Formation of Si nanorods and hollow nano-structures using high precision plasma-treated nanosphere lithography

Abstract

We report a novel method for the formation of hollow rod silicon nanostructures via plasma treated nanosphere lithography and reactive ion etching. The plasma treatment of polystyrene nanospheres (PS-NS) with O₂ and methanol–O₂ gases is studied as a function of nanosphere size, plasma power and treatment time on a silicon substrate covered by a thin layer of nickel. Applying hydrogen or HCl plasma to already treated polystyrenes, nano-rings and nano-dots can be obtained without any physical or sputtering damage. Large area silicon nano-pillars and hollow silicon nano-structures with inner diameters of 50 nm were obtained by using optimized plasma treatment. These nanostructures show a hexagonal order with good fidelity to the PS-NS patterns. Various characterization techniques such as AFM, SEM, TEM, XPS and FTIR spectroscopy have been exploited to study the samples. The hollow rod structures made using this method will have applications in low power high performance electronic devices, optoelectronic and lithium ion battery devices. Also, a model for the formation of hollow rings is presented.

---

Introduction

Nanostructured silicon performs key functions in nanoscale electronics,^1,2 optoelectronics,³ thermoelectrics,⁴ photovoltaics,⁵ biosensors,⁶ nanocapacitor arrays,⁷ as electrodes in Li-ion batteries^8–11 and vertical transistors.^12–16 Among various silicon nanostructures, silicon nanopillar arrays allow higher device densities and they are of particular interest because of their material and process compatibility with traditional integrated circuit devices. With the realization of silicon nanostructures in the sub 100 nm regime, various techniques have been proposed one of which is the anodized aluminum-oxide (AAO) method combined with metal assisted etching.¹⁷ In addition, a nano-imprint technique has been used to realize small features in the sub-100 nm regime.¹⁸

As an alternative approach, a combination of nanosphere lithography and etching can be used to fabricate high aspect ratio arrays of nanostructures. Compared to other nanofabrication processes such as electron beam lithography or focused ion beam patterning, colloidal lithography enjoys the advantage of being a parallel process that does not require expensive equipment to yield structures with comparable resolution.¹⁹ In addition to conventional nanopillar arrays, the evolution of hollow silicon-based structures can open up new opportunities to improve the performance of batteries, biological devices, transistors and sensors. For such applications, a controllable and reliable lithography approach is needed to allow inexpensive and reproducible formation of hollow nano-structure arrays. To arrive at such structures, the aforementioned techniques can be employed among which the colloidal lithography is found to be more suitable for inexpensive and large area applications. Despite the relative ease in forming small features, in common nanosphere lithography, nanoparticles are usually formed by filling the interstitial spacing between neighboring polystyrene beads, which are in physical contact and produce arrays of triangular shaped particles. The formation of undesired lines and larger spots as a result of defects in the nanosphere arrangement is another shortcoming of this technique. Although this method is applicable for the realization of nanopillars, it is not immediately suitable for hollow nano-structures. For such structures, highly ordered arrays of nano-rings are needed, for which the post processing of nano-spheres is found to be suitable.^20–23 In the past few years, the dimension and shape of PS nanoparticles have been adjusted by O₂ or mixture of O₂ and other gases plasma modification.^24–27 The evolution of necks between neighboring spheres as a consequence of this plasma processing is not desired for the formation of nanorings. Further O₂ plasma ashing, which is used to remove such necks, can disturb the order of spheres especially for smaller nanoparticles and achieving highly ordered fine structures becomes difficult.^25,28–30 Most recently, people have used an Ar plasma treatment plus solvent etching to arrive at ordered nano-rings.³¹ Despite the formation of PS nano-rings, the surface of silicon is slightly sputtered during Ar treatment which leads to undesired surface roughness.

Although polystyrene nanorings can be used as masks for subsequent etching steps, these masks are continuously damaged during deep etching of silicon. This side-effect is not desirable for the formation of high aspect ratio silicon hollow nanostructures. A metal hard mask is suitable for the etching purpose which needs to be patterned prior to the deep etching step. In this paper we report, for the first time, a method that the formation of nanorings is simultaneously achieved with the etching of nickel as the hard mask. This method is the combination of methanol–oxygen plasma to adjust the size and shape of polystyrenes and the PS pattern is transferred to nickel layer, acting as the hard mask for silicon etching during subsequent RIE processes. The desired nickel patterns (nano-dots and nano-rings) are created using dry HCl etching of a thin nickel layer, previously patterned by nano-spheres. This dry etching is sufficiently gentle that the ordered structure of PS mask remains undisturbed during creating nickel etching mask. The composition and morphology of polystyrenes in methanol–O₂ plasma and after HCl plasma will be discussed. Nanopillar formation has also been investigated to adjust both the lithography as well as the etching processes. To the best of our knowledge, this is the first time that the hollow rod silicon nanostructures are created by means of a nano-sphere colloidal lithography.

Results and discussion

1. Plasma modification of PS nanospheres

O₂ plasma is been regularly used to control the size and distribution of polystyrene nanospheres. Oxygen molecules are efficient in the decomposition process of the polymeric material as chemical decomposition by insertion of peroxo-radicals into the backbone followed by chain scission as the dominating process of plasma etching.³² It is known that during short O₂ plasma treatment, undesired necks are formed between neighboring spheres, which are completely removed by further plasma treatment. However, small polystyrene nanospheres (PS-NS) cannot tolerate extended O₂ plasma treatment and the hexagonal order and shape of them could disrupt, so their final application as the mask for further processing will be paralyzed. To circumvent this limitation we have used methanol–O₂ plasma instead of O₂ plasma for treatment of 300 and 460 nm PS nanospheres. Unlike mere oxygen plasma, the methanol–O₂ plasma showed gentle etching condition, providing possibility of hexagonally ordered 100 nm PS-NS without necks between them.

Fig. 1 compares various O₂ and methanol–O₂ plasma conditions for modification of 300 and 460 nm PS-NS. These ordered nanospheres are created on a nickel coated silicon wafer. First, 4 nm Ni thin film, coated by means of electron-beam evaporation, is treated under O₂/H₂ plasma for 30 s. This treatment is necessary for successful over-coating of PS nanospheres with a hexagonal order. Then, PS nanospheres are spin-coated on nickel layer. Fig. 1a and b show the modification of 300 and 460 nm PS nanospheres with O₂ and methanol–O₂ plasmas, respectively. Oxygen plasma obviously creates necks and protrusions around the spheroidal dots. However, by using methanol–oxygen plasma with the same times and powers these residues are not observed. Moreover, methanol–oxygen plasma does not disturb the order of PS nanospheres so we can obtain smaller ordered PS nanospheres with various structures compared to oxygen plasma. The difference between methanol–O₂ with mere oxygen plasma could originate from terminating dangling bonds and consuming oxygen radicals to produce CO₂ which can reduce aggressive conditions in plasma environment.

Fig. 1 SEM images of PS nanosphere arrays modified by O₂ and methanol–oxygen plasma. (a) 300 nm PS treated by O₂ (top images) and methanol–O₂ (bottom images) for 5 and 10 minutes (b) 460 nm PS treated by oxygen (top images) and methanol–oxygen (bottom images) for 10 and 16 minutes. All treatments have been done in 200 sccm of O₂ and 200 sccm of methanol vapor. (Scale bar is 400 nm for all images.)

Fig. 2 shows other SEM images of PS nanospheres after methanol–oxygen plasma treatment. Fig. 2a depicts a large area of ordered 460 nm PS nanospheres that were reduced to 150 nm using 50 Watt methanol–oxygen plasma for 32 minutes. It can be clearly seen that even long exposure times to methanol–O₂ plasma treatment preserves the nanospheres order very well. On the other hand, oxygen plasma treatment under the same power distorts the order of nanospheres even for 16 min treatment. Fig. 2b and c show the cross-sectional views of 460 nm PS nanospheres after methanol–oxygen treatment for 16 and 30 min, respectively, clearly demonstrating the anisotropic modification effect of plasma on PS nanospheres. The etching is more efficient on top of the particles, resulting in cone-shaped structures after several minutes of etching.

Fig. 2 SEM images (a) top view (b and c) cross section of methanol–oxygen plasma treated 460 nm ps nanosphere array using power of 50 Watt for 32, 16 and 30 minutes respectively. (The scale bar in (b and c) 400 nm.)

To further study the effect of methanol–oxygen plasma treatment on polystyrene nanospheres, we have used atomic force microscopy (AFM) and transmission electron microscopy (HR-TEM) for the analysis of the morphology modifications and FTIR and XPS for evaluation of chemical composition in 460 and 300 nm polystyrenes. It is well known that plasma can alter the chemical composition and structure of a polymer. Plasma treatment of polystyrenes in RIE effects on the polystyrene surface by the formation of highly cross-linked regions through formation of C–O–C bonds.³³ These regions can act as barrier for chain scission in further plasma treatment.

The AFM images, provided in Fig. 3, reveal the effect of long time plasma treatment on the morphology of polystyrene nanospheres. Fig. 3a shows topography of untreated polystyrenes array on silicon substrate. The topography and phase of polystyrene samples which have been treated for 15 minutes in 150 W plasma of methanol–O₂ are depicted in parts (b and c) of this figure. It can be seen that not only the spherical shapes of polystyrenes transformed into cone-shaped features, some wrinkles on the top surfaces of polystyrene were observed. The wrinkles are more obvious in Fig. 3d which relates to long time plasma treatment. We speculate that top surface of cone shaped polystyrene is more strongly affected by plasma ions rather than the area around their spherical shape, which in turn causes the formation of wrinkles on their top surface. These wrinkles have important roles in opening of polystyrene mask and nanoring creation in later steps.

Fig. 3 AFM analysis of 460 nm polystyrene arrays on silicon substrate showing (a) topography of untreated polystyrenes, (b) topography and (c) phase of 15 minutes 150 Watt methanol–O₂ plasma treated polystyrenes. (d) Topography of 22 minutes 150 Watt methanol–O₂ plasma treated polystyrenes. (Wrinkles are shown by arrows on polystyrene top surface after methanol–oxygen plasma treatment.)

For better demonstration of wrinkle creation mechanism, Fig. 4 shows schematically what happens to polystyrenes during plasma treatment. At initial stages of plasma treatment the sphere shape of polystyrenes transforms to cone shape (Fig. 4b). During further plasma treatment, competition between cross-linking and chain scissioning changes the shape of polystyrenes (Fig. 4c). Fine features on PS surface show the surface roughness which is attributed to polymer aggregation associated with cross-linking induced by plasma. Applying more plasma can create continuous crosslinks around the surface while the center of polystyrene from top surface is etched by chain scission mechanism due to direct exposure to plasma (Fig. 4d). The resulting shape achieved in Fig. 4d depends also on the plasma gas mixture. While O₂ plasma is found to enhance the cross-linking during PS etching, methanol can suppress cross-linking. Non-aggressive condition in methanol–oxygen plasma gives more time to cross-linked domain to be agglomerated to larger grain sizes around the PS. This area act as barrier to chain scission in further plasma time while allows polystyrenes on top surface to be etched. As a result deep wrinkles can be created on polystyrene top surface. Details of HCl plasma effects (Fig. 4e) will be discussed in part 3.

Fig. 4 The schematic of polystyrene structure during plasma treatment. The initial stages of plasma lead to conical formation of spheres (b) while the extended treatment results in the creation of crosslinks on the top surfaces (c). The process of aggregation of crosslinked features occurs after more extended plasma (d) and by means of HCl plasma, the nano-rings are formed (e). Hydrogen plasma could lead to the same ring formation.

Fig. 5a–f show bright and dark field TEM images of 300 nm polystyrene nanospheres after 5, 10 and 30 minutes (50 Watt) methanol–oxygen plasma treatment, respectively. TEM images illustrate that in methanol–oxygen plasma treatment first some cubic nanostructures come out from polystyrenes. These structures are assumed to be the cross-linked area, which are created to release the stress on PS structure due to crosslinking. This observation corroborates our hypothesis of consuming internal chains during crosslinking. Applying more plasma creates smaller size ordered polystyrene array. Fig. 5d illustrates that the depth of plasma diffusion is about 20 nm. The cross-linked area around PS and the evolved wrinkles on the polystyrene top surface after long time methanol–O₂ plasma can be observed in TEM images (Fig. 5e and f). The TEM images have been obtained using a Philips CM30 unit operated at 200 kV on carbon–Formvar grid as substrate.

Fig. 5 TEM images of 300 nm polystyrene arrays on carbon–Formvar grid after being treated in 50 Watt methanol–oxygen plasma. (a and b) Bright and dark field images of polystyrenes after 5 minutes treatment, (c) nanostructures produced from polystyrenes during treatment make the spheres hollow, (d) polystyrenes after 10 minutes treatment, (e) and (f) bright and dark field images of polystyrenes after 30 minutes treatment. It can be seen that the order of the array is preserved after long time treatment while nanostructures are totally removed.

Additional investigations on methanol–oxygen plasma and its effect on chemical composition of PS nanospheres are carried out by FTIR characterization of nanospheres before and after methanol, methanol–oxygen and mere oxygen plasma treatment (Fig. 6). When the untreated polystyrene spectrum is compared with the plasma treated one it is seen that a broad band range from 3100 cm⁻¹ to 3600 cm⁻¹ appears which is believed to correspond to the hydroxyl group (–OH) bonds. Also, the peaks in 1120 cm⁻¹ and 1300 cm⁻¹ appear for both oxygen and methanol–oxygen plasma which could be related to C–O aliphatic ether and carboxyl in PS nanospheres, respectively. These functional groups can act as active sites to start crosslinking. These peaks are not observed in mere methanol plasma treated spectrum. This shows methanol does not effect on PS backbone chemical composition and only makes gas mixture plasma environment less aggressive than mere O₂. Moreover, X-ray photoelectron spectroscopy (XPS) has been exploited to further investigate the bond formation of polystyrene prior and after such plasma treatments. The XPS spectra of polystyrene (Fig. 7) show that chemical environment of carbon on the PS surface has significantly changed after methanol–oxygen plasma treatment. As expected, the plasma-treated sample contains much oxygen on the surface compatible with FTIR results.

Fig. 6 FT-IR spectra of polystyrene before and after plasma treatment. The evolution of –OH groups is observed for plasma treated samples as opposed to un-treated ones. In addition, the presence of aliphatic groups is more obvious in oxygen treated sample while methanol treated structure do not show such groups.

Fig. 7 XPS spectrum of the ultra-thin film of PS surface (a) before and (b) after the methanol–oxygen plasma treatment.

Apart from topography and morphology alteration, AFM force–displacement spectroscopy has been investigated on polystyrenes before and after methanol–O₂ plasma treatment. This method is suitable to estimate the stiffness of nanostructures. The Young's modulus of polystyrene extracted from force–displacement curves (not shown) decreases remarkably on top surface of PS after plasma treatment. This can be related to consumption of polymer chains in the center to create cross-linked area around the PS. It can be a good confirmation for our hypothesis that during long time methanol–O₂ plasma treatment the spheres are etched from top surface and become hollow in the wrinkles.

To better understand the advantage of controllable RIE mask design using our modified polystyrene lithography, we have transferred various shapes of plasma treated polystyrene pattern to nickel and fabricated highly ordered silicon nanopillar arrays in large area and then obtained new silicon nanostructures which can improve performance of vertical transistors and lithium batteries. In the following section we describe the formation of silicon nano-rods and hollow nanostructures by a combination of plasma-treated nano-sphere lithography and high precision reactive ion etching.

2. Silicon nanopillars fabrication

The experimental procedure for the fabrication of large area silicon nanopillars is illustrated schematically in Fig. 8. First PS nanospheres are spin-coated on hydrophilic nickel thin film. Following this step, PS nanospheres are shrunk using methanol–oxygen plasma. Subsequently, the non-close packed nanosphere pattern is transferred into the nickel layer by using HCl vapor plasma. To obtain nickel nano-dots we have applied a short time and low energy methanol–oxygen and HCl plasma to prevent polystyrene and nickel nanoring creation. In this process wet nickel etching method cannot be used because it destroys the order of nanospheres and as a result, the underlying nickel is not favorably patterned.

Fig. 8 Schematic of silicon nanopillar fabrication process. (a) Spin coated PS on nickel layer (b) short time and low energy methanol–oxygen plasma treatment of PS (c) transferring PS pattern to nickel layer using HCl plasma (d) removing PS by sonication in dichloromethane (e) vertical etching of silicon wafer using RIE.

To fabricate silicon nanopillar arrays we have used vertical etching process that recently developed in our laboratory, which neither needs high density plasma nor uses external polymeric coating for the passivation. The vertical removal of silicon substrate is achieved with sequential etching and passivation sub cycles.³⁴ After lift-off of the polystyrene layer, the sample is placed in the RIE chamber and etched with SF₆ plasma. Deep etching with high precision features requires two sub-cycles of passivation and etching; in the passivation step a mixture of H₂/O₂ gases with a trace of SF₆ is used. While in the etching subsequence only SF₆ is used as the feed gas. This process is used to fabricate all other nanostructures besides nanopillars in this work. Fig. 9a shows a large-area SEM image of silicon nanopillar arrays that were fabricated using a 300 nm PS nanosphere beads modified by methanol–O₂ plasma and RIE. Also Fig. 9b demonstrates good quality large-area periodic silicon nanopillar arrays, obtained by this method. By optimizing the etching parameters and because of using a metal mask (nickel) high aspect silicon nanopillar arrays also can be produced (Fig. 9c and d). Fig. 9e shows cross-sectional view of high aspect ratio nanopillars. Please bear in mind that sidewalls are vertical and their scalloping is very low compared to common silicon nanopillars that were reported using DRIE methods.³⁵ All wires are etched to the same depth. These results demonstrate the capability of modifying the shape and size and distance of silicon nanopillar arrays by adjusting plasma treatment, nickel etching and sequential vertical etching conditions. In addition to solid nanopillars, we have observed interesting nanostructures by using protrusions around the spheroidal dots. Fig. 9f shows a tilted view SEM image of these silicon nanostructures. From this SEM image one ensures that silicon nanopillars are ordered in a hexagonal hole patterns.

Fig. 9 (a–d) 45° tilted SEM image of silicon nanopillar arrays fabricated by 15 and 25 sequences of etching and passivation in RIE respectively. (e) Cross section of nanorods (f) nanostructures created using protrusions around the PS spheres mask.

3. Hollow rod silicon nanostructures fabrication

To achieve nickel nanoring masks, the methanol–oxygen plasma treated polystyrenes with wrinkles on top surface were exposed to HCl plasma. HCl plasma is chosen because it can simultaneously act as opening reagent by H-radicals and etching of nickel layer by Cl radicals. HCl plasma opens up the polystyrene wrinkles by increasing chain scission compared to crosslinking rate on top surface. We have further corroborated our hypothesis of hydrogen effect in opening the PS, by exposing polystyrenes in mere hydrogen plasma. Fig. 10a and b show nanorings created successfully after hydrogen plasma treatment of 300 nm and 460 nm PS, respectively.

Fig. 10 PS nanorings created after hydrogen plasma treatment.

It is worth noting that the argon plasma treatment reported in ref. 31 to produce hollow polystyrenes and then PS nanoring, can damage both nanospheres and substrate in long time treatments. In contrast our process of methanol–oxygen plasma uses the advantage of preserving the polystyrenes and the substrate undamaged even in long time treatment. It can also produce smaller non-close-packed arrays by oxidation process. Moreover, using HCl plasma instead of solvent has advantage of simultaneous etching of nickel during PS etching and ring creation.

Taking advantage of the aforementioned method, we have been able to produce novel hollow rod silicon structures. These structures have been introduced as active building blocks in various research areas of modern nanotechnology such as lithium ion batteries and vertical transistors. The important challenge in silicon based ion lithium batteries is the volume change of silicon during charge and discharge.¹¹ Currently, there is substantial interest in the fabrication of silicon nanowire based lithium ion batteries to surpass this limitation. These novel hollow structures can be one of solutions due to their possibility of stress release during lithium diffusion into the silicon and having more active area compare to nanorods. On the other hand, recently many computer simulations have been performed to compare the performance and power consumption of classical planar FETs and all-around gate FETs fabricated by silicon nanowires and hollow cylindrical nanorods.^14–16 Such simulations implied that using silicon hollow cylindrical nanorods could combine the advantage of high drive current with superior off-state characteristics. To the best of our knowledge, this is the first time that vertical silicon hollow nano-rod arrays are realized.

Fabrication steps of novel hollow rod silicon structures are shown schematically in Fig. 11. Fig. 12a shows cross view of 300 nm PS spheres with nano-hole after modification with 50 W methanol–oxygen plasma for 33 min. It can be clearly seen that hexagonal order of PS beads has been preserved. Fig. 12b shows PS nanoring created by HCl plasma treatment. Fig. 12c and d demonstrate SEM images of hollow rod silicon nanostructure arrays after vertical etching of silicon by nanoring masks. Using SEM image analysis the average outer and inner diameters of hollow rod structures were found to be 150 nm and 70 nm, respectively for 300 nm polystyrenes (Fig. 12c). Using 460 nm PS, we have produced hollow rods with the size of 300 nm as outer and 200–250 nm as inner diameters (Fig. 12d). The hole size could be adjusted as a function of plasma power and etching period. This investigation is currently under progress. Part (e) of Fig. 12 shows the etching process allow fabrication of silicon nanostructures with their hollow tube has progressed to the bottom of the structure. Under etching of silicon hollow nanorods can be used to show the nickel nanoring mask (the inset of Fig. 12e). Occasionally the plasma conditions make polystyrenes as ring mask with a dot in the middle (inset of Fig. 12f). Transferring PS mask to nickel layer and etching the silicon produce nanotubes with rods in the middle (Fig. 12f). This case occurs when polystyrenes do not have enough time in HCl plasma to be opened up completely.

Fig. 11 Schematic of silicon hollow rod silicon nanostructures fabrication process. (a) Spin coated PS on nickel layer (b) fabrication of PS with wrinkles in top surface using long time methanol–oxygen plasma treatment (c) PS and nickel nanoring formation simultaneously using HCl plasma (d) removing PS by sonication in dichloromethane (e) vertical etching of silicon wafer using RIE.

Fig. 12 SEM images of (a) cross view of 300 nm ps spheres with nano-hole after modification in 50 Watt methanol–O₂ plasma for 33 minutes. (b) Shows PS nanorings produced after HCl plasma (c and d) 45° tilted view hollow rod silicon nanostructures created by etching of silicon with nickel nanoring masks in (b). (e) Scratched hollow rod silicon structures. (f) Nanotube structures with the nanorods in the middle. The inset shows PS mask of these structures.

Conclusions

In summary, we have developed a low cost and high throughput fabrication process for various kinds of silicon nanostructures with modification of the PS nanosphere plasma treatment and RIE method. The style, size and separation of the fabricated periodic nanostructures can be independently adjusted by selecting different diameters of polystyrene beads in the nanosphere lithography step and by using an optimum plasma modification process. By controlling the conditions of the methanol–O₂ and HCl plasma we have successfully created nano-dot and nano-ring nickel masks. Using metal masks instead of polymeric masks allows one to achieve higher aspect ratios. Also, with this modified fabrication process novel hollow rod silicon nanostructures were obtained that can have new applications in electronics, optoelectronics and ion lithium battery devices.

Experimental

To fabricate silicon nanostructures 1 × 1 cm² n-doped (100) silicon wafers were used. These silicon substrates were cleaned using RCA #1 solution and rinsed in D.I. water. First 4 nm nickel layer was deposited on substrate by electron beam evaporation. The layer was treated by O₂/H₂ plasma in RIE and polystyrene beads were spin-coated nickel. For this experiment, polystyrene beads with diameter of 460 nm and 300 nm have been brought from Sigma Aldrich Corporation. Next, PS spheres were modified in RIE. A gas mixture of 200 sccm methanol vapor and O₂ with 50 and 150 W RF power was used for PS spheres size reduction. Non-closed packed PS nanospheres pattern was transferred to nickel layer by HCl vapor plasma. After etching of nickel, the sample was placed in an RIE chamber and etched with SF₆ plasma. Etch conditions were as follows; etching process uses SF₆ with respective flows of 40 sccm, plasma power of 130 W and 8 s duration in around 160 mTorr in the etching step. For the passivation step, a gas mixture of SF₆/H₂/O₂ with respective flows of 230/4/620 sccm was used for a period of 50 seconds and plasma power of 210 W in pressure of 600 mTorr. As our patterns are dense for getting higher aspect ratio we should increase etching sub-cycle power from 130 W to 145 W. All SEM images were taken in a Hitachi scanning electron microscopy at 20 kV.

References

1. V. Schmidt, H. Riel, S. Senz, S. Karg, W. Riess and U. Gösele, Small, 2006, 2, 85.
2. J. Goldberger, A. I. Hochbaum, R. Fan and P. Yang, Nano Lett., 2006, 6, 973.
3. Y. Kanemitsu, Phys. Rep., 1995, 263, 1.
4. A. I. Hochbaum, R. Chen, R. D. Delgado, W. Liang, E. C. Garnett, M. Najarian, A. Majumdar and P. Yang, Nature, 2008, 451, 163.
5. B. Tian, X. Zheng, T. J. Kempa, Y. Fang, N. Yu, G. Yu, J. Huang and C. M. Lieber, Nature, 2007, 449, 885.
6. Y. Cui, Q. Wei, H. Park and C. M. Lieber, Science, 2001, 293, 1289.
7. S. W. Chang, J. Oh, S. T. Boles and C. V. Thompson, Appl. Phys. Lett., 2010, 96, 153108.
8. C. K. Chan, H. Peng, G. Liu, K. McIlwrath, X. F. Zhang, R. A. Huggins and Y. Cui, Nat. Nanotechnol., 2008, 3, 31.
9. A. M. Chockla, J. T. Harris, V. A. Akhavan, T. D. Bogart, V. C. Holmberg, C. Steinhagen, C. B. Mullins, K. J. Stevenson and B. A. Korgel, J. Am. Chem. Soc., 2011, 51, 20914.
10. J. K. Yoo, J. Kim, Y. S. Jung and K. Kang, Adv. Mater., 2012, 24, 5452.
11. T. Song, J. Xia, J. H. Lee, D. H. Lee, M. S. Kwon, J. M. Choi, J. Wu, S. K. Doo, H. Chang, W. Park, D. S. Zang, H. Kim, Y. Huang, K. C. Hwang, J. A. Rogers and U. Paik, Nano Lett., 2010, 10, 1710.
12. P. Nguyen, H. T. Ng, T. Yamada, M. K. Smith, J. Li, J. Han and M. Meyyappan, ibid., 2004, 4, 651.
13. T. Bryllert, L. E. Wernersson, L. E. Froberg and L. Samuelson, IEEE Electron Device Lett., 2006, 5, 323.
14. H. M. Fahad and M. M. Hussain, IEEE Trans. Electron Devices, 2013, 60, 1034.
15. H. M. Fahad, C. E. Smith, J. P. Rojas and M. M. Hussain, Nano Lett., 2011, 11, 4393.
16. H. M. Fahad and M. M. Hussain, Sci. Rep., 2012, 2, 475.
17. Z. Huang, X. Zhang, M. Reiche, L. Liu, W. Lee, T. Shimizu, S. Senz and U. Gösele, Nano Lett., 2008, 8, 3046.
18. K. J. Morton, G. Nieberg, S. Bai and S. Y. Chou, Nanotechnology, 2008, 19, 345301.
19. X. Ye and L. Qi, Nano Today, 2011, 6, 608.
20. A. Kosiorek, W. Kandulski, H. Glaczynska and M. Giersig, Small, 2005, 1, 439.
21. F. Yan and W. A. Goedel, Nano Lett., 2004, 4, 1193.
22. J. McLellan, M. Geissler and Y. Xia, J. Am. Chem. Soc., 2004, 126, 10830.
23. Y. Qingfeng, L. Fang, W. Likui, J. Yang Lee and X. S. Zhao, J. Mater. Chem., 2006, 16, 2132.
24. A. Plett, F. Enderle, M. Saitner, A. Manzke, C. Pfahler, S. Wiedemann and P. Ziemann, Adv. Funct. Mater., 2009, 19, 3279.
25. D. Choi, H. Yu, S. Jang and S. Yang, J. Am. Chem. Soc., 2004, 126, 7019.
26. M. Manso Silván, M. Arroyo Hernández, V. Torres Costa, R. J. Martín Palma and J. M. Martínez Duart, Europhys. Lett., 2006, 76, 690.
27. B. J.-Y. Tan, C.-H. Sow, K.-Y. Lim, F.-C. Cheong, G.-L. Chong, A. T.-S. Wee and C.-K. Ong, J. Phys. Chem. B, 2004, 108, 18575.
28. C. Haginoya, M. Ishibashi and K. Koike, Appl. Phys. Lett., 1997, 71, 2934.
29. T. Fujimura, T. Tamura, T. Itoh, C. Haginoya, Y. Komori and T. Koda, Appl. Phys. Lett., 2001, 78, 1478.
30. D. Gogel, M. Weinl, J. K. N. Lindner and B. Stritzker, J. Optoelectron. Adv. Mater., 2010, 12, 740.
31. Z. A. Lewicka, A. Bahloul, W. W. Yu and V. L. Colvin, Nanoscale, 2013, 5, 11078.
32. N. Vogel, S. Goerres, K. Landfester and C. K. Weiss, Macromol. Chem. Phys., 2011, 212, 1719.
33. Y. Ting, Ch. Liu, S. Park, H. Jiang 3, P. F. Nealey and A. E. Wendt, Polymers, 2010, 2, 649.
34. S. Azimi, M. Mehran, A. Amini, A. Vali, S. Mohajerzadeh and M. Fathipour, J. Vac. Sci. Technol., B: Nanotechnol. Microelectron.: Mater., Process., Meas., Phenom., 2010, 28, 1125.
35. A. A. Ayon, R. Braff, C. C. Lin, H. H. Sawin and M. A. Schmidt, J. Electrochem. Soc., 1999, 146, 339.

---