Plasma Deposition of Superhydrophobic Ag@TiO2 Core@shell Nanorods on Processable Substrates

- 1 - ((please add journal code and manuscript number, e.g., DOI: 10.1002/ppap.201100001)) Full Paper Plasma deposition of superhydrophobic Ag@TiO2 core@shell nanorods on processable substrates a Manuel Macias-Montero, Ana Borras,* Pablo Romero-Gomez, Jose Cotrino, Fabian Frutos and Agustin R. Gonzalez-Elipe ––––––––– M. Macias, Dr. Borras, Dr. Romero-Gomez, Prof. Cotrino, Prof. Gonzalez-Elipe Nanotechnology on Surfaces Laboratory, Materials Science Institute of Seville (ICMS), CSIC-University of Seville, C/ Americo Vespucio 49, Seville, E-41092, Spain E-mail: [email protected] Dr. F. Frutos Applied Physic Department, E.T.S. Ingenieria Informatica (University of Seville), Avd. Reina Mercedes s/n, Seville, E-41012, Spain ––––––––– This work reports the low temperature plasma formation of Ag@TiO2 nanorods consisting of a silver core and an external shell of TiO2 deposited on silver coated processable substrates. Layers of supported nanorods have been analyzed by scanning electron microscopy (SEM), transmission electron microscopy (TEM) and time of flight secondary ion mass spectroscopy (TOF-SIMS). The analysis of the nanorods morphologies as a function of the preparation conditions has revealed that the effect of the plasma sheath, the high mobility of the silver and the incoming direction of the precursor moieties are key factors determining the morphology, dimensions and tilting orientation of the core@shell nanostructures. Post-annealing of the asprepared samples at 673 K induced the crystallization of the amorphous TiO2 shell into the anatase phase. The as-grown amorphous and anatase Ag@TiO2 NRs surfaces present a superhydrophobic behavior with water contact angles higher than 150º, that turn into superhydrophilic through a two-step kinetic by irradiation with UV light. This wetting behavior is discussed within the framework of the Wenzel and Cassie-Baxter models, a Supporting Information is available at Wiley Online Library or from the author. ((Other reference to the authors can also appear here, such as Author-One and Author-Two contributed equally to this work.)) - 2 - including the effect of the water penetration in the inter-NR space during the light-mediated superhydrophobic-superhydrophilic transformation. Introduction During the last decade, the fabrication and processing of one dimensional (1D) supported TiO2 nanostructures have deserved an ever-increasing interest because of their high performance as photocatalysts, electrodes in photovoltaic cells, nanosensors, biomaterials, self-cleaning surfaces or microfluidic components.[1] In relation with these two latter applications, the singular properties of the 1D TiO2 nanostructures have prompted their use as model systems to investigate basic wetting phenomena. Among these features we can quote i) a high thermal and chemical stability, ii) a hydrophobic-superhydrophilic conversion by illumination with UV light or iii) the superhydrophobic character of high density 1D nanostructured-layers. In this context, a key feature for many applications is to succeed in reversibly controlling the wettability of surfaces (i.e., the change from hydrophobic to hydrophilic and back). This has been attempted by the application of electric fields,[2-5] heating[6-7] and light irradiation.[8] The reversible light induced superhydrophobicsuperhydrophylic transition of Ag@TiO2 core@shell nanofibers (NFs) is a clear example of the possibilities offered by these structures to light tuning the water contact angle (WCA) of a nanostructured surface.[9] Typically, TiO2 nanostructures comprising nanorods, nanofibers and nanotubes have been prepared by means of chemical or electrochemical routes including template methods.[10-17] High temperature annealing of Ti foils has been also intended for the preparation of this type of systems.[18] These methods involve successive drying, heating, or high temperature annealing steps that hamper their application when mild conditions are required or when the synthesis has to be carried out in a single experimental set-up. Plasma methods provide a way to circumvent part of these shortcomings[19-20] as clearly exemplified by the fabrication of - 3 - carbon nanotubes (CNTs) at temperatures much lower than with thermal CVD or hydrothermal processes.[21-23] Nanoand meso-porous TiO2 films have been previously deposited at low substrate temperatures by plasma-enhanced chemical vapour deposition (PECVD).[24-25] The aforementioned Ag@TiO2 core@shell NFs were also prepared by this technique at T > 403 K by depositing TiO2 on a plasma oxidized silver foil.[9,26-28] A special feature of those supported Ag@TiO2 core@shell NFs was that the inner silver core was formed by a 20 nm thick single crystalline thread of this metal. In the present work we have succeeded in extending the methodology of fabrication of Ag@TiO2 nanostructures from bulk silver to processable substrates, the first step required for their integration in actual devices such as solar cells or microfluidics circuits. This goal has been attempted by plasma deposition of TiO2 on a thin silver layer deposited on flat substrates by DC sputtering. Recently, an equivalent route has been fruitfully applied for the preparation of Ag-NPs@ZnO nanorods (NRs)[29,30] characterized by a polycrystalline hollow shell of ZnO decorated in its interior with Ag nanoparticles. Herein, to tailor the shape, density and other morphological characteristics of the Ag@TiO2 1D nanostructures, we have systematically varied experimental parameters such as the deposition geometry or the temperature during the plasma deposition of TiO2. In the quest for manageable wetting surfaces, we have also studied the surface wetting behavior of the nanorods-arrays either as prepared or after UV irradiation to induce a hydrophobic to hydrophilic transformation. The results obtained, rationalized by applying the Wenzel and Cassie-Baxter models,[25b,31] design a complex pattern where the surface nanostructure, the intrinsic photo-activity of TiO2 and the dependence of this latter property on its crystalline structure play critical roles. Experimental Section Ag@TiO2 Nanorods fabrication - 4 - Fused silica slides and Si(100) wafers were covered by silver via DC sputtering from a metal wire under a pressure of 1.5 Torr of Argon and an applied voltage of 400 V. Calibration of the deposition rate allowed us to control the amount of deposited silver and to determine the equivalent layer thickness (i.e., the thickness in case that silver would have formed a continuous and compact film). The sample holder with the silver covered substrates was heated by irradiation with quartz lamps up to a maximum temperature of 430 K. The reactor consisted of a stainless-steel chamber supplied with a microwave (MW) plasma source (SLAN from Plasma Consult, Germany) in a remote configuration and directly supplied with the plasma gas, in our case oxygen at a flow rate of 21.6 sccm. A detailed description of this reactor and its working conditions can be found elsewhere.[32] Herein, in order to account for the orientation of the deposited structures, it is important to specifically describe the inner reactor geometry and other working details. Figure 1 shows the reactor schematics highlighting the following features: i) a plasma source which is directly fed with the oxygen plasma gas, ii) a precursor dispenser placed above the substrate and iii) a pumping system, which placed below the substrate position, defines a preferential top-bottom flux of both the plasma gas and precursor fragments. The silver covered substrates were oxidized by exposure to oxygen plasma at a temperature of 407 K during 30 minutes. As it will be discussed later, such a treatment increases the roughness of the silver substrate and promotes the mobility of the silver clusters. Similar conditions of oxygen plasma have been previously applied to the fabrication of supported hybrid nanowires.[33] For the plasma enhanced chemical vapor deposition (PECVD) of TiO2, the reactor was supplied with oxygen and excited with a microwave power of 400 W. The Ti[OCH(CH3)2]4 precursor (Sigma-Aldrich) was dosed by passing and additional oxygen flow (0.5 – 2.5 sccm) through a stainless-steel recipient containing the precursor at 313 K. Total working pressure during deposition was 5 x 10-3 Torr. Both the line connecting this container with the chamber and the precursor shower-like dispenser were heated at T > 353 K - 5 - to avoid condensation of the metal-organic precursor. The oxygen flow through the precursor reservoir was used as an indirect way of controlling the amount of precursor dosed in each experiment. Conversion of the amorphous TiO2 shell into the anatase phase is achieved by the annealing of samples in air at 673 K for 4h.[34] Characterization Scanning electron microscopy (SEM) analysis of the specimen was carried out with a Hitachi S5200 field-emission microscope. The samples were examined without any specific manipulation. For transmission electron microscopy (TEM, Phillips CM200) analysis of the individual nanorods, they were scraped from the surface with a diamond tip and subsequently collected with a Holley-Carbon grid. Time-of-Flight Secondary Ion Mass Spectroscopy (TOF-SIMS) depth profiling analysis was performed on a TOF-SIMS type ION-TOF IV instrument (IONTOF GmbH, Münster, Germany), equipped with a Bi polyatomic primary ion source, a Cs/electron impact dual source column, and a low-energy electron flood gun (for charge compensation of insulating samples). Gracing incidence X-ray diffraction (GAXRD) studies were performed in an X`Pert Pro from Panalytical for X-Ray angles < 1º. Water Contact angles (WCA) were determined on aged samples (i.e., stored for 30 days in a desiccator after their preparation) by the Young method with a Data Physic Instruments using small droplets of mili-Q water. The samples were irradiated with a Xe lamp (photon intensity at the position of the samples ca. 2 W cm-2) and the WCA were measured after irradiation for given periods of time. Except for the original samples before irradiation where advancing and receding angles could be determined (the reported values correspond to the mean value of five measurements in different zones of the sample), only static contact angles from just one small water droplet of 5µl size were taken for the data as a function of the irradiation time. The - 6 - reason for that is that illuminated surfaces of TiO2 are not static and their wetting characteristics evolve with time and exposure to visible light. This precludes the reliability of any long essay and makes that the reported values must be taken as indicative of tendencies rather than as statistically exact. Results and Discussion Synthesis and characterization of Ag@TiO2 nanorods Ag@TiO2 nanorods (NRs) were grown on the selected processable substrates by following a three-step methodology: 1) silver deposition by DC sputtering on Si or quartz, 2) plasma oxidation of the deposited silver to get structured and chemically modified silver surfaces and 3) PECVD deposition of TiO2 at 407 K. An example of the as-prepared substrates and of the effect of plasma oxidation is shown in Figure 2, together with the images of the final NRs obtained by TiO2 deposition for 60 minutes on a silver layer of 50 nm nominal thickness and with an oxygen flow through the precursor dispenser of 0.5 sccm. The as-prepared silver layers (Figure 2a and d), consisted of a continuous and smooth film of agglomerated silver particles. In good agreement with previous studies,[27-28] after plasma oxidation the treated silver was no longer smooth neither continuous as shown in Figures 2 b) and e), where it is apparent that the oxidized substrates become highly rough and presented bare silicon areas not covered by silver. On these substrates plasma deposition of TiO2 rendered a large number of core@shell NRs with a length of 550 nm and a thickness comprised between 80 nm and 250 nm depending on the deposition time (c.f., Figures 2 c) and f)). It is worthy of note that despite the inhomogeneous silver distribution in the plasma pre-treated substrate the obtained NRs were quite homogeneous, both in length and thickness. This result suggests that silver undergoes additional morphological modifications during the PECVD of TiO2. We will be back to this discussion in the next section regarding the formation of tilted nanorods. TEM and time of flight secondary ion mass spectroscopy (ToFSIMS) analysis were carried out to, respectively, characterize the internal microstructure and - 7 - compositional depth distribution of silver through the nanorod layer (Figure 3). The bright field TEM micrograph of a single Ag@TiO2 NR in Figure 3 a) shows a high contrast between an inner core attributed to silver (30nm thick) and an external shell of titanium dioxide. Contrary to the single crystalline character of the silver threads formed in the Ag@TiO2 NFs developed on bulk silver substrates,[26-27] for the Ag@TiO2 NRs on processable supports the silver appears in the form of agglomerated particles mostly distributed in the core of the 1D heterostructure although isolated silver clusters can be also observed in the surface of the TiO2 shell (Figure 3 a). This latter fact further supports the key role of the mobility of the silver clusters upon plasma and temperature activation. On other hand, complementary ToFSIMS analysis (Figure 3 b) shows that both silver and titanium dioxide are distributed homogeneously from the top of the NRs up to their base. The appearance of silicon in the depth distribution profile before that silver and titanium signals go to zero is inherent to the open microstructure of the samples and indicates that in this NR microstructure the sputtering ions enter up to the silicon substrate before the complete removal of TiO2 and silver. Control of morphology and formation of tilted NRs Different NR morphologies could be obtained by varying some process parameters, particularly the precursor dose, the thickness of the silver layer and the deposition time. In the curse of this investigation it could be determined that a minimum equivalent thickness of 20 nm of silver was required for the formation of NRs. A selection of typical microstructures is shown in Figure 4 reporting first the effect of changing the flow of oxygen bubbled through the precursor to control its dosage. According to Figure 4 a), TiO2 deposition for 60 minutes with an oxygen flow of 0.5 sccm renders NRs of about 550 nm long and 100 nm thick. Deposition for the same time and an oxygen flow of 1.0 sccm yields NRs with a similar thickness and a mean length of 900 nm (Figure 4 b), a result indicating that it is possible to get high aspect ratio NRs by just increasing the amount of precursor dosed in the chamber. In a limit case, when the oxygen flow was augmented to 2.5 sccm (Figure 4 c), inhomogeneous - 8 - and irregular nanostructures possessing an even higher aspect ratio (i.e., lengths in the order of 1400 nm and approximate thickness of 100 nm) were obtained. These new forms might be useful for applications requiring large specific areas. Meanwhile, experiments carried out for 120 minutes (c.f., Figure 4 d) keeping constant the oxygen flow through the precursor at 0.5 sccm, lead to the formation of thick (100-250nm) and relatively low aspect ratio NRs. This result indicates that, once the NRs reach a critical height, they start to grow radially with an inverted-conical morphology characteristic of the self-shadowing effect presented on plasma made 1D core@shell systems,[35,36] more pronounced in the thicker nanorods (Figure 4 d). A close inspection of the SEM micrographs in Figure 4 also reveals that the long and thick NRs coexist with other shorter and thinner nanostructures. This heterogeneous distribution of NRs lengths might be related to both, deposition of TiO2 on small size metal seeds and the shadowing effects produced by the bigger NRs. Such a hint of the existence of shadowing effects during the growth process of the NRs prompted us to prepare tilted nanostructures by playing with the directionality of the TTIP precursor flow with respect to the normal direction to the substrate. Effectively, we found that a wide range of NRs tilting angles could be obtained by varying both the angle and separation of the substrate with respect to the precursor dispenser. Details of the procedure can be also found in reference 30 where the fabrication of tilted ZnO NRs by PECVD was reported. For TiO2, an example of the effect of the gas flow directionality is presented in Figure 4 e) where we show the cross section SEM micrograph of a nanorods-arrays forming 30º with the normal to the substrate. These NRs were obtained by placing the substrate at 45º and a distance of 7 cm with respect to the precursor dispenser (cf. Figure 1). The crystal structure of the NRs was analyzed by GA-XRD at an incident angle of 1º. Figure 5 b) shows the XRD pattern corresponding to the material prepared at 407 K, where no peaks from crystalline titanium oxide or silver can be detected. Since for many applications of 1D TiO2 nanostructures it is preferable to use the anatase phase of this - 9 - material, we subjected the obtained Ag@TiO2 amorphous samples to a post annealing treatment at 673 K. Secondary and backscattered SEM images in Figure 5 a) reveal that after this treatment part of the silver segregates as small nanoparticles on top of the NRs. Except for this modification of the metal distribution, the nanostructure morphology remained unaltered even if GAXRD (Figure 5 b) revealed the crystallization of both TiO2 and silver. Growth mechanism for supported Ag@TiO2 NRs Plasma deposition has been widely used for the controlled synthesis of carbon nanorods and nanotubes.[19-23] When used for this purpose, carbon species from a hydrocarbon plasma bind under the action of metal catalyst particles ending up with the formation of the 1D nanostructures.[37,38] Our original results concerning the formation of Ag@TiO2 NFs on bulk silver substrates[26-28] were accounted for by a “volcano” mechanism with no intervention of catalytic effects and where the high mobility of the silver oxide formed during the silver pretreatment with an oxygen plasma was the key process. Although the formation of NRs grown on processable substrates reported here is in part congruent with this model, two additional specific features should be taken into account: i) the possibility to form vertical and tilted nanostructures and ii) the formation of quite homogeneous NRs arrays despite the inhomogeneity of the pre-oxidized silver substrates. Figure 6 depicts a scheme to account for the formation of Ag@TiO2 NRs on processable substrates. Two situations are considered depending on whether the substrate is normal to the incoming flux of material or it forms an angle with respect to this direction. These two situations will be discussed separately. After plasma oxidation and heating, the silver seed layer (Figure 6 Left) (steps t0 and t1) forms irregular and randomly distributed structures of silver and silver oxide nuclei. Since the amount of nucleation points is limited and its distribution random, the distribution of NRs obtained by PECVD of TiO2 was rather homogeneous at a mesoscopic scale (step t2). According to the “volcano” mechanism the NRs growth takes place preferentially on top of silver oxide moieties which become reduced to metallic silver by reaction with the TTIP at - 16 - Received: ((will be filled in by the editorial staff)); Revised: ((will be filled in by the editorial staff)); Published online: ((please add journal code and manuscript number, e.g., DOI: 10.1002/ppap.201100001)) Keywords: nanorods; nanowires; superhydrophobic; titanium dioxide; wettability [1] a) X. Chen, Chin. J. Catal. 2009, 30, 839; b) C. Cao, C. Hu, X. Wang, S. Wang, Y. Tian, H. Zhang, Sens. Act. B 2011, 156, 114; c) K. S. Brammer, S. Oh , J. O. Gallagher, S. Jin, Nano Lett. 2008, 8, 786. [2] J. L. Campbell, M. Breedon, K. Latham, K. Kalantar-Zadeh, Langmuir 2008, 24, 5091. [3] B. Kakade, R. Mehta, A. Durge, S. Kulkarni and V. Pillai, Nano Lett. 2008, 8, 2693. [4] N. Verplanck, E. Galopin, J. C. Camart, V. Thomy, Y. Coffinier, R. Boukherroub, Nano Lett. 2007, 7, 813. [5] T. N. Krupenkin, J. A. Taylor, T. M. Schneider, S. Yang, Langmuir 2004, 20, 3824. [6] H. Z. Wang, Z. P. Huang, Q. J. Cai, K. Kulkarni, C.-L. Chen, D. Carnahan, Z. F. Ren, Carbon 2010, 48, 868. [7] J. Yang, Z. Zhang, X. Men, X. Xu, X. Zhu, Carbon 2011, 49, 19. [8] X. Feng, J. Zhai, L. Jiang, Angew. Chem. Int. Ed. 2005, 44, 5115. [9] A. Borras, A. Barranco, A. R. Gonzalez-Elipe, Langmuir 2008, 24, 8021. [10] S. Yoo, S. A. Akbar, K. H. Sandhage, Adv. Mater. 2004, 16, 260. [11] J. Joo, S. G. Kwon, T. Yu, J. Phys. Chem. B 2005, 109, 15297. [12] C. Dames, B. Poudel, W. Z. Wang, J. Y. Huang, Z. F. Ren, Y. Sun, J. I. Oh, C. Opeil, M. J. Naughton, G. Chen, Appl. Phys. Lett. 2005, 87, 031901. [13] P. Hoyer, Langmuir 1996, 12, 1411. - 17 - [14] D. Gong, C. A. Grimes, O. K. Varghese, W. Hu, R. S. Singh, Z. Chen, E. C. Dickey, J. Mater. Res. 2001, 16, 3331. [15] G. H. Du, Q. Chen, R. C. Che, Z. Y. Yuan, L. M. Peng, Appl. Phys. Lett. 2001, 79, 3702. [16] T. Kasuga, M. Hiramatsu, A. Hoson, T. Sekino, K. Niihara, Adv. Mater. 1999, 11, 1307. [17] J. Du, J. Zhang, Zh. Liu, B. Han, T. Jiang, Y. Huang, Langmuir 2006, 22, 1307. [18] J. Zhou, Y. Ding, S. Z. Deng, L. Gong, N. S. Xu, Z. L. Wang, Adv. Mater. 2005, 17, 2107. [19] K. Ostrikov, U. Cvelbar, A. B. Murphy, J. Phys. D: Appl. Phys. 2011, 44, 174001. [20] K. Ostrikov , E. C. Neyts, M. Meyyappan, Advances in Physics 2013, 62, 113. [21] G. Zhong, M. Tachiki, H. Umezawa, T. Fujisaki, H. Kawarada, Chem. Vap. Deposition 2004, 10, 125. [22] K. B. K. Teo, S.-B. Lee, M. Chhowalla, V. Semet, V. T. Binh, O. Groening, M. Castignolles, A. Loiseau, G. Pirio, P. Legagneux, D. Pribat, D. G. Hasko, H. Ahmed, G. A. J. Amaratunga, W. I. Milne, Nanotechnol. 2003, 14, 204. [23] H. S. Uh, S. S. Park, Thin Solid Films 2006, 504, 50. [24] a) A. Borras, J. R. Sanchez-Valencia, J. Garrido-Moliner, A. Barranco, A. R. GonzalezElipe, Micro. Meso. Mater. 2009, 118, 314; b) A. Borras, R. Alvarez, J. R. Sanchez-Valencia, J. Ferrer, A. R. Gonzalez-Elipe, Micro. Meso. Mater. 2012, 160, 1. [25] a) A. Borras, J. R. Sanchez-Valencia, R. Widmer, V. J. Rico, A. Justo, A. R. GonzalezElipe, Crys. Growth Design 2009, 9, 2868-2876; b) A. Borras. A. R. Gonzalez-Elipe, Langmuir 2010, 26, 15875 [26] A. Borras, A. Barranco, F. Yubero, A. R. Gonzalez-Elipe, Nanotechnol. 2006, 17, 3518. [27] A. Borras, A. Barranco, J. P. Espinos, J. Cotrino, J. P. Holgado, A. R. Gonzalez-Elipe, Plasma Process. Polym. 2007, 4, 515. [28] A. Borras, M. Macias-Montero, P. Romero-Gomez, A. R. Gonzalez-Elipe, J. Phys. D. 2011, 44, 174016. - 18 - [29] M. Macias-Montero, A. Borras, Z. Saghi, P. Romero-Gomez, J. R. Sanchez-Valencia, J. C. Gonzalez, A. Barranco, P. Midgley, J. Cotrino, A. R. Gonzalez-Elipe, J. Mater. Chem. 2012, 22, 1341. [30] M. Macias-Montero, A. Borras, Z. Saghi, J. P. Espinos, A. Barranco, J. Cotrino, A. R. Gonzalez-Elipe, Nanotechnol. 2012, 23, 255303. [31] M. Macias-Montero, A. Borras, R. Alvarez, A. R. Gonzalez-Elipe, Langmuir 2012, 28, 15047. [32] A. Barranco, J. Cotrino, F. Yubero, J. P. Espinos, J. Benitez, C. Clerc, A. R. GonzalezElipe, Thin Solid Films 2001, 401, 150. [33] M. Alcaire, J. R. Sanchez-Valencia, F. J. Aparicio, Z. Saghi, J. C. Gonzalez-Gonzalez, A. Barranco, Y. Oulad-Zian, A. R. Gonzalez-Elipe, P. Midgley, J. P. Espinos, P. Groening, A. Borras, Nanoscale 2011, 3, 4554. [34] F. Gracia , J. P. Holgado, A. Caballero, A. R. Gonzalez-Elipe, J. Phys. Chem. B, 2004, 108, 17466. [35] M. Macias-Montero, A. N. Filippin, Z. Saghi, F. J. Aparicio, A. Barranco, J. P. Espinos, F. Frutos, A. R. Gonzalez-Elipe, A. Borras, Adv. Funct. Mater. 2013, DOI: 10.1002/adfm.201301120. [36] K. Moore, C. B. Clemons, K. L. Kreider, G. W. Young, J. Appl. Phys.2007, 101, 064305. [37] G. Zhong, M. Tachiki, H. Umezawa, T. Fujisaki, H. Kawarada, Chem. Vapor Depos. 2004, 10, 125. [38] K. B. K. Teo, S. B. Lee, M. Chhowalla, V. Semet, V. T. Binh, O. Groening, M. Castignolles, A. Loiseau, G. Pirio, P. Legagneux, D. Pribat, D. G. Hasko, H. Ahmed, G. A. J. Amaratunga, W. I. Milne, Nanotechnol. 2003, 14, 204. - 19 - [39] L. Gonzalez-Garcia, J. Parra-Barranco, J. R. Sanchez-Valencia, A. Barranco, A. Borras, A. R. Gonzalez-Elipe, M. C. Garcia-Gutierrez, J. J. Hernandez, D. R. Rueda, T. A. Ezquerra, Nanotechnology 2012, 23, 255701. [40] A. Lafuma, D. Quere, Nat. Mater. 2003, 2, 457. [41] A. Borras, A. Barranco and A. R. Gonzalez-Elipe, Langmuir 2008, 24, 8021. [42] A. Borras, P. Groening, J. R. Sanchez-Valencia, A. Barranco, J. P. Espinos, A. R. Gonzalez-Elipe, Langmuir 2010, 26, 1487. [43] Y. Tak, K. Yong, C. Park, J. Electrochem. Soc. 2005, 152, G794. [44] M. Haupt, A. Ladenburger, R. Sauer, K. Thonke, R. Glass, W. Roos, J. P. Spatz, H. Rauscher, S. Riethmuller, M. Moller, J. Appl. Phys. 2003, 93, 6252. [45] R. Wang, K. Hashimoto, A. Fujishima, M. Chikuni, E. Kojima, A. Kitamura, M. Shimohigoshi, T. Watanabe, Nature 1997, 388, 431. [46] N. Stevens, C. I. Priest, R. Sedev, J. Ralston, Langmuir 2003, 19, 3272. [47] D. Wang, Y. Liu, X. Liu, F. Zhou, W. Liu, Q. Xue, Chem. Commun,. 2009, 7018. - 20 - Figure 1. Scheme of the experimental setup showing two different orientations of the sample substrates with respect to the precursor flow imposed by the pumping system. - 21 - 500nm 500nm 500nm 500nm 500nm 200nm a) b) c) d) e) f) Figure 2. Cross sections and normal view SEM micrographs of sputtered silver as prepared (a and d); sputtered silver after 30 minutes of oxygen plasma treatment at 407 K (b and e); Ag@TiO2 NRs formed by deposition at 407 K (c and f). - 22 - 0200 400 600 800 1000 1200 1400 Intensity (a.u.) Depth (nm) Ti Ag Si a) b) 50nm Ag core TiO2shell Figure 3. a) TEM micrograph of an isolated NR showing a silver core and a TiO2 shell; b) ToF-SIMS analysis of a characteristic vertical aligned Ag@TiO2 sample. - 23 - 500nm 500nm 500nm 500nm500nm a) b) c) d) e) Figure 4. a), b) and c) SEM micrographs of the NRs grown with different doses of precursor controlled by chaning the oxygen flow bubbled through it: 0.5, 1 and 2.5 sccm respectively; d) NRs formed after a long plasma deposition time (120 minutes); e) Tilted structures formed by changing the orientation of the substrates with respect of the precursor flow according to the scheme in Figure 1. - 24 - Figure 5. a) Secondary electron SEM image of annealed Ag@TiO2 NRs with backscattered electrons image as an inset, b) XRD spectra of the as-grown and annealed NRs. The peaks of anatase and silver are conveniently assigned in the diagrams. - 25 - Figure 6. Different stages of the Ag@TiO2 NRs formation based on the volcano-type mechanism for vertical (left) and tilted (right) growth. Thick arrows describe the preferential direction of the precursor flow towards the substrate. Thin arrows represent the electrical filed lines associated to the plasma sheath.