD by ultraviolet (UV) spectroscopy. The results show that the protein adsorption property was improved considering the design of micro pattern in VACNTs structures. On the other hand, in the second method, pillar-structured PDMS substrate was molded using a photoresist mold by photolithography and VACNTs was transferred on PDMS substrate by transfer-press equipment. The results indicate that VACNTs can be transferred on the top of micro pillar of PDMS substrate controlling the load of transfer press. Furthermore, it is indicated that micro-pillar VACNTs structures can be fabricated by molding and transfer press with lower cost than the combination of photolithography and thermal CVD. Key words : Micro bio-analysis device, Carbon nanotubes, Photolithography, Molding process, Transfer press 1. Introduction Micro bio-analysis devices (MBD), such as micro Total Analysis Systems ( -TAS) and Lab-on-a-chip (LOC), are demanded to achieve a rapid diagnosis of biomolecules (for instance, viruses and proteins). This is due to the diffusion distance of biomolecules becomes short by miniaturizing the detection part of MBD, as it is called a reaction field, from sizes of the conventional diagnosis device (Delamarche, et al., 2005, Matsunaga, 2009). However, it is difficult to detect biomolecules when the amount of regent solution containing biomolecules is reduced by downsizing a reaction field. The reaction field with micro structures in MBD can offer large specific surface area (surface areas per volume) and the reduction of the diffusion distance which improves the detection sensitivity and reaction efficiency (Han, et al., 2013). For the fabrication techniques of micro structures, there are two approaches which are top-down process and bottom-up process. Top-down process reduces a large piece of materials to produce the form with the desired shape and the size of micrometer scale. This process can control the form shape and size in a large area and has multiple steps Paper No.15-00567 J-STAGE Advance Publication date: 7 April, 2016 1

which induce high cost for mass production. On the other hand, bottom-up process easily builds up the form with nanometer scale from nano-components (for example, atom, molecular and nano-materials) by self-assembly although this process is difficult to control the desired shape and size (Moronuki, 2011). In the reaction field of MBD, the structural dimension of micrometer order is required. This is because the structural surface can contribute to biomolecules adsorption and reaction efficiency due to dynamic flow between structures (Jomeh and Hoofar, 2010). Therefore, the combination of top-down process and bottom-up process is important to fabricate micro structures with the desired shape and size effectively. This combination technique using the synthesis of vertically aligned carbon nanotubes (VACNTs) as bottom-up process has been attracting attention for the fabrication of micro structures due to the fact that CNTs are one of the nanomaterials with chemical stability and high aspect ratio which brings in high specific surface area (Hu, et al., 2009, Chu, et al., 2010). In this work, we performed the fabrication of the reaction field with micro structures which form is a pillar shape for high sensitivity of MBD by the combination of top-down process and the VACNTs synthesis. This work entails two methods as top-down process. The first method is the combination of photolithography and thermal chemical vapor deposition (CVD) as VACNTs synthesis. The second method is the molding process of polydimethylsiloxane (PDMS) substrate with micro pillars and the transfer press of VACNTs synthesized by thermal CVD on PDMS substrate for lower cost in mass production compared with photolithography process. In the first method, circular-patterned metal film on silicon (Si) substrate as the catalyst for VACNTs synthesis was fabricated by photolithography and VACNTs-pillars array was fabricated using the substrate with circular-patterned metal film by thermal CVD. Furthermore, the protein adsorption property of this structures array was evaluated as the reaction field of MBD by ultraviolet (UV) spectroscopy. In the second method, PDMS substrate with micro pillar array was structured using a mold made by photolithography and VACNTs was transferred on PDMS substrate by transfer-press equipment. 2. Experiments 2.1 VACNTs synthesis by thermal CVD VACNTs were synthesized by thermal CVD using iron (Fe) and aluminum (Al) as metal catalysts. Figure 1 shows the schematic image of the CVD system (MICROPHASE Co., Japan). Al film was deposited on silicon (Si) substrate by sputtering using electron cyclotron resonance (ECR) type ion shower equipment (ELI-200ER, ELIONIX Co., Japan) followed by Fe film. Ethanol was evaporated by the heater in this system for a carbon source of VACNTs. The conditions for deposition of metal catalysts and for VACNTs synthesis is presented in Table 1 and Table 2, respectively. This method was included in previous studies which indicated that synthesized VACNTs were thought to be multi-walled nanotubes with many disorders from the results of Raman spectroscopic analysis (Kobayashi and Yang, 2009, Yang, et al., 2013). VACNTs Substrate Heater Fe Al Si Ethanol Pump Power supply N 2 reservoir Fig. 1 Thermal CVD system for VACNTs synthesis. Before VACNTs synthesis, this vacuum chamber was purged by nitrogen gas. Al film was deposited on Si substrate by sputtering using electron cyclotron resonance type ion shower equipment followed by Fe film. Ethanol was evaporated by the heater in this system for a carbon source of VACNTs. 2

Table 1 Conditions for deposition of metal catalyst. Acceleration Voltage [V] 1250 Ion current density [ma/cm 2 ] 0.75 Vacuum [Pa] 7 10-3 Sputtering time of Al [min] 7 Sputtering time of Fe [min] 1 Sputtering atom Argon (Ar) Table 2 Conditions for VACNTs synthesis. Vacuum [MPa] 0.01 First heating time (810 o C) [s] 60 Second heating time (910 o C) [s] 30 2.2 VACNTs patterning by photolithography and evaluation of protein adsorption property Figure 2 shows the schematic image of the fabrication process of VACNTs-pillars array by the combination of photolithography and thermal CVD. In this process, positive photoresist (OFPR-800 LB, TOKYO OHKA KOGYO., LTD., Japan) coated on Si substrate was hole-structured using a hole-patterned photomask by photolithography. The metal catalysts which are Fe and Al for CVD process were deposited on the hole-structured photoresist film as a template. After removing photoresist film from the substrate, VACNTs-pillars array was synthesized by thermal CVD. Photolithography conditions were shown in Table 3. Conditions for deposition of metal catalysts and for VACNTs synthesis are as in Table 1 and Table 2. Furthermore, VACNTs-pillar array with different dimensions was fabricated using photomasks with different dimensions of the hole-pattern. Figure 3 and Table 4 present the schematic image of the hole-patterned photomask and hole-pattern dimensions, respectively. The form of VACNTs-pillars array was observed by scanning electron microscope (SEM, VE-9800, KEYENCE Co., Japan). The protein adsorption of this structures array was evaluated as the reaction field by UV spectroscopy. First, the VACNTs reaction field was washed by phosphate buffered saline (PBS, ph = 7.0, Wako Pure Chemical Industries, LTD., Japan). Then, 20 l bovine serum albumins (BSAs, Wako Pure Chemical Industries, LTD., Japan) solution (e.g., 1.0 g/l) suspended by PBS was dropped on the VACNTs reaction field and left for 30 min. After that, the sample was washed by PBS and set in the UV spectroscope (UV-2450, SHIMADZU CO., Japan). The amount of adsorbed BSAs was evaluated by measuring the light absorbance of the sample at 205 nm wavelength which brings in light adsorption of protein (Scopes, 1974). 1. Coating photoresist 2. Exposing UV 3. Developing 4. Depositing metal catalyst Fig. 2 5. Removing photoresist Power 6. Synthesizing CNTs VACNTs-pillars array Fabrication process of VACNTs-pillars array by the combination of photolithography and thermal CVD. Firstly, positive photoresist coated on Si substrate was hole-structured using a hole-patterned photomask by photolithography. The metal catalysts which are Fe and Al for CVD process were deposited on the hole-structured photoresist film as a template. After removing photoresist film from the substrate, VACNTs-pillars array was synthesized by thermal CVD. Table 3 Photolithography conditions for hole-structured photoresist film. Dose amount [mj/cm 2 ] 92.5 Developing time [s] 15 23

D Fig.3 G Hole-patterned photomask. D is a hole diameter. G is a gap between holes. Holes are hexagonal-arranged in a photomask. Table 4 Hole-pattern dimensions of a photomask. Hole diameter (D) [ m] Gap (G) [ m] (A) 10 10 (B) 10 30 (C) 10 50 2.3 Molding of PDMS substrate and transfer press of VACNTs Figure 4 shows the schematic image of molding and transfer press. To prepare the mold for the fabrication of pillar-structured PDMS substrate, negative photoresist (SU-8 50, MicroChem Corp., USA) coated on Si substrate was hole-structured using a circular-patterned photomask by photolithography. This photomask is the inverted one in Figure 3 and the dimensions of circular-pattern are as in Table 4. PDMS (Slygard 184, Dow Corning Corp., USA) was coated on this mold. After baking the sample at 60 o C for 1 h, pillar-structured PDMS substrate was peeled off from the mold. VACNTs on Si substrate synthesized by thermal CVD and pillar-structured PDMS substrate were set in transfer-press equipment (show Figure 4b), and VACNTs were transfer to PDMS substrate. Conditions for the fabrication of hole-structured mold and for transfer press are presented in Table 5 and Table 6, respectively. Conditions for deposition of metal catalysts and for VACNTs synthesis are as in Table 1 and Table 2. The form of pillar-structured PDMS substrate and VACNTs-pillars array was observed by SEM. (a) 1. photolithography Hole-structured mold 2. Coating PDMS Pillar-structured PDMS substrate (b) Fig. 4 Load cell Sample Z-axis stage Z -axis stage Si wafer PDMS VACNTs Fabrication process of VACNTs-pillars array by molding and transfer press. (a) Fabrication process of hole-structured mold and pillar-structured PDMS substrate. (b) Transfer-press equipment. To prepare the mold for the fabrication of pillar-structured PDMS substrate, negative photoresist coated on Si substrate was hole-structured using a circular-patterned photomask by photolithography. PDMS was coated on this mold. After baking the sample at 60 o C for 1 h, pillar-structured PDMS substrate was peeled off from the mold. VACNTs on Si substrate synthesized by thermal CVD and pillar-structured PDMS substrate were set in transfer-press equipment, and VACNTs were transfer to PDMS substrate. Press VACNTs-pillars array Table 5 Photolithography conditions for hole-structures mold. Dose amount [mj/cm 2 ] 600 Developing time [min] 10 Table 6 Conditions for transfer press of VACNTs Press load [N] 5, 10 Press time [min] 10 24

3. Results and discussions 3.1 VACNTs patterning by photolithography and evaluation of protein adsorption property Figure 5 shows optical images of hole-structured photoresist film and circular-patterned metal catalysts on Si substrate, and a SEM image of VACNTs-pillars array on Si substrate. It is clarified that VACNTs pillars were synthesized retaining the form of circular-patterned metal catalysts. VACNTs-pillars array with different dimensions are presented in Figure 6. The diameters and heights of VACNTs pillars are approximately 10 m. It is indicated that the dimensions of VACNTs-pillars array can be controlled using photomasks with different dimensions. In previous studies on the fabrication of the reaction field of clumped CNTs by self-assembly (Kobayashi and Yang, 2009, Yang, et al., 2013), the fabrication process involved simple and low steps. Moreover, this process can control the structural dimensions of nanometer magnitude. However, the reproducibility of this process is low because self-assembly process is difficult to control the desired shape and size. On the other hand, the process proposed in this work can fabricate and control the micro structures with high aspect ratio which are required in the reaction field with high sensitivity and reaction efficiency. It is furthermore reasonable to expect that this fabrication method is highly reproducible due to the ease of control in the structure dimensions although this process involves complicated and multiple steps. (a) (b) (c) Fig. 5 10 μm 10 μm 10 μm (a) Optical image of hole-structured photoresist film. (b) Optical image of circular-patterned metal catalysts. (c) SEM image of VACNTs-pillars array. The dimension of the used photomask is (A) in Table 4. (a) (b) (c) Fig. 6 10 μm 10 μm 10 μm SEM images of VACNTs-pillars array. The pillar diameter in all images is 10 m. The gaps between pillars are 10 m, 30 m and 50 m in (a) ~ (c), respectively. The light absorbance of BSAs on the VACNTs reaction field is presented in Figure 7. The light absorbance of gap 50 m is not shown due to the low reproducibility of the fabrication for the VACNTs reaction field with this particular gap. The light absorbance of gap 10 m increases 1.9 times of gap 30 m. Furthermore, the specific surface area S S1 of the VACNTs reaction field was calculated as follows: S S1 = S 1,f + 4S 1,p V 1,f 4V 1,p = 2 3(d + g)2 + 4πdh h(2 3(d + g) 2 πd 2 ) (1) where S 1,f and S 1,p are the surface area of the unit-cell substrate and a VACNTs pillar in the VACNTs reaction field, respectively, V 1,f and V 1,p are the space volume of the unit cell and a VACNTs pillar in the VACNTs reaction field, respectively, d is the diameter of a VACNTs pillar, g is the gap between VACNTs pillars, and h is the height of a VACNTs pillar (show Figure 8). Specific surface areas in gap 10 m and gap 30 m are 0.25 m -1 and 0.13 m -1, respectively, and the increase rate of gap 10 m to gap 30 m is 1.9. Hence, the results imply that VACNTs pillars effectively contribute to BSAs adsorption within the scope of the applied gap scale. In the future, the relationship 25

between the increase rates of light absorbance and specific surface area in the VACNTs reaction field will be further examined. Additionally, the variation of the structural dimension, shape and surface properties of micro structures in the reaction field can affect the dynamic flow of biomolecules solution and the interactions between biomolecules and the reaction field surface occurred by van der Waals force and electric double layer effect which are dominant factors in the detection sensitivity and the reaction efficiency (Kanda, et al., 2007). Therefore, these properties can be significantly improved by considering the effect of VACNTs-structures designs on these phenomena. 0.4 Light absorbance of BSAs [a.u.] 0.3 0.2 0.1 Fig. 7 0 Gap 10 μm Gap 30 μm Light absorbance of BSAs on VACNTs reaction fields with different gap. The light absorbance of gap 50 m is not shown due to the low reproducibility of the fabrication for the VACNTs reaction field with this particular gap. The light absorbance of gap 10 m increase 1.9 times of gap 30 m. Top view of VACTNs reaction field Unit cell of VACTNs reaction field h d Fig. 8 Definition of a unit cell in the VACNTs reaction field for the calculation of the specific surface area. d is the diameter of a VACNTs pillar, g is the gap between VACNTs pillars, and h is the height of a VACNTs pillar. 3.2 Molding of PDMS substrate and transfer press of VACNTs Figure 9 shows SEM images of pillar-structured PDMS substrate and VACNTs-pillar array after transfer press with press load 10 N. The heights of PDMS pillars and VACNTs pillars are approximately 20 m. In gap 10 m and 30 m, VACNTs are successfully transferred to the top of PDMS pillars. On the other hand, VACNTs are transferred to whole PDMS substrate in gap 50 m. This is due to the deformation of PDMS substrate in transfer press. It seems that PDMS pillars are more compressed to PDMS matrix because the stress in transfer press is more concentrated into pillars, decreasing the number of pillars per unit area. Hence, transfer press with press load 5 N was performed in gap 50 m. Figure 10 shows SEM images of VACNTs-pillar array in gap 50 m after transfer press with press load 5 N. This result indicates that VACNTs are transferred to the top of PDMS pillars which is similar to the case in gap 10 m and 30 m with press load 10 N. 26

(a) (b) (c) (d) 10 µm 10 µm 10 µm (e) (f) 25 µm 25 µm (g) (h) (i) 25 µm VACNTs Fig. 9 PDMS 5 µm 5 µm 5 µm SEM images of (a) ~ (c) pillar-structured PDMS substrate, (d) ~ (f) VACNTs-pillar array after transfer press with press load 10 N and (g) ~ (i) extended figures of (d) ~ (f). The diameter of PDMS pillars and VACNTs pillars is 10 m. (a), (d), (g) gap 10 m. (b), (e), (h) gap 30 m. (c), (f), (i) gap 50 m. (a) (b) Fig. 10 25 µm 5 µm (a) SEM images of VACNTs-pillar array in gap 50 m after transfer press with press load 5 N. (b) The extended figure of (a). However, the shape of a VACNTs pillar fabricated by molding and transfer press differs from that fabricated by the combination of photolithography and thermal CVD. Figure 11 shows schematic image of different shapes in VACNTs pillars obtained by the combination of photolithography and thermal CVD and by molding and transfer press. The specific surface area S S2 of the VACNTs reaction field fabricated by molding and transfer press was calculated as follows: S S2 = S 2,f + 4S 2,p V 2,f 4V 2,p = 2 3(d 3 + g) 2 + π {(d 1 + d 2 ) (d 1 d 2 ) 2 + 4h 2 1 + (d 3 + d 2 ) (d 3 d 2 ) 2 + 4h 2 2 + (d 2 1 d 2 3 )} 2 3(d 3 + g) 2 (h 1 + h 2 ) π 3 {h 1 (d 2 1 d 2 3 ) + h d 1 d 2 (d 2 3 d 3 2 )} 2 d 3 d 2 (2) where S 2,f and S 2,p are the surface area of the unit-cell substrate and a VACNTs pillar in the VACNTs reaction field, respectively, and V 2,f and V 2,p are the space volume of the unit cell and a VACNTs pillar in the VACNTs reaction field, respectively. d 1, d 2, and d 3 are the top, center, and bottom diameters of a VANCTs pillar. h 1 and h 2 are the heights of VACNTs and PDMS structure. d 1, d 2, d 3, h 1, and h 2 are approximately 9 m, 7.5 m, 10 m, 12 m, and 8 m, 27

respectively. Specific surface areas in gap 10 m and gap 30 m are 0.11 m -1 and 0.07 m -1, respectively. Although specific surface areas in the case of molding and transfer press decrease compared with the case of the combination of photolithography and thermal CVD, the aspect ratio of a VACNTs pillar in this case is twice larger than the case of the case of the combination of photolithography and thermal CVD. Therefore, this method can be utilized for fabricating the reaction field with high specific area controlled the aspect ratio of VACNTs pillars and the gap between pillars. Moreover, VACNTs are fixed on PDMS pillars by penetrating VACNTs to the pillars during transfer press. Hence, the stability of structures fabricated by this method is higher than the case of the combination of photolithography and thermal CVD. It is due to the low adhesion between VACNTs and Si substrate in the VACNTs synthesis. Finally, this advantage is also to enable the transfer of VACNTs with higher aspect ratio to PDMS substrate, which results in higher specific surface area. Furthermore, the cost efficiency of this method is expected to be lower than photolithography in terms of mass production because the mold for pillar-structured PDMS substrate can be reused after the fabrication of the substrate. d VACNTs pillar d 1 Fig. 11 h Silicon (a) PDMS (b) Schematic image of different shapes in VACNTs pillars in (a) the combination of photolithography and thermal CVD and (b) molding and transfer press. Mathematical symbols in (a) is as in Figure 8. In (b), d 1, d 2, and d 3 are the top, center, and bottom diameters of a VACNTs pillar, respectively. h 1 and h 2 are the heights of VACNTs and PDMS structure, respectively. d 2 d 3 h 1 h 2 4. Conclusions The fabrication of VACNTs-pillars array as the reaction field for high sensitivity of MBD by the combination of top-down process and VACNTs synthesis was performed. Two methods as top-down process were applied in the presented research. The first method is the combination of photolithography and thermal CVD as VACNTs synthesis. The second method is the molding process of pillar-structured PDMS substrate and the transfer press of VACNTs synthesized by thermal CVD on the substrate. In the first method, VACNTs-pillars array was fabricated by patterning metal catalysts on Si substrate. Using photomasks with different dimensions of the hole-pattern, the dimensions of VACNTs-pillars array was controlled. On the basis of the protein adsorption evaluation, it is indicated that the detection sensitivity of the reaction field would be much improved by considering the design of the form and the dimension in VACNTs structures. In the second method, pillar-structured PDMS substrate was molded using a photoresist mold by photolithography and VACNTs was transferred on this substrate by transfer-press equipment. The results show that VACNTs can be transferred on the top of micro pillars by controlling the load of transfer press. Furthermore, it is indicated that micro-pillar VACNTs structures can be fabricated by molding and transfer press with lower cost than the combination of photolithography and thermal CVD. References Chu, K., Wu, Q., Jia, C., Liang, X., Nie, J., Tian, W., Gai, G. and Guo, H., Fabrication and effective thermal conductivity of multi-walled carbon nanotubes reinforced Cu matrix composites for heat sink applications, Composites Science and Technology, Vol.70, No.2 (2010), pp.298-304. Delamarche, E., Juncker, D. and Schmid, H., Microfluidics for Processing Surfaces and Miniaturizing Biological Assays, Advanced Materials, Vol.17, No.24 (2005), pp.2911-2933. 28

Han, S. W., Lee, S., Hong, J., Jang, E., Lee, T. and Koh, W., Multiscale substrates based on hydrogel-incorporated silicon nanowires for protein patterning and microarray-based immunoassays, Biosensors and Bioelectronics, Vol.45, No.15 (2013), pp.129 135. Hu, L., Choi, J. W., Yang, Y., Jeong, S., Mantia, F. L., Cui, L. and Cui, Y., Highly conductive paper for energy-storage devices, Proceedings of the National Academy of Sciences of the United States of America, Vol.106, No.51 (2009), pp.21490-21494. Jomeh, S. and Hoorfar. M., Numerical modeling of mass transport in microfluidic biomolecule-capturing devices equipped with reactive surfaces, Chemical Engineering Journal, Vol.165, No.2 (2010), pp.668-677. Kanda, K., Ogata, S., Jingu, K. and Yang, M., Measurement of Particle distribution in Microchannel Flow Using 3D-TIRM Method, Journal of Visualization, Vol.10, No.2 (2007), pp.207-216. Kobayashi, R. and Yang, M., Nanostructured Surface by Self-Assembly of Carbon Nanotubes for Bio-Analysis, Journal of Solid Mechanics and Materials Engineering, Vol.3, No.2 (2009), pp.366-374. Matsunaga, T., Elemental Technology and Application for Biochip (2009), p.228, CMC Publishing Co., LTD. (in Japanese) Moronuki, N., Surface functions brought by surface micro structures (2011), p.151, Morikita Publishing Co., LTD. (in Japanese) Scopes, R. K., Measurement of protein by spectrophotometry at 205 nm, Analytical Biochemistry, Vol. 59, No.1 (1974), pp.277-282. Yang, M., Yabe, T. and Uchiyama K., Fabrication of Micro Device for Rapid and High-Sensitive Bio-Analysis, Journal of Solid Mechanics and Materials Engineering, Vol.7, No.2 (2013), pp.199-205. 29