Photoconductive CdSe Nanowire Arrays, Serpentines, and LoopsFormed by Electrodeposition on Self-Organized Carbon NanotubesX. Wendy Gu,†,§ Nitzan Shadmi,† Tohar S. Yarden,†,∥ Hagai Cohen,‡ and Ernesto Joselevich*,†

†Department of Materials and Interfaces, Weizmann Institute of Science, Rehovot 76100, Israel‡Chemical Research Support, Weizmann Institute of Science, Rehovot 76100, Israel

*S Supporting Information

ABSTRACT: Semiconducting nanowires frequently have enhanced proper-ties and unique functionality compared to their bulk counterparts.Controlling the geometry of nanowires is crucial for their integration intonanoscale devices because the shape of a device component can dictate itsfunctionality, such as in the case of a mechanical spring or an antenna. Wedemonstrate a novel synthetic method for making polycrystalline CdSenanowires with controlled geometries by using self-organized single-walledcarbon nanotubes as a template for the selective electrodeposition ofnanowires. Nanowires of up to hundreds of micrometers in length are formedas high-density straight arrays, as well as in the shape of serpentines and loops.These nanowires exhibit significant photoluminescence and photoconductiv-ity applicable to photodetectors and respond to illumination up to 2 orders ofmagnitude faster than single crystalline CdSe.

■ INTRODUCTIONOne-dimensional nanostructures, such as carbon nanotubes andinorganic nanowires, are currently regarded with intenseinterest because of their potential integration into nanoscaledevices.1−3 In particular, semiconductor nanowires are animportant class of materials because they can be incorporatedinto nanoelectronics,4 photodetectors,5,6 piezoelectrics,7 andsolar cells.8,9 A variety of methods have been developed forsynthesizing semiconductor nanowires, including vapor−liquid−solid (VLS) growth,10 solution growth,11 and electro-deposition into porous templates12,13 and on the step edges ofhighly oriented pyrolytic graphite.14 However, it is still achallenge to control the geometry of nanowires as the existingsynthetic approaches most commonly limit the shape ofnanowires to simple rods.11,12,14 Recently, nanowires withzigzag,15 helical,16,17 sawlike,10 and perfectly aligned18,19

geometries have also been achieved by controlling the growthconditions. Controlling the geometry of nanowires opens uppossibilities for their uses as nanoscale mechanical springs,lighting elements, and electronics components or in othergeometry-specific applications.Electrodeposition is a favorable method for creating

nanowires because of its low cost and scalability. In addition,electrodeposition onto a conductive scaffold leaves theelectrodeposited material as a conformal layer so that theshape of the electrodeposited material can be controlled by theunderlying electrode. Carbon nanotubes have been previouslyused as nanoelectrodes for the electrodeposition of noble metalnanoparticles.20−23 Other groups have combined the propertiesof semiconductor quantum dots in a one-dimensional structureby forming covalently bonded quantum-dot-decorated carbon

nanotube heterostructures.24 We have previously developedself-organized single-walled carbon nanotube (SWNT) patternsof complex geometries directed by crystal surfaces incombination with external forces.25−27 Recently, we haveshown that these self-organized carbon nanotube patterns canbe used as templates for the electrodeposition of continuousmetallic nanowires with the geometry of the nanotubes.28

In this paper, we demonstrate that this approach of “drawingwith nanotubes” can be extended to compound semiconductornanowires of various geometries, including perfectly alignedarrays of hundreds of CdSe nanowires, serpentines, and loops.We show that the nanowire CdSe stoichiometry can beprecisely controlled through the electrodeposition process, andwe evaluate the kinetics of the electrodeposition process. Thestructure of the nanowires is characterized with scanningelectron microscopy (SEM) and atomic force microscopy(AFM), and their composition is determined using energy-dispersive spectroscopy (EDS), and X-ray photoelectronspectroscopy (XPS). The photoluminescence and photo-conductivity of these new functional nanostructures ismeasured and found to be comparable to that of single-crystalline CdSe, although with improved photoconductiverecovery and response times.

■ EXPERIMENTAL METHODSElectrodeposition onto Carbon Nanotubes. SWNTs

were grown horizontally on a miscut quartz substrate using

Received: July 9, 2012Revised: August 24, 2012Published: August 30, 2012

Article

pubs.acs.org/JPCC

© 2012 American Chemical Society 20121 dx.doi.org/10.1021/jp306804y | J. Phys. Chem. C 2012, 116, 20121−20126

pubs.acs.org/JPCChttp://pubs.acs.org/action/showImage?doi=10.1021/jp306804y&iName=master.img-000.jpg&w=183&h=123

chemical vapor deposition (CVD) (Figure 1a,b). A grid patternof iron catalyst was deposited on the quartz substrate usingphotolithography and electron-beam evaporation of the metal.The grid pattern allows SWNTs to begin growth from the ironcatalyst and then grow horizontally onto bare quartz. Titanium(75 nm) was deposited over the grid pattern to obtain reliableelectrical contact between the carbon nanotubes and a wirelead. The Ti leads were protected with 25 nm of amorphousSiO2 to prevent both corrosion by the electrolyte solution andelectrodeposition onto the Ti instead of SWNTs. CVDconditions were changed to obtain growth of SWNTs asstraight arrays or as serpentines and loops, as describedpreviously.25,28,29

Electrodeposition was carried out in a 5 × 5 × 2 mm3 cell cutout of a layer of polydimethylsiloxane (PDMS) elastomer(DOW Corning Sylgard) with a bottom made of a glass slide(Figure 1c). A 100 nm layer of gold was deposited on the glassslide and served as the counter electrode. A silver wire wasthreaded across the electrochemical cell and reduced in thepresence of KCl to form a Ag/AgCl pseudoreferenceelectrode.28 The cell was filled with an aqueous solution of120 mM CdCl2 (Baker) and 1 mM SeO2 (Aldrich, 99.999%purity) in deionized water adjusted to pH 2−3 with HCl(Gadot) based on the solution used by Li et al.14 The SWNT-covered quartz slide was placed on top of the cell to serve as theworking electrode. This experimental cell can be placed on anoptical microscope stage such that electrodeposition can beobserved and monitored in real time (Figure 1d). Electro-chemical processes were controlled by a PAR model 263Apotentiostat and the accompanying Power Suite software.Characterization and Optical Measurements. Absorb-

ance spectra were obtained with a Varian Cary 50 series UV−vis spectrophotometer. Photoluminescence spectra wereobtained with a LabRam HR800-PL spectrofluorimeter micro-scope (Horiba Jobin-Yivon) and an excitation source at 457 nm(Ar-ion CW laser, ∼15mW/cm2) that was directed at thesample surface at an angle of 70° relative to the sample surface.SEM images and composition information were obtained with aLeo Supra 55VP microscope at a 2.50 kV accelerating voltagewith built-in EDS (Oxford). XPS measurements wereperformed on a Kratos AXIS-Ultra setup, at a base pressureof ∼10−9 Torr. A monochromatic X-ray source was used, anddetection pass energies ranged from 10 to 80 eV. Conductive

carbon tape was attached to the quartz surface and to the Tilines for controlled charging experiments. Charge compensa-tion was applied controllably by means of an electron flood gunat bias voltages up to 3.6 V and a filament current of 1.8 A.For the photocurrent measurements, electrodes were

fabricated on typical nanowire samples by standard photo-lithography and electron-beam evaporation of 5 nm of Ti (KurtJ. Lesker, 99.995%) and 80 nm of Au (Kurt J. Lesker, 99.999%).A HeNe laser (JDS Unipase 1122/P), with λ = 632.8 nm and aflux density of 0.23 W/cm2, was used to excite the samples,while applying dc voltages between −1.0 and 1.0 V. For the on/off switching of the laser light, an optical chopper (Thorlaboratories) was used at a frequency of 5 Hz. The signal wasamplified using a preamp, and recorded using an oscilloscope(Tektronix TDS 2012B). The sample was contacted usingprobes (Karl Suss PH100 and Micromanipulator 110/210).

■ RESULTS AND DISCUSSIONCdSe nanowires were deposited according to the overallreaction described in eq 1. It has been suggested thatdeposition of CdSe occurs through the reduction of H2SeO3to H2Se (eq 2), and then further reaction of H2Se with Cd

2+

takes place (eq 3).30 The formation of CdSe from H2Se (eq 3)competes with the formation of elemental selenium (eq 4).Alternately, it has been suggested that CdSe forms from directreduction without the formation of an H2Se intermediate (eq1).31 Cyclic voltammetry (CV) of our electrolyte solutionshows a first reduction process that begins at approximately−0.6 V and a second reduction process that starts at −0.85 V,but which shifts to slightly less negative potentials on thereverse scan and subsequent voltammetry cycles. We attributethe less-negative reduction process to the deposition of CdSethrough the two-step mechanism and possibly the formation ofelemental selenium (eqs 2−4) (Figure 2a). The directdeposition of CdSe and the deposition of elemental cadmium(eq 5) occur at a more negative potential than the two-stepformation of CdSe, so we assign these processes to the morenegative reduction peak. Indeed, a constant potentialdeposition at −1.0 V resulted in nanowires with a Cd/Seratio of 3 to 3.5, as measured by energy-dispersive spectroscopy(EDS). On the reverse scan, a large cadmium stripping peak isseen (eq 5).

Figure 1. Experimental setup for the electrodeposition of CdSe nanowires onto SWNTs. (a) Quartz sample with grid pattern of Fe catalyst/Ti-SiO2contact layer. (b) SEM image of typical straight aligned SWNTs growing out of grid pattern. (c) Electrochemical cell with working electrode (WE),counter electrode (CE), and reference electrode (RE). (d) Optical microscope image of deposited CdSe nanowires. An optical microscopepositioned above the electrochemical cell is used to observe the progress of electrodeposition in real time.

The Journal of Physical Chemistry C Article

dx.doi.org/10.1021/jp306804y | J. Phys. Chem. C 2012, 116, 20121−2012620122

http://pubs.acs.org/action/showImage?doi=10.1021/jp306804y&iName=master.img-001.jpg&w=311&h=155

+ + + → ++ − +H SeO Cd 6e 4H CdSe 3H O2 32

2 (1)

+ + → +− +H SeO 6e 6H H Se 3H O2 3 2 2 (2)

+ → ++ +H Se Cd CdSe 2H22

(3)

+ → +2H Se H SeO 3Se 3H O2 2 3 2 (4)

+ →+ −Cd 2e Cd2 (5)

Cyclic deposition and stripping was used to deposit smoothCdSe nanowires of controllable stoichiometry.14,32 The electro-deposition window was kept as narrow as possible in order toreduce roughness due to diffusion-limited growth caused by alarge overpotential. The electrodeposition scheme shown inFigure 2b results in stoichiometric CdSe nanowires. CdSe andcadmium are deposited through a cyclic pulse from −0.8 to−0.825 V to −0.8 V at 10 mV/s, and excess cadmium isstripped from −0.8 to −0.775 V to −0.8 V at 10 mV/s (Figure2b). Because of the large amount of cadmium species insolution relative to selenium species, we assume that anydeposited elemental selenium reacts and forms CdSe withexcess deposited cadmium.32 Increasing the number ofdeposition/stripping cycles was found to increase the size ofdeposits while maintaining nanowire composition (see theSupporting Information), with continuous nanowires generallyformed after 10 cycles.We analyzed the nucleation kinetics in our system by fitting

current transients to the Scharifker−Hills (S-H) model,33 as we

did previously with Au electrodeposition on SWNTs.28

However, for the electrodeposition of CdSe, no satisfactoryfitting of the transient currents to the theoretical equationscould be found. The nucleation mechanism could thus not beclassified as a simple instantaneous nucleation or progressivenucleation, because the current rose slowly before reaching itsmaximum, as in progressive nucleation, and then decayedslowly, as in instantaneous nucleation for some of the voltagesexplored (see the Supporting Information). The initialprogressive-like behavior could be due to adsorption ofselenium onto the carbon nanotube,14 which was also seen ingold nucleation studies.34 The S-H model assumes randomnucleation on a 2D surface, while it has been shown thataligned or randomly oriented SWNTs do not provide asufficient density of nucleation sites to assume a 2D surfaceunder the conditions that allow nanowire synthesis. Therefore,this assumption may contribute to the discrepancy betweenobserved and theoretical current transients.23,28 The transientsobserved for CdSe deposition also show slow, non-Cottrellian,current decay that may be an additional signature of preferentialnucleation along nanotubes (a 1D surface)23,28,35 or point tosome role of charge-transfer limitation. The nucleation kineticsmay also be complicated by the fact that different chemicalspecies, such as H2Se, Se, Cd, and CdSe, are deposited atsimilar potentials, but through different mechanisms.CdSe nanowires with straight, serpentine, and looped

geometries are formed by electrodeposition on SWNTs ofthe corresponding shape (Figure 3). Virtually all SWNTs incontact with the electrolyte solution were covered with CdSeafter the electrochemical scheme was applied. As a result, thedensity of the CdSe nanowires was determined by the densityof the underlying SWNTs. Both samples of straight nanowiresand samples of nanowires with curved geometries containnanowires that are continuous along the length of the carbonnanotube, as well as nanotubes with gaps in the CdSe coverage.Some continuous nanowires were found to extend from onetitanium line to the next one on the grid pattern, a distance ofabout 450 μm. We speculate that the extent of CdSe coveragemay be affected by the semiconducting or metallic nature of theunderlying SWNT, which would agree with previouslydeveloped theory.36 About 1 in 20 nanowires were found totaper to a smaller diameter at the end of the nanowire (see theSupporting Information). These nanowires are possiblydeposited over thin semiconducting nanotubes, which have arelatively larger band gap, causing a drop in potential andincreased resistance along the nanotube. Using atomic forcemicroscopy (AFM), straight nanowires were found to rangefrom 70 to 200 nm in height and curved nanowires ranged from140 to 480 nm in height. A 13% variation in height on averageoccurs along every micrometer of distance along the length of ananowire. Nanowires appear to be about twice as wide as theyare high in SEM and AFM images, respectively, suggesting ahemicylindrical shape.EDS measurements on a straight aligned nanowire sample

showed a Cd/Se stoichiometric ratio of 1.0 ± 0.2. Details aboutthe surface chemistry of the nanowires were obtained with X-ray photoelectron spectroscopy (XPS). The XPS-derivedatomic concentrations, measured from a macroscopic analysisspot, ∼0.5 mm in diameter, yielded 1.58 ± 1% and 1.27 ± 4%of Cd and Se, respectively, with the remainder of O, Si, and C,as expected from measurements on quartz substrates. Thecarbon was most likely deposited during SEM imaging. Thenonstoichiometric Cd/Se ratio is believed to reflect Cd

Figure 2. Electrodeposition processes used to produce and character-ize the formation of nanowires. (a) Cyclic voltammogram of 120 mMCdCl2 and 1 mM SeO2 at 10 mV/s. (b) Trace for CdSe cyclicdeposition/stripping from −0.8 to −0.825 V to −0.775 to −0.8 V at10 mV/s (3 cycles are shown, although 10 cycles are used to formcontinuous nanowires).

The Journal of Physical Chemistry C Article

dx.doi.org/10.1021/jp306804y | J. Phys. Chem. C 2012, 116, 20121−2012620123

http://pubs.acs.org/action/showImage?doi=10.1021/jp306804y&iName=master.img-002.jpg&w=184&h=302

oxidation at the CdSe surface. XPS can easily resolve theselenide and oxidized selenium peaks, but not those of theselenide versus oxidized cadmium signals. However, curvefitting applied to the cadmium peak (assuming a stoichiometricratio of unoxidized cadmium and unoxidized selenium)suggests that oxidized cadmium is approximately 71% of theoxide layer, most likely in the form of CdO or Cd(OH)2.

37 Theremaining 29% of the oxide layer is oxidized selenium, mostlikely SeO2. The unequal content of oxidized selenium andcadmium is associated with desorption of SeO2. The overalloxide layer at the nanowire surface is estimated to be ∼0.23 nmin thickness.The optical properties of the CdSe nanowires electro-

deposited on SWNTs are similar to those of bulk CdSe. Theabsorption spectra of the CdSe nanowires show an onset ofabsorbance at about 750 nm (Figure 4a). The photo-luminescence spectrum of a few nanowires shows a peakcentered at 720 nm (Figure 4b). Both absorbance andphotoluminescence appear to be dominated by band-edgetransitions corresponding to a band gap of 1.75 eV, similar tothat of bulk CdSe. The lack of blue shift indicates that the grainsize of the polycrystalline CdSe is larger than the Bohr excitondiameter in CdSe (11 nm).38 In principle, SWNTs have theability to quench the photoluminescence of attached CdSenanoparticles.39,40 However, although there may be somequenching, our nanowires exhibit a significant yield ofphotoluminescence. This result can be attributed to the factthat the diameter of the CdSe nanowires is 2 orders of

magnitude greater than that of the SWNTs so that most of theCdSe volume is too far from the SWNT to be quenched beforeradiative recombination occurs.Photoconductivity measurements on a few nanotube-

templated CdSe nanowires are shown in Figure 5. Uponillumination at 632.8 nm, the conductivity of these nanowiresincreases by an on/off ratio of 5.5 on average. A range of on/offratios from 1 to 10 is observed for different nanowires, whichmay reflect variations in nanowire size, and degree of contactbetween CdSe particles along a nanowire, and between themetallic contact pads and nanowires. Time-dependent measure-ments (Figure 5b) show response and recovery times of τon =9.1 ± 0.3 ms and τoff = 3.2 ± 0.1 ms, respectively. Forcomparison with other related systems, the on−off ratio of ourSWNTs-templated CdSe nanowires is about 4 times smallerthan that reported for single-crystalline CdSe nanowires,5 butthe response and recovery times are 2 orders of magnitudefaster. On the other hand, the on/off ratio and response timesfor our SWNT-templated CdSe nanowires are still significantlysmaller and slower than those reported for lithographicallypatterned nanocrystalline CdSe nanowires.41 We may speculatethat quenching by the SWNT present at the core of the CdSenanowire attenuates the on/off ratio, but can also make theresponse faster by competing with the slow recombination ofseparated electron−hole pairs, which is usually responsible forthe slow photocurrent recovery in single-crystal semiconduct-ing nanowires.42 It has also been shown that the response/recovery time of nanocrystalline semiconductor films and

Figure 3. Continuous CdSe nanowires of various geometries. SEM images show geometries such as (a, b, f) straight arrays, (c, d) thin serpentinewith continuous curved sections, (e, h) serpentine, and (i) coiled. AFM image of (g) straight nanowire arrays with marked sections with heights of87 nm (blue markers), 84 nm (red), and 68 nm (green), and (j) serpentine nanowire with marked sections with heights of 450 nm (blue), 540 nm(red), and 500 nm (green).

The Journal of Physical Chemistry C Article

dx.doi.org/10.1021/jp306804y | J. Phys. Chem. C 2012, 116, 20121−2012620124

http://pubs.acs.org/action/showImage?doi=10.1021/jp306804y&iName=master.img-003.jpg&w=359&h=323

nanowires is faster than that of single-crystalline ones.41 In thisrespect, the fact that the nanotube-templated semiconductingnanowires are nanocrystalline can actually be advantageous.Additional insight into the electrical properties of the CdSe

nanowires was obtained by applying controlled surface chargingduring XPS measurements.43 The CdSe nanowires were foundto be well connected electrically to the ground, because theCdSe signal did not shift significantly (less than 100 mV) underpositive and negative charging conditions, in marked contrastwith the surrounding quartz (shifts > 3 V). This result confirmsthe high selectivity of CdSe deposition on the nanotubetemplate with no residual CdSe on the quartz substrate. TheCdSe nanowires also showed a reversible photovoltage whenexposed to light during XPS measurements (see the SupportingInformation). An 80 meV line shift was measured in responseto light, indicating p-type behavior, where excited holes arepreferentially captured at the nanowire surface.44

■ CONCLUSIONSIn summary, we have developed a novel method for creatingsemiconducting nanowires with complex geometries by electro-deposition onto self-organized patterns of single-walled carbonnanotubes. This approach of “drawing with nanotubes” isdemonstrated here with the creation of CdSe nanowires withperfectly parallel, serpentine, and looped geometries. Theserpentine shape has interesting geometric properties because itcovers a maximum area with a single line; hence, it iscommonly seen in macroscopic functional systems of heating,cooling, lighting, collection, and irrigation. One can envisage a

vast array of analogous functional nanosystems enabled by thebottom-up assembly of semiconducting nanowires withserpentine shapes, as demonstrated herein for the first time.CdSe is an interesting semiconductor for its photoluminescent,photoconductive, and photovoltaic properties in the visiblelight spectrum. We have characterized the optical properties ofSWNT-templated CdSe nanowires, showing photoconductivitywith an up to 1 order of magnitude on/off ratio at a relativelyfast response/recovery time. The optical properties of the CdSenanowires could be further enhanced by passivating them withan inorganic shell layer with a wider band gap or with organicligands. Our ability to easily organize CdSe nanowires intoordered arrays indicates that this material can potentially beintegrated into devices such as photodetectors and photovoltaicelements. We have also shown that cyclic electrodeposition andstripping onto a carbon nanotube template is a low-cost, simple,and versatile technique that allows control over stoichiometryand shape of nanowires in high yield. In the future, thistechnique may be used to produce nanowires of other inorganicmaterials as well as organic molecules, and combine the uniquegeometry of self-organized carbon nanotube templates with theproperties of the deposited material for the achievement of newfunctionalities.

Figure 4. Optical properties of CdSe nanowires. (a) Absorptionspectrum. (b) Photoluminescence spectrum. Inset shows that the 457nm laser used for the photoluminescence measurement was focusedon a few nanowires and not a large ensemble.

Figure 5. Photoconductivity of CdSe nanowires. (a) I−V curves with(red) and without (black) illumination (laser wavelength λ = 632.8nm, intensity = 0.23 W/cm2). Inset shows an SEM image of the CdSewires, measured between two electrodes. (b) Electrical current (blackline) measured during the switching of the laser on and of f (red line)successively (bias voltage, 1.0 V; response time, τon = 9.1 ± 0.3 ms;recovery time, τoff = 3.2 ± 0.1 ms).

The Journal of Physical Chemistry C Article

dx.doi.org/10.1021/jp306804y | J. Phys. Chem. C 2012, 116, 20121−2012620125

http://pubs.acs.org/action/showImage?doi=10.1021/jp306804y&iName=master.img-004.jpg&w=178&h=308http://pubs.acs.org/action/showImage?doi=10.1021/jp306804y&iName=master.img-005.jpg&w=189&h=301

■ ASSOCIATED CONTENT*S Supporting InformationThe Supporting Information includes SEM images of additionalnanowire geometries, SEM images of nanowire growth process,plot showing kinetic analysis of the nucleation mechanism forCdSe electrodepostion and comparison to the Sharifker−Hillsmodel, and XPS data. This material is available free of chargevia the Internet at http://pubs.acs.org.

■ AUTHOR INFORMATIONCorresponding Author*E-mail: ernesto.joselevich@weizmann.ac.il. Phone: +972-8-9342350. Fax: +972-8-9344138.Present Addresses§Division of Chemistry and Chemical Engineering, CaliforniaInstitute of Technology, 1200 E. California Blvd., Pasadena, CA91125.∥Department of Neurobiology, Hebrew University, Jerusalem91904, Israel.NotesThe authors declare no competing financial interest.

■ ACKNOWLEDGMENTSWe thank O. Kedem and I. Rubinstein for help with theabsorbance measurements, G. Gotesman for help withphotoluminescence spectra, and G. Hodes for helpfuldiscussions. X.W.G. is grateful for funding from the FulbrightProgram. E.J. acknowledges support from the Israel ScienceFoundation, the U.S.−Israel Binational Science Foundation, theKimmel Center for Nanoscale Science, and the Djanogly,Alhadeff, and Perlman Family Foundations.

■ REFERENCES(1) Xia, Y. N.; Yang, P. D.; Sun, Y. G.; Wu, Y. Y.; Mayers, B.; Gates,B.; Yin, Y. D.; Kim, F.; Yan, Y. Q. Adv. Mater. 2003, 15, 353.(2) Iijima, S. Nature 1991, 354, 56.(3) Morales, A. M.; Lieber, C. M. Science 1998, 279, 208.(4) Cui, Y.; Lieber, C. M. Science 2001, 291, 851.(5) Fan, Z. Y.; Ho, J. C.; Jacobson, Z. A.; Razavi, H.; Javey, A. Proc.Natl. Acad. Sci. U.S.A. 2008, 105, 11066.(6) Kind, H.; Yan, H. Q.; Messer, B.; Law, M.; Yang, P. D. Adv.Mater. 2002, 14, 158.(7) Wang, Z. L.; Song, J. H. Science 2006, 312, 242.(8) Law, M.; Greene, L. E.; Johnson, J. C.; Saykally, R.; Yang, P. D.Nat. Mater. 2005, 4, 455.(9) Tian, B. Z.; Zheng, X. L.; Kempa, T. J.; Fang, Y.; Yu, N. F.; Yu, G.H.; Huang, J. L.; Lieber, C. M. Nature 2007, 449, 885.(10) Ma, C.; Wang, Z. L. Adv. Mater. 2005, 17, 2635.(11) Greene, L. E.; Law, M.; Goldberger, J.; Kim, F.; Johnson, J. C.;Zhang, Y. F.; Saykally, R. J.; Yang, P. D. Angew. Chem., Int. Ed. 2003,42, 3031.(12) Peng, X. S.; Zhang, J.; Wang, X. F.; Wang, Y. W.; Zhao, L. X.;Meng, G. W.; Zhang, L. D. Chem. Phys. Lett. 2001, 343, 470.(13) Pena, D. J.; Mbindyo, J. K. N.; Carado, A. J.; Mallouk, T. E.;Keating, C. D.; Razavi, B.; Mayer, T. S. J. Phys. Chem. B 2002, 106,7458.(14) Li, Q.; Brown, M. A.; Hemminger, J. C.; Penner, R. M. Chem.Mater. 2006, 18, 3432.(15) Tian, B. Z.; Xie, P.; Kempa, T. J.; Bell, D. C.; Lieber, C. M. Nat.Nanotechnol. 2009, 4, 824.(16) Kong, X. Y.; Wang, Z. L. Nano Lett. 2003, 3, 1625.(17) Gao, P. X.; Ding, Y.; Mai, W. J.; Hughes, W. L.; Lao, C. S.;Wang, Z. L. Science 2005, 309, 1700.(18) Tsivion, D.; Schvartzman, M.; Popovitz-Biro, R.; von Huth, P.;Joselevich, E. Science 2011, 333, 1003.

(19) Tsivion, D.; Schvartzman, M.; Popovitz-Biro, R.; Joselevich, E.ACS Nano 2012, 6, 6433.(20) Dudin, P. V.; Snowden, M. E.; Macpherson, J. V.; Unwin, P. R.ACS Nano 2011, 5, 10017.(21) Dudin, P. V.; Unwin, P. R.; Macpherson, J. V. J. Phys. Chem. C2010, 114, 13241.(22) Quinn, B. M.; Dekker, C.; Lemay, S. G. J. Am. Chem. Soc. 2005,127, 6146.(23) Day, T. M.; Unwin, P. R.; Macpherson, J. V. Nano Lett. 2007, 7,51.(24) Banerjee, S.; Wong, S. S. J. Am. Chem. Soc. 2003, 125, 10342.(25) Geblinger, N.; Ismach, A.; Joselevich, E. Nat. Nanotechnol. 2008,3, 195.(26) Ismach, A.; Joselevich, E. Nano Lett. 2006, 6, 1706.(27) Ismach, A.; Segev, L.; Wachtel, E.; Joselevich, E. Angew. Chem.,Int. Ed. 2004, 43, 6140.(28) Yarden, T. S.; Joselevich, E. Nano Lett. 2010, 10, 4742.(29) Ding, L.; Yuan, D. N.; Liu, J. J. Am. Chem. Soc. 2008, 130, 5428.(30) Skyllaskazacos, M.; Miller, B. J. Electrochem. Soc. 1980, 127, 869.(31) Mishra, K. K.; Rajeshwar, K. J. Electroanal. Chem. 1989, 273,169.(32) Kressin, A. M.; Doan, V. V.; Klein, J. D.; Sailor, M. J. Chem.Mater. 1991, 3, 1015.(33) Scharifker, B.; Hills, G. Electrochim. Acta 1983, 28, 879.(34) Li, Y. G.; Chrzanowski, W.; Lasia, A. J. Appl. Electrochem. 1996,26, 843.(35) Day, T. M.; Unwin, P. R.; Wilson, N. R.; Macpherson, J. V. J.Am. Chem. Soc. 2005, 127, 10639.(36) Heller, I.; Kong, J.; Williams, K. A.; Dekker, C.; Lemay, S. G. J.Am. Chem. Soc. 2006, 128, 7353.(37) Sarkar, S. K.; Cohen, H.; Hodes, G. J. Phys. Chem. B 2005, 109,182.(38) Dabbousi, B. O.; RodriguezViejo, J.; Mikulec, F. V.; Heine, J. R.;Mattoussi, H.; Ober, R.; Jensen, K. F.; Bawendi, M. G. J. Phys. Chem. B1997, 101, 9463.(39) Pan, B. F.; Cui, D. X.; Ozkan, C. S.; Ozkan, M.; Xu, P.; Huang,T.; Liu, F. T.; Chen, H.; Li, Q.; He, R.; Gao, F. J. Phys. Chem. C 2008,112, 939.(40) Peng, X. H.; Sfeir, M. Y.; Zhang, F.; Misewich, J. A.; Wong, S. S.J. Phys. Chem. C 2010, 114, 8766.(41) Kung, S. C.; van der Veer, W. E.; Yang, F.; Donavan, K. C.;Penner, R. M. Nano Lett. 2010, 10, 1481.(42) Calarco, R.; Marso, M.; Richter, T.; Aykanat, A. I.; Meijers, R.;Hart, A. V.; Stoica, T.; Luth, H. Nano Lett. 2005, 5, 981.(43) Shabtai, K.; Rubinstein, I.; Cohen, S. R.; Cohen, H. J. Am. Chem.Soc. 2000, 122, 4959.(44) Sarkar, S. K.; Hodes, G.; Kronik, L.; Cohen, H. J. Phys. Chem. C2008, 112, 6564.

The Journal of Physical Chemistry C Article

dx.doi.org/10.1021/jp306804y | J. Phys. Chem. C 2012, 116, 20121−2012620126

http://pubs.acs.org

1

Supporting Information

for

Photoconductive CdSe Nanowire Arrays,

Serpentines and Loops Formed by Electrodeposition

on Self-Organized Carbon Nanotubes

X. Wendy Gu†§

, Nitzan Shadmi†, Tohar S. Yarden

†‖, Hagai Cohen‡ and Ernesto Joselevich*

†

†Department of Materials and Interfaces, Weizmann Institute of Science, Rehovot 76100, Israel.

‡Department of Chemical Research Support; Weizmann Institute of Science, Rehovot 76100,

Israel.

E-mail: Ernesto.Joselevich@weizmann.ac.il

Phone: 972-8-934 2350

Fax: 972-8-934 4138

RECEIVED DATE

§Present address: Division of Chemistry and Chemical Engineering, California Institute of

Technology, 1200 E. California Blvd., Pasadena, CA 91125, USA.

‖Present address: Department of Neurobiology, Hebrew University, Jerusalem 91905, Israel.

2

Figure S1. Additional nanowire geometries. a) SEM image of curved, and b) EKG-shaped nanowires. c)

SEM image of a staircase geometry nanowire. A section of this nanowire is broken, revealing the

underlying carbon nanotube. d) SEM image of a coiled nanowire that tapers to a smaller diameter at the

end of the nanowire.

3

Figure S2. Deposits on SWNTs increase in size and eventually merge to form continuous nanowires with

increased pulses. SEM images of crossed nanowires after a) 1 cycle, and b) 5 cycles. SEM images of a

straight nanowire after c) 5 cycles, and d) 10 cycles.

4

Figure S3. Kinetic analysis for the elucidation of the nucleation mechanism by comparing dimensionless

current transients at different applied potentials to theoretical curves for instantaneous (continuous blue

line) and progressive (solid green line) current transients, according to the Scharifker- Hills model.1,2

5

Figure S4. CdSe nanowires show a reversible photovoltage during XPS measurements when exposed to

light. The signal shifts by 80 meV as expected for p-type behavior. The black line is the XPS signal when

dark, blue line is the signal when light is shown on the nanowires, and red line is signal when light is

turned off again.

References for Supporting Information

(1) Scharifker, B.; Hills, G. Electrochim. Acta 1983, 28, 879.

(2) Yarden, T. S.; Joselevich, E. Nano Lett 2010, 10, 4742.