Article pubs.acs.org/Langmuir
From Nanowires to Nanopores: A Versatile Method for Electroless Deposition of Nanostructures on Micropatterned Organic Substrates Ashley A. Ellsworth† and Amy V. Walker*,†,‡ †
Department of Chemistry and ‡Department of Materials Science and Engineering, University of Texas at Dallas, Richardson, Texas 75802, United States S Supporting Information *
ABSTRACT: We demonstrate a fast, flexible, parallel, and highly controllable method by which to synthesize a variety of nanoscale and mesoscale structures. This method addresses one of the most significant challenges in nanoscience: the in situ parallel placement and synthesis of nano-objects over the mesoscale. The method is based on electroless nanowire deposition on micropatterned substrates (ENDOM). In ENDOM nanostructures are produced at the boundary between two unlike materials if two conditions are met: (a) deposition is kinetically preferred on one of the materials while (b) transport of reactants is favored on the other. In this study, copper structures were deposited on patterned −OH/−CH3-terminated alkanethiolate self-assembled monolayers (SAMs) by exploiting the different reaction rates of electroless deposition on these using the reducing agent dimethylamine borane (DMAB). We demonstrate production of nanowires (width < 100 nm), mesowires (100 nm < width < ∼3000 nm), nanorings, nanopores, and nanochannels. We show that a variety of experimental conditions can be employed, making this method compatible with many substrates. We have also studied the nucleation and growth kinetics of the ENDOM process. The width of the deposit grows exponentially with deposition time and can be modeled using classical nucleation theory. Although the deposit width increases, the height and grain size of the copper deposit is constant (to within experimental uncertainty) with deposition time. These observations indicate that the minimum deposit width is controlled by the nanoparticle dimensions and so can be controlled using the reaction conditions.
1. INTRODUCTION Nanowires, nanopores, nanorings, and nanochannels have technological applications in electronics,1 sensing,2−5 optoelectronics,6 and nonlinear optics.7 These structures are often created by complex processes that are not easily controlled. For example, nanowires can be formed in situ using electrodeposition8 or chemical vapor deposition,9 but these methods can require many different lithographic steps8 and growth can be slow (≤1 μm/min).8 Similarly, available techniques for the preparation of nanopores, nanorings, and nanochannels are complex processes requiring multiple lithographic, deposition, and etching steps.1−7,10 In this paper we demonstrate a single method by which to synthesize in situ nano- and mesostructures in parallel over square-centimeter areas. We show that by understanding the nucleation and growth processes we are able to control the formation of a wide variety of structures with excellent dimensional uniformity. We demonstrate the synthesis and precise placement of nanowires (width < 100 nm), mesowires (100 nm < width < ∼3000 nm), nanorings, nanopores, and nanochannels. Finally we note that this technique can be performed on the laboratory bench, is fast, and does not require the use of complex lithography or high temperatures. Our method is based on our recently introduced technique, electroless nanowire deposition on micropatterned substrates © XXXX American Chemical Society
(ENDOM). We have previously demonstrated ENDOM using the synthesis of nickel,11 copper,12 silver,12 and palladium12 nanowires with widths from 200 to ∼2000 nm on patterned self-assembled monolayers (SAMs). In ENDOM, a nanowire forms at the boundary between two different materials if the following requirements are met.11,12 Deposition must be kinetically preferred on one material due to the adsorption of the reducing agent. Second, transport of reactants must be favored on the surface with the slower deposition rate. The shape of the nanowire is controlled by the underlying pattern of the two dissimilar materials, while the dimensions of the wire are controlled by the deposition time. ENDOM is compatible with many technologically relevant substrates13,14 and lithographic techniques, including microcontact printing,15 dip pen nanolithography,15 electron beam lithography,16 and photolithographies.16 In this paper we investigate the nucleation and growth processes of nanostructures and mesostructures produced using Cu ENDOM. Through a detailed understanding of the nucleation and growth process, we show that ENDOM can be used to synthesize nanowires (widths < 100 nm), nanopores, Received: December 21, 2015 Revised: February 22, 2016
A
DOI: 10.1021/acs.langmuir.5b04674 Langmuir XXXX, XXX, XXX−XXX
Article
Langmuir
Figure 1. Formation of nanostructures using electroless nanowire deposition on micropatterned substrates (ENDOM). (1) A pattern is produced in SAM1 (−OH-terminated SAM) using UV photopatterning. (2) In the photooxidized SAM1 areas, SAM2 (−CH3-terminated SAM) is adsorbed. (3) The sample is then placed in an electroless deposition bath. At the interface between SAM1 and SAM2, nanostructures and mesostructures are electrolessly deposited. Initially nanowires and nanorings form. At longer deposition times, these grow into mesostructures, and eventually nanochannels and nanopores form. Example SEM images of nanowires, mesowires, nanochannels, nanorings, and nanopores are shown.
methyl-terminated SAMs can be accounted for in the following way.11 The BH3 fragment acts as a Lewis acid and so has a slight negative charge (δ−). The C−OH terminal bond of the hydroxyl-terminated SAM is covalent and polar with the −OH group having a small negative charge (δ−). In contrast the C−H bonds of the methyl terminal group are not polar. Thus, BH3 adsorption is preferred on the −CH3-terminated SAM. In agreement with this proposed mechanism, we have demonstrated that the metal deposition begins at slightly shorter times on −CH3-terminated SAMs than on −OH-terminated SAMs using DMAB reducing agents.25 Further, this effect is magnified if the reducing agent is the borohydride ion.25 Initially nanowires (width < 100 nm) are produced at the interface between −OH- and −CH3-terminated SAMs. If an enclosed substrate pattern is employed, nanorings are synthesized. At longer deposition times these nanostructures develop into mesostructures, and eventually nanopores or nanochannels form as the deposit nearly fills the −CH3terminated SAM area (Figure 1).
and nanochannels. A schematic of the ENDOM method is displayed in Figure 1. An image is first created in an −OHterminated SAM (SAM1) using UV photopatterning (step 1). A −CH3-terminated SAM (SAM2) is then adsorbed in the areas where SAM1 has been photooxidized (step 2). Finally the patterned SAM1/SAM2 sample is immersed in a deposition bath (step 3). The deposit is formed using seedless electroless deposition (ELD). ELD processes are REDOX reactions that are used to deposit overlayers including metals,17−19 semiconductors,20,21 and even insulators.22 It is compatible with a range of reducing agents including Lewis acids/bases17 and ionic compounds,17 which allows for great synthetic flexibility. In this study copper is deposited by the reduction of Cu2+ dimethylamine borane (DMAB): 3Cu 2 +(aq) + (CH3)2 NHBH3(aq) + 3H 2O(l) → 3Cu(s) + (CH3)2 NH 2+(aq) + H3BO3(aq) + 5H+(aq)
(1)
2. EXPERIMENTAL SECTION
Initially copper deposition begins at the interface between the −OH- and −CH3-terminated SAMs because DMAB is preferentially adsorbed on the −CH3-terminated SAM while the transport of reactants is favored on the hydrophilic −OHterminated SAM. One of the initial steps in the ELD reaction is the adsorption and decomposition of DMAB:23,24
2.1. Materials. Gold (99.995%), chromium (99.995%), and triethanolamine (98+%) were obtained from Alfa Aesar, Inc. (Ward Hill, MA). Copper(II) sulfate pentahydrate (CuSO4·5H2O, 98+%), ethylenediaminetetraacetic acid (EDTA) (98%), dimethylamine borane complex (97%), hexadecanethiol (HDT) (99+%), and 16hydroxy-1-hexadecanethiol (MHL) (99+%) were acquired from Sigma-Aldrich, Inc. (St. Louis, MO). Concentrated sulfuric acid (95%) was purchased from BDH Aristar, Inc. (Chester, PA). All reactants were used without further purification. Silicon wafers (⟨111⟩ orientation) were acquired from Addison Engineering, Inc. (San Jose, CA) and cleaned using piranha etch (H2SO4/H2O2 = 3:1) prior to use. To produce the patterned hydrophilic/hydrophobic gold surface, the following procedure was employed, based on the method to produce hydrophilic gold substrates described by King.26 A mask was placed atop a freshly evaporated gold surface. The sample was then placed in a UV/ozone cleaner (UV/Ozone ProCleaner, Bioforce
(CH3)2 NHBH3(aq) → (CH3)2 NHBH3(ads) H+
⎯→ ⎯ (CH3)2 NH 2+(aq) + BH3(ads)
(2)
The dimethylamine group ((CH3)2NH) acts as a Lewis base and so reacts with protons in solution to form soluble (CH3)2NH2+ species. The preferential adsorption of BH3 on B
DOI: 10.1021/acs.langmuir.5b04674 Langmuir XXXX, XXX, XXX−XXX
Article
Langmuir Nanosciences) and exposed to UV/ozone for 20 min. In the areas exposed to UV/ozone the gold became hydrophilic, while in the unexposed areas the gold remained hydrophobic. 2.2. Preparation and Photopatterning of Self-Assembled Monolayers. The preparation of self-assembled monolayers (SAMs) is well-known.27,28 In brief, chromium (∼50 Å) and then gold (∼1000 Å) were thermally deposited onto clean Si wafers. A well-ordered, −OH-terminated SAM (MHL) was then formed by immersing the gold substrate into a 1 mM ethanolic solution of MHL for 24 h at ambient temperature (21 ± 2 °C). After SAM formation the substrate was then removed from the solution, washed with ethanol, and then dried under N2 gas. The MHL SAM was then UV photopatterned using the procedure described in refs 29 and 30. A mask (copper transmission electron microscopy (TEM) grid of the appropriate pattern; Electron Microscopy, Inc., Hatfield, PA) was placed on top of the MHL SAM. The sample was then placed ∼50 mm from a 500 W Hg arc lamp equipped with a narrow band-pass UV filter (280−400 nm) and a dichroic mirror (Thermal Oriel, Spectra Physics, Inc., Stratford, CT). To ensure that photooxidation was complete, the MHL SAM was then exposed to the UV light for 3 h. After photopatterning the SAM was rinsed with ethanol and then placed in a 1 mM ethanolic solution of HDT for 24 h at room temperature. In the areas exposed to UV light, the photooxidized MHL was displaced by a HDT, creating a patterned −OH/−CH3-patterned SAM surface. The samples were then washed with ethanol, dried with N2 gas, and used immediately for deposition. For each batch, one SAM sample was taken and characterized using single-wavelength ellipsometry (Gaertner Scientific Corp., Skokie, IL) and time-of-flight secondary ion mass spectrometry (TOF SIMS) to ensure that the SAMs were free of significant chemical contamination. Similarly one photopatterned −OH/−CH3 SAM surface was characterized using TOF SIMS using the methods in refs 29 and 30. 2.3. Copper Seedless Electroless Deposition. The standard copper electroless deposition solution (“100 % Cu concentration”) was composed of 0.032 M copper(II) sulfate pentahydrate, 0.24 M triethanolamine (TEA), 0.037 M EDTA, and 0.067 M dimethylamine borane (DMAB, (CH3)2NHBH3). To investigate the growth of the nanowires, the concentrations of copper(II) sulfate pentahydrate and DMAB were altered together so that the ratio between the two remained constant, and they are reported as a percentage of the standard concentration. For all depositions, the TEA and EDTA concentrations were kept constant. The pH of the deposition bath was adjusted to 9 before addition of DMAB. The deposition temperature was 22 ± 1 °C. After deposition each substrate was rinsed with DI water and ethanol. The resulting constructs were examined using TOF SIMS, scanning electron microscopy (SEM), and atomic force microscopy (AFM). 2.4. Time-of-Flight Secondary Ion Mass Spectrometry. TOF SIMS measurements were performed using an ION TOF IV spectrometer (ION TOF, Inc., Chestnut Hill, NY) which has a Bi liquid metal ion gun (LMIG) primary ion source. The instrument has three chambers for sample introduction, preparation, and analysis. The pressure of the analysis and preparation chambers is typically = − The reported dimensions of the wires are the average of these measurements (at least 50 data points), and the reported error is the standard deviation.
3. RESULTS AND DISCUSSION 3.1. Nanowire and Mesowire Formation. Isolated nanoparticles initially form at the boundary between the −OH and −CH3-terminated SAMs (Figure 2). As deposition
Figure 2. AFM images of nanowires formation with deposition time. Reaction conditions: 0.032 M copper(II) sulfate pentahydrate, 0.24 M TEA, 0.037 M EDTA, and 0.067 DMAB; deposition temperature 22 ± 1 °C. Image size: (20 × 20) μm2.
continues, the number of nanoparticles increases until a continuous polycrystalline nanowire forms (Figure 2). The nanowire grows in width and at later deposition times becomes a mesowire (Figure 1). These observations suggest that wire formation begins with heterogeneous nucleation of nanoparticles at the interface between the −CH3- and −OHterminated SAM areas. Time-of-flight secondary ion mass spectra confirm this hypothesis. When the nanowires initially form, cluster ions of the form [Cu2(MHL) (HDT)OH]− (MHL = −S(CH2)15CH2OH; HDT = −S(CH2)15CH3) are observed (Figure 3). Because in SIMS cluster ions form from species in close proximity to each other (≤6 Å apart), this cluster ion indicates that the copper is deposited at the −OH/−CH3-terminated SAM interface. Subsequently, additional nanoparticles appear to nucleate at the interface between the −CH3-terminated SAM and the growing Cu deposit due to the preferential adsorption of DMAB on the methyl-terminated SAM. As the deposition continues, eventually a continuous polycrystalline wire is formed (Figure 2). The wire growth continues into the −CH3-terminated SAM area until eventually copper covers the whole area. C
DOI: 10.1021/acs.langmuir.5b04674 Langmuir XXXX, XXX, XXX−XXX
Article
Langmuir
Figure 3. High-resolution negative ion spectra from m/z 660 to m/z 685. Reaction conditions: 0.032 M copper(II) sulfate pentahydrate, 0.24 M TEA, 0.037 M EDTA, and 0.067 M DMAB; deposition temperature 22 ± 1 °C; deposition time 10 min.
We have demonstrated nanowire and mesowire formation over a variety of reaction conditions including concentrations of copper(II) sulfate ranging from 0.016 to 0.032 M and DMAB ranging from 0.17 to 0.67 M, bath pH from 6 to 11, and bath temperatures from room temperature up to 50 °C (see Supporting Information, Figure S1). The upper temperature is limited only by the stability of the substrate. Above room temperature the density of defects within the SAM substantially increases.31,32 Thus, to prevent the formation of a very defective substrate film, only temperatures up to 50 °C were employed. We also note that an ENDOM-like process may occur between hydrophilic and hydrophobic surfaces without the presence of SAMs. (The formation of mesowires has also been observed at the junction of hydrophobic and hydrophilic gold areas (Figure 4).) However, these wires are not as uniform as those formed using a patterned SAM substrate, and it is unclear that nanowires (width < 100 nm) can be produced on such surfaces.
Figure 5. (a) Variation of wire width with reaction time at room temperature, 22 ± 1 °C. The dotted lines are the best fits to the data, which were fit using a nonlinear regression to an exponential. (b) Plot of ln w (ln(width)) versus deposition time. The dotted lines show the fit to the data with slope B = 0.85 ± 0.08 nm/min. Reaction conditions: the composition of the “100 % concentration” bath was 0.032 M copper(II) sulfate pentahydrate, 0.24 M TEA, 0.037 M EDTA, and 0.067 M DMAB. The concentrations of copper(II) sulfate and DMAB were altered together so that the ratio between the two remained constant, and they are reported as a percentage of the standard concentration (“100 % concentration”). For all depositions, the TEA and EDTA concentrations were kept constant.
5b). We note that the standard deviations in the wire widths are 15−20%, which are comparable to other methods to make nanowires8,34−36 (Figure 5a). It is expected that this variation can be significantly reduced by using commercially available deposition solutions34 and better lithography. However, the induction time decreases with increasing Cu2+ (and DMAB) concentration (Figures 5 and 6). Here the induction time is defined as the time at which complete nanowires are first observed. For precipitation from a supersaturated solution, classical nucleation theory predicts that the nucleation rate is given by37
Figure 4. SEM images of a copper mesowire formed at the interface of hydrophobic and hydrophilic gold.
3.2. Nucleation and Growth Process. As expected for an electroless deposition reaction,33 the wire width increases exponentially with reaction time (Figure 5a): w = A exp(Bt )
⎛ 16πγ 3Vm 2 ⎞ ⎟ K = K 0 exp⎜ − 3 2 ⎝ 3(kT ) (ln S) ⎠
(3)
where S = C/C* is the supersaturation constant, C is the Cu2+ concentration, C* is the critical concentration, Vm is the molecular volume, k is the Boltzmann constant, T is the temperature, and γ is the surface free energy. The critical (saturation) concentration of metallic copper in aqueous solution is not well-known; a value of C* = 1.58 × 10−8 ppb at room temperature is used, here taken from Palmer and Petrov.38 Because the induction time, tind, is inversely proportional to the nucleation rate,37
where w is the deposit width in nanometers, A is a constant, t is the deposition time, and B is the rate constant. At constant temperature, ln w = ln A + Bt
(5)
(4)
i.e., a plot of ln w versus t should give a straight line with slope B. At constant [Cu2+]/[DMAB] and temperature, for different Cu2+ concentrations the growth rate is the same (to within experimental certainty) with B = 0.85 ± 0.08 nm/min (Figure D
DOI: 10.1021/acs.langmuir.5b04674 Langmuir XXXX, XXX, XXX−XXX
Article
Langmuir
Figure 6. (a) Variation of induction time with Cu2+ concentration; (b) ln tind versus 1/(ln S)2 where S is the degree of supersaturation. (b) The dotted line is the best fit to the data and was a straight line. The goodness of fit, r2, is 0.998. The composition of the “100 % concentration” bath was 0.032 M copper(II) sulfate pentahydrate, 0.24 M TEA, 0.037 M EDTA, and 0.067 M DMAB. The concentrations of copper(II) sulfate and DMAB were altered together so that the ratio between the two remained constant, and they are reported as a percentage of the standard concentration (“100 % concentration”). For all depositions, the TEA and EDTA concentrations were kept constant.
⎛ 16πγ 3V 2 ⎞ m ⎟ t ind = A exp⎜ 3 3( kT ) (ln S)2 ⎠ ⎝
Figure 7. (a) Variation of wire height with wire width with different Cu2+ concentrations. (b) Variation of grain size with wire width at 100% Cu2+ concentrations. The standard deviation in the wire height and grain size is 20−24%. The composition of the “100 % concentration” bath was 0.032 M copper(II) sulfate pentahydrate, 0.24 M TEA, 0.037 M EDTA, and 0.067 M DMAB. The concentrations of copper(II) sulfate and DMAB were altered together so that the ratio between the two remained constant, and they are reported as a percentage of the standard concentration (“100 % concentration”). For all depositions, the TEA and EDTA concentrations were kept constant. Please note the scale of the nanowire height and grain size axes.
on the −CH3-terminated SAM. Thus, the interface moves into the −CH3-terminated areas and the wire grows in width but not height. The grain size of the copper also does not change (to within experimental uncertainty) with deposition time (Figure 7b). Interestingly, the grain size is ∼2× larger than the measured wire height, suggesting that the deposits are composed of nanoparticles that are wider than they are tall. These observations indicate that the minimum nanowire width is determined by the nanoparticle dimensions. From classical nucleation theory, if the concentration of copper is increased, the activation barrier to nucleation is reduced, resulting in the production of a larger number of smaller nuclei.39 Hence, narrower nanowires should form if the rate of copper production is increased. To test this hypothesis, the [DMAB] to [Cu2+] ratio was doubled, which should increase the reaction rate. The nanoparticle diameter was observed to decrease by ∼30%, and the minimum nanowire width also decreased by ∼30% from 97 ± 10 to 71 ± 7 nm. We note that similar strategies have also been employed in the synthesis of ultrasmall nanoparticles.40 3.3. Formation of Nanopores and Nanochannels. Because the wires nucleate at the boundary between the −OH and −CH3-terminated SAMs, their shape is controlled by the underlying substrate pattern. This has multiple important consequences for the formation of wires. The wires can have complex shapes such as arcs and right-angle bends (see
(6)
at constant temperature, ⎛ 16πγ 3V 2 ⎞ B m ⎟ = ln A + ln t ind = ln A + ⎜ 3 2 (ln S)2 ⎝ 3(kT ) (ln S) ⎠
where B =
16πγ 3Vm 2 3(kT )3
(7)
, i.e., a plot of ln tind versus 1/(ln S)2 should
give a straight line with gradient B. The induction time was determined from Figure 5a by finding the time at which a 100nm-wide nanowire was initially observed (Figure 6a). The plot of ln tind versus 1/(ln S)2 is quite linear (Figure 6b), indicating that classical nucleation theory can be employed to predict the induction time for deposition of the wires. Because for different Cu2+ concentrations the growth rate is identical (to within experimental uncertainty) and the induction time is known, the deposit width can be precisely controlled from the nanometer to micron scale by controlling the deposition time using a stopwatch. Although the wire width increases exponentially with deposition time, the wire height remains constant to within experimental uncerta1inty (Figure 7a). This is because as the polycrystalline wire grows, additional copper ions are reduced on the surface at the interface between the copper and −CH3terminated SAM due to the preferential adsorption of DMAB E
DOI: 10.1021/acs.langmuir.5b04674 Langmuir XXXX, XXX, XXX−XXX
Article
Langmuir
formation of copper nanoscale and mesoscale structures to demonstrate and study the technique. In ENDOM, initially isolated nanoparticles heterogeneously nucleate at the boundary between −CH3- and −OH-terminated SAMs. The growing deposit eventually forms a polycrystalline wire. Because the shape of the deposit is determined by the underlying substrate pattern, nanorings are formed if the substrate pattern is a series of enclosed shapes. As the deposition continues, the copper deposits at the Cu/−CH3terminated SAM interface eventually fill the −CH3-terminated SAM area, leaving a nanochannel or nanopore. The width of the deposit grows exponentially with deposition time, which is expected for an ELD process. At constant [Cu2+]/[DMAB], the deposition rate is the same (to within experimental uncertainty) for all Cu2+ concentrations studied. In contrast, the induction time for deposition decreases with increasing Cu2+ concentration. Further, classical nucleation theory can be used to predict the induction time for the deposition of nanowires. Because the reaction rate and induction time are known, this allows the deposit width to be precisely controlled using the deposition time. In contrast to the deposit width, the height and grain size of the copper deposit is constant (to within experimental uncertainty). These observations indicate that the minimum deposit width is controlled by the nanoparticle dimensions, which from classical nucleation theory is controlled by the reaction rate and thus the reagent concentrations. By doubling the [DMAB] to [Cu2+] ratio, the reaction rate increased. Further, the nanoparticle diameter decreased by ∼30%, and the minimum nanowire width also reduced by ∼30% from 97 ± 10 to 71 ± 7 nm. Finally we note that ENDOM is a very flexible method. It is compatible with a wide range of lithographies and technologically relevant substrates. Furthermore, we have observed an ENDOM-like process at the interface of hydrophilic and hydrophobic gold surfaces, suggesting that a patterned SAM substrate may not even be required. This method therefore enables the development of new and cheaper nanotechnological devices with possible applications in nanoelectronics, nanofluidics, nanobiotechnology, and sensing.
Supporting Information, Figures S2 and S3). Second, the process is parallel, so many nanostructures can be formed simultaneously over large areas (square-centimeters) (see Supporting Information, Figure S3). However, perhaps the most important consequence is that ENDOM can be employed to create a variety of different structures using the same basic approach. By changing the substrate pattern and controlling the deposition time, nanorings, nanochannels, and nanopores can be produced. At early deposition times, nanorings are formed if the substrate pattern is a series of enclosed shapes (Figures 1 and 8a).
Figure 8. SEM images of the formation of a nanopore from a nanoring. Initially a nanoring is formed (a). As the deposition time increases, the deposit thickens in the direction of the white arrows (b). At longer times the deposit nearly covers the −CH3-terminated SAM (hexagonal area), forming a pore in the center (c). Finally, after continued deposition a nanopore is formed in the center of the ring (d). Reaction conditions: 0.0304 M copper(II) sulfate pentahydrate, 0.24 M TEA, 0.037 M EDTA, and 0.0637 M DMAB; deposition temperature 45 ± 2 °C. Note that the scale bars vary between images.
■
ASSOCIATED CONTENT
S Supporting Information *
Continued copper deposition leads to the thickening of the ring as the deposit fills the hydrophobic −CH3-terminated SAM (hexagonal) area (Figure 8b). At longer times the deposit nearly covers the −CH3-terminated SAM forming a micronsized pore in the center (Figure 8c). Finally, at later times after continued deposition, a nanopore is formed in the center of the ring (Figures 1 and 8d). We note that, at later deposition times, the copper layer becomes thicker and rougher as deposition occurs atop the copper area. However, the grain size remains approximately constant. (In Figure 8 the grain size is different to that shown in Figure 7 because different deposition conditions were used.) Similarly if a parallel line pattern is employed, nanowires form first (Figure 1). As the deposition continues, mesowires form and then nanochannels as the deposit widens and almost fills the −CH3-terminated SAM areas.
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.langmuir.5b04674. SEM images of wires formed under different experimental conditions; large-area SEM and optical images of nano- and mesostructures produced using the ENDOM process; fits of wire width to exponential growth rate (PDF)
■
AUTHOR INFORMATION
Corresponding Author
*E-mail: [email protected]. Phone: +1 972 883 5780. Fax: +1 972 883 5725. Notes
The authors declare no competing financial interest.
■
4. CONCLUSIONS In summary we have developed and investigated a simple, fast, and versatile method by which to deposit a wide range of nanostructures in parallel over the mesoscale. We use the
ACKNOWLEDGMENTS The authors gratefully acknowledge support from the National Science Foundation (CHE1213546 and DMR1156423). F
DOI: 10.1021/acs.langmuir.5b04674 Langmuir XXXX, XXX, XXX−XXX
Article
Langmuir Brianna L. Dermody (B.L.D.) obtained the “110% concentration” data shown in Figure 2. B.L.D. was supported by a Research Experience for Undergraduates program funded by the National Science Foundation (DMR1156423).
■
Media Containing Zinc Nitrate and Dimethylamineborane. J. Electrochem. Soc. 2006, 153, C735−C740. (22) Chigane, M.; Izaki, M.; Hatanaka, Y.; Shinagawa, T.; Ishikawa, M. Effect of dimethylamine borane concentration on antireflection properties of silica thin films via redox deposition. Thin Solid Films 2006, 515, 2513−2518. (23) Lelental, M. Dimethylamine Borane as the Reducing Agent in Electroless Plating Systems. J. Electrochem. Soc. 1973, 120, 1650−1654. (24) Lelental, M. Effect of amine borane structure on activity in electroless plating. J. Catal. 1974, 32, 429−433. (25) Shi, Z.; Walker, A. V. Investigation of the Mechanism of Nickel Electroless Deposition on Functionalized Self-Assembled Monolayers. Langmuir 2011, 27, 6932−6939. (26) King, D. E. Oxidation of gold by ultraviolet light and ozone at 25°C. J. Vac. Sci. Technol., A 1995, 13, 1247−1253. (27) Nuzzo, R. G.; Dubois, L. H.; Allara, D. L. Fundamental Studies of Microscopic Wetting on Organic Surfaces. 1. Formation and Structural Characterization of a Self-Consistent Series of Polyfunctional Organic Monolayers. J. Am. Chem. Soc. 1990, 112, 558−569. (28) Fisher, G. L.; Walker, A. V.; Hooper, A. E.; Tighe, T. B.; Bahnck, K. B.; Skriba, H. T.; Reinard, M. D.; Haynie, B. C.; Opila, R. L.; Winograd, N.; Allara, D. L. Bond Insertion, Complexation and Penetration Pathways of Vapor-Deposited Aluminum Atoms with HO- and CH3O- Terminated Organic Monolayers. J. Am. Chem. Soc. 2002, 124, 5528−5541. (29) Zhou, C.; Walker, A. V. Dependence of Patterned Binary Alkanethiolate Self-Assembled Monolayers on “UV-Photopatterning” Conditions and Evolution with Time, Terminal Group, and Methylene Chain Length. Langmuir 2006, 22, 11420−11425. (30) Zhou, C.; Walker, A. V. UV Photooxidation and Photopatterning of Alkanethiolate Self-Assembled Monolayers (SAMs) on GaAs (001). Langmuir 2007, 23, 8876−8881. (31) Yamada, Y.; Wano, H.; Uosaki, K. Effect of Temperature on Structure of the Self-Assembled Monolayer of Decanethiol on Au(111) Surface. Langmuir 2000, 16, 5523−5525. (32) Chidsey, C. E. D.; Liu, G.-Y.; Rowntree, P.; Scoles, G. Molecular Order at the Surface of an Organic Monolayer Studied by Low Energy Helium Diffraction. J. Chem. Phys. 1989, 91, 4421−4423. (33) Mallory, G. O.; Hajdu, J. B. Electroless plating: Fundamentals and applications; American Electroplaters and Surface Finishers Society: 1990. (34) Kung, S. C.; Xing, W.; Donavan, K. C.; Yang, F.; Penner, R. M. Photolithographically patterned silver nanowire electrodeposition. Electrochim. Acta 2010, 55, 8074−8080. (35) Xiang, C.; Thompson, M. A.; Yang, F.; Menke, E. J.; Yang, L.-M. C.; Penner, R. M. Lithographically patterned nanowire electrodeposition. Phys. Phys. Status Solidi C 2008, 5, 3503−3505. (36) Koenigsmann, C.; Santulli, A. C.; Sutter, E.; Wong, S. S. Ambient Surfactantless Synthesis, Growth Mechanism, and SizeDependent Electrocatalytic Behavior of High-Quality, Single Crystalline Palladium Nanowires. ACS Nano 2011, 5, 7471−7487. (37) Mullin, J. W. Crystallization, 3rd ed.; Butterworth-Heinemann: Oxford, U.K., 1993. (38) Palmer, D.; Petrov, A. Evaluation of Models for Solubility and Volatility of Copper Compounds Under Steam Generation Conditions. In International Conference on the Properties of Water and Steam; Tremaine, P. R., Hill, P. G., Irish, D. E., Balakrishnan, P. V., Eds.; National Research Council of Canada: Toronto, CA, 1999; pp 821−827. (39) Turkevich, J. Colloidal Gold. Part I: Historical and Preparative Aspects, Morphology and Structure. Gold Bulletin 1985, 18, 86−91. (40) Kim, B. H.; Hackett, M. J.; Park, J.; Hyeon, T. Synthesis, Characterization, and Application of Ultrasmall Nanoparticles. Chem. Mater. 2014, 26, 59−71.
REFERENCES
(1) Chen, J.; Reed, M. A.; Rawlett, A. M.; Tour, J. M. Large On-Off Ratios and Negative Differential Resistance in a Molecular Electronic Device. Science 1999, 286, 1550−1552. (2) Miles, B. N.; Ivanov, A. P.; Wilson, K. A.; Doğan, F.; Japrung, D.; Edel, J. B. Single Molecule Sensing with Solid-State Nanopores: Novel Materials, Methods, and Applications. Chem. Soc. Rev. 2013, 42, 15− 28. (3) Dekker, C. Solid-State Nanopores. Nat. Nanotechnol. 2007, 2, 209−215. (4) Hou, X.; Guo, W.; Jiang, L. Biomimetic Smart Nanopores and Nanochannels. Chem. Soc. Rev. 2011, 40, 2385−2401. (5) Reisner, W.; Pedersen, J. N.; Austin, R. H. DNA confinement in nanochannels: Physics and biological applications. Rep. Prog. Phys. 2012, 75, 106601. (6) Halpern, A. R.; Corn, R. M. Lithographically Patterned Electrodeposition of Gold, Silver, and Nickel Nanoring Arrays with Widely Tunable Near-Infrared Plasmonic Resonances. ACS Nano 2013, 7, 1755−1762. (7) Junesch, J.; Sannomiya, T. Ultrathin Suspended Nanopores with Surface Plasmon Resonance Fabricated by Combined Colloidal Lithography and Film Transfer. ACS Appl. Mater. Interfaces 2014, 6, 6322−6331. (8) Menke, E. J.; Thompson, M. A.; Xiang, C.; Yang, L. C.; Penner, R. M. Lithographically patterned nanowire electrodeposition. Nat. Mater. 2006, 5, 914−919. (9) Decker, C. A.; Solanki, R.; Freeouf, J. L.; Carruthers, J. R.; Evans, D. R. Directed growth of nickel silicide nanowires. Appl. Phys. Lett. 2004, 84, 1389−1391. (10) Whitney, A. V.; Myers, B. D.; Van Duyne, R. P. Sub-100 nm Triangular Nanopores Fabricated with the Reactive Ion Etching Variant of Nanosphere Lithograpy and Angle-Resolved Nanosphere Lithography. Nano Lett. 2004, 4, 1507−1511. (11) Shi, Z.; Walker, A. V. Synthesis of Nickel Nanowires via Electroless Nanowire Deposition on Micropatterned Substrates. Langmuir 2011, 27, 11292−11295. (12) Ellsworth, A. A.; Borner, K.; Yang, J.; Walker, A. V. Towards Molecular Electronics: Using Solution-Based Methods to Deposit Nanowires. ECS Trans. 2014, 58, 1−10. (13) Chaudhury, M. K. Self-assembled monolayers on polymer surfaces. Biosens. Bioelectron. 1995, 10, 785−788. (14) Ulman, A. Formation and Structure of Self-Assembled Monolayers. Chem. Rev. 1996, 96, 1533−1554. (15) Smith, R. K.; Lewis, P. A.; Weiss, P. S. Patterning SelfAssembled Monolayers. Prog. Surf. Sci. 2004, 75, 1−68. (16) Zhou, C.; Trionfi, A.; Jones, J. C.; Hsu, J. W. P.; Walker, A. V. Comparison of Chemical Lithography Using Alkanethiolate SelfAssembled Monolayers on GaAs (001) and Au. Langmuir 2010, 26, 4523−4528. (17) Watanabe, T. Film Formation Mechanism in Electrodeposition. In Nano PlatingMicrostructure Formation Theory of Plated Films and a Database of Plated Films; Elsevier: Oxford, U.K., 2004; pp 95−139. (18) Schmeckenbecher, A. F.; Lindholm, J. A. Potentiometric determination of dimethylamine borane and hypophosphite. Anal. Chem. 1967, 39, 1014−1016. (19) Yokoshima, T.; Kaneko, D.; Akahori, M.; Nam, H.-S.; Osaka, T. Electroless CoNiFeB soft magnetic thin films with high corrosion resistance. J. Electroanal. Chem. 2000, 491, 197−202. (20) Izaki, M.; Omi, T. Transparent Zinc Oxide Films Chemically Prepared from Aqueous Solution. J. Electrochem. Soc. 1997, 144, L3− L5. (21) Murase, K.; Tada, H.; Shinagawa, T.; Izaki, M.; Awakura, Y. QCM Studies of Chemical Solution Deposition of ZnO in Aqueous G
DOI: 10.1021/acs.langmuir.5b04674 Langmuir XXXX, XXX, XXX−XXX
From Nanowires to Nanopores: A Versatile Method for Electroless Deposition of Nanostructures on Micropatterned Organic Substrates Ashley A. Ellsworth† and Amy V. Walker*,†,‡ †
Department of Chemistry and ‡Department of Materials Science and Engineering, University of Texas at Dallas, Richardson, Texas 75802, United States S Supporting Information *
ABSTRACT: We demonstrate a fast, flexible, parallel, and highly controllable method by which to synthesize a variety of nanoscale and mesoscale structures. This method addresses one of the most significant challenges in nanoscience: the in situ parallel placement and synthesis of nano-objects over the mesoscale. The method is based on electroless nanowire deposition on micropatterned substrates (ENDOM). In ENDOM nanostructures are produced at the boundary between two unlike materials if two conditions are met: (a) deposition is kinetically preferred on one of the materials while (b) transport of reactants is favored on the other. In this study, copper structures were deposited on patterned −OH/−CH3-terminated alkanethiolate self-assembled monolayers (SAMs) by exploiting the different reaction rates of electroless deposition on these using the reducing agent dimethylamine borane (DMAB). We demonstrate production of nanowires (width < 100 nm), mesowires (100 nm < width < ∼3000 nm), nanorings, nanopores, and nanochannels. We show that a variety of experimental conditions can be employed, making this method compatible with many substrates. We have also studied the nucleation and growth kinetics of the ENDOM process. The width of the deposit grows exponentially with deposition time and can be modeled using classical nucleation theory. Although the deposit width increases, the height and grain size of the copper deposit is constant (to within experimental uncertainty) with deposition time. These observations indicate that the minimum deposit width is controlled by the nanoparticle dimensions and so can be controlled using the reaction conditions.
1. INTRODUCTION Nanowires, nanopores, nanorings, and nanochannels have technological applications in electronics,1 sensing,2−5 optoelectronics,6 and nonlinear optics.7 These structures are often created by complex processes that are not easily controlled. For example, nanowires can be formed in situ using electrodeposition8 or chemical vapor deposition,9 but these methods can require many different lithographic steps8 and growth can be slow (≤1 μm/min).8 Similarly, available techniques for the preparation of nanopores, nanorings, and nanochannels are complex processes requiring multiple lithographic, deposition, and etching steps.1−7,10 In this paper we demonstrate a single method by which to synthesize in situ nano- and mesostructures in parallel over square-centimeter areas. We show that by understanding the nucleation and growth processes we are able to control the formation of a wide variety of structures with excellent dimensional uniformity. We demonstrate the synthesis and precise placement of nanowires (width < 100 nm), mesowires (100 nm < width < ∼3000 nm), nanorings, nanopores, and nanochannels. Finally we note that this technique can be performed on the laboratory bench, is fast, and does not require the use of complex lithography or high temperatures. Our method is based on our recently introduced technique, electroless nanowire deposition on micropatterned substrates © XXXX American Chemical Society
(ENDOM). We have previously demonstrated ENDOM using the synthesis of nickel,11 copper,12 silver,12 and palladium12 nanowires with widths from 200 to ∼2000 nm on patterned self-assembled monolayers (SAMs). In ENDOM, a nanowire forms at the boundary between two different materials if the following requirements are met.11,12 Deposition must be kinetically preferred on one material due to the adsorption of the reducing agent. Second, transport of reactants must be favored on the surface with the slower deposition rate. The shape of the nanowire is controlled by the underlying pattern of the two dissimilar materials, while the dimensions of the wire are controlled by the deposition time. ENDOM is compatible with many technologically relevant substrates13,14 and lithographic techniques, including microcontact printing,15 dip pen nanolithography,15 electron beam lithography,16 and photolithographies.16 In this paper we investigate the nucleation and growth processes of nanostructures and mesostructures produced using Cu ENDOM. Through a detailed understanding of the nucleation and growth process, we show that ENDOM can be used to synthesize nanowires (widths < 100 nm), nanopores, Received: December 21, 2015 Revised: February 22, 2016
A
DOI: 10.1021/acs.langmuir.5b04674 Langmuir XXXX, XXX, XXX−XXX
Article
Langmuir
Figure 1. Formation of nanostructures using electroless nanowire deposition on micropatterned substrates (ENDOM). (1) A pattern is produced in SAM1 (−OH-terminated SAM) using UV photopatterning. (2) In the photooxidized SAM1 areas, SAM2 (−CH3-terminated SAM) is adsorbed. (3) The sample is then placed in an electroless deposition bath. At the interface between SAM1 and SAM2, nanostructures and mesostructures are electrolessly deposited. Initially nanowires and nanorings form. At longer deposition times, these grow into mesostructures, and eventually nanochannels and nanopores form. Example SEM images of nanowires, mesowires, nanochannels, nanorings, and nanopores are shown.
methyl-terminated SAMs can be accounted for in the following way.11 The BH3 fragment acts as a Lewis acid and so has a slight negative charge (δ−). The C−OH terminal bond of the hydroxyl-terminated SAM is covalent and polar with the −OH group having a small negative charge (δ−). In contrast the C−H bonds of the methyl terminal group are not polar. Thus, BH3 adsorption is preferred on the −CH3-terminated SAM. In agreement with this proposed mechanism, we have demonstrated that the metal deposition begins at slightly shorter times on −CH3-terminated SAMs than on −OH-terminated SAMs using DMAB reducing agents.25 Further, this effect is magnified if the reducing agent is the borohydride ion.25 Initially nanowires (width < 100 nm) are produced at the interface between −OH- and −CH3-terminated SAMs. If an enclosed substrate pattern is employed, nanorings are synthesized. At longer deposition times these nanostructures develop into mesostructures, and eventually nanopores or nanochannels form as the deposit nearly fills the −CH3terminated SAM area (Figure 1).
and nanochannels. A schematic of the ENDOM method is displayed in Figure 1. An image is first created in an −OHterminated SAM (SAM1) using UV photopatterning (step 1). A −CH3-terminated SAM (SAM2) is then adsorbed in the areas where SAM1 has been photooxidized (step 2). Finally the patterned SAM1/SAM2 sample is immersed in a deposition bath (step 3). The deposit is formed using seedless electroless deposition (ELD). ELD processes are REDOX reactions that are used to deposit overlayers including metals,17−19 semiconductors,20,21 and even insulators.22 It is compatible with a range of reducing agents including Lewis acids/bases17 and ionic compounds,17 which allows for great synthetic flexibility. In this study copper is deposited by the reduction of Cu2+ dimethylamine borane (DMAB): 3Cu 2 +(aq) + (CH3)2 NHBH3(aq) + 3H 2O(l) → 3Cu(s) + (CH3)2 NH 2+(aq) + H3BO3(aq) + 5H+(aq)
(1)
2. EXPERIMENTAL SECTION
Initially copper deposition begins at the interface between the −OH- and −CH3-terminated SAMs because DMAB is preferentially adsorbed on the −CH3-terminated SAM while the transport of reactants is favored on the hydrophilic −OHterminated SAM. One of the initial steps in the ELD reaction is the adsorption and decomposition of DMAB:23,24
2.1. Materials. Gold (99.995%), chromium (99.995%), and triethanolamine (98+%) were obtained from Alfa Aesar, Inc. (Ward Hill, MA). Copper(II) sulfate pentahydrate (CuSO4·5H2O, 98+%), ethylenediaminetetraacetic acid (EDTA) (98%), dimethylamine borane complex (97%), hexadecanethiol (HDT) (99+%), and 16hydroxy-1-hexadecanethiol (MHL) (99+%) were acquired from Sigma-Aldrich, Inc. (St. Louis, MO). Concentrated sulfuric acid (95%) was purchased from BDH Aristar, Inc. (Chester, PA). All reactants were used without further purification. Silicon wafers (⟨111⟩ orientation) were acquired from Addison Engineering, Inc. (San Jose, CA) and cleaned using piranha etch (H2SO4/H2O2 = 3:1) prior to use. To produce the patterned hydrophilic/hydrophobic gold surface, the following procedure was employed, based on the method to produce hydrophilic gold substrates described by King.26 A mask was placed atop a freshly evaporated gold surface. The sample was then placed in a UV/ozone cleaner (UV/Ozone ProCleaner, Bioforce
(CH3)2 NHBH3(aq) → (CH3)2 NHBH3(ads) H+
⎯→ ⎯ (CH3)2 NH 2+(aq) + BH3(ads)
(2)
The dimethylamine group ((CH3)2NH) acts as a Lewis base and so reacts with protons in solution to form soluble (CH3)2NH2+ species. The preferential adsorption of BH3 on B
DOI: 10.1021/acs.langmuir.5b04674 Langmuir XXXX, XXX, XXX−XXX
Article
Langmuir Nanosciences) and exposed to UV/ozone for 20 min. In the areas exposed to UV/ozone the gold became hydrophilic, while in the unexposed areas the gold remained hydrophobic. 2.2. Preparation and Photopatterning of Self-Assembled Monolayers. The preparation of self-assembled monolayers (SAMs) is well-known.27,28 In brief, chromium (∼50 Å) and then gold (∼1000 Å) were thermally deposited onto clean Si wafers. A well-ordered, −OH-terminated SAM (MHL) was then formed by immersing the gold substrate into a 1 mM ethanolic solution of MHL for 24 h at ambient temperature (21 ± 2 °C). After SAM formation the substrate was then removed from the solution, washed with ethanol, and then dried under N2 gas. The MHL SAM was then UV photopatterned using the procedure described in refs 29 and 30. A mask (copper transmission electron microscopy (TEM) grid of the appropriate pattern; Electron Microscopy, Inc., Hatfield, PA) was placed on top of the MHL SAM. The sample was then placed ∼50 mm from a 500 W Hg arc lamp equipped with a narrow band-pass UV filter (280−400 nm) and a dichroic mirror (Thermal Oriel, Spectra Physics, Inc., Stratford, CT). To ensure that photooxidation was complete, the MHL SAM was then exposed to the UV light for 3 h. After photopatterning the SAM was rinsed with ethanol and then placed in a 1 mM ethanolic solution of HDT for 24 h at room temperature. In the areas exposed to UV light, the photooxidized MHL was displaced by a HDT, creating a patterned −OH/−CH3-patterned SAM surface. The samples were then washed with ethanol, dried with N2 gas, and used immediately for deposition. For each batch, one SAM sample was taken and characterized using single-wavelength ellipsometry (Gaertner Scientific Corp., Skokie, IL) and time-of-flight secondary ion mass spectrometry (TOF SIMS) to ensure that the SAMs were free of significant chemical contamination. Similarly one photopatterned −OH/−CH3 SAM surface was characterized using TOF SIMS using the methods in refs 29 and 30. 2.3. Copper Seedless Electroless Deposition. The standard copper electroless deposition solution (“100 % Cu concentration”) was composed of 0.032 M copper(II) sulfate pentahydrate, 0.24 M triethanolamine (TEA), 0.037 M EDTA, and 0.067 M dimethylamine borane (DMAB, (CH3)2NHBH3). To investigate the growth of the nanowires, the concentrations of copper(II) sulfate pentahydrate and DMAB were altered together so that the ratio between the two remained constant, and they are reported as a percentage of the standard concentration. For all depositions, the TEA and EDTA concentrations were kept constant. The pH of the deposition bath was adjusted to 9 before addition of DMAB. The deposition temperature was 22 ± 1 °C. After deposition each substrate was rinsed with DI water and ethanol. The resulting constructs were examined using TOF SIMS, scanning electron microscopy (SEM), and atomic force microscopy (AFM). 2.4. Time-of-Flight Secondary Ion Mass Spectrometry. TOF SIMS measurements were performed using an ION TOF IV spectrometer (ION TOF, Inc., Chestnut Hill, NY) which has a Bi liquid metal ion gun (LMIG) primary ion source. The instrument has three chambers for sample introduction, preparation, and analysis. The pressure of the analysis and preparation chambers is typically = − The reported dimensions of the wires are the average of these measurements (at least 50 data points), and the reported error is the standard deviation.
3. RESULTS AND DISCUSSION 3.1. Nanowire and Mesowire Formation. Isolated nanoparticles initially form at the boundary between the −OH and −CH3-terminated SAMs (Figure 2). As deposition
Figure 2. AFM images of nanowires formation with deposition time. Reaction conditions: 0.032 M copper(II) sulfate pentahydrate, 0.24 M TEA, 0.037 M EDTA, and 0.067 DMAB; deposition temperature 22 ± 1 °C. Image size: (20 × 20) μm2.
continues, the number of nanoparticles increases until a continuous polycrystalline nanowire forms (Figure 2). The nanowire grows in width and at later deposition times becomes a mesowire (Figure 1). These observations suggest that wire formation begins with heterogeneous nucleation of nanoparticles at the interface between the −CH3- and −OHterminated SAM areas. Time-of-flight secondary ion mass spectra confirm this hypothesis. When the nanowires initially form, cluster ions of the form [Cu2(MHL) (HDT)OH]− (MHL = −S(CH2)15CH2OH; HDT = −S(CH2)15CH3) are observed (Figure 3). Because in SIMS cluster ions form from species in close proximity to each other (≤6 Å apart), this cluster ion indicates that the copper is deposited at the −OH/−CH3-terminated SAM interface. Subsequently, additional nanoparticles appear to nucleate at the interface between the −CH3-terminated SAM and the growing Cu deposit due to the preferential adsorption of DMAB on the methyl-terminated SAM. As the deposition continues, eventually a continuous polycrystalline wire is formed (Figure 2). The wire growth continues into the −CH3-terminated SAM area until eventually copper covers the whole area. C
DOI: 10.1021/acs.langmuir.5b04674 Langmuir XXXX, XXX, XXX−XXX
Article
Langmuir
Figure 3. High-resolution negative ion spectra from m/z 660 to m/z 685. Reaction conditions: 0.032 M copper(II) sulfate pentahydrate, 0.24 M TEA, 0.037 M EDTA, and 0.067 M DMAB; deposition temperature 22 ± 1 °C; deposition time 10 min.
We have demonstrated nanowire and mesowire formation over a variety of reaction conditions including concentrations of copper(II) sulfate ranging from 0.016 to 0.032 M and DMAB ranging from 0.17 to 0.67 M, bath pH from 6 to 11, and bath temperatures from room temperature up to 50 °C (see Supporting Information, Figure S1). The upper temperature is limited only by the stability of the substrate. Above room temperature the density of defects within the SAM substantially increases.31,32 Thus, to prevent the formation of a very defective substrate film, only temperatures up to 50 °C were employed. We also note that an ENDOM-like process may occur between hydrophilic and hydrophobic surfaces without the presence of SAMs. (The formation of mesowires has also been observed at the junction of hydrophobic and hydrophilic gold areas (Figure 4).) However, these wires are not as uniform as those formed using a patterned SAM substrate, and it is unclear that nanowires (width < 100 nm) can be produced on such surfaces.
Figure 5. (a) Variation of wire width with reaction time at room temperature, 22 ± 1 °C. The dotted lines are the best fits to the data, which were fit using a nonlinear regression to an exponential. (b) Plot of ln w (ln(width)) versus deposition time. The dotted lines show the fit to the data with slope B = 0.85 ± 0.08 nm/min. Reaction conditions: the composition of the “100 % concentration” bath was 0.032 M copper(II) sulfate pentahydrate, 0.24 M TEA, 0.037 M EDTA, and 0.067 M DMAB. The concentrations of copper(II) sulfate and DMAB were altered together so that the ratio between the two remained constant, and they are reported as a percentage of the standard concentration (“100 % concentration”). For all depositions, the TEA and EDTA concentrations were kept constant.
5b). We note that the standard deviations in the wire widths are 15−20%, which are comparable to other methods to make nanowires8,34−36 (Figure 5a). It is expected that this variation can be significantly reduced by using commercially available deposition solutions34 and better lithography. However, the induction time decreases with increasing Cu2+ (and DMAB) concentration (Figures 5 and 6). Here the induction time is defined as the time at which complete nanowires are first observed. For precipitation from a supersaturated solution, classical nucleation theory predicts that the nucleation rate is given by37
Figure 4. SEM images of a copper mesowire formed at the interface of hydrophobic and hydrophilic gold.
3.2. Nucleation and Growth Process. As expected for an electroless deposition reaction,33 the wire width increases exponentially with reaction time (Figure 5a): w = A exp(Bt )
⎛ 16πγ 3Vm 2 ⎞ ⎟ K = K 0 exp⎜ − 3 2 ⎝ 3(kT ) (ln S) ⎠
(3)
where S = C/C* is the supersaturation constant, C is the Cu2+ concentration, C* is the critical concentration, Vm is the molecular volume, k is the Boltzmann constant, T is the temperature, and γ is the surface free energy. The critical (saturation) concentration of metallic copper in aqueous solution is not well-known; a value of C* = 1.58 × 10−8 ppb at room temperature is used, here taken from Palmer and Petrov.38 Because the induction time, tind, is inversely proportional to the nucleation rate,37
where w is the deposit width in nanometers, A is a constant, t is the deposition time, and B is the rate constant. At constant temperature, ln w = ln A + Bt
(5)
(4)
i.e., a plot of ln w versus t should give a straight line with slope B. At constant [Cu2+]/[DMAB] and temperature, for different Cu2+ concentrations the growth rate is the same (to within experimental certainty) with B = 0.85 ± 0.08 nm/min (Figure D
DOI: 10.1021/acs.langmuir.5b04674 Langmuir XXXX, XXX, XXX−XXX
Article
Langmuir
Figure 6. (a) Variation of induction time with Cu2+ concentration; (b) ln tind versus 1/(ln S)2 where S is the degree of supersaturation. (b) The dotted line is the best fit to the data and was a straight line. The goodness of fit, r2, is 0.998. The composition of the “100 % concentration” bath was 0.032 M copper(II) sulfate pentahydrate, 0.24 M TEA, 0.037 M EDTA, and 0.067 M DMAB. The concentrations of copper(II) sulfate and DMAB were altered together so that the ratio between the two remained constant, and they are reported as a percentage of the standard concentration (“100 % concentration”). For all depositions, the TEA and EDTA concentrations were kept constant.
⎛ 16πγ 3V 2 ⎞ m ⎟ t ind = A exp⎜ 3 3( kT ) (ln S)2 ⎠ ⎝
Figure 7. (a) Variation of wire height with wire width with different Cu2+ concentrations. (b) Variation of grain size with wire width at 100% Cu2+ concentrations. The standard deviation in the wire height and grain size is 20−24%. The composition of the “100 % concentration” bath was 0.032 M copper(II) sulfate pentahydrate, 0.24 M TEA, 0.037 M EDTA, and 0.067 M DMAB. The concentrations of copper(II) sulfate and DMAB were altered together so that the ratio between the two remained constant, and they are reported as a percentage of the standard concentration (“100 % concentration”). For all depositions, the TEA and EDTA concentrations were kept constant. Please note the scale of the nanowire height and grain size axes.
on the −CH3-terminated SAM. Thus, the interface moves into the −CH3-terminated areas and the wire grows in width but not height. The grain size of the copper also does not change (to within experimental uncertainty) with deposition time (Figure 7b). Interestingly, the grain size is ∼2× larger than the measured wire height, suggesting that the deposits are composed of nanoparticles that are wider than they are tall. These observations indicate that the minimum nanowire width is determined by the nanoparticle dimensions. From classical nucleation theory, if the concentration of copper is increased, the activation barrier to nucleation is reduced, resulting in the production of a larger number of smaller nuclei.39 Hence, narrower nanowires should form if the rate of copper production is increased. To test this hypothesis, the [DMAB] to [Cu2+] ratio was doubled, which should increase the reaction rate. The nanoparticle diameter was observed to decrease by ∼30%, and the minimum nanowire width also decreased by ∼30% from 97 ± 10 to 71 ± 7 nm. We note that similar strategies have also been employed in the synthesis of ultrasmall nanoparticles.40 3.3. Formation of Nanopores and Nanochannels. Because the wires nucleate at the boundary between the −OH and −CH3-terminated SAMs, their shape is controlled by the underlying substrate pattern. This has multiple important consequences for the formation of wires. The wires can have complex shapes such as arcs and right-angle bends (see
(6)
at constant temperature, ⎛ 16πγ 3V 2 ⎞ B m ⎟ = ln A + ln t ind = ln A + ⎜ 3 2 (ln S)2 ⎝ 3(kT ) (ln S) ⎠
where B =
16πγ 3Vm 2 3(kT )3
(7)
, i.e., a plot of ln tind versus 1/(ln S)2 should
give a straight line with gradient B. The induction time was determined from Figure 5a by finding the time at which a 100nm-wide nanowire was initially observed (Figure 6a). The plot of ln tind versus 1/(ln S)2 is quite linear (Figure 6b), indicating that classical nucleation theory can be employed to predict the induction time for deposition of the wires. Because for different Cu2+ concentrations the growth rate is identical (to within experimental uncertainty) and the induction time is known, the deposit width can be precisely controlled from the nanometer to micron scale by controlling the deposition time using a stopwatch. Although the wire width increases exponentially with deposition time, the wire height remains constant to within experimental uncerta1inty (Figure 7a). This is because as the polycrystalline wire grows, additional copper ions are reduced on the surface at the interface between the copper and −CH3terminated SAM due to the preferential adsorption of DMAB E
DOI: 10.1021/acs.langmuir.5b04674 Langmuir XXXX, XXX, XXX−XXX
Article
Langmuir
formation of copper nanoscale and mesoscale structures to demonstrate and study the technique. In ENDOM, initially isolated nanoparticles heterogeneously nucleate at the boundary between −CH3- and −OH-terminated SAMs. The growing deposit eventually forms a polycrystalline wire. Because the shape of the deposit is determined by the underlying substrate pattern, nanorings are formed if the substrate pattern is a series of enclosed shapes. As the deposition continues, the copper deposits at the Cu/−CH3terminated SAM interface eventually fill the −CH3-terminated SAM area, leaving a nanochannel or nanopore. The width of the deposit grows exponentially with deposition time, which is expected for an ELD process. At constant [Cu2+]/[DMAB], the deposition rate is the same (to within experimental uncertainty) for all Cu2+ concentrations studied. In contrast, the induction time for deposition decreases with increasing Cu2+ concentration. Further, classical nucleation theory can be used to predict the induction time for the deposition of nanowires. Because the reaction rate and induction time are known, this allows the deposit width to be precisely controlled using the deposition time. In contrast to the deposit width, the height and grain size of the copper deposit is constant (to within experimental uncertainty). These observations indicate that the minimum deposit width is controlled by the nanoparticle dimensions, which from classical nucleation theory is controlled by the reaction rate and thus the reagent concentrations. By doubling the [DMAB] to [Cu2+] ratio, the reaction rate increased. Further, the nanoparticle diameter decreased by ∼30%, and the minimum nanowire width also reduced by ∼30% from 97 ± 10 to 71 ± 7 nm. Finally we note that ENDOM is a very flexible method. It is compatible with a wide range of lithographies and technologically relevant substrates. Furthermore, we have observed an ENDOM-like process at the interface of hydrophilic and hydrophobic gold surfaces, suggesting that a patterned SAM substrate may not even be required. This method therefore enables the development of new and cheaper nanotechnological devices with possible applications in nanoelectronics, nanofluidics, nanobiotechnology, and sensing.
Supporting Information, Figures S2 and S3). Second, the process is parallel, so many nanostructures can be formed simultaneously over large areas (square-centimeters) (see Supporting Information, Figure S3). However, perhaps the most important consequence is that ENDOM can be employed to create a variety of different structures using the same basic approach. By changing the substrate pattern and controlling the deposition time, nanorings, nanochannels, and nanopores can be produced. At early deposition times, nanorings are formed if the substrate pattern is a series of enclosed shapes (Figures 1 and 8a).
Figure 8. SEM images of the formation of a nanopore from a nanoring. Initially a nanoring is formed (a). As the deposition time increases, the deposit thickens in the direction of the white arrows (b). At longer times the deposit nearly covers the −CH3-terminated SAM (hexagonal area), forming a pore in the center (c). Finally, after continued deposition a nanopore is formed in the center of the ring (d). Reaction conditions: 0.0304 M copper(II) sulfate pentahydrate, 0.24 M TEA, 0.037 M EDTA, and 0.0637 M DMAB; deposition temperature 45 ± 2 °C. Note that the scale bars vary between images.
■
ASSOCIATED CONTENT
S Supporting Information *
Continued copper deposition leads to the thickening of the ring as the deposit fills the hydrophobic −CH3-terminated SAM (hexagonal) area (Figure 8b). At longer times the deposit nearly covers the −CH3-terminated SAM forming a micronsized pore in the center (Figure 8c). Finally, at later times after continued deposition, a nanopore is formed in the center of the ring (Figures 1 and 8d). We note that, at later deposition times, the copper layer becomes thicker and rougher as deposition occurs atop the copper area. However, the grain size remains approximately constant. (In Figure 8 the grain size is different to that shown in Figure 7 because different deposition conditions were used.) Similarly if a parallel line pattern is employed, nanowires form first (Figure 1). As the deposition continues, mesowires form and then nanochannels as the deposit widens and almost fills the −CH3-terminated SAM areas.
The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.langmuir.5b04674. SEM images of wires formed under different experimental conditions; large-area SEM and optical images of nano- and mesostructures produced using the ENDOM process; fits of wire width to exponential growth rate (PDF)
■
AUTHOR INFORMATION
Corresponding Author
*E-mail: [email protected]. Phone: +1 972 883 5780. Fax: +1 972 883 5725. Notes
The authors declare no competing financial interest.
■
4. CONCLUSIONS In summary we have developed and investigated a simple, fast, and versatile method by which to deposit a wide range of nanostructures in parallel over the mesoscale. We use the
ACKNOWLEDGMENTS The authors gratefully acknowledge support from the National Science Foundation (CHE1213546 and DMR1156423). F
DOI: 10.1021/acs.langmuir.5b04674 Langmuir XXXX, XXX, XXX−XXX
Article
Langmuir Brianna L. Dermody (B.L.D.) obtained the “110% concentration” data shown in Figure 2. B.L.D. was supported by a Research Experience for Undergraduates program funded by the National Science Foundation (DMR1156423).
■
Media Containing Zinc Nitrate and Dimethylamineborane. J. Electrochem. Soc. 2006, 153, C735−C740. (22) Chigane, M.; Izaki, M.; Hatanaka, Y.; Shinagawa, T.; Ishikawa, M. Effect of dimethylamine borane concentration on antireflection properties of silica thin films via redox deposition. Thin Solid Films 2006, 515, 2513−2518. (23) Lelental, M. Dimethylamine Borane as the Reducing Agent in Electroless Plating Systems. J. Electrochem. Soc. 1973, 120, 1650−1654. (24) Lelental, M. Effect of amine borane structure on activity in electroless plating. J. Catal. 1974, 32, 429−433. (25) Shi, Z.; Walker, A. V. Investigation of the Mechanism of Nickel Electroless Deposition on Functionalized Self-Assembled Monolayers. Langmuir 2011, 27, 6932−6939. (26) King, D. E. Oxidation of gold by ultraviolet light and ozone at 25°C. J. Vac. Sci. Technol., A 1995, 13, 1247−1253. (27) Nuzzo, R. G.; Dubois, L. H.; Allara, D. L. Fundamental Studies of Microscopic Wetting on Organic Surfaces. 1. Formation and Structural Characterization of a Self-Consistent Series of Polyfunctional Organic Monolayers. J. Am. Chem. Soc. 1990, 112, 558−569. (28) Fisher, G. L.; Walker, A. V.; Hooper, A. E.; Tighe, T. B.; Bahnck, K. B.; Skriba, H. T.; Reinard, M. D.; Haynie, B. C.; Opila, R. L.; Winograd, N.; Allara, D. L. Bond Insertion, Complexation and Penetration Pathways of Vapor-Deposited Aluminum Atoms with HO- and CH3O- Terminated Organic Monolayers. J. Am. Chem. Soc. 2002, 124, 5528−5541. (29) Zhou, C.; Walker, A. V. Dependence of Patterned Binary Alkanethiolate Self-Assembled Monolayers on “UV-Photopatterning” Conditions and Evolution with Time, Terminal Group, and Methylene Chain Length. Langmuir 2006, 22, 11420−11425. (30) Zhou, C.; Walker, A. V. UV Photooxidation and Photopatterning of Alkanethiolate Self-Assembled Monolayers (SAMs) on GaAs (001). Langmuir 2007, 23, 8876−8881. (31) Yamada, Y.; Wano, H.; Uosaki, K. Effect of Temperature on Structure of the Self-Assembled Monolayer of Decanethiol on Au(111) Surface. Langmuir 2000, 16, 5523−5525. (32) Chidsey, C. E. D.; Liu, G.-Y.; Rowntree, P.; Scoles, G. Molecular Order at the Surface of an Organic Monolayer Studied by Low Energy Helium Diffraction. J. Chem. Phys. 1989, 91, 4421−4423. (33) Mallory, G. O.; Hajdu, J. B. Electroless plating: Fundamentals and applications; American Electroplaters and Surface Finishers Society: 1990. (34) Kung, S. C.; Xing, W.; Donavan, K. C.; Yang, F.; Penner, R. M. Photolithographically patterned silver nanowire electrodeposition. Electrochim. Acta 2010, 55, 8074−8080. (35) Xiang, C.; Thompson, M. A.; Yang, F.; Menke, E. J.; Yang, L.-M. C.; Penner, R. M. Lithographically patterned nanowire electrodeposition. Phys. Phys. Status Solidi C 2008, 5, 3503−3505. (36) Koenigsmann, C.; Santulli, A. C.; Sutter, E.; Wong, S. S. Ambient Surfactantless Synthesis, Growth Mechanism, and SizeDependent Electrocatalytic Behavior of High-Quality, Single Crystalline Palladium Nanowires. ACS Nano 2011, 5, 7471−7487. (37) Mullin, J. W. Crystallization, 3rd ed.; Butterworth-Heinemann: Oxford, U.K., 1993. (38) Palmer, D.; Petrov, A. Evaluation of Models for Solubility and Volatility of Copper Compounds Under Steam Generation Conditions. In International Conference on the Properties of Water and Steam; Tremaine, P. R., Hill, P. G., Irish, D. E., Balakrishnan, P. V., Eds.; National Research Council of Canada: Toronto, CA, 1999; pp 821−827. (39) Turkevich, J. Colloidal Gold. Part I: Historical and Preparative Aspects, Morphology and Structure. Gold Bulletin 1985, 18, 86−91. (40) Kim, B. H.; Hackett, M. J.; Park, J.; Hyeon, T. Synthesis, Characterization, and Application of Ultrasmall Nanoparticles. Chem. Mater. 2014, 26, 59−71.
REFERENCES
(1) Chen, J.; Reed, M. A.; Rawlett, A. M.; Tour, J. M. Large On-Off Ratios and Negative Differential Resistance in a Molecular Electronic Device. Science 1999, 286, 1550−1552. (2) Miles, B. N.; Ivanov, A. P.; Wilson, K. A.; Doğan, F.; Japrung, D.; Edel, J. B. Single Molecule Sensing with Solid-State Nanopores: Novel Materials, Methods, and Applications. Chem. Soc. Rev. 2013, 42, 15− 28. (3) Dekker, C. Solid-State Nanopores. Nat. Nanotechnol. 2007, 2, 209−215. (4) Hou, X.; Guo, W.; Jiang, L. Biomimetic Smart Nanopores and Nanochannels. Chem. Soc. Rev. 2011, 40, 2385−2401. (5) Reisner, W.; Pedersen, J. N.; Austin, R. H. DNA confinement in nanochannels: Physics and biological applications. Rep. Prog. Phys. 2012, 75, 106601. (6) Halpern, A. R.; Corn, R. M. Lithographically Patterned Electrodeposition of Gold, Silver, and Nickel Nanoring Arrays with Widely Tunable Near-Infrared Plasmonic Resonances. ACS Nano 2013, 7, 1755−1762. (7) Junesch, J.; Sannomiya, T. Ultrathin Suspended Nanopores with Surface Plasmon Resonance Fabricated by Combined Colloidal Lithography and Film Transfer. ACS Appl. Mater. Interfaces 2014, 6, 6322−6331. (8) Menke, E. J.; Thompson, M. A.; Xiang, C.; Yang, L. C.; Penner, R. M. Lithographically patterned nanowire electrodeposition. Nat. Mater. 2006, 5, 914−919. (9) Decker, C. A.; Solanki, R.; Freeouf, J. L.; Carruthers, J. R.; Evans, D. R. Directed growth of nickel silicide nanowires. Appl. Phys. Lett. 2004, 84, 1389−1391. (10) Whitney, A. V.; Myers, B. D.; Van Duyne, R. P. Sub-100 nm Triangular Nanopores Fabricated with the Reactive Ion Etching Variant of Nanosphere Lithograpy and Angle-Resolved Nanosphere Lithography. Nano Lett. 2004, 4, 1507−1511. (11) Shi, Z.; Walker, A. V. Synthesis of Nickel Nanowires via Electroless Nanowire Deposition on Micropatterned Substrates. Langmuir 2011, 27, 11292−11295. (12) Ellsworth, A. A.; Borner, K.; Yang, J.; Walker, A. V. Towards Molecular Electronics: Using Solution-Based Methods to Deposit Nanowires. ECS Trans. 2014, 58, 1−10. (13) Chaudhury, M. K. Self-assembled monolayers on polymer surfaces. Biosens. Bioelectron. 1995, 10, 785−788. (14) Ulman, A. Formation and Structure of Self-Assembled Monolayers. Chem. Rev. 1996, 96, 1533−1554. (15) Smith, R. K.; Lewis, P. A.; Weiss, P. S. Patterning SelfAssembled Monolayers. Prog. Surf. Sci. 2004, 75, 1−68. (16) Zhou, C.; Trionfi, A.; Jones, J. C.; Hsu, J. W. P.; Walker, A. V. Comparison of Chemical Lithography Using Alkanethiolate SelfAssembled Monolayers on GaAs (001) and Au. Langmuir 2010, 26, 4523−4528. (17) Watanabe, T. Film Formation Mechanism in Electrodeposition. In Nano PlatingMicrostructure Formation Theory of Plated Films and a Database of Plated Films; Elsevier: Oxford, U.K., 2004; pp 95−139. (18) Schmeckenbecher, A. F.; Lindholm, J. A. Potentiometric determination of dimethylamine borane and hypophosphite. Anal. Chem. 1967, 39, 1014−1016. (19) Yokoshima, T.; Kaneko, D.; Akahori, M.; Nam, H.-S.; Osaka, T. Electroless CoNiFeB soft magnetic thin films with high corrosion resistance. J. Electroanal. Chem. 2000, 491, 197−202. (20) Izaki, M.; Omi, T. Transparent Zinc Oxide Films Chemically Prepared from Aqueous Solution. J. Electrochem. Soc. 1997, 144, L3− L5. (21) Murase, K.; Tada, H.; Shinagawa, T.; Izaki, M.; Awakura, Y. QCM Studies of Chemical Solution Deposition of ZnO in Aqueous G
DOI: 10.1021/acs.langmuir.5b04674 Langmuir XXXX, XXX, XXX−XXX
