Lithographically Defined Porous Carbon Electrodes - Semantic Scholar
communications Pyrolytic carbon electrodes
Lithographically Defined Porous Carbon Electrodes** D. Bruce Burckel,* Cody M. Washburn, Alex K. Raub, Steven R. J. Brueck, David R. Wheeler, Susan M. Brozik, and Ronen Polsky* The special nature of the CC bond can lead to various polymorphic forms of carbon such as graphite, glassy-carbon, fullerenes (such as buckyballs), carbon nanotubes, and diamond. Electrodes made from carbon exhibit many useful properties including wide potential windows, low background capacitance, resistance to fouling, and catalytic activity for many analytes compared to solid metal electrodes.[1] In addition to the intrinsic material properties of carbon, functionalized films can be produced through chemical modification using a wide range of chemistries. Because of this flexibility and utility, fabrication of both macro- and microporous carbon films, with their commensurate increase in surface area, continues to receive significant research interest.[2] Some of the specific applications for porous carbon materials include fuel cells, electrochemical double layer capacitors, high surface area catalytic supports, water purification, and gas separation. Recently, it has been found that pyrolyzed photoresist films (PPFs) have the same unique properties of carbon electrodes with an advantage that they can be lithographically defined. The goal of this work was to create lithographically defined porous pyrolyzed carbon electrodes and characterize the deposition and electrochemical properties of metal nanoparticles on these electrodes. We report a robust fabrication method capable of producing large area (100 s cm2) submicrometer porous carbon films. In our approach, interferometric lithography (IL) is used to pattern thick photoresist films into 3D periodic lattices. These structures are then converted to carbon via pyrolysis under flowing forming gas.[3] During pyrolysis, the non-carbon species in the resist polymer backbone are removed, while the bulk of the carbon remains. The patterned structures undergo significant shrinkage, but remarkably maintain their morphology and adhesion to the substrate. The degree of carbonization is a function of the pyrolysis temperature, which has a profound [] Dr. D. B. Burckel, Dr. R. Polsky, Dr. C. M. Washburn, Dr. D. R. Wheeler, Dr. S. M. Brozik Sandia National Laboratories Albuquerque, NM 87185 (USA) E-mail: [email protected]; [email protected] A. K. Raub, Prof. S. R. J. Brueck Center for High Technology Materials 1313 Goddard S.E. Albuquerque, NM 87106 (USA) [] The authors thank Bonnie McKenzie for SEM imaging. Sandia is a multiprogram laboratory operated by the Sandia Corporation, a Lockheed Martin company, for the US Department of Energy’s Nuclear Security Administration under Contract DE-AC0494AL85000. DOI: 10.1002/smll.200901084
2792
effect on the direct current (DC) conductivity. The resulting porous carbon structures proved to be conductive and suitable for the electrochemical deposition of ultrasmall (1–3 nm) gold nanoparticles (AuNPs) with high catalytic surface area. Second, palladium was electrolessly deposited on the previously deposited AuNPs supported on the surface of the porous carbon electrode as a demonstration of the potential scope of the methodology. The palladium-modified electrodes exhibit a catalytic response for methanol oxidation with potential application towards microfabricated fuel cells. This Communication details the fabrication and electrochemical characterization of these carbon electrodes. IL is a maskless lithography approach where coherent planewaves are combined forming an interference pattern.[4] Because it is maskless, IL can be used to generate volumetric exposures. The geometry of the exposed resist is controlled by the number and relative angles of the interfering planewaves. 3D face-centered cubic (fcc) structures can be created in a single exposure with four interfering beams, however, control over polarization is critical in order to ensure maximum contrast between each of the four beams. Formation of the 3D porous carbon structures begins by coating a silicon wafer with a bottom antireflection coating (BARC), i-CON 7, followed by an adhesion layer composed of thin, crosslinked, chemically amplified phenolic resin i-Line negative resist (NR-7). Next a thick layer (6 mm) of NR-7 is spun onto the wafer and softbaked. For the work presented here three separate two-beam exposures were used, with an in-plane sample rotation of 1208 between exposures.[5] The exposure geometry illustrated in the schematic of Figure 1A results in a volumetric interference pattern that causes crosslinking in the thick negative resist in regions of high intensity. A post exposure bake (PEB) completes the crosslinking process. The resist is then developed in commercial developer (TMAH) to yield the final resist structure. When careful control over the relative spatial phase between the three exposures is exercised, IL is capable of producing defect-free 3D lattices with a sub-wavelength periodicity uniformly over samples in excess of 2 cm per side. Figure 1B contains a low-magnification scanning electron microscopy (SEM) image of the developed photoresist pattern. Figure 1C contains a higher magnification image identifying the relevant geometrical dimensions. The structure consists of nominally spherical nodes connected by spokes in a triangular lattice. A slight variation in the pore diameter is observed, but the resist structure is uniform to this scale of variation over the entire 2.5 cm 2.5 cm sample. The interference pattern exists throughout the volume where the interfering planewaves overlap, making it possible to create high-aspect-ratio
ß 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
small 2009, 5, No. 24, 2792–2796
structures in the thick resist. Alternatively, fewer layers can be achieved by simply starting with a thinner resist stack. The development time for this thick structure was 120 s. With such a long development cycle, the top layer of the structure was exposed to the developer longer than the bottom layer. As a result, the diameter of the spokes on the top layer is noticeably
Figure 1. A) Schematicof the IL exposuregeometry.B) Low-magnification SEM image of 3D photoresist structure after development. C) Highmagnification SEM image of photoresist structure. small 2009, 5, No. 24, 2792–2796
smaller than the diameter of the spokes on the subsequent layers. The fully formed resist structure with the 3D periodic geometry of Figure 1 was then placed in a tube furnace with flowing forming gas and heated to 1050 8C at a ramp rate of 5 8C min1 and held isothermal for 1 h before cooling to room temperature at a similar ramp rate. Heating the nonconducting resist in this reducing atmosphere (5% H2/95% N2) causes pyrolysis of the photoresist, where non-carbon species in the polymer are driven off, leaving predominately carbon.[6] Raman spectroscopy of these pyrolyzed films confirms that the carbon is in an amorphous state with only localized graphitic crystallization, consistent with other accounts in the literature.[7] Previous work indicates that the patterns shrink as much as 80% in the vertical direction and 50% in the horizontal direction.[8] Figure 2 shows SEM images of the pyrolyzed resist structure. The seemingly dense photoresist structure of Figure 1 is now a sparse carbon structure, but remarkably, the triangular symmetry is largely maintained. There is some obvious distortion of the matrix, but the resultant lattice dimension only changes by 5% while the diameter of the nodes shrinks by 40%. The spokes on the top layer of the structure shrink by 80% while the spokes on the second layer shrink by 60%. Throughout the pyrolysis process, the structures maintain adhesion to the substrate, so that the final carbon structures possess a rigid support. Silicon and glassy carbon both possess similar thermal expansion coefficients (TCESi ¼ 3.9 106 8C1[9]; TCEglassy-C ¼ 3.2–6.92 106 8C1[10]), however, we have performed similar pyrolysis on both fused silica and Al2O3 with similar results. The combination of lateral shrinkage and simultaneous adhesion to the substrate results in formation of a ‘‘foot’’ at the interface between the carbon structure and the substrate.[8] Although the work presented here uses a chemically amplified phenolic resin negative resist, we have also pyrolyzed patterned positive tone novolac resin resists as well as the photo-epoxy SU-8 with similar results. Electrochemical deposition was used to decorate the porous carbon structures with AuNPs. Modification of carbon surfaces with nanoparticles can be used to create highly active electrodes with electrocatalytic characteristics and have found applications as supports to immobilize a wide range of various ligands and biomolecules.[11,12] AuNPs were deposited onto the porous PPF structures from a 0.1 mM HAuCl4/0.5 M H2SO4 solution using a pulsed deposition stepped from þ1.055 V to 0.05 V. SEM images (Figure 3A and B) show a uniform deposition of nanoparticles with particle sizes ranging from 1–3 nm. The pyrolysis of photoresist structures has been reported to result in extremely smooth near-atomically flat surfaces with minimum defects.[13] It is believed that the smooth surface, characteristic of PPFs, results in homogeneous nucleation sites responsible for the small and uniform particle sizes. The cross-sectional image shown in Figure 4 indicates that the AuNPs decorate only the top 2.5 layers, with the highest density of nanoparticles depositing on the top layer. The reason for the gradient of particle deposition is most likely due to poor solution penetration to the bottom layers, possibly due to hydrophobicity of the carbon and thus incomplete wetting of the underlying surfaces. Figure 4 also demonstrates the disparity in shrinkage between the lateral and vertical
ß 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.small-journal.com
2793
communications directions. The initial resist structure had a thickness of 6000 nm and a vertical periodicity of 1400 nm (cross section not shown), while the pyrolyzed pattern of Figure 4 has a periodicity of 450 nm and a thickness of 1700 nm.
Figure 3. A) Low-magnification SEM image of electrochemically deposited AuNPs (1–3 nm in diameter) on the surface of the 3D pyrolyzed carbon structure. B) High-magnification SEM image of deposited AuNPs. C) Cyclic voltammetry curves from the Au particle deposition.
Figure 2. A) Low-magnification SEM image of 3D pyrolyzed carbon structure. B.) Medium-magnification SEM image of 3D pyrolyzed carbon structure. C) High-magnification SEM image of pyrolyzed carbon structure.
2794 www.small-journal.com
Cyclic voltammetry can be used to assess the electrochemically active surface area of the deposited AuNPs. Cyclic voltammetry in N2-saturated 0.05 M H2SO4 results in the formation of surface oxides on the forward scan. During the reverse scan reduction of the formed Au surface oxide monolayer occurs. Calculating the amount of charge consumed during the reduction of the Au surface oxide monolayer at 850 mV and using a reported value[14] of 400 mC cm2 indicates the real surface area for the AuNP-coated PPF to be 0.033 cm2 (Figure 3C). Metal-nanoparticle-decorated carbon supports are the material of choice for fuel cell electrodes, with Pt or Pd particles being the most efficient catalysts. In order to demonstrate the possibility of using the porous PPF structures for fuel cell applications, Pd shells were grown on the
ß 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
small 2009, 5, No. 24, 2792–2796
Figure 4. Cross-section SEM image of Au-decorated electrode with obvious particle density gradient from top layer to bottom layer.
deposited AuNPs from a solution of 0.85 mM Pdþ2 using 0.014 mM ascorbic acid as a reducing agent. Figure 5 shows that the electroless deposition results in high density, asymmetrical growth of Pd clusters after 2 min and 5 min (Figure 5A and B, respectively). As a proof of concept demonstration of the catalytic behavior of the porous PPF electrodes, the response of the porous PPF towards methanol oxidation is shown in Figure 5C. As expected, the bare and AuNP-modified porous PPF show no activity towards methanol oxidation (black and green curves, respectively) while a catalytic anodic wave is
observed to increase after 2 and 5 min of Pd deposition (red and blue lines, respectively). The high density and catalytic activity of the Pd modification along with the ability of the porous material to be lithographically defined show promise as a route to fabrication of high-surface-area microfuel cell electrodes. In summary, we have demonstrated a new lithographic method for fabrication of porous carbon electrodes with high surface area and controllable dimensions providing enormous flexibility to tailor electrodes toward specific applications. The electrodes are rugged, electrically conductive, and show excellent electrochemical behavior. The AuNPs electrochemically deposited on the surface of the porous carbon electrodes exhibit ultrasmall dimensions with uniform size distribution, attributed to the smooth carbon surface. The prepared samples possess structural dimensions that span seven orders of magnitude, from the 2-cm chip dimension to the 1-nm lattice periodicity, 250-nm nodes, 25-nm spokes, and 2-nm AuNPs. The ability to create structures over these wide-ranging size scales offers the potential to harness nanoscale behaviors in measurable, macroscale devices. The lithographically formed carbon structures presented here have not been optimized to maximize either the Au deposition or the Pd deposition. Structural parameters such as pore size and 3D morphology can be expected to have an impact on our ability to control the electrochemical behavior of the electrode. One advantage of this approach for fabrication of the electrode is that these parameters can be controlled at the lithography step, and hence can be rationally engineered for optimal performance.
Experimental Section
Figure 5. A) High-magnification SEM image of 2-min electroless Pd growth on Au-decorated electrode. B) High-magnification SEM image of 5-min electroless Pd growth on Au-decorated electrode. C) Current–voltage curve obtained during methanol oxidation. small 2009, 5, No. 24, 2792–2796
Materials: All solutions were prepared with 18 MV water using a Barnstead Nanopure water purifier (Boston, MA). Isopropanol, H2SO4, and HAuCl4 were purchased from Sigma and used as received. All electrochemical measurements were performed on an Autolab PGSTAT 12 EcoChemie (Netherlands) and were measured versus an Ag/AgCl (aqueous solutions) reference and a Pt counter electrode from Bioanalytical Systems (West Lafayette, IN). SEM imaging was performed on a Zeiss Supra 55VP field-emission gun SEM using a 5 kV accelerating voltage. IL: A BARC (iCON-7; Brewer Science) was spun onto silicon wafers at 3000 rpm and baked on a vacuum hotplate at 205 8C for 60 s. Negative-tone NR-7 from Futurrex Inc. was used in all of the experiments. A thin layer of NR-7 100P (100 nm) was deposited and spun at 3000 rpm to create an adhesion layer and soft-baked at 130 8C for 120 s on a vacuum hotplate. A flood exposure and post exposure bake at 85 8C on a vacuum hotplate completed the adhesion layer. A thick layer (6 mm) of NR-7 6000P was deposited and spun at 3000 rpm and soft-baked at 130 8C for 120 s. The frequency tripled 355-nm line of a Q-switched Nd:YAG laser was used to form the inference pattern. The beam was expanded and split into two separate beams and interfered with an angle of 328 between the planewave propagation vectors. The plane of incidence contains both propagation vectors as well as the angle bisector of the propagation vectors. The angle bisector is tilted with respect to the sample surface normal by 458. After each exposure the sample was rotated in the plane by 1208 and the
ß 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.small-journal.com
2795
communications process is repeated a total of three times. After exposure the sample received a post-exposure bake of 85 8C for 2 min on the vacuum hotplate. A 120 s puddle development, using RD-6 (Futurrex, Inc), and spin-drying completed fabrication of the resist structures. Pyrolysis: The samples were baked on a hotplate at 180 8C for 30 min. The samples were then placed in a tube furnace and heated with a ramp rate of 5 8C min1 to 1050 8C and held for 1 h in forming gas (5% H2/95% N2) before cooling at a similar ramp rate to room temperature. Electrochemical Deposition: PPF electrodes were rinsed with isopropanol and water and dried under a stream of N2. Using a N2-saturated 0.1 mM HAuCl4/0.5 M H2SO4 solution the electrodes were treated with five 5 s symmetrical pulses stepping from þ1.055 V to 0.05 V followed by rinsing with water and drying under a stream of N2.
Keywords:
carbon . electrodes . lithography . nanoparticles . porous materials
[1] R. L. McCreery, Chem. Rev. 2008, 108, 2646. [2] a) A. Stein, Z. Y. Wang, M. A. Fierke, Adv. Mater. 2009, 21, 265; b) C. D. Liang, Z. J. Li, S. Dai, Angew. Chem. Int. Ed. 2008, 31, 3696; c) J. Lee, J. Kim, T. Hyeon, Adv. Mater. 2006, 18, 2073.
2796 www.small-journal.com
[3] S. Ranganathan, R. L. McCreery, S. M. Majji, M. Madou, J. Electrochem. Soc. 2000, 147, 277. [4] S. R. J. Brueck, Proc. IEEE 2005, 93, 1704. [5] Y. Liu, S. Liu, X. Zhang, Photonic Crystal Materials and Nanostructures, (Ed.: R. M. De La Rue ), Proc. of SPIE Vol. 5450, SPIE, Bellingham, WA, 2004, p 473. [6] a) A. Singh, J. Jayaram, M. Madou, S. Akbar, J. Electrochem. Soc. 2002, 149, E78; b) C. L. Wang, L. Taherabadi, G. Y. Jia, M. Madou, Y. T. Yeh, B. Dunn, Electrochem. Solid-State Lett. 2004, 7, A435. [7] R. Kostecki, B. Schnyder, D. Alliata, X. Song, K. Kinoshita, R. Kotz, Thin Solid Films 2001, 396, 36. [8] D. B. Burckel, H. Y. Fan, G. Thaler, D. D. Koleske, J. Crystal Growth 2008, 310, 3113. [9] S. K. Ghandhi, in VLSI Fabrication Principles, Wiley, New York 1983, Ch. 1, p. 45. [10] S. T. Iacono, M. W. Perpall, P. G. Wapner, W. P. Hoffman, D. W. Smith, Jr, Carbon 2007, 45, 931. [11] a) D. Hernandez-Santos, M. B. Gonzales-Garcia, A. Costa-Garcia, Electroanalysis 2000, 12, 1461; b) T. M. Lee, I. M. Hsing, Analyst 2005, 130, 364. [12] A. N. Shipway, E. Katz, I. Willner, ChemPhysChem 2000, 1, 18. [13] a) A. M. Nowak, R. L. McCreery, Anal. Chem. 2004, 76, 1089; b) S. Ranganathan, R. McCreery, Anal. Chem. 2001, 73, 893. [14] M. S. El-Deab, T. Okajima, T. Ohsaka, J. Electrochem. Soc. 2003, 150, A851.
ß 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Received: June 23, 2009 Revised: August 10, 2009 Published online: October 12, 2009
small 2009, 5, No. 24, 2792–2796
Lithographically Defined Porous Carbon Electrodes** D. Bruce Burckel,* Cody M. Washburn, Alex K. Raub, Steven R. J. Brueck, David R. Wheeler, Susan M. Brozik, and Ronen Polsky* The special nature of the CC bond can lead to various polymorphic forms of carbon such as graphite, glassy-carbon, fullerenes (such as buckyballs), carbon nanotubes, and diamond. Electrodes made from carbon exhibit many useful properties including wide potential windows, low background capacitance, resistance to fouling, and catalytic activity for many analytes compared to solid metal electrodes.[1] In addition to the intrinsic material properties of carbon, functionalized films can be produced through chemical modification using a wide range of chemistries. Because of this flexibility and utility, fabrication of both macro- and microporous carbon films, with their commensurate increase in surface area, continues to receive significant research interest.[2] Some of the specific applications for porous carbon materials include fuel cells, electrochemical double layer capacitors, high surface area catalytic supports, water purification, and gas separation. Recently, it has been found that pyrolyzed photoresist films (PPFs) have the same unique properties of carbon electrodes with an advantage that they can be lithographically defined. The goal of this work was to create lithographically defined porous pyrolyzed carbon electrodes and characterize the deposition and electrochemical properties of metal nanoparticles on these electrodes. We report a robust fabrication method capable of producing large area (100 s cm2) submicrometer porous carbon films. In our approach, interferometric lithography (IL) is used to pattern thick photoresist films into 3D periodic lattices. These structures are then converted to carbon via pyrolysis under flowing forming gas.[3] During pyrolysis, the non-carbon species in the resist polymer backbone are removed, while the bulk of the carbon remains. The patterned structures undergo significant shrinkage, but remarkably maintain their morphology and adhesion to the substrate. The degree of carbonization is a function of the pyrolysis temperature, which has a profound [] Dr. D. B. Burckel, Dr. R. Polsky, Dr. C. M. Washburn, Dr. D. R. Wheeler, Dr. S. M. Brozik Sandia National Laboratories Albuquerque, NM 87185 (USA) E-mail: [email protected]; [email protected] A. K. Raub, Prof. S. R. J. Brueck Center for High Technology Materials 1313 Goddard S.E. Albuquerque, NM 87106 (USA) [] The authors thank Bonnie McKenzie for SEM imaging. Sandia is a multiprogram laboratory operated by the Sandia Corporation, a Lockheed Martin company, for the US Department of Energy’s Nuclear Security Administration under Contract DE-AC0494AL85000. DOI: 10.1002/smll.200901084
2792
effect on the direct current (DC) conductivity. The resulting porous carbon structures proved to be conductive and suitable for the electrochemical deposition of ultrasmall (1–3 nm) gold nanoparticles (AuNPs) with high catalytic surface area. Second, palladium was electrolessly deposited on the previously deposited AuNPs supported on the surface of the porous carbon electrode as a demonstration of the potential scope of the methodology. The palladium-modified electrodes exhibit a catalytic response for methanol oxidation with potential application towards microfabricated fuel cells. This Communication details the fabrication and electrochemical characterization of these carbon electrodes. IL is a maskless lithography approach where coherent planewaves are combined forming an interference pattern.[4] Because it is maskless, IL can be used to generate volumetric exposures. The geometry of the exposed resist is controlled by the number and relative angles of the interfering planewaves. 3D face-centered cubic (fcc) structures can be created in a single exposure with four interfering beams, however, control over polarization is critical in order to ensure maximum contrast between each of the four beams. Formation of the 3D porous carbon structures begins by coating a silicon wafer with a bottom antireflection coating (BARC), i-CON 7, followed by an adhesion layer composed of thin, crosslinked, chemically amplified phenolic resin i-Line negative resist (NR-7). Next a thick layer (6 mm) of NR-7 is spun onto the wafer and softbaked. For the work presented here three separate two-beam exposures were used, with an in-plane sample rotation of 1208 between exposures.[5] The exposure geometry illustrated in the schematic of Figure 1A results in a volumetric interference pattern that causes crosslinking in the thick negative resist in regions of high intensity. A post exposure bake (PEB) completes the crosslinking process. The resist is then developed in commercial developer (TMAH) to yield the final resist structure. When careful control over the relative spatial phase between the three exposures is exercised, IL is capable of producing defect-free 3D lattices with a sub-wavelength periodicity uniformly over samples in excess of 2 cm per side. Figure 1B contains a low-magnification scanning electron microscopy (SEM) image of the developed photoresist pattern. Figure 1C contains a higher magnification image identifying the relevant geometrical dimensions. The structure consists of nominally spherical nodes connected by spokes in a triangular lattice. A slight variation in the pore diameter is observed, but the resist structure is uniform to this scale of variation over the entire 2.5 cm 2.5 cm sample. The interference pattern exists throughout the volume where the interfering planewaves overlap, making it possible to create high-aspect-ratio
ß 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
small 2009, 5, No. 24, 2792–2796
structures in the thick resist. Alternatively, fewer layers can be achieved by simply starting with a thinner resist stack. The development time for this thick structure was 120 s. With such a long development cycle, the top layer of the structure was exposed to the developer longer than the bottom layer. As a result, the diameter of the spokes on the top layer is noticeably
Figure 1. A) Schematicof the IL exposuregeometry.B) Low-magnification SEM image of 3D photoresist structure after development. C) Highmagnification SEM image of photoresist structure. small 2009, 5, No. 24, 2792–2796
smaller than the diameter of the spokes on the subsequent layers. The fully formed resist structure with the 3D periodic geometry of Figure 1 was then placed in a tube furnace with flowing forming gas and heated to 1050 8C at a ramp rate of 5 8C min1 and held isothermal for 1 h before cooling to room temperature at a similar ramp rate. Heating the nonconducting resist in this reducing atmosphere (5% H2/95% N2) causes pyrolysis of the photoresist, where non-carbon species in the polymer are driven off, leaving predominately carbon.[6] Raman spectroscopy of these pyrolyzed films confirms that the carbon is in an amorphous state with only localized graphitic crystallization, consistent with other accounts in the literature.[7] Previous work indicates that the patterns shrink as much as 80% in the vertical direction and 50% in the horizontal direction.[8] Figure 2 shows SEM images of the pyrolyzed resist structure. The seemingly dense photoresist structure of Figure 1 is now a sparse carbon structure, but remarkably, the triangular symmetry is largely maintained. There is some obvious distortion of the matrix, but the resultant lattice dimension only changes by 5% while the diameter of the nodes shrinks by 40%. The spokes on the top layer of the structure shrink by 80% while the spokes on the second layer shrink by 60%. Throughout the pyrolysis process, the structures maintain adhesion to the substrate, so that the final carbon structures possess a rigid support. Silicon and glassy carbon both possess similar thermal expansion coefficients (TCESi ¼ 3.9 106 8C1[9]; TCEglassy-C ¼ 3.2–6.92 106 8C1[10]), however, we have performed similar pyrolysis on both fused silica and Al2O3 with similar results. The combination of lateral shrinkage and simultaneous adhesion to the substrate results in formation of a ‘‘foot’’ at the interface between the carbon structure and the substrate.[8] Although the work presented here uses a chemically amplified phenolic resin negative resist, we have also pyrolyzed patterned positive tone novolac resin resists as well as the photo-epoxy SU-8 with similar results. Electrochemical deposition was used to decorate the porous carbon structures with AuNPs. Modification of carbon surfaces with nanoparticles can be used to create highly active electrodes with electrocatalytic characteristics and have found applications as supports to immobilize a wide range of various ligands and biomolecules.[11,12] AuNPs were deposited onto the porous PPF structures from a 0.1 mM HAuCl4/0.5 M H2SO4 solution using a pulsed deposition stepped from þ1.055 V to 0.05 V. SEM images (Figure 3A and B) show a uniform deposition of nanoparticles with particle sizes ranging from 1–3 nm. The pyrolysis of photoresist structures has been reported to result in extremely smooth near-atomically flat surfaces with minimum defects.[13] It is believed that the smooth surface, characteristic of PPFs, results in homogeneous nucleation sites responsible for the small and uniform particle sizes. The cross-sectional image shown in Figure 4 indicates that the AuNPs decorate only the top 2.5 layers, with the highest density of nanoparticles depositing on the top layer. The reason for the gradient of particle deposition is most likely due to poor solution penetration to the bottom layers, possibly due to hydrophobicity of the carbon and thus incomplete wetting of the underlying surfaces. Figure 4 also demonstrates the disparity in shrinkage between the lateral and vertical
ß 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.small-journal.com
2793
communications directions. The initial resist structure had a thickness of 6000 nm and a vertical periodicity of 1400 nm (cross section not shown), while the pyrolyzed pattern of Figure 4 has a periodicity of 450 nm and a thickness of 1700 nm.
Figure 3. A) Low-magnification SEM image of electrochemically deposited AuNPs (1–3 nm in diameter) on the surface of the 3D pyrolyzed carbon structure. B) High-magnification SEM image of deposited AuNPs. C) Cyclic voltammetry curves from the Au particle deposition.
Figure 2. A) Low-magnification SEM image of 3D pyrolyzed carbon structure. B.) Medium-magnification SEM image of 3D pyrolyzed carbon structure. C) High-magnification SEM image of pyrolyzed carbon structure.
2794 www.small-journal.com
Cyclic voltammetry can be used to assess the electrochemically active surface area of the deposited AuNPs. Cyclic voltammetry in N2-saturated 0.05 M H2SO4 results in the formation of surface oxides on the forward scan. During the reverse scan reduction of the formed Au surface oxide monolayer occurs. Calculating the amount of charge consumed during the reduction of the Au surface oxide monolayer at 850 mV and using a reported value[14] of 400 mC cm2 indicates the real surface area for the AuNP-coated PPF to be 0.033 cm2 (Figure 3C). Metal-nanoparticle-decorated carbon supports are the material of choice for fuel cell electrodes, with Pt or Pd particles being the most efficient catalysts. In order to demonstrate the possibility of using the porous PPF structures for fuel cell applications, Pd shells were grown on the
ß 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
small 2009, 5, No. 24, 2792–2796
Figure 4. Cross-section SEM image of Au-decorated electrode with obvious particle density gradient from top layer to bottom layer.
deposited AuNPs from a solution of 0.85 mM Pdþ2 using 0.014 mM ascorbic acid as a reducing agent. Figure 5 shows that the electroless deposition results in high density, asymmetrical growth of Pd clusters after 2 min and 5 min (Figure 5A and B, respectively). As a proof of concept demonstration of the catalytic behavior of the porous PPF electrodes, the response of the porous PPF towards methanol oxidation is shown in Figure 5C. As expected, the bare and AuNP-modified porous PPF show no activity towards methanol oxidation (black and green curves, respectively) while a catalytic anodic wave is
observed to increase after 2 and 5 min of Pd deposition (red and blue lines, respectively). The high density and catalytic activity of the Pd modification along with the ability of the porous material to be lithographically defined show promise as a route to fabrication of high-surface-area microfuel cell electrodes. In summary, we have demonstrated a new lithographic method for fabrication of porous carbon electrodes with high surface area and controllable dimensions providing enormous flexibility to tailor electrodes toward specific applications. The electrodes are rugged, electrically conductive, and show excellent electrochemical behavior. The AuNPs electrochemically deposited on the surface of the porous carbon electrodes exhibit ultrasmall dimensions with uniform size distribution, attributed to the smooth carbon surface. The prepared samples possess structural dimensions that span seven orders of magnitude, from the 2-cm chip dimension to the 1-nm lattice periodicity, 250-nm nodes, 25-nm spokes, and 2-nm AuNPs. The ability to create structures over these wide-ranging size scales offers the potential to harness nanoscale behaviors in measurable, macroscale devices. The lithographically formed carbon structures presented here have not been optimized to maximize either the Au deposition or the Pd deposition. Structural parameters such as pore size and 3D morphology can be expected to have an impact on our ability to control the electrochemical behavior of the electrode. One advantage of this approach for fabrication of the electrode is that these parameters can be controlled at the lithography step, and hence can be rationally engineered for optimal performance.
Experimental Section
Figure 5. A) High-magnification SEM image of 2-min electroless Pd growth on Au-decorated electrode. B) High-magnification SEM image of 5-min electroless Pd growth on Au-decorated electrode. C) Current–voltage curve obtained during methanol oxidation. small 2009, 5, No. 24, 2792–2796
Materials: All solutions were prepared with 18 MV water using a Barnstead Nanopure water purifier (Boston, MA). Isopropanol, H2SO4, and HAuCl4 were purchased from Sigma and used as received. All electrochemical measurements were performed on an Autolab PGSTAT 12 EcoChemie (Netherlands) and were measured versus an Ag/AgCl (aqueous solutions) reference and a Pt counter electrode from Bioanalytical Systems (West Lafayette, IN). SEM imaging was performed on a Zeiss Supra 55VP field-emission gun SEM using a 5 kV accelerating voltage. IL: A BARC (iCON-7; Brewer Science) was spun onto silicon wafers at 3000 rpm and baked on a vacuum hotplate at 205 8C for 60 s. Negative-tone NR-7 from Futurrex Inc. was used in all of the experiments. A thin layer of NR-7 100P (100 nm) was deposited and spun at 3000 rpm to create an adhesion layer and soft-baked at 130 8C for 120 s on a vacuum hotplate. A flood exposure and post exposure bake at 85 8C on a vacuum hotplate completed the adhesion layer. A thick layer (6 mm) of NR-7 6000P was deposited and spun at 3000 rpm and soft-baked at 130 8C for 120 s. The frequency tripled 355-nm line of a Q-switched Nd:YAG laser was used to form the inference pattern. The beam was expanded and split into two separate beams and interfered with an angle of 328 between the planewave propagation vectors. The plane of incidence contains both propagation vectors as well as the angle bisector of the propagation vectors. The angle bisector is tilted with respect to the sample surface normal by 458. After each exposure the sample was rotated in the plane by 1208 and the
ß 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.small-journal.com
2795
communications process is repeated a total of three times. After exposure the sample received a post-exposure bake of 85 8C for 2 min on the vacuum hotplate. A 120 s puddle development, using RD-6 (Futurrex, Inc), and spin-drying completed fabrication of the resist structures. Pyrolysis: The samples were baked on a hotplate at 180 8C for 30 min. The samples were then placed in a tube furnace and heated with a ramp rate of 5 8C min1 to 1050 8C and held for 1 h in forming gas (5% H2/95% N2) before cooling at a similar ramp rate to room temperature. Electrochemical Deposition: PPF electrodes were rinsed with isopropanol and water and dried under a stream of N2. Using a N2-saturated 0.1 mM HAuCl4/0.5 M H2SO4 solution the electrodes were treated with five 5 s symmetrical pulses stepping from þ1.055 V to 0.05 V followed by rinsing with water and drying under a stream of N2.
Keywords:
carbon . electrodes . lithography . nanoparticles . porous materials
[1] R. L. McCreery, Chem. Rev. 2008, 108, 2646. [2] a) A. Stein, Z. Y. Wang, M. A. Fierke, Adv. Mater. 2009, 21, 265; b) C. D. Liang, Z. J. Li, S. Dai, Angew. Chem. Int. Ed. 2008, 31, 3696; c) J. Lee, J. Kim, T. Hyeon, Adv. Mater. 2006, 18, 2073.
2796 www.small-journal.com
[3] S. Ranganathan, R. L. McCreery, S. M. Majji, M. Madou, J. Electrochem. Soc. 2000, 147, 277. [4] S. R. J. Brueck, Proc. IEEE 2005, 93, 1704. [5] Y. Liu, S. Liu, X. Zhang, Photonic Crystal Materials and Nanostructures, (Ed.: R. M. De La Rue ), Proc. of SPIE Vol. 5450, SPIE, Bellingham, WA, 2004, p 473. [6] a) A. Singh, J. Jayaram, M. Madou, S. Akbar, J. Electrochem. Soc. 2002, 149, E78; b) C. L. Wang, L. Taherabadi, G. Y. Jia, M. Madou, Y. T. Yeh, B. Dunn, Electrochem. Solid-State Lett. 2004, 7, A435. [7] R. Kostecki, B. Schnyder, D. Alliata, X. Song, K. Kinoshita, R. Kotz, Thin Solid Films 2001, 396, 36. [8] D. B. Burckel, H. Y. Fan, G. Thaler, D. D. Koleske, J. Crystal Growth 2008, 310, 3113. [9] S. K. Ghandhi, in VLSI Fabrication Principles, Wiley, New York 1983, Ch. 1, p. 45. [10] S. T. Iacono, M. W. Perpall, P. G. Wapner, W. P. Hoffman, D. W. Smith, Jr, Carbon 2007, 45, 931. [11] a) D. Hernandez-Santos, M. B. Gonzales-Garcia, A. Costa-Garcia, Electroanalysis 2000, 12, 1461; b) T. M. Lee, I. M. Hsing, Analyst 2005, 130, 364. [12] A. N. Shipway, E. Katz, I. Willner, ChemPhysChem 2000, 1, 18. [13] a) A. M. Nowak, R. L. McCreery, Anal. Chem. 2004, 76, 1089; b) S. Ranganathan, R. McCreery, Anal. Chem. 2001, 73, 893. [14] M. S. El-Deab, T. Okajima, T. Ohsaka, J. Electrochem. Soc. 2003, 150, A851.
ß 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Received: June 23, 2009 Revised: August 10, 2009 Published online: October 12, 2009
small 2009, 5, No. 24, 2792–2796
