Assembly and Densification of Nanowire Arrays via Shrinkage

May 16, 2014 - commercially available, polystyrene substrate (K&B Inovations,. Inc.) using a .... locations on a large polystyrene substrate (2.5 × 6...
0 downloads 0 Views 5MB Size
Download PDF
Letter pubs.acs.org/NanoLett

Assembly and Densification of Nanowire Arrays via Shrinkage Jaehoon Bang,† Jonghyun Choi,† Fan Xia,‡ Sun Sang Kwon,‡ Ali Ashraf,† Won Il Park,‡ and SungWoo Nam*,†,§ †

Department of Mechanical Science and Engineering, University of Illinois, Urbana−Champaign, Urbana, Illinois 61801, United States ‡ Division of Materials Science and Engineering, Hanyang University, Seoul 133-791, Republic of Korea § Department of Materials Science and Engineering, University of Illinois, Urbana−Champaign, Urbana, Illinois 61801, United States S Supporting Information *

ABSTRACT: Chemically synthesized semiconductor nanowires (NWs) have demonstrated substantial promise for nanoelectronics, nanoenergy, and nanobiotechnology, but the lack of an effective and controllable assembly process has limited the wide adoption of NWs in these areas. Here we demonstrate a facile, robust, and controllable approach to assembling and densifying a parallel array of NWs using shrinkable shape memory polymers. Using thermal-induced shrinkage of polystyrene, we were able to successfully assemble and densify NW arrays up to close-packing and, furthermore, achieve tunable density (up to ∼300% amplification of density) by controlling the shrinkage process. We also demonstrate scalable assembly and densification of NWs on a 2.5 × 6 inch scale to explore the manufacturability of the shrink-induced assembly process. Finally, we demonstrate the successful transfer of the shrink-assembled NW arrays onto various 2-dimensional and 3-dimensional substrates without compromising the integrity of NW assembly and density. KEYWORDS: Nanowire, shape memory polymer, shrinkage, assembly, transfer

C

between NWs and a target substrate. Contact printing methods have realized reliable assembly quality and scalability but have yet to achieve the ultrahigh density (close-packing) assembly of NWs. Here we report a shrink-induced assembly and densification of NWs, which overcomes the limitations of previously reported NW assembly methods. We preassemble NWs on a shape memory polymer substrate, to demonstrate significant densification (up to ∼300%) and improvement of assembly quality of the NWs following the shrinkage of the substrate. By analogy, our approach could be viewed as a solid-phase LB method where uniaxial compression of the NWs on the shape memory polymer assembles and densifies the NWs. This approach allowed us to realize a closely packed nanowire array as well as tunable amplification of NW density. More notably, we demonstrated the scalability of our assembly approach on a substrate up to 2.5 × 6 inches in size while maintaining uniform densification and assembly of the NWs. Lastly, we explored a solution transfer of shrink-assembled NWs from the shape memory polymer substrate to various 2-dimensional (2D) and 3-dimensional (3D) substrates for future electronic applications.

hemically synthesized semiconductor nanowires (NWs)1,2 have been extensively investigated as nanoscale building blocks for advanced nanoelectronics,3−7 nanoenergy,8−15 and nanobiotechnology.16−22 These 1-dimensional (1D) nanoscale semiconductors could be synthesized with various materials,23,24 and furthermore, the diameter, length, morphology, and doping characteristics of NWs could be precisely controlled.2,25−28 Despite all of the merits of semiconductor NWs, there has not been an effective method to realize NW arrays with controlled assembly and ultrahigh density, to yield novel, large-scale applications by leveraging the unique properties that individual NWs could provide. To address such challenges, diverse nanowire assembly techniques have been developed. Flow-assisted method29 and bubble-blown method30 use shear force of the medium where the NWs are dispersed. Both techniques, however, provide a relatively low density of NWs (∼1 NW per μm) and are far from the ideal density of NWs for high-performance electronics applications. Langmuir−Blodgett (LB) method31 uniaxially compresses a NW−surfactant monolayer on an aqueous subphase to assemble NWs and is able to obtain high density nanowire arrays. Nonetheless, the LB method requires extensive preparation of surface-modified NW solutions and, furthermore, suffers from reliability of assembly quality and lack of uniformity due to limitations in solution processing. In contrast, contact printing or nanocombing of NWs32−34 is a direct mechanical printing method using the shear friction © 2014 American Chemical Society

Received: February 25, 2014 Revised: April 30, 2014 Published: May 16, 2014 3304

dx.doi.org/10.1021/nl500709p | Nano Lett. 2014, 14, 3304−3308

Nano Letters

Letter

Figure 1. Concept of shrink-induced NW assembly process. (a) Process flow of shrink-induced NW assembly. (b) Macroscopic comparison of before (bottom) and after (top) uniaxial shrinkage (scale bar, 2 cm). Figure 2. Optical and scanning electron microscope (SEM) characterizations and analysis. (a) Dark-field optical microscope image of the NW array after the shrink-induced assembly (scale bar, 100 μm). (b) SEM image of NW arrays after shrink-induced closepacked assembly (scale bar, 1 μm). (c) Density increase of NW arrays related to different shrinking rates of polystyrene substrate. Inset SEM images are taken from the samples as indicated (scale bars, 1 μm). The error bars represent one standard deviation. (d) Improvement of NW assembly quality (i.e., misalignment angle) via shrink-assembly. Inset drawing shows how the angle of individual NW was defined.

Figure 1 illustrates the process of shrink-induced NW assembly. Silicon NWs were synthesized using Au-catalyzed vapor−liquid−solid method and then transferred onto a commercially available, polystyrene substrate (K&B Inovations, Inc.) using a contact printing preassembly method (Figure S1, Supporting Information).32,33 Shrink polymers are widely used in the shrink-wrap materials industry and also in microfluidic and metallic film applications35,36 due to their low cost and large amount of shrinkage upon heating. Next, the polystyrene substrate was clamped with a customized holder on two sides to induce uniaxial shrinkage perpendicular to the prealigned NW direction. By applying heat above a glass transition temperature (150 °C) to the clamped polystyrene substrate (Figure 1a), the restriction caused the polystyrene substrate to shrink only in one direction. Since the clamping ensures that shrinkage of the substrate occurs only in the direction that is perpendicular to the prealigned NWs, the density of NWs increases without exhibiting any significant buckling37 following shrinkage. Figure 1b shows how the mechanical clamping decides the shrinkage direction of the polystyrene polymer. As shown in Figure 1b, the “I” mark shrinks only in the horizontal direction and retains its length in the vertical direction. In the absence of clamping, the polystyrene substrate is designed to uniformly reduce its length down to 1/3 in both horizontal and vertical directions. Optical and scanning electron microscope (SEM) characterizations were carried out to systematically study the shrinkinduced assembly of NWs. Figure 2a shows a dark-field optical microscope image of the NW array after the shrink-induced assembly step. Directional contrast pattern confirms that the direction of the NW assembly remains intact after the shrinkage of the substrate. A SEM image of NW arrays after the shrinkinduced assembly step (Figure 2b) clearly shows a close-packed array of NWs with a NW diameter of 100 nm. The average density of the NWs assembled by our shrink-induced assembly is ∼9 NWs/μm (with a NW diameter of 100 nm), and furthermore, the density amplification factor was around 300%, as initial density of NWs was ∼3 NWs/μm. Even when initial NW density was relatively low (∼1.25 NWs/μm), similar trends of density amplification (∼300%) could be observed upon shrinkage (Figure S2, Supporting Information). We explore differing processing conditions and their impact on NW density (Figure 2c). In particular, we employed

differing geometrical constraints to control the shrinkage of the polystyrene substrate. More specifically, we varied the aspect ratios of the polystyrene substrate (while keeping the clamp length the same) to realize differing shrinkage rates. We use aspect ratios of 0.5:1, 1:1 , and 1.5:1 (shrinking direction: clamped direction) (Figure S3, Supporting Information), which resulted in shrinkage rates of 190%, 250%, and 310% of the polystyrene substrate, respectively. To correlate the influence of macroscopic shrinking on NW density amplification, we performed extensive investigations of NW density before and after the shrinkage step at each shrinkage rate. Such NW density amplification is correlated with uniaxial shrinkage rates. We observed that as the uniaxial shrinkage rate increases the NW density amplification factor also shows an increase (Figure 2c). We note that at higher shrinkage rates NW density amplification showed a slower increase as compared with macroscopic shrinkage rate, which we attribute to NW sliding during the shrinkage process.38 Furthermore, to quantitatively investigate the improvement of NW assembly quality (i.e., using the misalignment angle) after shrinkage, we analyzed changes in misalignment angle before and after the shrinkage assembly step. Figure 2d demonstrates such trends; NW misalignment angle showed a clear decrease from 5.4° (no shrinkage) to 0.6° (310% shrinkage). In addition, the full width at half-maximum (fwhm) of misalignment angles showed a sizable change from 13.8° (no shrinkage) to 3.9° (310% shrinkage), further substantiating the assembly quality improvement. Such results are consistent with other investigations39 where strain-release assembly was used to improve the quality of the NW assembly. Furthermore, we investigated the effect of initial misalignment on final NW assembly quality (Figure S4, Supporting Information). Our results showed that NWs with an initial misalignment angle of 30° (Figure S4a, Supporting Informa3305

dx.doi.org/10.1021/nl500709p | Nano Lett. 2014, 14, 3304−3308

Nano Letters

Letter

Figure 3. Investigation of scalability of shrink-induced assembly. Image in the center is the large size polystyrene substrate (2.5 × 6 inches; scale bar, 1 inch). Black dash lines indicate the width of the original polystyrene (2.5 inches). White dash line indicates the outline of shrunken polystyrene substrate (∼300% shrinkage). Red boxes indicate the region where shrink-induced NW assemblies exist. SEM images and Gaussian plots on the sides are related to each region as indicated (scale bars, 1 μm). The x-axis of each plot indicates the density of NW assembly (NWs/μm), and the y-axis indicates count.

the solution after dissolution. The metal film was then transferred to another fresh toluene bath to remove the residual polymer. After rinsing with fresh toluene, the metal film/NW array was transferred on to a bare silicon oxide wafer. SEM analyses were carried out (Figure 4a,b) after the removal

tion) can be reassembled to nearly perfect alignment, which clearly demonstrates the strength of our assembly approach. The scalability of our approach was further investigated by carrying out a shrink-induced NW assembly on a 2.5 × 6 inch size substrate. We contact-printed NW arrays on 6 different locations on a large polystyrene substrate (2.5 × 6 inches) and then carried out the shrinkage step on the substrate (∼300%). By undertaking SEM analysis of NW densities at 6 differing locations of the shrunken substrate, we were able to identify that the average NW density of each region was around 4−5 NWs/μm with a similar distribution from region to region (Figure 3). We note that there is no intrinsic limit on the size of the sample for density amplification, and our results clearly suggest that scalability and uniformity of densification and assembly of NWs are achievable with our method. We explored the transferability of the assembled and densified NW arrays onto various target substrates. Although our approach demonstrates a clear advance in NW density amplification and assembly quality improvement, the polystyrene substrate on which the shrinkage process is carried out is not compatible with conventional microfabrication processes and thus restricts potential applications. As such, we developed a solution transfer approach to detach the NW array from the shrunken polystyrene substrate to transfer it to a silicon wafer with a thermally grown silicon oxide layer (300 nm, NOVA Electronic Materials, LLC). To perform the transfer step, thin films of Cr/Au (5 nm/100 nm) were deposited onto the NW arrays on the shrunken polystyrene substrate. After the metal layer deposition step, the NW sample was immersed in a toluene solution for approximately 1 h. Since polystyrene dissolves in toluene, only the metal film/NW array remains in

Figure 4. Transfer of shrink-induced NW arrays onto various types and dimensionalities of substrates. (a) SEM image of NW array transferred onto a SiO2/Si substrate (scale bar, 1 μm). (b) Gaussian plots of NW density before and after the transfer. (c) Reflective dark field microscope image of shrink-induced NW assembly transferred onto a flexible Kapton film (scale bar, 50 μm). (d) Reflective dark field microscope image of shrink-induced NW assembly transferred onto a 3D glass flask (scale bar, 50 μm). 3306

dx.doi.org/10.1021/nl500709p | Nano Lett. 2014, 14, 3304−3308

Nano Letters

Letter

alignment quality could be precisely controlled to achieve higher reliability. Furthermore, the large scale assembly results ensure that our shrink-induced assembly process is highly scalable and could be potentially adopted by manufacturing industry for developing NW-based electronics and other application areas. Finally, we also demonstrated a simple solution transfer method to interface the shrink-assembled NW array from the polystyrene substrate onto substrates of various types and dimensionalities including Kapton plastics and 3D surfaces. Our approach to assembling and densifying NW arrays via shrinkage will facilitate the realization of NW integrated circuits based on multiple NWs. Furthermore, we envision that the close-packed NW arrays obtained by our reliable and scalable approach can be used to realize assembled mesoscale materials where unique nanoscale surface morphologies of NW close-packed arrays can lead to enhanced optoelectronic functions in the future.14,40

of metal layers using etchants. X-ray photoelectron spectroscopy (XPS) result (Figure S5, Supporting Information) confirms that the amount of Au metal residue after the metal etching process is negligible. The optical and SEM images and analyses of NW density (Figures 4a and S6, Supporting Information) confirm that the arrangement and density of the NW assembly remain intact after the solution transfer step. The density before and after the transfer onto the silicon oxide substrate was 4.4 and 4.2 NWs/ μm, respectively, and the Gaussian distribution shows that the assembly is well preserved during the solution transfer process (Figure 4b). Note that for the ease of the SEM statistical analyses, we intentionally chose lower initial densities of NWs, and thus, the NW density after the shrink-induced assembly was relatively low in both Figures 3 and 4. Furthermore, we carried out the fabrication of NW field-effect transistors (FETs) (Figure S7, Supporting Information) using the transferred NW arrays on SiO2/Si substrates. The results clearly demonstrated that the shrink assembled NW FETs (300% shrinkage) showed at least 2-fold increase in transconductance levels compared with NW FETs without shrink assembly, consistent with the NW density increase confirmed by SEM (Figure S7b,c, Supporting Information). We further investigate the transferability of the shrinkinduced NW arrays by expanding the types and dimensionalities of the target substrates. The capability to transfer the NW arrays onto flexible and 3D structured substrates in addition to conventional Si wafers will enable new opportunities for flexible, 3D nanoelectronics.4,32 We transferred our shrink-induced NW arrays from the polystyrene substrate to Kapton film as well as 3D glass flasks (Figure 4c,d). For both cases, we confirmed by use of dark-field optical microscopy that the NWs were successfully transferred and that their density and alignment were preserved. We believe that our shrink-induced assembly of NWs has several key advantages as compared with earlier works.37−39 First, the large (∼300%) and tunable densification of NWs by shrinkage will allow deterministic control of NW density and, furthermore, offers a reliable method to realize close-packed NW arrays. To the best of our knowledge, such large degrees of densification of NWs have not been demonstrated to date and could provide a unique and reliable approach toward closepacked NW assembly. Second, our shrink-induced assembly/ processing could be easily applied to other one-dimensional nanomaterials (e.g., carbon nanotubes) for densification and close-packed assembly. Third, the scalability of our shrink assembly process will allow large scale integration of NWs with controlled density and could lead to low cost manufacturing. Finally, the low cost and wide commercial availability of shrink polymers used in our shrink assembly offers a technology that is readily adoptable by the manufacturing community and, furthermore, enables a manufacturable pathway toward assembled NW-based applications.2,3 In conclusion, we demonstrate the assembly and densification of NW arrays via shrinkage of a shape memory polymer substrate. Our assembly technique realized close-packing of NW arrays with a high degree of controllability and uniformity. We controlled the shrinkage rates by imposing geometrical constraints on polystyrene substrate and thus achieved tunable densification of our NW arrays. Simultaneously, shrink-induced densification of NWs further enhanced the quality of the assembly. We believe that our assembly approach could enable manufacturable NW assembly processes where the density and

■

ASSOCIATED CONTENT

S Supporting Information *

Methods and supporting figures. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*(S.N.) E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS S.N. acknowledges support from the Air Force Office of Scientific Research/Asian Office of Aerospace Research Development (AFOSR/AOARD) Nano Bio Info Technology (NBIT) Phase III Program (AOARD-13-4125) and by the University of Illinois, Urbana−Champaign (UIUC) Campus Research Board (CRB) Program (RB13191) and UIUC startup funding. W.I.P. acknowledges support by National Research Foundation of Korea (NRF-2013K1A3A1A32035393). The SEM imaging and NW FET device fabrication were carried out in the Frederick Seitz Materials Research Laboratory Central Facilities and the Micro and Nano Technology Laboratory at the University of Illinois at Urbana−Champaign.

■

REFERENCES

(1) Morales, A. M.; Lieber, C. M. Science 1998, 279, 208−211. (2) Lieber, C. M.; Wang, Z. L. MRS Bull. 2007, 32, 99−108. (3) Lu, W.; Lieber, C. M. Nat. Mater. 2007, 6, 841−850. (4) Nam, S.; Jiang, X.; Xiong, Q.; Ham, D.; Lieber, C. M. Proc. Natl. Acad. Sci. U.S.A. 2009, 106, 21035−21038. (5) Takei, K.; Takahashi, T.; Ho, J. C.; Ko, H.; Gillies, A. G.; Leu, P. W.; Fearing, R. S.; Javey, A. Nat. Mater. 2010, 9, 821−826. (6) Yan, H.; Choe, H. S.; Nam, S.; Hu, Y.; Das, S.; Klemic, J. F.; Ellenbogen, J. C.; Lieber, C. M. Nature 2011, 470, 240−244. (7) Yao, J.; Yan, H.; Das, S.; Klemic, J. F.; Ellenbogen, J. C.; Lieber, C. M. Proc. Natl. Acad. Sci. U.S.A. 2014, 111, 2431−2435. (8) Tian, B.; Zheng, X.; Kempa, T. J.; Fang, Y.; Yu, N.; Yu, G.; Huang, J.; Lieber, C. M. Nature 2007, 449, 885−890. (9) Qin, Y.; Wang, X.; Wang, Z. L. Nature 2008, 451, 809−813. (10) Hochbaum, A. I.; Chen, R.; Delgado, R. D.; Liang, W.; Garnett, E. C.; Najarian, M.; Majumdar, A.; Yang, P. Nature 2008, 451, 163− 167. (11) Fan, Z.; Razavi, H.; Do, J.; Moriwaki, A.; Ergen, O.; Chueh, Y.L.; Leu, P. W.; Ho, J. C.; Takahashi, T.; Reichertz, L. A.; Neale, S.; Yu, K.; Wu, M.; Ager, J. W.; Javey, A. Nat. Mater. 2009, 8, 648−653. 3307

dx.doi.org/10.1021/nl500709p | Nano Lett. 2014, 14, 3304−3308

Nano Letters

Letter

(12) Xu, S.; Qin, Y.; Xu, C.; Wei, Y.; Yang, R.; Wang, Z. L. Nat. Nanotechnol. 2010, 5, 366−373. (13) Tang, J.; Huo, Z.; Brittman, S.; Gao, H.; Yang, P. Nat. Nanotechnol. 2011, 6, 568−572. (14) Kim, S.-K.; Day, R. W.; Cahoon, J. F.; Kempa, T. J.; Song, K.-D.; Park, H.-G.; Lieber, C. M. Nano Lett. 2012, 12, 4971−4976. (15) Kempa, T. J.; Day, R. W.; Kim, S.; Park, H.; Lieber, C. M. Energy Environ. Sci. 2013, 6, 719−733. (16) Patolsky, F.; Timko, B. P.; Yu, G.; Fang, Y.; Greytak, A. B.; Zheng, G.; Lieber, C. M. Science 2006, 313, 1100−1104. (17) Kim, W.; Ng, J. K.; Kunitake, M. E.; Conklin, B. R.; Yang, P. J. Am. Chem. Soc. 2007, 129, 7228−7229. (18) Tian, B.; Cohen-Karni, T.; Qing, Q.; Duan, X.; Xie, P.; Lieber, C. M. Science 2010, 329, 830−834. (19) Robinson, J. T.; Jorgolli, M.; Shalek, A. K.; Yoon, M.-H.; Gertner, R. S.; Park, H. Nat. Nanotechnol. 2012, 7, 180−184. (20) Xie, C.; Lin, Z. L.; Hanson, L.; Cui, Y.; Cui, B. X. Nat. Nanotechnol. 2012, 7, 185−190. (21) Tian, B.; Liu, J.; Dvir, T.; Jin, L.; Tsui, J. H.; Qing, Q.; Suo, Z.; Langer, R.; Kohane, D. S.; Lieber, C. M. Nat. Mater. 2012, 11, 986− 994. (22) Duan, X.; Fu, T.-M.; Liu, J.; Lieber, C. M. Nano Today 2013, 8, 351−373. (23) Duan, X.; Lieber, C. M. Adv. Mater. 2000, 12, 298−302. (24) Li, Y.; Qian, F.; Xiang, J.; Lieber, C. M. Mater. Today 2006, 9, 18−27. (25) Gudiksen, M. S.; Lauhon, L. J.; Wang, J.; Smith, D. C.; Lieber, C. M. Nature 2002, 415, 617−620. (26) Lauhon, L. J.; Gudiksen, M. S.; Wang, D.; Lieber, C. M. Nature 2002, 420, 57−61. (27) Bierman, M. J.; Lau, Y. K. A.; Kvit, A. V.; Schmitt, A. L.; Jin, S. Science 2008, 320, 1060−1063. (28) Tian, B.; Xie, P.; Kempa, T. J.; Bell, D. C.; Lieber, C. M. Nat. Nanotechnol. 2009, 4, 824−829. (29) Huang, Y.; Duan, X.; Wei, Q.; Lieber, C. M. Science 2001, 291, 630−633. (30) Yu, G.; Cao, A.; Lieber, C. M. Nat. Nanotechnol. 2007, 2, 372− 377. (31) Whang, D.; Jin, S.; Wu, Y.; Lieber, C. M. Nano Lett. 2003, 3, 1255−1259. (32) Javey, A.; Nam, S.; Friedman, R. S.; Yan, H.; Lieber, C. M. Nano Lett. 2007, 7, 773−777. (33) Fan, Z.; Ho, J. C.; Jacobson, Z. A.; Yerushalmi, R.; Alley, R. L.; Razavi, H.; Javey, A. Nano Lett. 2007, 8, 20−25. (34) Yao, J.; Yan, H.; Lieber, C. M. Nat. Nanotechnol. 2013, 8, 329− 335. (35) Grimes, A.; Breslauer, D. N.; Long, M.; Pegan, J.; Lee, L. P.; Khine, M. Lab Chip 2008, 8, 170−172. (36) Fu, C.-C.; Grimes, A.; Long, M.; Ferri, C. G. L.; Rich, B. D.; Ghosh, S.; Ghosh, S.; Lee, L. P.; Gopinathan, A.; Khine, M. Adv. Mater. 2009, 21, 4472−4476. (37) Ryu, S. Y.; Xiao, J.; Park, W. I.; Son, K. S.; Huang, Y. Y.; Paik, U.; Rogers, J. A. Nano Lett. 2009, 9, 3214−3219. (38) Qin, Q.; Zhu, Y. ACS Nano 2011, 5, 7404−7410. (39) Xu, F.; Durham, J. W.; Wiley, B. J.; Zhu, Y. ACS Nano 2011, 5, 1556−1563. (40) Garnett, E.; Yang, P. Nano Lett. 2010, 10, 1082−1087.

3308

dx.doi.org/10.1021/nl500709p | Nano Lett. 2014, 14, 3304−3308