畢 業(yè) 設(shè) 計(jì)（論 文）外 文 參 考 資 料 及 譯 文
譯文題目：由表面應(yīng)力引起的納米多孔金 懸臂梁的宏觀彎曲
學(xué)生姓名：
專(zhuān) 業(yè)：
所在學(xué)院：
指導(dǎo)教師：
職 稱(chēng)：
Surface-Stress Induced Macroscopic
Bending of Nanoporous Gold
Cantilevers
Dominik Kramer,*,2 Raghavan Nadar Viswanath,2 and Jo1rg Weissmu1ller2,3
Institut fuè r Nanotechnologie, Forschungszentrum Karlsruhe GmbH, 76021 Karlsruhe, Germany, and Fachrichtung Technische Physik, UniVersitaè t des Saarlandes, 66041 Saarbruè cken, Germany
Received January 13, 2004
ABSTRACT
NANO
LETTERS
2004 Vol. 4, No. 5
793-796
We report the preparation of composite foils consisting of two layers, one solid gold and one nanoporous gold. Tip displacements of several millimeters are observed when the foils are immersed in aqueous electrolytes and the electrochemical potential varied. This suggests that nanoporous metals could be used as the active component in actors, and it demonstrates for the first time that changes in the surface stress f of the metal-electrolyte interface can induce a macroscopic strain, orders of magnitude larger than the amplitudes which are reached in conventional cantilever bending experiments used to measure f.
Changes of the shape of liquid mercury electrodes in response to changes of the electrical potential have been observed as early as the 19th century. In 1872, Gabriel Lippmann invented his capillary electrometer in which small voltage differences can be measured by observation of the displacement of a mercury meniscus. The Lippmann equation relates the surface tension of a liquid electrode to the electrode potential, and it is
also a good approximation for solids.1 However, the surface stress f of a solid is not even approximately equal to its surface tension γ , and it exhibits a different (generally, stronger)2
dependence on the potential.1 Furthermore, due to the stiffness of solids, potential-dependent changes in the position or shape of solid surfaces are much smaller than those of liquid
electrodes. Highly sensitive extensometers3 were used to monitor the strain, and in the past decade, surface stress changes have been measured using atomic force microscope type techniques: thin metal films on the cantilevers are used as electrodes, and techniques as, for instance, laser beam deflection allow the tip displacement (in the lower nanometer range, e.g., in ref 4) induced by changes in the surface stress to
be measured.5-10 Because the surface stress in solids could hitherto only be detected in a laboratory environment using sophisticated equipment, it might be considered as an “exotic” phenomenon of little practical relevance. Even in thin film growth, where the
* Corresponding author. E-mail: Dominik.Kramer@int.fzk.de. Ad-dress: Dr. Dominik Kramer, Forschungszentrum Karlsruhe GmbH, Institut fuè r Nanotechnologie, PO Box 3640, D-76021 Karlsruhe, Germany, Tel. +49-(0)7247 82 6379, Fax +49-(0)7247 82 6369.
2 Forschungszentrum Karlsruhe. 3 Universitaèt des Saarlandes.
10.1021/nl049927d CCC: $27.50 ? 2004 American Chemical Society
interface-induced stress may be large, its importance
remains the subject to current research.11
More recently, surface stress induced length changes of 1.5 ím have been observed in nanoporous mm-sized platinum cubes, an indication that the capillary effects can be enhanced
by increasing the surface-to-volume ratio α,12 which takes on exceptionally large values in porous nano-structures. Since the pressure in the bulk required to balance the surface stress scales with α independent of the geometry of the
microstructure,13 large volume changes and a considerable mechanical work density result from changes in the surface stress of the nanoporous metal.12 Therefore, it has been suggested that such materials may be attractive for use as actuators.14 However, integration of the porous metal into a device requires that it can be precisely and reproducibly shaped, and that it can be bonded to the parts of the device that transmit displacement and load. It has not been demonstrated so far how this can be achieved using nano-powder compacts; furthermore, while powder compacts support a considerable hydrostatic pressure, their resistance to shear stress may be poor. Here, we show that nanoporous metals prepared by dealloying a bulk solid solution exhibit similarly large strain amplitudes as nanopowder compacts, and that the porous material can be joined to solid metal foils to form a composite cantilever beam actuator. The charge-induced expansion or contraction of the porous metal gives rise to a biaxial stress component that results in a large bending of the foil. In this way, the effect of the interface-induced stress is amplified so that the deflection becomes visible to the naked eye: the tip moves by 3 mm, an increase
Published on Web 03/31/2004
Figure 1. Scanning electron micrograph of the nanoporous gold structure obtained by etching silver-gold alloy in perchloric acid.
by the factor 106 compared to previous cantilever bending experiments using a planar surface. This demonstrates that changes in the surface stress of nanoporous metals can be exploited to do work in cantilever bending, analogously to
what was recently reported for carbon nanotubes,15 vanadium oxide nanofibers,16 and conducting polymers.17
Dealloying, the selective dissolution of the less noble component from a solid solution, is well-known to result in
nanoporous structures.18 Dealloying is attractive as a tech-nique for preparing nanoporous solids since it can be applied irrespective of the shape of the active part of a device - including, conceivably, lithographically shaped miniaturized
components. Our samples were obtained by the dealloying
of Ag75Au25 master alloy sheets (see Methods). Figure 1 shows a scanning electron microscopy image of the nano-porous gold microstructure. The ligament size is ca. 20 nm.
Cuboids of porous gold of dimension 1.2×1.2× 1 mm3 were investigated in a commercial dilatometer equipped with an in-situ electrochemical cell. Figure 2A is the cyclic voltammogram (current vs potential curve) of a nanoporous gold sample immersed in 50 mM sulfuric acid, recorded in-situ in the dilatometer cell. The potential limits are given by the onset of hydrogen evolution (ca. -0.25 V) and gold oxidation (above 1 V). The voltammogram in Figure 2A is typical of a polycrystalline gold surface: The current is almost constant over the entire potential range, indicating a continuous capacitive double-layer charging and
discharging, in agreement with the known tendency of SO4-anions to interact only weakly with Au.19
Figure 2B shows the change L in sample length versus the time as the potential is cycled between -0.26 and +1.05 V in 50 mM H2SO4. The length changes periodically and reversibly with the potential, with a small irreversible shrinking superimposed to that. When the reversible part of ￠L(zhǎng) is plotted versus the potential (Figure 2C), it is apparent that the length of the sample can be changed reproducibly by controlling the potential, with a small hysteresis of 0.1 V (or 0.02 μ m). The charge was obtained by integration of the current of Figure 2A and by setting the potential of zero charge (pzc) to 0.25 V [compare ref 20]. The graph of strain versus charge (Figure 2E) is highly reversible and linear both
Figure 2. In-situ dilatometry using 15 succesive cycles of the potential of a cuboid nanoporous gold sample in 50 mM sulfuric acid. (A)
Cyclic voltammogram (current I versus the electrochemical potential E). (B) Length change ￠L(zhǎng) versus time t during the 15 cycles of (A).
(C) Reversible part of ￠L(zhǎng) versus E, obtained by subtraction of an constant arbitrary value for each cycle. (D) Total charge Q versus E.
(E) ￠L(zhǎng)/L0 versus Q. (A) and (C)-(E) display results of all 15 cycles superimposed.
794 Nano Lett., Vol. 4, No. 5, 2004
in the negatively and positively charged regimes; it exhibits a change in slope near the pzc. A similar linear correlation has been observed for a Au(111) surface by STM,21,22 but the break near the pzc of Figure 2E was not resolved there.
It is a matter of debate in how far the potential dependence of the surface stress reflects the details of the bonding of adsorbates to the surface (see ref 10 and references therein). We have carried out experiments using perchloric acid as the electrolyte, and found the results to be in qualitative agreement with Figure 2 (see Supporting Information). Since
the ClO4- ion adsorbs even more weakly than SO4-, this finding is compatible with the notion that the potential-induced strain does not intrinsically require the formation of the chemical bonds involved in specific adsorption; this would imply that the change in surface stress reflects the modified
bonding in the space-charge layer within the metal surface.2,12 Two further observations in support of this notion are: (i) whereas we find Au to contract at negative potential, carbon nanotubes show the opposite effect, expansion upon negative
charging,15 which indicates that the change in surface stress is strongly related to the nature of the bonding in the solid; and (ii) in-situ X-ray adsorption near edge spectroscopy (XANES) data show a significant change in d-band occupancy in Pt nanoparticles as the Pt-electrolyte interface is charged, confirming that the superficial electronic structure of the solid
can be changed.23 If the change in surface stress and the surface-induced strain in our samples are indeed a consequence of the modified bonding in the metal, then the
results provide support for a more general concept:24 by controlling the net charge in space-charge layers at metal surfaces, one can modify the electronic density of states and, thereby, the local properties of the matter at the surface. In nanomaterials, which have a large surface-to-volume ratio, this will result in changes of the overall properties, opening a way for tuning all those materials properties that depend on the density of states.
The action of the surface stress can be amplified by use of bilayer foils. Each of the foils consists of a layer of porous Au bonded to a layer of solid Au, see the cross-sections in Figure 3A. When the foil is immersed in an electrolyte, and its potential varied, then the porous layer will tend to expand or contract, whereas the solid layer will tend to maintain its dimensions. This will result in shear stress at the interface between the two layers, and in a bending of the foil, quite analogous to the effect of the differential thermal expansion used in bimetal thermometers. A similar arrangement has also
been used to produce carbon nanotube actuators.15
To make the bilayer foils, a 2 mm thick sheet of silver-gold alloy was cold-welded to a 0.5 mm thick sheet of pure gold by rolling. After reducing the thickness of the stack to 30 ím by further rolling, the resulting foil was annealed for stress relief and strips 35-40 mm long and 2 mm wide cut from it. Dealloying resulted in a composite foil consisting of a 6 ím thick layer of solid Au covered with 24 ím of porous Au. Two
foils were immersed in 1 M HClO4 and wired as the working and the counter electrodes, respectively. Figures 3A and 3B show a schematic drawing and a photograph of the experimental setup. Both foils undergo a
Nano Lett., Vol. 4, No. 5, 2004
Figure 3. Illustration of the operation of the composite foils. (A) schematic cross-section through an electrochemical cell comprising two identical foils that serve (interchangeably) as working electrode and counter electrode. (B-D) photographs of an electrochemical cell with two bimetallic stripes (nanoporous gold on gold), similar to the schematic in (A). The electrolyte is 1 M perchloric acid. The lower scale of the ruler is calibrated in mm. (C, D) Two enlarged views of the cell in (B), showing the tip of one of the foils with two different voltages applied between the two foils, +1 V (C) and -1 V (D). It is seen, that when the voltage is inversed, the tip moves by ca. 3 mm. The arrows serve as reference markers, emphasizing the tip displacement.
reversible bending as the voltage is changed. Figures 3C,D show enlarged views of the tip of one of the foils before and after inverting the applied voltage. When the potential difference between the electrodes is switched from -1 V to +1 V, the tip moves by as much as 3 mm. Thus, compared to cantilever bending experiments using planar surfaces, the displacement resulting from surface stress changes has increased from few nanometers to the millimeter regime,
that is, by about a factor of 106. A video clip showing the actuator operation is displayed as Supporting Information. For the first time, the effects potential-induced changes of the interface stress, which had previously required sophisticated experimental equipment, have become visible to the naked eye.
For actuator applications, the response time is important. Figure 4A shows the time-dependent ￠L(zhǎng) during a series of
795
Figure 4. (A) Length change ￠L(zhǎng) of the sample in 50 mM sulfuric acid versus time, measured in the dilatometer during a series of potential jumps from -0.2 to 1 V and back (dashed line: potential).
(B) Frequency dependence of the amplitude during potential jumps (rectangular wave) in sulfuric acid. Large squares: Amplitude of the charge curve. Small circles: Amplitude of the length change as measured in the dilatometer. The dilatometers
maximum sampling rate of 10 s-1 limits the experimental strain amplitude at high frequency.
potential jumps from -0.2 V to 1 V and back. The half-times of the jumps in current and strain are 220 and 270 ms.
Because of the limited sampling rate (10 s-1) of the dilatometer, the time constant obtained from the charging curves is considered more accurate. The strain amplitude at a frequency of 0.3 Hz is almost identical to that during slower switching (Figure 4B), which is consistent with the response time given above. The bilayer foils react similarly fast, despite the drag of the electrolyte. The intrinsic time scale is given by the time constant of the charging current, which was determined as 25 ms, considerably faster than in the thicker dilatometer samples. This agrees qualitatively with the expectation that the drift of ions into the pores will be accelerated as the path is shortened.
The large mechanical response induced by changes in the surface stress predestines porous gold as an active component in sensors, especially if its surface is modified by adsorption, e.g., of molecules functionalized by thiol groups. These can be chosen to react selectively with specific molecules, for instance, antibodies; the reaction changes the surface stress, e.g., by steric repulsion of the product, and sensors detecting
these changes have been proposed and tested.8,25-27 Their sensitivity may be significantly enhanced by using nano-porous layers instead of planar surfaces.
In addition to its performance as a simple actor producing reversible strain controlled by an applied voltage, the device
shown in Figure 3 can also be regarded as a primitive voltmeter. If the tip displacement was observed with an optical microscope as in Lippmann's device, it would be suited to measure small voltage differences. Thus, Lipp-mann's 19th century voltmeter based on changes of the surface tension of liquid mercury interface has found a modern equivalent based on changes in the surface stress of a solid metal.
Acknowledgment. Stimulating discussions with H. Gle-iter and support by DFG (Center for Functional Nanostruc-tures) are gratefully acknowledged.
Supporting Information Available: Experimental de-tails, two additional Figures (S1, S2), and a video showing the movement of the bilayer foils. This material is available free of charge via the Internet at http://pubs.acs.org.
References
(1) Lipkowski, J.; Schmickler, W.; Kolb, D. M.; Parsons, R. J. Elec-troanal. Chem. 1998, 452, 193-197.
(2) Schmickler, W.; Leiva, E. J. Electroanal. Chem. 1998, 453, 61-67.
(3) Lin, K. F.; Beck, T. R. J. Electrochem. Soc. 1976, 123, 1145-1151.
(4) Haiss, W.; Sass, J. K. J. Electronanl. Chem. 1995, 386, 267-270.
(5) Ibach, H. Surf. Sci. Rep. 1997, 29, 195-263. Ibach, H. Surf. Sci. Rep.
1999, 35, 71-73.
(6) Raiteri, R.; Butt, H.-J. J. Phys. Chem. 1995, 99, 15728-15732.
(7) Miyatani, T.; Fujihira, M. J. Appl. Phys. 1997, 81, 7099-7115.
(8) Ibach, H.; Bach, C. E.; Giesen, M.; Grossmann, A. Surf. Sci. 1997, 375, 107-119.
(9) Raiteri, R.; Butt, H.-J.; Grattarola, M. Electrochim. Acta 2000, 46, 157-163.
(10) Friesen, C.; Dimitrov, N.; Cammarata, R. C.; Sieradzki, K.
Langmuir 2001, 17, 807-815.
(11) Cammarata, R. C.; Trimble, T. M.; Srolovitz, D. J. J. Mater. Res. 2000, 15, 2468-2474.
(12) Weissmuè ller, J.; Viswanath, R. N.; Kramer, D.; Zimmer, P.; Wuè rschum, R.; Gleiter, H. Science 2003, 300, 312-315.
(13) Weissmuè ller, J.; Cahn, J. W. Acta Mater. 1997, 45, 1899-1906.
(14) Baughman, R. H. Science 2003, 300, 268-269.
(15) Baughman, R. H.; Cui, C.; Zakhidov, A. A.; Iqbal, Z.; Barisci, J. N.; Spinks, G. M.; Wallace, G. G.; Mazzoldi, A.; De Rossi, D.; Rinzler, A. G.; Jaschinski, O.; Roth, S.; Kertesz, M. Science 1999, 284, 1340-1344.
(16) Gu, G.; Schmid, M.; Chiu, P.-W.; Minett, A.; Fraysse, J.; Kim, G.-T.; Roth, S.; Kozlov, M.; Mun? oz, E.; Baughman, R. H. Nature Mater. 2003, 2, 316-319.
(17) Baughman, R. H. Synth. Met. 1996, 78, 339-353.
(18) Erlebacher, J.; Aziz, M. J.; Karma, A.; Dimitrov, N.; Sieradzki, K.
Nature 2001, 410, 450-453.
(19) Kolb, D. M. Prog. Surf. Sci. 1996, 51, 109-173.
(20) Kramer, D., Thesis, University of Ulm, Germany, 2000.
(21) Haiss, W.; Nichols, R. J.; Sass, J. K.; Charle, K. P J. Electroanal. Chem. 1998, 452, 199-202.
(22) Nichols, R. J.; Nouar, T.; Lucas, C. A.; Haiss, W.; Hofer, W. A.
Surf. Sci. 2002, 513, 263-271.
(23) Mukerjee, S.; Srinivasan, S.; Soriaga, M. P.; McBreen, J. J. Electrochem. Soc. 1995, 142, 1409-1422.
(24) Gleiter, H.; Weissmuè ller, J.; Wollersheim, O.; Wuè rschum, R.
Acta Mater. 2001, 49, 737-745.
(25) Berger, R.; Delamarche, E.; Lang, H. P.; Gerber, Ch.; Gimzewski, J. K.; Meyer, E.; Guè ntherodt, H.-J. Science 1997, 276, 2021-2024.
(26) Chen, G. Y.; Thundat, T.; Wachter, E. A.; Warmack, R. J. J. Appl. Phys. 1995, 77, 3618-3622.
(27) Fritz, J.; Baller, M. K.; Lang, H. P.; Strunz, T.; Meyer, E.; Guè ntherodt, H.-J.; Delamarche, E.; Gerber, Ch.; Gimzewski, J. K. Langmuir 2000, 16, 9694-9696.
NL049927D
796 Nano Lett., Vol. 4, No. 5, 2004
本文研究了一種雙層復(fù)合材料箔的制備，其中一層為固體金，另一層為納米多孔金。當(dāng)箔浸入電解質(zhì)水溶液中，并改變?nèi)芤旱碾娀瘜W(xué)勢(shì)，可以觀察到數(shù)毫米大小的端部位移。這一現(xiàn)象表明納米多孔金屬可以作為作動(dòng)器中的應(yīng)激組件，還首次揭露了金屬-電解液界面上表面應(yīng)力f的改變將引起宏觀應(yīng)變，該應(yīng)變的大小比用于測(cè)量表面應(yīng)力f的傳統(tǒng)懸臂梁彎曲實(shí)驗(yàn)中所能達(dá)到的振幅要高幾個(gè)數(shù)量級(jí)。
早在19世紀(jì)，人們就已觀察到，在電勢(shì)改變時(shí)液態(tài)汞電極將會(huì)產(chǎn)生形狀改變。1872年，Gabriel Lippmann（加布里埃爾·李普曼，外國(guó)人的名字不必翻譯）發(fā)明了毛細(xì)靜電計(jì)，可通過(guò)觀察一個(gè)水銀彎曲面的位移來(lái)測(cè)得微小的電壓改變。李普曼方程建立了液態(tài)電極的表面張力和電極電勢(shì)的相關(guān)關(guān)系，而且對(duì)固體電極也可以獲得令人滿(mǎn)意的近似結(jié)果1。然而，固體的表面應(yīng)力f并不近似等于其表面張力γ，并表現(xiàn)出對(duì)電勢(shì)的不同依賴(lài)性1（一般而言依賴(lài)性更強(qiáng)）2。進(jìn)一步講，由于固體具有一定的剛度，其表面依賴(lài)于電勢(shì)的位置和形狀改變遠(yuǎn)小于液體電極。高敏延展計(jì)3用于監(jiān)測(cè)應(yīng)變，在過(guò)去的十年間，利用原子力顯微鏡技術(shù)已可測(cè)量表面應(yīng)力的改變：將薄金屬片置于懸臂梁上作為電極，然后采取激光束偏轉(zhuǎn)分析技術(shù)，就可以得到由表面應(yīng)力改變引起的懸臂梁末端位移（可達(dá)到納米級(jí)，如參4）5-10。然而由于目前僅能在實(shí)驗(yàn)室環(huán)境內(nèi)依托精密儀器檢測(cè)到固體的表面應(yīng)力，所以這一發(fā)現(xiàn)還只被視作一種缺少實(shí)用價(jià)值的“獨(dú)特”現(xiàn)象。即使在界面導(dǎo)致應(yīng)力較大的薄膜生長(zhǎng)方面，其重要性也有待進(jìn)一步研究11。
最近，在具有毫米量級(jí)尺寸的納米多孔鉑立方體上觀測(cè)到了由于表面應(yīng)力引起的長(zhǎng)度變化，達(dá)到1.5μm。這說(shuō)明提高表體比α能夠放大毛細(xì)管效應(yīng)12，而多孔納米結(jié)構(gòu)的表體比非常大。由于塊體材料中平衡表面應(yīng)力所需的體積應(yīng)力會(huì)隨α變化，而α不依賴(lài)于微觀結(jié)構(gòu)的幾何特性，所以納米多孔金屬的表面應(yīng)力變化將引起可觀的機(jī)械功密度及大的體積改變12。因此，這類(lèi)金屬被認(rèn)為在作動(dòng)器制造方面前景誘人14。然而，將多孔金屬集成在設(shè)備上需要精確且可重復(fù)化的多空金屬成型工藝，并且能與設(shè)備上傳遞位移和荷載的部分緊密聯(lián)結(jié)。目前為止，使用納米粉末一體化壓實(shí)技術(shù)還不能達(dá)到這些要求，因?yàn)殡m然一體化納米粉末抗壓能力優(yōu)良，但其抗剪能力很差。本文中展示了利用固熔塊體脫合金法制備的納米多孔金屬不僅具有與一