- Email: [email protected]
Chirality assignment to carbon nanotubes integrated in MEMS by tilted-view transmission electron microscopy
Sensors and Actuators B 154 (2011) 155–159
Contents lists available at ScienceDirect
Sensors and Actuators B: Chemical journal homepage: www.elsevier.com/locate/snb
Chirality assignment to carbon nanotubes integrated in MEMS by tilted-view transmission electron microscopy M. Muoth a,∗ , F. Gramm b , K. Asaka c , L. Durrer a , T. Helbling a , C. Roman a , S.-W. Lee a , C. Hierold a a b c
Micro and Nanosystems, Department of Mechanical and Process Engineering, ETH Zurich, Tannenstrasse 3, CH-8092 Zurich, Switzerland Electron Microscopy ETH Zurich, ETH Zurich, Wolfgang-Pauli-Strasse 16, CH-8093 Zurich, Switzerland Department of Quantum Engineering, Nagoya University, Nagoya 464-8603, Japan
a r t i c l e
i n f o
Article history: Available online 6 December 2009 Keywords: Carbon nanotube Electron diffraction TEM Chirality Integration MEMS
a b s t r a c t Front side etching combined with sample tilting – instead of wafer through etching – allows for transmission electron microscopy (TEM) investigations on nanostructures integrated in microelectromechanical systems (MEMS). We present electron diffraction of an individual single-walled carbon nanotube (SWNT) suspended between sharp polycrystalline silicon tips as far as 165 m away from the MEMS chip edge. This novel approach for transmission-beam characterization avoids complex wafer backside processing and facilitates alignment of the SWNTs to the focal plane using tips defined directly by photolithographic means. The demonstration of chirality assignment to the integrated SWNT paves the way for correlating experimentally measured response of the SWNT sensing element upon stimuli with the response predicted by theory. © 2009 Elsevier B.V. All rights reserved.
1. Introduction Single-walled carbon nanotubes (SWNTs) exhibit extraordinary electronic and mechanical properties that render them promising candidates for sensor applications [1]. The structure and properties of SWNTs are uniquely determined by the chiral indices which define, for example, whether they are metallic or semiconducting. For physical or chemical sensors based on SWNTs, chirality is expected to strongly influence sensor response [2–4]. Experimentally, correlating chirality with sensor sensitivity is difficult to achieve within functional devices. TEM electron diffraction, as one of the most reliable chirality assignment methods, requires electron beam transparency, commonly provided via costly through holes [5–9]. In some cases, even substrate thinning has to be applied [5]. In addition, through holes require backside etching which is often incompatible with suspended structures. An alternative method [10,11] that avoids wafer through etching consists of accessing
Abbreviations: TEM, transmission electron microscopy; MEMS, microelectromechanical systems; SWNT, single-walled carbon nanotube; ED, electron diffraction; RBM, radial breathing mode. ∗ Corresponding author at: Micro and Nanosystems, ETH Zurich, CLA H3, Tannenstrasse 3, CH-8092 Zurich, Switzerland. Tel.: +41 44 632 4705; fax: +41 44 632 1462. E-mail address: [email protected] (M. Muoth). 0925-4005/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.snb.2009.12.002
devices close to the edge of the chip. Etching the bulk support of integrated, metal-clamped SWNTs establishes suitable electron transparency, but it exposes the nanostructures to etchants and requires critical point drying. Due to the necessity to place the SWNT under investigation close to the edge of the chip, integration in more complex micromachined structures is challenging as well. Another approach [12–14] which avoids wet processing after SWNT synthesis involves large arrays of pillars. After cleaving the chips, suspended SWNTs grown on the pillar rows in the vicinity of the edge can be accessed in TEM by mounting the substrate vertically and tilting it by a few degrees. However, to allow for characterization of functional SWNT integrated in MEMS devices, control on the edge’s position is needed. Moreover, the restriction in accessing only the very edge of the chip has to be overcome to have enough area for e.g. actuators or inertial masses. In previous work, transparency was obtained by sliding large micromachined structures beyond the edge [15] which imposes limitations in MEMS design as mechanical anchoring of actuators is difficult under the constraint that the structures have to be moved. Here, we show chirality measurements without backside etching on freestanding electrically contacted SWNTs integrated in MEMS structures of high design complexity. TEM imaging of the SWNT-support interfaces located far away from the chip edge is demonstrated. In the future, electrical measurements on suspended and integrated structures will lead to data, which allow direct model verification by chirality assignment.
156
M. Muoth et al. / Sensors and Actuators B 154 (2011) 155–159
2. Material and methods 2.1. Surface micromachined MEMS for SWNT integration Fig. 1a shows a MEMS structure with two opposing polysilicon tips fabricated by the PolyMUMPsTM [MEMSCAP] process. A close-up of the tip region is shown in the inset of the figure. One tip is connected to an electrically heated thermomechanical actuator while the other is fixed. This structure allows for straining nanomaterials connected between the tips. Electromechanical characterization of SWNTs will also be possible on the structure by electrically contacting the SWNTs. Assuming the SWNT growing on top of one of the polysilicon tips and landing on top of its counterpart, the nanotube’ elevation is 3 m above the substrate. Only for locations close to the dicing edge, this elevation is enough to enable transmission-beam investigations in tilted manner. However, MEMS actuators or proof masses occupy relatively large area and strained functional nanostructures are preferably placed in the symmetry axis of moving MEMS structures. Therefore the location of integrated SWNTs is typically too far away from the dicing edge to be accessed by TEM in tilted-view. 2.2. Front side notch for transmission-beam access In our novel approach, a notch towards the edge of the chip is formed by front side etching allowing for TEM inspection on the bulk of a MEMS chip as schematically depicted in Figs. 1b and 1c. The anyway available, and thus inexpensive, front side etching is an overlay of all standard etching masks of the PolyMUMPsTM process made up of dimple, anchor and via etch masks. Initially, these etching steps are meant to fabricate dimple structures lowering the risk of stiction, for anchoring structural polysilicon layers to the substrate and to provide interconnections between the polysil-
icon layers. Neglecting the original purpose of all these etching masks, a notch can be created in parallel to the MEMS fabrication without any additional processing effort. The resulting notch penetrates the substrate consisting of 600 nm silicon nitride on Si for more than 4 m. Chips of sufficiently small width, e.g. 1 mm, can be mounted vertically on standard Cu support platelets for TEM samples. With the notch starting nearby the functional SWNT and reaching into the dicing path, a small tilt is sufficient to get access to the nanostructure of interest located 165 m away from the chip edge.
2.3. Sharpened tips solely by photolithographic means Concerning the fabrication of the sharp polysilicon tips, shown in the inset of Figs. 1a (inset) and e and 2c, we report on tips thinned down in lateral dimensions below 500 nm formed solely by photolithographic means. Having a standard photolithography process with a specified minimum feature size of 2 m, much sharper structures were achieved by violating the design rules. As illustrated in Fig. 2a, two circles with radii of 7 m, separated from each other by 0.25 m only, were subtracted from a bar in the mask design of the polysilicon layer. The mask layout after conversion to polygons is depicted in Fig. 2b, exhibiting a bridge narrowed below the minimum feature size towards its middle. As the lithography is not capable of reproducing this waisted bridge design completely, desired sharp polysilicon tips are formed within the standard processing. This is an important improvement with respect to our previous approach, where sharp tips had to be fabricated by a HF etch of thermally oxidized polysilicon bridges [16]. Avoiding oxidation and SiO2 removal for tip formation, the initial structural layer thickness can be maintained, buckling due to heavy oxidation is eliminated and processing steps are circumvented.
Fig. 1. (a) Field-emission SEM image of a MEMS structure consisting of two opposing polysilicon tips, a thermomechanical actuator and the front side etched notch towards the edge of the chip allowing for TEM inspection at tilting angles close to vertical sample orientation. The notch is fabricated by overlaying dimple, anchors and via etching steps which penetrates the substrate (600 nm silicon nitride on silicon) for more than 4 m. The inset shows the two polysilicon tips facing each other. The nanotube is 3 m above the substrate spanning the polysilicon tips; (b) schematic cross-sectional illustration of electron beam access to nanostructures located on the bulk area of a MEMS chip and (c) perspective representation of the beam path; (d) TEM micrograph of a vertically mounted chip at a tilt of 2◦ where the front side etched notch enables electron beam access; direction of view corresponds to the arrow in (b); (e) top view field-emission SEM micrograph of a suspended SWNT between partially metalized polysilicon tips. The sharp tip pair was achieved within standard photolithography-based polysilicon processing.
M. Muoth et al. / Sensors and Actuators B 154 (2011) 155–159
157
Fig. 2. (a) Mask layout for sharp tip formation by photolithographic means only. Circles of 7 m in radius were subtracted from a bar of the polysilicon mask layer. (b) After transformation to polygons, a symmetric bridge is designed being narrower than the minimum feature size towards the middle of the bridge. (c) Top view SEM image of the resulting sharp pair of opposing polysilicon tips.
2.4. Instrumentation for TEM and Raman spectroscopy TEM micrographs and electron diffraction (ED) patterns were obtained operating a FEI CM12 at an acceleration voltage of 100 kV. For ED, using a condenser aperture of 30 m in size, a 100–300 nm long segment of a straight, suspended SWNT was illuminated. The diffraction patterns were recorded on imaging plates during 300 s at a camera length of 770 mm. Data readout was performed at enhanced sensitivity of the scanner. The sample holder was cleaned thoroughly in Ar/O2 (25%/75%) plasma, such as to minimize carbonaceous deposition during the long exposure times. Confocal Raman spectroscopy was conducted using a WiTec CRM 200 setup. Excitation was provided by a laser of 532 nm wavelength focused through a 100× objective of 0.8 numerical aperture. 3. Results and discussion Fig. 1e shows a SWNT grown on the sharp micromachined polysilicon tips by ferritin-based Fe-catalysed chemical vapour deposition [17,18]. Although the SWNT spanning the tips is located
165 m away from the chip border, the notch enabled TEM imaging as shown in Fig. 3a. For this configuration, the notch allowed for an accessible tilt range of 2.5◦ . Thus, the geometry of metal contacts to nanotubes could be investigated and the absence of SWNT bundles could be verified. Electron diffraction [11,19–21] of the same individual singlewalled carbon nanotube is presented in Fig. 3b. The chiral indices (17,17) were unambiguously assigned to the integrated nanotube. This corresponds to an armchair-type nanotube with a diameter of 2.31 nm. While chirality assignment could also be derived within some limitations from such techniques as Rayleigh scattering or Raman spectroscopy without sample transparency, the detailed imaging capabilities of TEM inevitably require transmissibility as provided here. Taking advantage of the fact that SWNTs tend to bridge to the nearest neighbouring objects during growth [16], the nanotubes are aligned to the tips and hence the SWNTs can be designed to lie parallel to the focal plane, facilitating significantly the TEM observations. As shown in Fig. 3a and more pronounced in Fig. 4a, the sharp tail of the polysilicon tips observed in tilted-view becomes
Fig. 3. (a) Bright field TEM micrograph showing a SWNT integrated on polysilicon supports. The arrows and the metal deposits act as guide to the eyes is seeing the SWNT. The inset shows a close-up of the SWNT. The amorphous carbon contamination around the nanotube and the metal deposit is mainly induced during observation. The dotted rectangle indicates exemplarily a region similar to the one shown in Fig. 4a. (b) Electron diffraction pattern of the same SWNT shown in (a) and Fig. 1e. The chiral indices (17,17) have been assigned, which corresponds to a nanotube diameter of 2.31 nm, armchair-type and hence metallic. Recording conditions at FEI CM12: 100 kV, 30 m condenser aperture, 300 s exposure time on imaging plate, 770 mm camera length.
158
M. Muoth et al. / Sensors and Actuators B 154 (2011) 155–159
Fig. 4. (a) TEM micrograph of the side face of the polysilicon tip (the area of the image was chosen similar to the area surrounded by the dotted rectangle in Fig. 3a) which thins down towards the right becoming sufficiently thin to allow for imaging SWNTs and catalyst nanoparticles in interaction with the substrate. Solely standard photolithography-based processing was applied. (b) TEM image of a freestanding SWNT reaching the polysilicon support at the right hand side.
Fig. 5. (a) Confocal Raman spectrum of an integrated SWNT with radial breathing mode at 163 rel. cm−1 . Integration time was 12 s. (b) Electron diffraction pattern of the same (14,7) SWNT where suboptimal tilting conditions lead to clipping (accessible tilt range 2.5◦ ). Image contrast is inverted for better visibility. (c) At proper tilt, ED pattern of a (24,6) SWNT located at a distance to the edge of the chip of only 25 m instead of 165 m which results in an enlarged accessible tilting range of 11◦ . A 150 nm long straight section of the nanotube was illuminated at 100 kV. Camera length was 770 mm.
sufficiently thin towards its apex such that TEM images can be collected of SWNTs in interaction with the substrate. For the SWNT of Fig. 4b, a diameter of ∼2.6 nm was derived in the freely suspended part whereas the imaged diameter increased to ∼2.7 nm in the polysilicon-supported part. Such side face view is not only an enabling platform to investigate growth catalyst nanoparticles it may also be a beneficial approach to prepare thin or pre-thinned samples for TEM in general. Cumbersome ion milling and demanding lift-out of TEM lamellae cut by focused ion beam might be bypassed. To emphasize the importance of setting the tilt properly, Fig. 5b shows an ED pattern recorded for the same device type and the same fabrication procedure as for the SWNT of Fig. 3b, but at inappropriate tilting conditions and therefore highlights pattern degeneration due to clipping. Still, the chiral indices (14,7) were assigned while (15,8) would be the second closest candidates. Raman spectroscopy confirmed this assignment by a measured radial breathing mode (RBM) frequency of ωRBM = 163 rel. cm−1 , shown in Fig. 5a. Making use of the relation d (nm) = 223.5/(ωRBM − 12.5) [22] a diameter of d = 1.49 nm is deduced. This corresponds reasonably well with the theoretical diameter of 1.45 nm for a (14,7) nanotube. More convincing, the RBM of (14,7) is expected to be in resonance for the employed excitation laser line whereas the (15,8) and others are not anticipated to exhibit a RBM peak at this excitation energy. Probably due to the previous ED pattern recording, the feature of the defect mode (D) is pronounced. As the G/D ratio of our pristine SWNTs is usually above one hundred the decrease of G/D to 15 seems to support the foreseen necessity to perform electrical sensor measurements prior to chirality assignment. Although our SWNTs endured routinely long electron beam exposures at 100 keV which is above the knock-on damage threshold of 87 keV [23] for carbon nanotubes, degradation cannot be excluded. Furthermore, carbonaceous deposition is likely which can originate from residual hydrocarbons polymerized under electron irradiation [24]. Another reason for the exceptional
high D peak may arise from metal deposition for electrical integration. In Fig. 5c, at an enlarged accessible tilt range of 11◦ because of a modified device type, the diffraction pattern for a (24,6) SWNT separated by only 25 m from the dicing edge exhibits undisturbed electron diffraction spots. 4. Conclusion We present for the first time ED patterns for a SWNT integrated in complex MEMS structures as far as 165 m away from the edge of the substrate without the need for a through hole. Instead of through holes, electron beam access is provided in a tilted manner via a notch towards the dicing edge. While forming the notch by existing front side etching masks of the MEMS fabrication process, backside mask alignment and backside patterning becomes superfluous. Such we readily enable TEM investigations on integrated nanostructures. The demonstrated chirality assignment paves the way for correlating device response and theory. Acknowledgements Funding by the Swiss National Science Foundation (200021108059/1, 200020-121831) and by ETH Zurich (13/05-3) as well as technical support by the ETH FIRST Lab, especially from Sandro Bellini and Otte Homan, are kindly acknowledged. Prof. Ph. Lambin, Les Facultés Universitaires Notre-Dame de la Paix, Belgium, is acknowledged for providing the FORTRAN code for diffraction simulation. References [1] C. Hierold, A. Jungen, C. Stampfer, T. Helbling, Nano electromechanical sensors based on carbon nanotubes, Sens. Actuators A 136 (2007) 51–61.
M. Muoth et al. / Sensors and Actuators B 154 (2011) 155–159 [2] E.D. Minot, Y. Yaish, V. Sazonova, J.Y. Park, M. Brink, P.L. McEuen, Tuning carbon nanotube band gaps with strain, Phys. Rev. Lett. 90 (2003) 4. [3] J. Kong, N.R. Franklin, C. Zhou, M.G. Chapline, S. Peng, K. Cho, H. Dai, Nanotube molecular wires as chemical sensors, Science 287 (2000) 622–625. [4] V. Sazonova, Y. Yaish, H. Ustunel, D. Roundy, T.A. Arias, P.L. McEuen, A tunable carbon nanotube electromechanical oscillator, Nature 431 (2004) 284–287. [5] H.D. Espinosa, Y. Zhu, N. Moldovan, Design and operation of a MEMS-based material testing system for nanomechanical characterization, J. Microelectromech. Syst. 16 (2007) 1219–1231. [6] T. Ishida, K. Kakushima, M. Mita, H. Toshiyoshi, H. Fujita, In-situ TEM observation of crystal-facet-dependent self-rearranging gold atoms under tensile stress controlled by MEMS nanoprobe positioner, in: Transducers’07 & Eurosensors XXI, Lyon, France, 2007, pp. 2505–2508. [7] Y. Ganesan, L. Yang, L. Jun, L. Hao, In situ mechanical characterization of one dimensional nanoscale building blocks using novel microfabricated devices, in: 2008 8th IEEE Conference on Nanotechnology (NANO), Arlington, USA, 2008, pp. 783–786. [8] M.A. Haque, M.T.A. Saif, In-situ tensile testing of nano-scale specimens in SEM and TEM, Exp. Mech. 42 (2002) 123–128. [9] Q.H. Jin, T. Li, Y.L. Wang, X.X. Li, P. Zhou, F.F. Xu, In-situ TEM tensile test of 90 nm-thick 1 1 0 SCS beam using MEMS chip, in: 2008 IEEE Sensors, Lecce, Italy, 2008, pp. 1116–1118. [10] J.C. Meyer, M. Paillet, S. Roth, Single-molecule torsional pendulum, Science 309 (2005) 1539–1541. [11] J.C. Meyer, Structure and Properties of Carbon Nanotubes, Fakultät für Mathematik und Physik, Eberhard-Karls-Universität, Tübingen, 2006. [12] A.M. Cassell, N.R. Franklin, T.W. Tombler, E.M. Chan, J. Han, H.J. Dai, Directed growth of free-standing single-walled carbon nanotubes, J. Am. Chem. Soc. 121 (1999) 7975–7976. [13] D. Takagi, Y. Homma, Y. Kobayashi, Selective growth of individual single-walled carbon nanotubes suspended between pillar structures, Phys. E: Low-Dim. Syst. Nanostruct. 24 (2004) 1–5. [14] Y. Homma, D. Takagi, S. Suzuki, K. Kanzaki, Y. Kobayashi, Electron-microscopic imaging of single-walled carbon nanotubes grown on silicon and silicon oxide substrates, in: International Symposium on Creation of Novel Nanomaterials, Osaka, Japan, 2004, pp. I3–I7. [15] A. Jungen, L. Durrer, C. Stampfer, C. Hierold, Nanoscale straining of individual carbon nanotubes by micromachined transducers, in: Transducers’07 & Eurosensors XXI, Lyon, France, 2007, pp. 1561–1564. [16] A. Jungen, S. Hofmann, J.C. Meyer, C. Stampfer, S. Roth, J. Robertson, C. Hierold, Synthesis of individual single-walled carbon nanotube bridges controlled by support micromachining, J. Micromech. Microeng. 17 (2007) 603–608. [17] L. Durrer, T. Helbling, C. Zenger, A. Jungen, C. Stampfer, C. Hierold, SWNT growth by CVD on Ferritin-based iron catalyst nanoparticles towards CNT sensors, Sens. Actuators B 132 (2008) 485–490. [18] L. Durrer, J. Greenwald, T. Helbling, M. Muoth, R. Riek, C. Hierold, Narrowing SWNT diameter distribution using size-separated ferritin-based Fe catalysts, Nanotechnology 20 (2009) 355601–355607. [19] S. Iijima, T. Ichihashi, Single-shell carbon nanotubes of 1-nm diameter, Nature 363 (1993) 603–605. [20] J.C. Meyer, M. Paillet, G.S. Duesberg, S. Roth, Electron diffraction analysis of individual single-walled carbon nanotubes, Ultramicroscopy 106 (2006) 176–190. [21] P. Lambin, A.A. Lucas, Quantitative theory of diffraction by carbon nanotubes, Phys. Rev. B 56 (1997) 3571–3574. [22] S.M. Bachilo, M.S. Strano, C. Kittrell, R.H. Hauge, R.E. Smalley, R.B. Weisman, Structure-assigned optical spectra of single-walled carbon nanotubes, Science 298 (2002) 2361–2366. [23] B.W. Smith, D.E. Luzzi, Electron irradiation effects in single wall carbon nanotubes, J. Appl. Phys. 90 (2001) 3509–3515. [24] R.F. Egerton, P. Li, M. Malac, Radiation damage in the TEM and SEM, Micron 35 (2004) 399–409.
Biographies Matthias Muoth is pursuing his research as a PhD student in the field of SWNT growth and integration at the Micro and Nanosystems group of ETH Zurich (Swiss
159
Federal Institute of Technology Zurich). In 2007, he received his Master of Science ETH in Mechanical Engineering with a focus on micro- and nanosystems. As part of his master thesis he attached carbon nanotubes to atomic force microscopy tips. His semester thesis investigated electrochemical formation of metal nanoparticles for SWNT synthesis. At the Nanotechnology group of ETH Zurich, during his bachelor thesis he dealt with glucose-based fuel cells. Fabian Gramm graduated at ETH Zurich in Earth Science. During his PhD he combined transmission electron microscopy and powder diffraction data to solve complex crystal structures in the laboratory of crystallography at ETH Zurich. In the following years he did a postdoc in solid-state physics on semiconducting nanowires. Since 2007 he is a staff member at EMEZ (Electron Microscopy ETH Zurich) and is focused on specimen preparation and TEM analysis on material sciences samples (nanowires and -tubes, nanoparticles, semiconducting materials, etc.). Koji Asaka received his bachelor, master, and doctor of engineering degrees from Osaka University, Japan in 1996, 1998, and 2001, respectively. He then worked at Fujitsu Laboratories Ltd., Japan. In 2003, he joined the Special Research Project on Nanoscience, Graduate School of Pure and Applied Sciences, University of Tsukuba. He is an assistant professor of Department of Quantum Engineering, Nagoya University since 2006. He was visiting the Micro and Nanosystems group of Prof. Hierold at ETH Zurich in 2008. His current research interests include the atomic structure dynamics and the mechanical, electrical, and optical properties of nano-carbon materials. Lukas Durrer studied chemistry at the University of Bern. During his study, he joined the Group of Prof. Riek at the Salk Institute in La Jolla, USA, where he determined 3D structures of Somatostatin analogues. For his diploma theses he investigated the electrochemical deposition mechanisms of selenium on single-crystal silver electrodes in the group of Prof. Siegenthaler. He joined the Micro and Nanosystems group as a PhD student to investigate a process for controlled CNT growth on silicon substrates. He is a founder member of the ETH spin-off company greenTEG which develops, produces and sales thermoelectric generators. Thomas Helbling received his MSc at ETH Zurich in Electrical Engineering in 2005. He carried out his master thesis at the Micro and Nanosystems group where he is currently also carrying out his PhD thesis. He focuses on design, fabrication and electrical characterization of carbon nanotube field effect transistors for the use as sensing elements in chemical and electromechanical sensors. During his thesis he fabricated and characterized the smallest, high sensitive piezoresistive pressure sensors with single-walled carbon nanotube transistors as piezoresistive sensing elements. Cosmin Roman has acquired various competences ranging from programming to condensed matter physics and mathematical physics. His research interests as a post doctoral researcher at the Micro and Nanosystems group at ETH Zurich include transport phenomena, mechanics, or more generally the physics of nanostructures. He did his doctoral studies at TIMA laboratory, National Polytechnic Institute of Grenoble (INPG), working on modelling quantum transport and mechanical properties of carbon nanotube-based devices for bio-sensing applications. He obtained his Electrical Engineering diploma from the Polytechnic University in Bucharest in 2002, with a specialization in Computer Science. Shih-Wei Lee focuses on the research field of tunable carbon nanotube resonators in the Micro and Nanosystems group as a PhD student. He studied at the Department of Power Mechanical Engineering at National Tsing Hua University, Taiwan, where he received his MSc in Mechanical Engineering. His research was focused on micromechanical system engineering at the Micro Device Lab, where he investigated the characteristics of micro-grippers as his master thesis. Christofer Hierold is a professor for Micro and Nanosystems at ETH Zurich since April 2002 and recently he became head of the Department of Mechanical and Process Engineering. His major research at ETH Zurich is focused on the field of nanotransducers, evaluation of new materials for MEMS and advanced microsystems. He was eleven years with Siemens AG, Corporate Research, and Infineon Technologies AG in Munich, Germany, working mainly on CMOS compatible microsystems. He has been serving in program committees of numerous scientific conferences. He is member of the Swiss Academy of Engineering Sciences.
Contents lists available at ScienceDirect
Sensors and Actuators B: Chemical journal homepage: www.elsevier.com/locate/snb
Chirality assignment to carbon nanotubes integrated in MEMS by tilted-view transmission electron microscopy M. Muoth a,∗ , F. Gramm b , K. Asaka c , L. Durrer a , T. Helbling a , C. Roman a , S.-W. Lee a , C. Hierold a a b c
Micro and Nanosystems, Department of Mechanical and Process Engineering, ETH Zurich, Tannenstrasse 3, CH-8092 Zurich, Switzerland Electron Microscopy ETH Zurich, ETH Zurich, Wolfgang-Pauli-Strasse 16, CH-8093 Zurich, Switzerland Department of Quantum Engineering, Nagoya University, Nagoya 464-8603, Japan
a r t i c l e
i n f o
Article history: Available online 6 December 2009 Keywords: Carbon nanotube Electron diffraction TEM Chirality Integration MEMS
a b s t r a c t Front side etching combined with sample tilting – instead of wafer through etching – allows for transmission electron microscopy (TEM) investigations on nanostructures integrated in microelectromechanical systems (MEMS). We present electron diffraction of an individual single-walled carbon nanotube (SWNT) suspended between sharp polycrystalline silicon tips as far as 165 m away from the MEMS chip edge. This novel approach for transmission-beam characterization avoids complex wafer backside processing and facilitates alignment of the SWNTs to the focal plane using tips defined directly by photolithographic means. The demonstration of chirality assignment to the integrated SWNT paves the way for correlating experimentally measured response of the SWNT sensing element upon stimuli with the response predicted by theory. © 2009 Elsevier B.V. All rights reserved.
1. Introduction Single-walled carbon nanotubes (SWNTs) exhibit extraordinary electronic and mechanical properties that render them promising candidates for sensor applications [1]. The structure and properties of SWNTs are uniquely determined by the chiral indices which define, for example, whether they are metallic or semiconducting. For physical or chemical sensors based on SWNTs, chirality is expected to strongly influence sensor response [2–4]. Experimentally, correlating chirality with sensor sensitivity is difficult to achieve within functional devices. TEM electron diffraction, as one of the most reliable chirality assignment methods, requires electron beam transparency, commonly provided via costly through holes [5–9]. In some cases, even substrate thinning has to be applied [5]. In addition, through holes require backside etching which is often incompatible with suspended structures. An alternative method [10,11] that avoids wafer through etching consists of accessing
Abbreviations: TEM, transmission electron microscopy; MEMS, microelectromechanical systems; SWNT, single-walled carbon nanotube; ED, electron diffraction; RBM, radial breathing mode. ∗ Corresponding author at: Micro and Nanosystems, ETH Zurich, CLA H3, Tannenstrasse 3, CH-8092 Zurich, Switzerland. Tel.: +41 44 632 4705; fax: +41 44 632 1462. E-mail address: [email protected] (M. Muoth). 0925-4005/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.snb.2009.12.002
devices close to the edge of the chip. Etching the bulk support of integrated, metal-clamped SWNTs establishes suitable electron transparency, but it exposes the nanostructures to etchants and requires critical point drying. Due to the necessity to place the SWNT under investigation close to the edge of the chip, integration in more complex micromachined structures is challenging as well. Another approach [12–14] which avoids wet processing after SWNT synthesis involves large arrays of pillars. After cleaving the chips, suspended SWNTs grown on the pillar rows in the vicinity of the edge can be accessed in TEM by mounting the substrate vertically and tilting it by a few degrees. However, to allow for characterization of functional SWNT integrated in MEMS devices, control on the edge’s position is needed. Moreover, the restriction in accessing only the very edge of the chip has to be overcome to have enough area for e.g. actuators or inertial masses. In previous work, transparency was obtained by sliding large micromachined structures beyond the edge [15] which imposes limitations in MEMS design as mechanical anchoring of actuators is difficult under the constraint that the structures have to be moved. Here, we show chirality measurements without backside etching on freestanding electrically contacted SWNTs integrated in MEMS structures of high design complexity. TEM imaging of the SWNT-support interfaces located far away from the chip edge is demonstrated. In the future, electrical measurements on suspended and integrated structures will lead to data, which allow direct model verification by chirality assignment.
156
M. Muoth et al. / Sensors and Actuators B 154 (2011) 155–159
2. Material and methods 2.1. Surface micromachined MEMS for SWNT integration Fig. 1a shows a MEMS structure with two opposing polysilicon tips fabricated by the PolyMUMPsTM [MEMSCAP] process. A close-up of the tip region is shown in the inset of the figure. One tip is connected to an electrically heated thermomechanical actuator while the other is fixed. This structure allows for straining nanomaterials connected between the tips. Electromechanical characterization of SWNTs will also be possible on the structure by electrically contacting the SWNTs. Assuming the SWNT growing on top of one of the polysilicon tips and landing on top of its counterpart, the nanotube’ elevation is 3 m above the substrate. Only for locations close to the dicing edge, this elevation is enough to enable transmission-beam investigations in tilted manner. However, MEMS actuators or proof masses occupy relatively large area and strained functional nanostructures are preferably placed in the symmetry axis of moving MEMS structures. Therefore the location of integrated SWNTs is typically too far away from the dicing edge to be accessed by TEM in tilted-view. 2.2. Front side notch for transmission-beam access In our novel approach, a notch towards the edge of the chip is formed by front side etching allowing for TEM inspection on the bulk of a MEMS chip as schematically depicted in Figs. 1b and 1c. The anyway available, and thus inexpensive, front side etching is an overlay of all standard etching masks of the PolyMUMPsTM process made up of dimple, anchor and via etch masks. Initially, these etching steps are meant to fabricate dimple structures lowering the risk of stiction, for anchoring structural polysilicon layers to the substrate and to provide interconnections between the polysil-
icon layers. Neglecting the original purpose of all these etching masks, a notch can be created in parallel to the MEMS fabrication without any additional processing effort. The resulting notch penetrates the substrate consisting of 600 nm silicon nitride on Si for more than 4 m. Chips of sufficiently small width, e.g. 1 mm, can be mounted vertically on standard Cu support platelets for TEM samples. With the notch starting nearby the functional SWNT and reaching into the dicing path, a small tilt is sufficient to get access to the nanostructure of interest located 165 m away from the chip edge.
2.3. Sharpened tips solely by photolithographic means Concerning the fabrication of the sharp polysilicon tips, shown in the inset of Figs. 1a (inset) and e and 2c, we report on tips thinned down in lateral dimensions below 500 nm formed solely by photolithographic means. Having a standard photolithography process with a specified minimum feature size of 2 m, much sharper structures were achieved by violating the design rules. As illustrated in Fig. 2a, two circles with radii of 7 m, separated from each other by 0.25 m only, were subtracted from a bar in the mask design of the polysilicon layer. The mask layout after conversion to polygons is depicted in Fig. 2b, exhibiting a bridge narrowed below the minimum feature size towards its middle. As the lithography is not capable of reproducing this waisted bridge design completely, desired sharp polysilicon tips are formed within the standard processing. This is an important improvement with respect to our previous approach, where sharp tips had to be fabricated by a HF etch of thermally oxidized polysilicon bridges [16]. Avoiding oxidation and SiO2 removal for tip formation, the initial structural layer thickness can be maintained, buckling due to heavy oxidation is eliminated and processing steps are circumvented.
Fig. 1. (a) Field-emission SEM image of a MEMS structure consisting of two opposing polysilicon tips, a thermomechanical actuator and the front side etched notch towards the edge of the chip allowing for TEM inspection at tilting angles close to vertical sample orientation. The notch is fabricated by overlaying dimple, anchors and via etching steps which penetrates the substrate (600 nm silicon nitride on silicon) for more than 4 m. The inset shows the two polysilicon tips facing each other. The nanotube is 3 m above the substrate spanning the polysilicon tips; (b) schematic cross-sectional illustration of electron beam access to nanostructures located on the bulk area of a MEMS chip and (c) perspective representation of the beam path; (d) TEM micrograph of a vertically mounted chip at a tilt of 2◦ where the front side etched notch enables electron beam access; direction of view corresponds to the arrow in (b); (e) top view field-emission SEM micrograph of a suspended SWNT between partially metalized polysilicon tips. The sharp tip pair was achieved within standard photolithography-based polysilicon processing.
M. Muoth et al. / Sensors and Actuators B 154 (2011) 155–159
157
Fig. 2. (a) Mask layout for sharp tip formation by photolithographic means only. Circles of 7 m in radius were subtracted from a bar of the polysilicon mask layer. (b) After transformation to polygons, a symmetric bridge is designed being narrower than the minimum feature size towards the middle of the bridge. (c) Top view SEM image of the resulting sharp pair of opposing polysilicon tips.
2.4. Instrumentation for TEM and Raman spectroscopy TEM micrographs and electron diffraction (ED) patterns were obtained operating a FEI CM12 at an acceleration voltage of 100 kV. For ED, using a condenser aperture of 30 m in size, a 100–300 nm long segment of a straight, suspended SWNT was illuminated. The diffraction patterns were recorded on imaging plates during 300 s at a camera length of 770 mm. Data readout was performed at enhanced sensitivity of the scanner. The sample holder was cleaned thoroughly in Ar/O2 (25%/75%) plasma, such as to minimize carbonaceous deposition during the long exposure times. Confocal Raman spectroscopy was conducted using a WiTec CRM 200 setup. Excitation was provided by a laser of 532 nm wavelength focused through a 100× objective of 0.8 numerical aperture. 3. Results and discussion Fig. 1e shows a SWNT grown on the sharp micromachined polysilicon tips by ferritin-based Fe-catalysed chemical vapour deposition [17,18]. Although the SWNT spanning the tips is located
165 m away from the chip border, the notch enabled TEM imaging as shown in Fig. 3a. For this configuration, the notch allowed for an accessible tilt range of 2.5◦ . Thus, the geometry of metal contacts to nanotubes could be investigated and the absence of SWNT bundles could be verified. Electron diffraction [11,19–21] of the same individual singlewalled carbon nanotube is presented in Fig. 3b. The chiral indices (17,17) were unambiguously assigned to the integrated nanotube. This corresponds to an armchair-type nanotube with a diameter of 2.31 nm. While chirality assignment could also be derived within some limitations from such techniques as Rayleigh scattering or Raman spectroscopy without sample transparency, the detailed imaging capabilities of TEM inevitably require transmissibility as provided here. Taking advantage of the fact that SWNTs tend to bridge to the nearest neighbouring objects during growth [16], the nanotubes are aligned to the tips and hence the SWNTs can be designed to lie parallel to the focal plane, facilitating significantly the TEM observations. As shown in Fig. 3a and more pronounced in Fig. 4a, the sharp tail of the polysilicon tips observed in tilted-view becomes
Fig. 3. (a) Bright field TEM micrograph showing a SWNT integrated on polysilicon supports. The arrows and the metal deposits act as guide to the eyes is seeing the SWNT. The inset shows a close-up of the SWNT. The amorphous carbon contamination around the nanotube and the metal deposit is mainly induced during observation. The dotted rectangle indicates exemplarily a region similar to the one shown in Fig. 4a. (b) Electron diffraction pattern of the same SWNT shown in (a) and Fig. 1e. The chiral indices (17,17) have been assigned, which corresponds to a nanotube diameter of 2.31 nm, armchair-type and hence metallic. Recording conditions at FEI CM12: 100 kV, 30 m condenser aperture, 300 s exposure time on imaging plate, 770 mm camera length.
158
M. Muoth et al. / Sensors and Actuators B 154 (2011) 155–159
Fig. 4. (a) TEM micrograph of the side face of the polysilicon tip (the area of the image was chosen similar to the area surrounded by the dotted rectangle in Fig. 3a) which thins down towards the right becoming sufficiently thin to allow for imaging SWNTs and catalyst nanoparticles in interaction with the substrate. Solely standard photolithography-based processing was applied. (b) TEM image of a freestanding SWNT reaching the polysilicon support at the right hand side.
Fig. 5. (a) Confocal Raman spectrum of an integrated SWNT with radial breathing mode at 163 rel. cm−1 . Integration time was 12 s. (b) Electron diffraction pattern of the same (14,7) SWNT where suboptimal tilting conditions lead to clipping (accessible tilt range 2.5◦ ). Image contrast is inverted for better visibility. (c) At proper tilt, ED pattern of a (24,6) SWNT located at a distance to the edge of the chip of only 25 m instead of 165 m which results in an enlarged accessible tilting range of 11◦ . A 150 nm long straight section of the nanotube was illuminated at 100 kV. Camera length was 770 mm.
sufficiently thin towards its apex such that TEM images can be collected of SWNTs in interaction with the substrate. For the SWNT of Fig. 4b, a diameter of ∼2.6 nm was derived in the freely suspended part whereas the imaged diameter increased to ∼2.7 nm in the polysilicon-supported part. Such side face view is not only an enabling platform to investigate growth catalyst nanoparticles it may also be a beneficial approach to prepare thin or pre-thinned samples for TEM in general. Cumbersome ion milling and demanding lift-out of TEM lamellae cut by focused ion beam might be bypassed. To emphasize the importance of setting the tilt properly, Fig. 5b shows an ED pattern recorded for the same device type and the same fabrication procedure as for the SWNT of Fig. 3b, but at inappropriate tilting conditions and therefore highlights pattern degeneration due to clipping. Still, the chiral indices (14,7) were assigned while (15,8) would be the second closest candidates. Raman spectroscopy confirmed this assignment by a measured radial breathing mode (RBM) frequency of ωRBM = 163 rel. cm−1 , shown in Fig. 5a. Making use of the relation d (nm) = 223.5/(ωRBM − 12.5) [22] a diameter of d = 1.49 nm is deduced. This corresponds reasonably well with the theoretical diameter of 1.45 nm for a (14,7) nanotube. More convincing, the RBM of (14,7) is expected to be in resonance for the employed excitation laser line whereas the (15,8) and others are not anticipated to exhibit a RBM peak at this excitation energy. Probably due to the previous ED pattern recording, the feature of the defect mode (D) is pronounced. As the G/D ratio of our pristine SWNTs is usually above one hundred the decrease of G/D to 15 seems to support the foreseen necessity to perform electrical sensor measurements prior to chirality assignment. Although our SWNTs endured routinely long electron beam exposures at 100 keV which is above the knock-on damage threshold of 87 keV [23] for carbon nanotubes, degradation cannot be excluded. Furthermore, carbonaceous deposition is likely which can originate from residual hydrocarbons polymerized under electron irradiation [24]. Another reason for the exceptional
high D peak may arise from metal deposition for electrical integration. In Fig. 5c, at an enlarged accessible tilt range of 11◦ because of a modified device type, the diffraction pattern for a (24,6) SWNT separated by only 25 m from the dicing edge exhibits undisturbed electron diffraction spots. 4. Conclusion We present for the first time ED patterns for a SWNT integrated in complex MEMS structures as far as 165 m away from the edge of the substrate without the need for a through hole. Instead of through holes, electron beam access is provided in a tilted manner via a notch towards the dicing edge. While forming the notch by existing front side etching masks of the MEMS fabrication process, backside mask alignment and backside patterning becomes superfluous. Such we readily enable TEM investigations on integrated nanostructures. The demonstrated chirality assignment paves the way for correlating device response and theory. Acknowledgements Funding by the Swiss National Science Foundation (200021108059/1, 200020-121831) and by ETH Zurich (13/05-3) as well as technical support by the ETH FIRST Lab, especially from Sandro Bellini and Otte Homan, are kindly acknowledged. Prof. Ph. Lambin, Les Facultés Universitaires Notre-Dame de la Paix, Belgium, is acknowledged for providing the FORTRAN code for diffraction simulation. References [1] C. Hierold, A. Jungen, C. Stampfer, T. Helbling, Nano electromechanical sensors based on carbon nanotubes, Sens. Actuators A 136 (2007) 51–61.
M. Muoth et al. / Sensors and Actuators B 154 (2011) 155–159 [2] E.D. Minot, Y. Yaish, V. Sazonova, J.Y. Park, M. Brink, P.L. McEuen, Tuning carbon nanotube band gaps with strain, Phys. Rev. Lett. 90 (2003) 4. [3] J. Kong, N.R. Franklin, C. Zhou, M.G. Chapline, S. Peng, K. Cho, H. Dai, Nanotube molecular wires as chemical sensors, Science 287 (2000) 622–625. [4] V. Sazonova, Y. Yaish, H. Ustunel, D. Roundy, T.A. Arias, P.L. McEuen, A tunable carbon nanotube electromechanical oscillator, Nature 431 (2004) 284–287. [5] H.D. Espinosa, Y. Zhu, N. Moldovan, Design and operation of a MEMS-based material testing system for nanomechanical characterization, J. Microelectromech. Syst. 16 (2007) 1219–1231. [6] T. Ishida, K. Kakushima, M. Mita, H. Toshiyoshi, H. Fujita, In-situ TEM observation of crystal-facet-dependent self-rearranging gold atoms under tensile stress controlled by MEMS nanoprobe positioner, in: Transducers’07 & Eurosensors XXI, Lyon, France, 2007, pp. 2505–2508. [7] Y. Ganesan, L. Yang, L. Jun, L. Hao, In situ mechanical characterization of one dimensional nanoscale building blocks using novel microfabricated devices, in: 2008 8th IEEE Conference on Nanotechnology (NANO), Arlington, USA, 2008, pp. 783–786. [8] M.A. Haque, M.T.A. Saif, In-situ tensile testing of nano-scale specimens in SEM and TEM, Exp. Mech. 42 (2002) 123–128. [9] Q.H. Jin, T. Li, Y.L. Wang, X.X. Li, P. Zhou, F.F. Xu, In-situ TEM tensile test of 90 nm-thick 1 1 0 SCS beam using MEMS chip, in: 2008 IEEE Sensors, Lecce, Italy, 2008, pp. 1116–1118. [10] J.C. Meyer, M. Paillet, S. Roth, Single-molecule torsional pendulum, Science 309 (2005) 1539–1541. [11] J.C. Meyer, Structure and Properties of Carbon Nanotubes, Fakultät für Mathematik und Physik, Eberhard-Karls-Universität, Tübingen, 2006. [12] A.M. Cassell, N.R. Franklin, T.W. Tombler, E.M. Chan, J. Han, H.J. Dai, Directed growth of free-standing single-walled carbon nanotubes, J. Am. Chem. Soc. 121 (1999) 7975–7976. [13] D. Takagi, Y. Homma, Y. Kobayashi, Selective growth of individual single-walled carbon nanotubes suspended between pillar structures, Phys. E: Low-Dim. Syst. Nanostruct. 24 (2004) 1–5. [14] Y. Homma, D. Takagi, S. Suzuki, K. Kanzaki, Y. Kobayashi, Electron-microscopic imaging of single-walled carbon nanotubes grown on silicon and silicon oxide substrates, in: International Symposium on Creation of Novel Nanomaterials, Osaka, Japan, 2004, pp. I3–I7. [15] A. Jungen, L. Durrer, C. Stampfer, C. Hierold, Nanoscale straining of individual carbon nanotubes by micromachined transducers, in: Transducers’07 & Eurosensors XXI, Lyon, France, 2007, pp. 1561–1564. [16] A. Jungen, S. Hofmann, J.C. Meyer, C. Stampfer, S. Roth, J. Robertson, C. Hierold, Synthesis of individual single-walled carbon nanotube bridges controlled by support micromachining, J. Micromech. Microeng. 17 (2007) 603–608. [17] L. Durrer, T. Helbling, C. Zenger, A. Jungen, C. Stampfer, C. Hierold, SWNT growth by CVD on Ferritin-based iron catalyst nanoparticles towards CNT sensors, Sens. Actuators B 132 (2008) 485–490. [18] L. Durrer, J. Greenwald, T. Helbling, M. Muoth, R. Riek, C. Hierold, Narrowing SWNT diameter distribution using size-separated ferritin-based Fe catalysts, Nanotechnology 20 (2009) 355601–355607. [19] S. Iijima, T. Ichihashi, Single-shell carbon nanotubes of 1-nm diameter, Nature 363 (1993) 603–605. [20] J.C. Meyer, M. Paillet, G.S. Duesberg, S. Roth, Electron diffraction analysis of individual single-walled carbon nanotubes, Ultramicroscopy 106 (2006) 176–190. [21] P. Lambin, A.A. Lucas, Quantitative theory of diffraction by carbon nanotubes, Phys. Rev. B 56 (1997) 3571–3574. [22] S.M. Bachilo, M.S. Strano, C. Kittrell, R.H. Hauge, R.E. Smalley, R.B. Weisman, Structure-assigned optical spectra of single-walled carbon nanotubes, Science 298 (2002) 2361–2366. [23] B.W. Smith, D.E. Luzzi, Electron irradiation effects in single wall carbon nanotubes, J. Appl. Phys. 90 (2001) 3509–3515. [24] R.F. Egerton, P. Li, M. Malac, Radiation damage in the TEM and SEM, Micron 35 (2004) 399–409.
Biographies Matthias Muoth is pursuing his research as a PhD student in the field of SWNT growth and integration at the Micro and Nanosystems group of ETH Zurich (Swiss
159
Federal Institute of Technology Zurich). In 2007, he received his Master of Science ETH in Mechanical Engineering with a focus on micro- and nanosystems. As part of his master thesis he attached carbon nanotubes to atomic force microscopy tips. His semester thesis investigated electrochemical formation of metal nanoparticles for SWNT synthesis. At the Nanotechnology group of ETH Zurich, during his bachelor thesis he dealt with glucose-based fuel cells. Fabian Gramm graduated at ETH Zurich in Earth Science. During his PhD he combined transmission electron microscopy and powder diffraction data to solve complex crystal structures in the laboratory of crystallography at ETH Zurich. In the following years he did a postdoc in solid-state physics on semiconducting nanowires. Since 2007 he is a staff member at EMEZ (Electron Microscopy ETH Zurich) and is focused on specimen preparation and TEM analysis on material sciences samples (nanowires and -tubes, nanoparticles, semiconducting materials, etc.). Koji Asaka received his bachelor, master, and doctor of engineering degrees from Osaka University, Japan in 1996, 1998, and 2001, respectively. He then worked at Fujitsu Laboratories Ltd., Japan. In 2003, he joined the Special Research Project on Nanoscience, Graduate School of Pure and Applied Sciences, University of Tsukuba. He is an assistant professor of Department of Quantum Engineering, Nagoya University since 2006. He was visiting the Micro and Nanosystems group of Prof. Hierold at ETH Zurich in 2008. His current research interests include the atomic structure dynamics and the mechanical, electrical, and optical properties of nano-carbon materials. Lukas Durrer studied chemistry at the University of Bern. During his study, he joined the Group of Prof. Riek at the Salk Institute in La Jolla, USA, where he determined 3D structures of Somatostatin analogues. For his diploma theses he investigated the electrochemical deposition mechanisms of selenium on single-crystal silver electrodes in the group of Prof. Siegenthaler. He joined the Micro and Nanosystems group as a PhD student to investigate a process for controlled CNT growth on silicon substrates. He is a founder member of the ETH spin-off company greenTEG which develops, produces and sales thermoelectric generators. Thomas Helbling received his MSc at ETH Zurich in Electrical Engineering in 2005. He carried out his master thesis at the Micro and Nanosystems group where he is currently also carrying out his PhD thesis. He focuses on design, fabrication and electrical characterization of carbon nanotube field effect transistors for the use as sensing elements in chemical and electromechanical sensors. During his thesis he fabricated and characterized the smallest, high sensitive piezoresistive pressure sensors with single-walled carbon nanotube transistors as piezoresistive sensing elements. Cosmin Roman has acquired various competences ranging from programming to condensed matter physics and mathematical physics. His research interests as a post doctoral researcher at the Micro and Nanosystems group at ETH Zurich include transport phenomena, mechanics, or more generally the physics of nanostructures. He did his doctoral studies at TIMA laboratory, National Polytechnic Institute of Grenoble (INPG), working on modelling quantum transport and mechanical properties of carbon nanotube-based devices for bio-sensing applications. He obtained his Electrical Engineering diploma from the Polytechnic University in Bucharest in 2002, with a specialization in Computer Science. Shih-Wei Lee focuses on the research field of tunable carbon nanotube resonators in the Micro and Nanosystems group as a PhD student. He studied at the Department of Power Mechanical Engineering at National Tsing Hua University, Taiwan, where he received his MSc in Mechanical Engineering. His research was focused on micromechanical system engineering at the Micro Device Lab, where he investigated the characteristics of micro-grippers as his master thesis. Christofer Hierold is a professor for Micro and Nanosystems at ETH Zurich since April 2002 and recently he became head of the Department of Mechanical and Process Engineering. His major research at ETH Zurich is focused on the field of nanotransducers, evaluation of new materials for MEMS and advanced microsystems. He was eleven years with Siemens AG, Corporate Research, and Infineon Technologies AG in Munich, Germany, working mainly on CMOS compatible microsystems. He has been serving in program committees of numerous scientific conferences. He is member of the Swiss Academy of Engineering Sciences.